Two proteins with seven transmembrane-spanning domains typical of guanosine-nucleotide-bindingprotein-coupled receptors have been identified as cannabinoid receptors ; the central cannabinoid receptor, CB1, and the peripheral cannabinoid receptor, CB2, initially described in rat brain and spleen, respectively. Here, we report the distribution patterns for both CB1 and CB2 transcripts in human immune cells and in several human tissues, as analysed using a highly sensitive and quantitative PCR-based method. CB1 was mainly expressed in the central nervous system and, to a lower extent, in several peripheral tissues such as adrenal gland, heart, lung, prostate, uterus, ovary, testis, bone marrow, thymus and tonsils. In contrast, the CB2 gene, which is not expressed in the brain, was particularly abundant in immune tissues, with an expression level 10-100-fold higher than that of CB1. Although CB2 mRNA was also detected in some other peripheral tissues, its level remained very low. In spleen and tonsils, the CB2 mRNA content was equivalent to that of CB1 mRNA in the central nervous system. Among the main human blood cell subpopulations, the distribution pattern of the CB2 mRNA displayed important variations. The rank order of CB2 mRNA levels in these cells was B-cells > natural killer cells S monocytes > polymorphonuclear neutrophil cells > T8 cells > T4 cells. The same rank order was also established in human cell lines belonging to the inyeloid, monocytic and lymphoid lineages. The prevailing expression of the CB2 gene in immune tissues was confirmed by Northern-blot analysis. In addition, the expression of the CB2 protein was demonstrated by an immunohistological analysis performed on tonsil sections using specific anti-(human CB2) IgG; this experiment showed that CB2 expression was restricted to Blymphocyte-enriched areas of the mantle of secondary lymphoid follicles. These results suggest that (a) CB1 and CB2 can be considered as tissue-selective antigens of the central nervous system and immune system, respectively, and (b) cannabinoids may exert specific receptor-mediated actions on the immune system through the CB2 receptor.Keywords: cannabinoid ; cannabinoid receptors (CB1; CB2) ; human immune system ; B cells; natural killer cells.A-9-Tetrahydrocannabinol, the major active component of cannabis, as well as other cannabinoids, are known to exert a wide range of physiological effects such as drowsiness, alterations in cognition and memory, analgesia, orexigenic effects, anti-emetic effects, a decrease in intra-ocular pressure, anti-inflammatory effects and immunosuppression [l]. Many studies have been conducted to decipher the cannabinoid system. First attributed to non-specific cell membrane disruption, the major cannabinoid effects are now thought to be mediated through specific cannabinoid receptors. A guanosine-nucleotide-bindingprotein-coupled receptor of 472 amino-acid residues, CB1, was initially characterized in rat brain [Z] and further cloned both in rat [3] and human [4]. As the expression of the c...

Abstract

The discovery of new cytokines normally relies on a prior knowledge of at least one of their biological effects, which is used as a criterion either for the purification of the protein or for the isolation of the complementary DNA by expression cloning. However, the redundancy of cytokine activities complicates the discovery of novel cytokines in this way, and the pleiotropic nature of many cytokines means that the principal activities of a new cytokine may bear little relation to that used for its isolation. We have adopted an alternative approach which relies on differential screening of an organized subtracted cDNA library from activated peripheral blood mononuclear cells, using the inducibility of lymphokine messenger RNAs by anti-CD28 as a primary screening criterion. The ligation of the CD28 antigen on the T lymphocyte by a surface antigen, B7/BB-1, expressed on activated B lymphocytes and monocytes is a key step in the activation of T lymphocytes and the accumulation of lymphokine mRNAs. Here we report the discovery by molecular cloning of a new interleukin (interleukin-13 or IL-13) expressed in activated human T lymphocytes. Recombinant IL-13 protein inhibits inflammatory cytokine production induced by lipopolysaccharide in human peripheral blood monocytes. Moreover, it synergizes with IL-2 in regulating interferon-gamma synthesis in large granular lymphocytes. Recent mapping of the IL-13 gene shows that it is closely linked to the IL-4 gene on chromosome 5q 23-31 (ref. 4). Interleukin-13 may be critical in regulating inflammatory and immune responses.

Marijuana and many of its constituent cannabinoids influence the central nervous system (CNS), probably through the cannabinoid receptor, which has recently been cloned in rat and human. While numerous reports have also described effects of cannabinoids on the immune system, the observation of both mRNA and cannabinoid receptor has hitherto been exclusively confined to the brain, a reported detection in the testis being the sole example of its presence at the periphery. Here we report the expression of the cannabinoid receptor on human immune tissues using a highly sensitive polymerase-chain-reaction-based method for mRNA quantification. We show that, although present in a much lower abundance than in brain, cannabinoid receptor transcripts are found in human spleen, tonsils and peripheral blood leukocytes. The distribution pattern displays important variations of the mRNA level for the cannabinoid receptor among the main human blood cell subpopulations. The rank order of mRNA levels in these cells is B cells > natural killer cells 2 polymorphonuclear neutrophils 2 T8 cells > monocytes > T4 cells. Cannabinoid-receptor mRNA, which is also found in monocytic, as well as T and B leukemia cell lines but not in Jurkat cells, presents a great diversity of expression on these cells as well, B-cell lines expressing a much higher level than T-cell lines. The cannabinoid receptor PCR products from leukocytes and brain are identical both in size and sequence suggesting a strong similarity between central and peripheral cannabinoid receptors. The expression of this receptor was demonstrated on membranes of the myelomonocytic U937 cells using the synthetic cannabinoid [3H]CP-55940 as ligand. The Kd determined from Scatchard analysis was 0.1 nM and the B,, for membranes was 525 fmol/mg protein. The demonstration of cannabinoid-receptor expression at both mRNA and protein levels on human leukocytes provides a molecular basis for cannabinoid action on these cells.

The cDNA sequences encoding the central cannabinoid receptor, CB1, are known for two species, rat and human. However, little information concerning the flanking, noncoding regions is presently available. We have isolated two overlapping clones from a human lung cDNA library with CB1 cDNA inserts. One of these, cann7, contains a short stretch of the CB1 coding region and 4 kilobase pairs (kb) of the 3'-untranslated region (UTR), including two polyadenylation signals. The other, cann6, is identical to cann7 upstream from the first polyadenylation signal, and in addition, it contains the whole coding region and extends for 1.8 kb into the 5'-UTR. Comparison of cann6 with the published sequence (Gérard, C. M., Mollereau, C., Vassart, G., and Parmentier, M. (1991) Biochem. J. 279, 129-134) shows the coding regions to be identical, but reveals important differences in the flanking regions. Notably, the cann6 sequence appears to be that of an immature transcript, containing 1.8 kb of an intronic sequence in the 5'-UTR. In addition, polymerase chain reaction amplification of the CB1 coding region in the IM-9 cell line cDNA resulted in two fragments, one containing the whole CB1 coding region and the second lacking a 167-base pair intron within the sequence encoding the amino-terminal tail of the receptor. This alternatively spliced form would translate to an NH2-terminal modified isoform (CB1A) of the receptor, shorter than CB1 by 61 amino acids. In addition, the first 28 amino acids of the putative truncated receptor are completely different from those of CB1, containing more hydrophobic residues. Rat CB1 mRNA is similarly alternatively spliced. A study of the distribution of the human CB1 and CB1A mRNAs by reverse transcription-polymerase chain reaction analysis showed the presence of both CB1 and CB1A throughout the brain and in all the peripheral tissues examined, with CB1A being present in amounts of up to 20% of CB1.