ADP ribosyl-cyclases (CD38/CD157), social skills and friendship
Introduction
Without friends no one would choose to live, though he had all other goods
−Aristotle: Nichomachean Ethics
Oxytocin (OT) is the predominant human and mammalian social hormone modulating social cognition and affiliative behaviors (Feldman et al., 2016) yet the valence of the oxytocin effect may be positive or negative depending on context (Shamay-Tsoory and Abu-Akel, 2016). Recently CD38, a transmembrane glycoprotein present in the brain (Yamada et al., 1997), was shown to mediate central release of OT (Jin et al., 2007). CD38 has multiple functions as receptor, ectoenzyme (Quarona et al., 2013) and second messenger acting as a universal calcium-signaling molecule (Higashida et al., 2007). CD38 catalyzes the formation of cyclic ADP-ribose (cADPR) and nicotinic acid adenine dinucleotide phosphate (NAADP) by ADP-ribosyl cyclase from NAD+ and NAD phosphate (Lee, 2001). CD38 is an immune cell marker where its expression appears increased in some circumstances e.g. chronic lymphocytic leukemia (CLL) (Malavasi et al., 2011). An important product of CD38, NADase activity from NAD+ pathway, is the generation of adenosine, a pleiotropic nucleoside involved in the modulation of immunity and inflammation that also plays role throughout the central nervous system (Skaper et al., 2010). The CD38 family may have evolved from a cytoplasmic soluble enzyme to surface molecule, simultaneously acting on the immune system and behavior while enabling bidirectional cross talk between them (Quarona et al., 2013).
Due to its vital role in brain OT release, research on CD38 has focused on its role in social behaviors. Intriguingly, a mouse knockout of the CD38 gene is characterized by deficits in social cognition responses (Jin et al., 2007). Further animal model studies highlighted behavioral deficits such as impaired social memory of conspecifics, reduced maternal instincts, lower ultrasonic vocalization frequency by pups in distress and higher locomotor activity away from mothers (Higashida et al., 2012). CD38 polymorphisms have also been linked to a variety of behaviors in nonclinical subjects including early care giving (Feldman et al., 2013), personality traits (Haram et al., 2014), social bonding (Algoe and Way, 2013), and parental behaviors (Lopatina et al., 2012). All these behaviors are (i) moderated by CD38 expression and DNA sequence variation across the CD38/CD157 gene region, (ii) observed in normal populations and (iii) represent salient elements of social cognition. Furthermore, we have shown that lymphoblastoid cell lines obtained from Israeli ASD patients express lower levels of surface CD38 relative to those from unaffected relatives. CD38 expression also significantly correlated with IQ and Vineland Adaptive Behavioral Scores (Lerer et al., 2010, Riebold et al., 2011). Importantly, the current study greatly expands these first observations by correlating CD38 expression in peripheral blood leukocytes (PBL) obtained from nonclinical Singaporean Chinese undergraduates with scores on the Autism Quotient (AQ) (Baron-Cohen et al., 2001). The AQ evaluates the extent of autistic traits, including proficiency of social engagements, in both clinical and non-clinical subjects (Austin, 2005). Besides CD38 expression in the periphery, the current investigation also reports the results of genotype analysis of CD38 and its paralogue CD157/BST1 (Quarona et al., 2013) for a much larger sample of undergraduates recruited at the same time. The interest for examining CD157 was prompted by a first report that genetic variations of the CD157 gene confer risk to ASD (Yokoyama et al., 2015). Both CD38 and CD157 play complementary roles as receptors and ectoenzymes, possessing complex activities related to signaling and cell homeostasis (Quarona et al., 2013). Importantly, CD157 is contiguous with CD38 on human chromosome 4p15, suggesting a common origin of CD157 from an ancient gene duplication of CD38 (Malavasi et al., 2008). Since accumulating evidence convincingly shows that cis-regulatory elements (enhancers, silencers and insulators) act at even long distances of tens to hundreds of kilobases upstream or downstream of their targets to regulate gene expression (Cowie et al., 2015), CD157 non-coding SNPs feasibly regulate CD38 expression. In clinical research these molecules have evolved from simple leukocyte activation markers, to multifunctional ectoenzymes and receptors important in health and disease (Quarona et al., 2013). Similar to CD38, CD157 contributes to CNS pathologies specifically Parkinson’s disease (Chen et al., 2014). In line with these observations, a fusion transcript resulting from a chromosomal deletion involving CD38 and CD157, is associated with autism (Ceroni et al., 2014). A CD157 knockout mouse is characterized by anxiety-like and depression-like behaviors, social avoidance, a less developed amygdala and lower plasma OT levels(Lopatina et al., 2014). Finally, it is suggested that CD157 either directly or indirectly affects the central axonal release of OT (Lopatina et al., 2014).
In the current study we examine the contribution of ADP ribosyl-cyclases (CD38/CD157) to human sociality and friendship starting with the relationship between PBL CD38 expression and social skills captured by the AQ. Our first hypothesis is (1) that CD38 expression is correlated with AQ scores. We extend our study by examining sequence variation in the CD38/CD157 gene region with CD38 expression and (2) hypothesize that common sequence variations in this region correlates with CD38 expression. Towards better understanding the relationship between CD157 eQTLs and social skills the identified eQTLs are then tested primarily for association with AQ and also plasma oxytocin immunoreactive products (POIP) levels in a larger group of students. We hypothesize (3) that CD38/CD157 eQTLs are mainly correlated with AQ scores and (4) possibly with POIP. Finally, the ecological validity of the in vitro and genetic association (CD38 expression, AQ and POIP) findings are vetted by looking at the relationship between self-reported number of friends, which is the real-world outcome of individual differences in pencil and paper measured social skills, and the Friendship Questionnaire that explores the contours of friendship from the individual perspective (Baron-Cohen and Wheelwright, 2003) Finally, we hypothesize (5) CD38 expression and CD38/CD157 sequence variations correlate with the Friendship Questionnaire and actual number of self-reported friends. The five hypotheses are tested through a series of step-by-step detailed analyses given in Results and Table S1 in Supplementary information (SI).
Section snippets
Participants
1327 students (727 females, Mage = 21.4, SD = 1.5) of Han Chinese descent from National University of Singapore were recruited via email advertisements for this study in which they were reimbursed an average of S$25 per hour. Written informed consent approved by the Institution Review Board of the National University of Singapore was obtained prior to participation. AQ and sequence variation was available for all participants, and plasma OT available in 1065 participants. We measure CD38 expression
Results
First, we examine AQ and CD38 gene expression in a randomly selected subsample (N = 214) of the full group of 1327 participants, (Step 1, Table S1, SI), followed by testing association (N = 229) between SNPs in the CD157/CD38 gene region and CD38 expression (Step 2). We next test for association with significant eQTL SNPs from step 2 with AQ in the full sample (N = 1327) (Step 3). In Step 4, we further test functionality of significant eQTL SNPs for association with POIP levels (N = 1065) in those
Discussion
The current investigation shows that PBL CD38 expression predicts AQ scores in 214 nonclinical Singaporean Han Chinese undergraduates, greatly extending our previous findings in lymphoblastoid cell lines derived from Israeli ASD subjects (Lerer et al., 2010, Riebold et al., 2011). In the current study, CD38 expression acted to identify eQTLs across the CD38/CD157 gene region and notably CD157 eQTL SNPs independently predict AQ scores in a sample of 1327 undergraduates. Further evidence, albeit
Conclusion
Our findings contribute to understanding the relationship between human sociability and a universal signaling molecule, CD38, obligatory in the release of brain oxytocin. We now show that CD38/CD157, by releasing oxytocin, not only regulate social life at the cellular level but also contribute to human social skills important in social bonds and friendship in nonclinical participants. We examine leukocyte CD38 mRNA levels in 214 participants and find that CD38 expression and CD157 (a paralogue
Contributors
Anne Chong collected, analyzed the data and wrote the paper. Mikhail Monakhov collected and analyzed the data. Fabio Malavasi and Salomon Israel contributed to the manuscript. Von Bing Yap provided statistical expertise and assisted with the analyses. Chiea Chuen Khor and Poh San Lai contributed reagents and analytical tools. Soo Hong Chew designed study. Richard Ebstein designed the study and wrote the paper.
Financial disclosures
The authors have neither biomedical financial interests nor any conflicts of interests with regards to this article.
Acknowledgements
The authors thank Xing Zhang, Yunfeng Lu, Yushi Jiang, Ha My Bui and Aileen Pang for their assistance in the study. Funding was provided by AXA Research Fund, John Templeton Foundation and the Ministry of Education of Singapore.
Glossary of acronyms
- AQ
- Autism Quotient measures the extent of five autistic traits of attention switching skill, attention to details, imagination, communication and social skills
- CD38
- An ectoenzyme involved in metabolism, immune functions and OT release
- CD157
- A paralogue of CD38 and implicated in immune dysfunction and autism
- eQTL
- Expression quantitative trait loci are loci in the genome that contributes to variation in expression levels of mRNA (messenger RNA: transcribed portions of DNA)
- FQ
- Friendship Questionnaire
References (65)
Personality correlates of the broader autism phenotype as assessed by the Autism Spectrum Quotient (AQ)
Pers. Individ. Diff.
(2005)- et al.
Controlling the false discovery rate in behavior genetics research
Behav. Brain Res.
(2001) - et al.
The effect of massage on cellular immunity, endocrine and psychological factors in women with breast cancer – a randomized controlled clinical trial
Auton. Neurosci.
(2008) - et al.
BST1 rs11724635 interacts with environmental factors to increase the risk of Parkinson’s disease in a Taiwanese population
Parkinsonism Relat. Disord.
(2014) - et al.
Plasma and CSF oxytocin levels after intranasal and intravenous oxytocin in awake macaques
Psychoneuroendocrinology
(2016) - et al.
Diminished plasma oxytocin in schizophrenic patients with neuroendocrine dysfunction and emotional deficits
Schizophr. Res.
(2008) - et al.
Cyclic ADP-ribose as a universal calcium signal molecule in the nervous system
Neurochem. Int.
(2007) - et al.
Social memory, amnesia, and autism: brain oxytocin secretion is regulated by NAD+ metabolites and single nucleotide polymorphisms of CD38
Neurochem. Int.
(2012) - et al.
Intranasal oxytocin: myths and delusions
Biol. Psychiatry
(2016) - et al.
Problems with measuring peripheral oxytocin: can the data on oxytocin and human behavior be trusted?
Neurosci. Biobehav. Rev.
(2013)