CD38, as many molecules described in this encyclopedia, was described and characterized in the period 1980–1990, using a mouse monoclonal antibody recognizing human CD38 (mAb) (OKT-10) developed by Ellie L. Reinherz et al. (1980). Through the use of this and many other mAbs, it was determined that CD38 is expressed in activated human T cells, plasma cells, and several lymphoid and myeloid cells as well as many cell malignancies. CD38 was rediscovered in mice by the mAb NIM-R5 developed by Mike Parkhouse and his group at the National Institute of Medical Research, Mill Hill, London, UK (Santos-Argumedo et al. 1993), and in conjunction with other tools, allowed its biochemical and biological characterization. Despite being a useful marker, very little was known about its function. The NIM-R5 antibody helped to demonstrate that CD38 was able to induce activation and proliferation of naive B cells. The cloning and sequencing of CD38 suggested a functional similarity with an enzyme from the sea slug Aplysia californica that catalyzes the modification of nicotinamide adenine dinucleotide (NAD+) to produce cyclic adenosine diphosphate ribose (cADPr). Soon it was demonstrated that CD38 has a similar cyclase activity but also a hydrolase function producing ADPr as the main product (Howard et al. 1993). It was known that cADPr mobilize Ca2+ from internal stores by an IP3 independent mechanism. A major surprise was the finding that the catalytic domain of CD38 is oriented to the extracellular milieu. This finding raised several questions that have not been fully solved. For example: What are the sources of extracellular NAD+? If CD38 is internalized, still its catalytic domain is orientated to the lumen of the vesicle, then, how is cADPr transported to the cytosol? The CD38-deficient mice did not provide solution to many important questions on the role of CD38 in the immune system but instead opened new avenues for research. Probably, the most surprising function for CD38 has been its role in regulating social behavior (maternal nurturing and social recognition) in CD38-deficient mice by regulating oxytocin release by hypothalamic neurons (Jin et al. 2007), but discussion of this subject is beyond the scope of this review.
CD38 Gene and Protein
Human Cd38 gene is encoded at chromosome 4 p15.32; it has a size of 74,956 bases with a plus strand orientation. The gene consists of eight exons with seven introns. The sequence includes a very long intronic sequence between exon 1 and 2 important for the regulation of its expression. Orthologues of CD38 have been described in mouse, rat, cattle, dog, buffalo, rabbit, wild yak, Atlantic herring, bison, seal, domestic cat, chimpanzee, chicken, zebra fish, tropical clawed frog, Sumatran orangutan, pig, Rhesus monkey, crab-eating macaque, and domestic ferret. Mouse Cd38 gene is located at chromosome 5 with a similar gene structure as described for its human counterpart (GenAtlas®). Cd38 belongs to a family that includes Cd157 gene with similar functions, both genes probably arose from a gene duplication event.
The crystal structure of CD38 has shown a conformation similar to Aplysia cyclase. The structures form dimers and tetramers previously identified by biochemical analysis. The catalytic pocket is buried between the middle cleft of the protein. Under pH neutral conditions CD38 uses NAD+ to generate cAPDr and ADPr, but under acidic conditions in the presence of nicotinic acid and NADP, NAADP is produced (Liu et al. 2005).
CD38 was first described as a leukocyte-associated antigen, for that reason was included as a “cluster of differentiation” (CD) by the Human Leucocyte Differentiation Antigens workshops. CD38 is found in all hematopoietic compartments with an expression that varies according to the maturation state of the cell. The most prominent compartments and linages expressing CD38 are hematopoietic stem cells, B, NK, and T lymphocytes and plasma cells (GenAtlas®). There are some discrepancies in the expression of CD38 between human and mouse; the differences are mainly related at when CD38 is expressed during the developmental states of lymphoid cells. However, it is useful to remember that the comparison is not always appropriate, because human lymphoid cells are largely studied in peripheral blood, tonsils, and bone marrow, whereas in mice the cells are usually identified in bone marrow, thymus, lymph nodes, and spleen.
CD38 as an Enzyme
CD38 as a Receptor
Anti-CD38 antibodies have been the most important tools to demonstrate the receptor activity of CD38. For example, naïve B cells from mice proliferate when treated with the mAbs NIM-R5 and Ab 90. Goat, rabbit, or rat polyclonal antibodies also have agonistic properties, comparable to mAbs, indicating that at least in mice, the recognition of different epitopes can trigger the activation of cells carrying CD38 (Moreno-García et al. 2005). Stimulation with these antibodies also prevents apoptosis of mature B cells but induces cell death in immature B lymphocytes. Activated murine B cells respond differentially to CD38 stimulation; thus, TLR4, 7, 8, or 9 stimulation induces the proliferation and subsequent plasma cell differentiation of naïve B cells; costimulation with anti-CD38 maintains the proliferative signals but inhibits the differentiation to plasma cells (Manjarrez-Orduño et al. 2007). CD38 induced maturation of Transitional (T)2 B cells, contrasting with T1 B cells which die by apoptosis. A defective differentiation is observed when CD38 is engaged with antibodies in T2 B cells from Btk-, Lyn-, or Fyn-deficient mice or when drugs interfering with PI3K or ERK function are used (Rodríguez-Alba et al. 2008). In humans, anti-CD38 antibodies seem to be more restricted in terms of activation. Despite that there are several clones of mAbs directed against human CD38, the only antibody with agonistic activities, reported so far, is the clone 1B4 that induces activation and proliferation of human B and T cells and leukemic cell lines (reviewed by Quarona et al. 2013). The receptorial activities of CD38 seem to be independent of its enzymatic activity because many of the assays described in the literature have been performed in the presence of inhibitory drugs that interfere with the enzymatic activity or with mutant CD38 lacking enzymatic activity (Lund et al. 2006).
The search for a ligand for CD38 has been elusive, CD31 has been the only candidate reported in the literature for human CD38 (Deaglio et al. 1998); unfortunately, in mice, it seems that CD31 has not role, and the searching for another ligand is still an open question. Hyaluronic acid has also been proposed as a ligand, giving CD38 the possibility of acting as an adhesion molecule, but there is no consensus on this function nor in the ability of hyaluronic acid to trigger the receptorial properties of CD38.
Signaling Pathways of CD38
The signaling can be divided accordingly to the described functions of CD38, although that division is arbitrary. As an enzyme, the formation of cADPr and NAADP give CD38 a role as a mobilizer of Ca2+. As it was stated above, if these products are produced outside the cell, they require to be internalized though different mechanisms to reach the Ca2+ reservoirs in the interior of the cell.
On the other hand, ADPr, the main product of the catalysis of NAD+ by CD38, affects the function of the transient receptor potential channel M2 (TRP-M2) (Partida-Sanchez et al. 2007). This channel has been implicated in the secretion of insulin by the beta pancreatic cells, as a receptor mediating inflammation, as a coreceptor of chemokine receptors in cells from the immune response system, etc. The link with CD38 is that this channel is activated through intracellular ADPr and free calcium in the cytosol. Thus, ADPr also needs to be transported from outside or produced inside the cell. ADPr is not only produced by CD38 but is also manufactured intracellularly by the poly ADP ribose polymerase (PARP) enzymes (Wei et al. 2014).
As a receptor, stimulation by anti-CD38 antibodies induces the phosphorylation of several substrates indicating the functional association with one or more tyrosine kinases. In human T cells, CD38 is associated with CD3 epsilon, CD247 (TCR zeta chain), Lck, and Zap70 within lipid rafts, thus CD38 could be acting as a coreceptor upon engagement with CD31 or any other putative ligand available (Muñoz et al. 2003). In NK cells, CD38 is associated with CD16 (FcRg3a). Through this association it may participate as a coreceptor for ADCC killing functions. In myeloid cells, CD38 is associated with c-Cbl a protein with an E3 ubiquitin-protein ligase. c-Cbl has a role in the retinoic acid induced differentiation of myeloid leukemias. It seems that CD38 and c-Cbl cooperate inducing differentiation of these cells through a MAPK pathway (Shen and Yen 2008).
CD38 in Health and Disease
As it was described above, the multiple functions of CD38 in different tissues and cells positions CD38 an important target in health and disease. The main obstacle to developing specific drugs to interfere with both its receptorial and its enzymatic activity is the lack of a detailed knowledge of its role in different processes. Most of the information about the biological role of this molecule came from observations done in CD38-deficient mice. For example, its absence produces some abnormalities in the humoral immune response, in pancreas it has been linked to the development of diabetes. In humans, it has been reported that the presence of antibodies against CD38 correlates with type I and II diabetes, as well with systemic lupus erythematous (OMIM®). The molecule has been implicated in regulating inflammation and regulating the outcome of an immune response by the fact that CD38 is expressed on regulatory T and B cells as well as some myeloid regulatory cells. There are also several polymorphisms in CD38 associated with the development of diabetes, autoimmunity, autism, some cardio vascular diseases, osteoporosis, etc. (OMIM®). The expression of CD38 is prominently high in multiple myeloma (MM), chronic lymphocytic leukemia (CLL), and acute promyelocytic leukemia (APL). For more than two decades, the expression of CD38 has been used as a marker to predict the progression of HIV infections and to follow their treatment with antiretrovirals, because CD38 is high on CD8 T cells during progression and is reduced with a successful treatment, concomitant with an increase of CD4 T cell counts. The recent findings that CD38 is important in the secretion of oxytocin by hypothalamic neurons, affecting social behaviors and some other functions such as memory, has generated expectation of new targets to treat autism and several neurological disorders.
The fact that CD38 is highly expressed by several malignancies such as MM, CLL, and APL makes it an attractive target for treatment. In the last few years, several mAbs have been analyzed with very promising results. One such antibody recently got FDA approval and was initially use in MM. The same mAb has shown encouraging results in the treatment of other hematological malignancies. The antibodies reported so far act through complement-dependent cytotoxicity, ADCC, opsonization, inducing programmed cell death, by modulating its enzymatic activity and more generally, through modulation of the immune response mediated by CD38+ regulatory lymphoid and myeloid cells (van de Donk et al. 2016).
CD38 is a molecule originally defined as an antigen on the surface of human-activated T lymphocytes by the monoclonal antibody OKT10. Further studies demonstrated that this molecule is also present on B cells, monocytes, and many other cellular lineages. In humans, it has been used as a marker for plasma cell differentiation. The molecule is a single pass transmembrane type II glycoprotein. Two main biochemical roles have been assigned to CD38: first, the extracellular domain contains a catalytic activity being NAD+, NADP+ and nicotinic acid their substrates and cyclic-ADP-ribose, ADP-ribose and nicotinic acid adenine dinucleotide phosphate (NAADP) their main products; the second activity is its role as a cellular receptor, CD31 being one of its ligands. Because CD38 is highly expressed in several B cell malignancies, it has become a natural target for treatment through the use of several monoclonal antibodies with very promising results. The study of CD38 has been extended to areas beyond immunity to its activity in the CNS and its role in controlling behavior of mice (opening the possibility of a similar role in humans). Thus, CD38 still represents a box full of unexplored surprises for future generations of researchers.
- GenAtlas®. http://genatlas.medecine.univ-paris5.fr/imagin/go_gene.php#Google Scholar
- Jin D, Liu HX, Hirai H, Torashima T, Nagai T, Lopatina O, Shnayder NA, Yamada K, Noda M, Seike T, Fujita K, Takasawa S, Yokoyama S, Koizumi K, Shiraishi Y, Tanaka S, Hashii M, Yoshihara T, Higashida K, Islam MS, Yamada N, Hayashi K, Noguchi N, Kato I, Okamoto H, Matsushima A, Salmina A, Munesue T, Shimizu N, Mochida S, Asano M, Higashida H. CD38 is critical for social behaviour by regulating oxytocin secretion. Nature. 2007;446(7131):41–5.CrossRefPubMedGoogle Scholar
- Lund FE, Muller-Steffner H, Romero-Ramirez H, Moreno-García ME, Partida-Sánchez S, Makris M, Oppenheimer NJ, Santos-Argumedo L, Schuber F. CD38 induces apoptosis of a murine pro-B leukemic cell line by a tyrosine kinase-dependent but ADP-ribosyl cyclase- and NAD glycohydrolase-independent mechanism. Int Immunol. 2006;18(7):1029–42.CrossRefPubMedGoogle Scholar
- Moreno-García ME, López-Bojórques LN, Zentella A, Humphries LA, Rawlings DJ, Santos-Argumedo L. CD38 signaling regulates B lymphocyte activation via a phospholipase C (PLC)-gamma 2-independent, protein kinase C, phosphatidylcholine-PLC, and phospholipase D-dependent signaling cascade. J Immunol. 2005;174(5):2687–95.CrossRefPubMedGoogle Scholar
- Partida-Sanchez S, Gasser A, Fliegert R, Siebrands CC, Dammermann W, Shi G, Mousseau BJ, Sumoza-Toledo A, Bhagat H, Walseth TF, Guse AH, Lund FE. Chemotaxis of mouse bone marrow neutrophils and dendritic cells is controlled by adp-ribose, the major product generated by the CD38 enzyme reaction. J Immunol. 2007;179(11):7827–39.CrossRefPubMedGoogle Scholar
- UniProtKB®. http://www.uniprot.org/uniprot/P28907