Cerebrospinal Fluid Arginine Vasopressin Is a Predictive Biomarker of Social Impairments in Male Rhesus Monkeys

Background: Autism spectrum disorder (ASD) is characterized by core social impairments, but its basic biology remains poorly understood. Early and accurate diagnoses are difficult and there are no laboratory-based diagnostic tests to detect, nor medications that effectively treat, the social features of ASD. Scientific progress has been impeded by the difficulty of obtaining relevant tissue samples from patients and matched controls. In mouse models, while tissue is available, there is frequently discordance between complex human behavior and laboratory-based mouse behavior, even with shared genetic etiologies. These limitations underscore the tremendous value in developing a monkey model of social impairments with more reliable behavioral and biological homology to the human disease.

Objectives: Like humans, rhesus monkeys (Macaca mulatta) are highly social, and both species display stable and pronounced individual differences in social functioning. At the behavioral extremes, low-social compared to high-social male monkeys initiate fewer affiliative interactions and display more inappropriate social behavior, suggesting both lower social motivation and poorer social skills. The aim of this study was to develop a novel primate model using ethological observations to identify low-social male rhesus macaques and test whether they demonstrate abnormalities in key neuropeptide pathways [e.g., arginine-vasopressin (AVP) and oxytocin (OXT)], previously implicated in ASD.

Methods: Study subjects were male rhesus monkeys aged 1-5 years. In cohort 1, N=42 monkeys (selected from a pool of N=222) were identified on the basis of archived behavioral data thought to predict later low (N=21) or high (N=21) social functioning. Quantitative social behavior data and personality assessments were collected by blinded observers using established protocols. After completion of behavioral phenotyping, monkeys with the most extreme scores (N=15 low-social monkeys and N=15 high-social monkeys) were selected for biological sample collection. In a second, independent cohort, N=164 male monkeys were observed and social behavior observations obtained using a higher-throughput scan sampling-based method to identify a subset of N=25 low-social and N=25 high-social monkeys. Cerebrospinal fluid (CSF) and blood samples were concomitantly drawn on two separate occasions and later quantified for AVP and OXT concentrations using established enzyme immunoassays.

Results: Logistic regression showed that CSF AVP concentration (LR ChiSq=16.55; p<0.0001) robustly classified monkeys by social group. Using a general linear model, we found that low-social monkeys also had lower CSF AVP concentrations compared to high-social monkeys (F_1,18=9.236; P=0.0071). We then sought to replicate these findings in the second cohort. As before, CSF AVP concentration strongly predicted social classification (LR ChiSq=7.969; p<0.0048), with low-social monkeys showing lower CSF AVP concentrations compared to high-social monkeys (F_1,24=8.847; P=0.0066). CSF OXT concentrations did not differ by social group.

Conclusions: Although the majority of research has focused on OXT, rather than AVP, as a biomarker of social deficits in ASD, our findings suggest the provocative notion that the AVP signaling pathway may be a more promising therapeutic target by which to enhance social functioning in male primates.