Genetic and Environmental Determinants of Primate Biobehavioral Development
- Stephen J. Suomi, PhD, Head, Comparative Behavior Genetics Section
- Annika Paukner, PhD, Senior Visiting Fellow
As in previous years, we conducted detailed longitudinal studies on the behavioral and biological consequences of differential early social rearing, most notably by comparing rhesus monkey infants reared by their biological mothers in pens containing adult males and other mothers with same-age infants for their first 6–7 months of life (MR) with monkeys separated from their mothers at birth, hand-reared in the lab's neonatal nursery for their first month, and then raised in small groups of same-age peers for the next six months, or housed in individual cages containing an inanimate surrogate mother and given two hours of daily interaction with similarly reared peers (NR). At 7–8 months of age, MR and NR infants were all moved into one large pen, where they lived together until puberty. Thus, the differential social rearing occurred only for the first 7–8 months; thereafter MR and NR monkeys shared the same physical and social environment. We previously demonstrated that NR monkeys cling more, play less, tend to be more impulsive and aggressive, and exhibit much greater behavioral and biological disruption during and immediately following short-term social separation at six months of age than do MR monkeys, and that they also exhibit deficits in serotonin metabolism (as indexed by chronically low values of cerebrospinal fluid [CSF] 5-HIAA [5-hydroxy-3-indole acetic acid, a serotonin metabolite]). Additionally, they exhibit significantly lower levels of 5-HTT (serotonin transporter) binding throughout many brain regions than do MR subjects. Many of these differences between MR and NR monkeys persist throughout the juvenile years in the absence of experimental interventions. For example, we recently published data extending these rearing condition differences to measures of social dominance status, maternal competence, telomere length, and physical health during childhood, adolescence, and adulthood. However, our most recent studies indicate that many of these rearing condition differences in behavioral, biological, and health outcomes appear to be largely reversible following specific social interventions.
Another major focus of recent research has been to characterize interactions between differential early social rearing and polymorphisms in several candidate genes (G x E interactions), most notably the 5-HTTLPR (serotonin transporter polymorphic region) gene. During the past two years, we expanded the range of outcomes for which G x E interactions involving the 5-HTTLPR polymorphism and early rearing condition differences appear, including in social play and behavioral reactions to a variety of social stressors, and in epigenetic regulation of brain activity. In addition, we recently reported significant G x E interactions between early MR vs. PR rearing and polymorphisms for several other candidate genes including: DRD1, which encodes the dopamine receptor D1; NPY, which encodes neuropeptide Y;OPRM1, which encodes the mu opioid receptor 1; BDNF, which encodes brain-derived neurotrophic factor; NOS-1, which encodes nitric oxide synthase 1 neuronal; and a single-nucleotide polymorphism (SNP) in the glucocorticoid gene, with outcome measures including aggression, play behavior, social buffering, behavioral and HPA (hypothalamic-pituitary-adrenal axis) reaction to an unfamiliar conspecific, naloxone treatment, alcohol consumption, and plasma BDNF concentrations. In virtually every case, we observed a similar pattern, i.e., the less efficient (from a transcriptional point of view) allele was associated with a negative outcome among PR–reared monkeys but with a neutral or, in some cases, even with an optimal outcome for MR–reared subjects carrying that same less efficient allele, suggesting an overall buffering effect of MR rearing for individuals carrying these so-called risk alleles.
Additionally, we recently published the results of two sets of studies investigating the effects of differences in early social rearing (MR vs. NR) on genome-wide patterns of mRNA expression in leukocytes and on methylation patterns in the prefrontal cortex (PFC) and in T cell lymphocytes. Our research involving mRNA expression, carried out in collaboration with Steven Cole and James Heckman, examined expression patterns in differentially reared 4-month-old monkey infants. In all, 521 different genes were significantly more expressed in MR infants than in NR infants, whereas the reverse was the case for another 717 genes. In general, NR–reared infants showed enhanced expression of genes involved in inflammation, T lymphocyte activation, and cell proliferation and suppression of antiviral and antibacterial responses. Since that initial study, we completed a prospective longitudinal study in which differentially reared subjects were sampled at 14 days, 30 days, 6–7 months, and every three months thereafter until they reached puberty. Data analyzed to date revealed that the above rearing-condition differences in genome-wide patterns of mRNA expression in leukocytes persist throughout development in the absence of any changes in the social environment but change dramatically whenever the social environment is altered during the juvenile years. These new findings are currently being prepared for publication.
The other set of studies, carried out in collaboration with Moshe Szyf and his lab, involved genome-wide analyses of DNA methylation patterns in differentially reared monkeys when they were adults. The initial study compared such patterns in PFC tissue and T cell lymphocytes obtained from 8-year-old monkeys differentially reared for the first 6–7 weeks of life and thereafter maintained under identical conditions until adulthood. Comparing the two groups, the analyses revealed that (1) more than 4,400 genes were differentially methylated in both the PFC and lymphocytes; (2) however, approximately 25% of the affected genes were identical in both PFC and lymphocytes, although there was considerable tissue specificity; and (3) in both the PFC and lymphocytes, methylated promoters tended to cluster both by chromosomal region and gene function. This past year, we completed a prospective longitudinal study of genome-wide methylation patterns in lymphocytes, collecting samples from exactly the same MR and NR monkeys at exactly the same time points as in the aforementioned longitudinal study of mRNA expression. We published the results of a long-term longitudinal study detailing genome-wide epigenetic changes in in MR and NR monkeys over their first two years of life. We found dramatic changes in methylation patterns of lymphocytes from infancy to 6 months in both males and females affecting wide swaths of the genome, but sex differences were largely reversed prior to weaning. The differences continued after weaning, albeit with some attenuation, but increased again by two years of age. Each sex of NR monkeys exhibited very different developmental trajectories over the same developmental period. In sum, genome-wide patterns of methylation in lymphocytes were highly dynamic throughout pre-pubertal development and varied dramatically as a function of both sex and early rearing history.
In another collaboration with the Szyf lab, we examined the epigenetic consequences of high vs. low ranking in established social groups of adult female monkeys and in offspring whose relative social dominance status matched that of their mothers [Reference 5]. It appeared that the cross-generational transmission of social status was mediated, at least in part, by the placenta, in that the genome-wide pattern of DNA methylation in tissues collected from placentas immediately after birth differed dramatically between offspring of high- and low-ranking females. Not only did the order of magnitude of these differences match that of the above-mentioned early social rearing condition differences, but also many of the same genes were involved, suggesting the existence of a subset of "early adversity" genes, i.e., genes sensitive to a range of different early life adversities.
Human mothers interact emotionally with their newborns through exaggerated facial expressions and mutual gaze, a capacity that has long been considered uniquely human. We previously began a research program on early face-to-face interactions in rhesus monkeys after we made the serendipitous discovery that very young rhesus monkey infants did, in fact, engage in extensive face-to-face interactions with their mothers, but only during the first month of life. This past year, we further characterized face-to-face interactions between monkey mothers and their newborn infants in a naturalistic setting. We found large individual variability in rates of maternal/infant face-to-face interactions, in that mothers who had only one or two infants engaged in mutual gazing/lip-smacking in the first 30 days of life significantly more than mothers who had had three or more infants, whereas the more experienced mothers let their infants out of arms’ reach significantly more in the first 30 days of life than newer mothers. Overall, mothers tended to engage in more face-to-face interactions with their male infants.
We also discovered that, during their first week, some (but not all) monkey infants could accurately match certain facial gestures produced by a human experimenter, even after a delay. For those infants who could imitate in this fashion, the capability was evident on the first postnatal day. We finished our initial investigation of brain activity during periods of imitation, using scalp electrodes to record EEG activity, and found a distinctive EEG signature involving significant suppression of mu rhythm activity at low frequencies in frontal and parietal brain regions exclusively during periods of imitation. We also reported that this pattern of EEG activity intensified through that first week and was significantly stronger in mother-reared than in nursery-reared neonates. The findings demonstrate similarities between infant human and infant monkey EEG during periods of imitation.
Using eye-tracking technology, we also demonstrated that week-old monkey infants readily respond to a computer-generated dynamic monkey avatar, and that those infants who imitate tend to focus on different aspects of the avatar's face (eyes and mouth) compared with those that do not imitate (mouth only). We also compared neonatal imitation abilities in mother-reared and nursery-reared monkeys, focusing on day 3 performance only. We reported that, even though NR infants show an imitation effect when tested over the first week, they do not exhibit imitation specifically on day 3. In contrast, MR monkeys responded to facial gestures with more gestures themselves, consistent with our previous EEG findings that MR infants show larger mu suppression than NR infants when viewing facial gestures.
Given the potential impact of neonatal imitation on infants’ social, cognitive, and emotional development, we devised one intervention whereby NR infants either received additional facial gesturing from a human caretaker, received additional handling (but did not see facial gestures), or remained in standard nursery rearing. We found that only the group that had received facial gesturing showed improved performance on the standard neonatal imitation task on day 7 as well as greater sensitivity to facial identity of others in a standardized stranger task. Infants from the facial gesturing group also showed higher preference for a social video at day 30 and again at day 40, had better memory for social stimuli when tested at day 60, and had higher levels of social contact with peers from day 40 to day 60 than did infants in the handling and standard rearing groups. Similar differences in social behavior persisted well into the second year of life.
A second intervention designed to increase infants’ social perception and social sensitivity looked at the effects of oxytocin on monkey infants’ social interactions. NR infants were nebulized with either oxytocin or saline and then tested in an imitation recognition task. We reported increased time spent looking at faces following oxytocin, but not saline, treatment. Salivary assays confirmed increased levels of oxytocin, and infants also showed more affiliative gesturing towards a human experimenter following oxytocin administration.
We completed a project begun last year involving the analysis of mothers’ milk in rhesus monkeys with respect to parity and early life history (i.e., rearing condition). In collaboration with Katie Hinde, we collected milk samples from monkey mothers over the first 30 days of their infant's life and analyzed the samples for cortisol content and nutrient composition. Similar to Hinde's studies on human mothers' milk in older infants, we found that parity predicted milk yield volume (MYE) in the first month of life. Our findings also indicated that mothers with higher hair cortisol during pregnancy had a higher MYE in the first 30 days of life. Additionally, we found that cortisol levels in mothers’ milk predicted infant cognitive functioning and social behavior later in life. Infants who ingested milk with higher cortisol content were less impulsive in a cognitive task but also initiated social behaviors with peers less frequently.
We used hair cortisol as a measure of chronic hypothalamic-pituitary axis (HPA) activity in two additional studies completed this past year. First, hair cortisol levels measured shortly after birth, which presumably reflect prenatal HPA activity from mid-gestation onward, predicted cognitive performance capabilities and infant temperament in the first postnatal months. Second, changes in hair cortisol concentrations during the juvenile years predicted differences in social dominance status among adult female monkeys.
We continued our research program on personality and facial characteristics with our capuchin monkeys, focusing on five personality dimensions (assertiveness, openness, neuroticism, sociability, and attentiveness), and found that the monkeys’ facial width-to-height ratio, as well as their face width/lower face height, are positively and significantly associated with assertiveness. A lower face width/face height ratio was also associated with neuroticism. This past year, we also provided some of our capuchins with stone tools and observed, for the first time in our colony, spontaneous use of those tools to crack open walnuts. Nut-cracking has been observed in a few isolated wild populations of this species but is clearly far from universal.
Publications
- Massert R, Nemoda Z, Sunderman MJ, Sutti S, Ruggiero AM, Dettmer AM, Suomi SJ, Szyf M. Early life adversity alters normal sex-dependent developmental dynamics of DNA methylation. Dev Psychopathol 2016 28(4pt2):1259-1272.
- Dettmer AM, Wooddell LJ, Rosenberg KL, Kaburu SS, Novak MA, Meyer JS, Suomi SJ. Associations between early life experience, chronic HPA axis activity and adult social rank in rhesus monkeys. Soc Neurosci 2017 12:92-101.
- Kaburu SS, Paukner A, Simpson EA, Suomi SJ, Ferrari PF. Neonatal imitation predicts infant rhesus macaque (Macaca mulatta) social and anxiety-related behaviours at one year. Sci Rep 2016 6:34997.
- Schneper LM, Brooks-Gunn J, Notterman DA, Suomi SJ. Early life experiences affect telomere length in rhesus monkeys: an exploratory study. Psychosom Med 2016 78:1066-1071.
- Massert R, Suderman MJ, Nemoda Z, Sutti S, Ruggiero AM, Dettmer AM, Suomi SJ, Szyf M. The signature of maternal social rank in placenta DNA methylation profiles in rhesus monkeys. Child Dev 2017 88(3):900-918.
Collaborators
- Enrico Alleva, MD, Istituto Superiore di Sanità, Rome, Italy
- Christina Barr, PhD, DVM, Laboratory of Clinical Sciences, NIAAA, Bethesda, MD
- Allyson J. Bennett, PhD, University of Wisconsin-Madison, Madison, WI
- Igor Brachi, PhD, Istituto Superiore di Sanità, Rome, Italy
- Sarah Brosnan, PhD, Georgia State University, Atlanta, GA
- Hannah Buchanan-Smith, PhD, University of Stirling, Stirling, United Kingdom
- Francesa Cirulli, PhD, Istituto Superiore di Sanità, Rome, Italy
- Steven W. Cole, PhD, University of California Los Angeles, Los Angeles, CA
- Jennifer Essler, MS, Bucknell University, Lewisburg, PA
- Pier F. Ferrari, PhD, Università di Parma, Parma, Italy
- David Goldman, MD, Laboratory of Neurogenetics, NIAAA, Bethesda, MD
- James J. Heckman, PhD, University of Chicago, Chicago, IL
- J.D. Higley, PhD, Brigham Young University, Provo, UT
- Katie Hinde, PhD, Harvard University, Cambridge, MA
- Stafano Kaburu, PhD, University of California Davis, Davis, CA
- Phyllis Lee, PhD, University of Stirling, Stirling, United Kingdom
- Jerrold S. Meyer, PhD, University of Massachusetts, Amherst, MA
- Melinda A. Novak, PhD, University of Massachusetts, Amherst, MA
- Daniel A. Notterman, MD, PhD, Princeton University, Princeton, NJ
- Andreas Reif, PhD, Universität Würzburg, Würzburg, Germany
- David X. Reiss, MD, Yale University, New Haven, CT
- Helena Rutherford, PhD, Yale University, New Haven, CT
- Lisa M. Schneper, PhD, Princeton University, Princeton, NJ
- Melanie L. Schwandt, PhD, Laboratory of Clinical and Translational Studies, NIAAA, Bethesda, MD
- Valentina Sclafani, PhD, Reading University, Reading, United Kingdom
- Alan Silberberg, PhD, American University, Washington, DC
- Elizabeth A. Simpson, PhD, University of Miami, Miami, FL
- Moshe Szyf, PhD, McGill University, Montreal, Canada
- Bernard Thierry, PhD, Centre d'Écologie, Physiologie et Éthologie, CNRS, Strasbourg, France
- Ross E. Vanderwert, PhD, Cardiff University, Cardiff, United Kingdom
- Elisabetta Visalberghi, PhD, Istituto de Scienze e Technologie della Cognizione, CNR, Rome, Italy
- Alexander Weiss, PhD, University of Edinburgh, Edinburgh, United Kingdom
- Jane Widness, PhD, Yale University, New Haven, CT
Contact
For more information, email suomis@mail.nih.gov.
