Introduction

Brain development involves spatiotemporally complex microstructural changes. To obtain optimal cognitive and behavioral functions, proper structural brain development is essential. Especially for abilities to construct and maintain smooth social relationships between individuals, the orbitofrontal-limbic regions of the brain are important. These brain regions include the orbitofrontal cortex, cingulate cortex, amygdala, and hippocampus, which induce and regulate emotion, do self-monitoring and self-control, associate memories, expect outcomes from actions, and enhance memories for emotionally-arousing events. A detailed understanding of the typical developmental patterns and connectivity of these orbitofrontal-limbic regions may help to clarify the anatomical mechanisms underlying their functions. In addition, it would be a normal reference for the disorders and diseases that impair the region-related functions described above. Thus, in this study, we assessed development of the orbitofrontal cortex, cingulate cortex, amygdala, and hippocampus in a non-human primate species, the common marmoset (Callithrix jacchus) by using longitudinal MRI data. We used common marmoset because 1) they reaches adult size and sexually gets matured at around 24 months ,and 2) like human beings, they are raised and grown by family.

Method

The 23 animals were obtained from healthy five pairs. They had grown in family, including parents, either/both elder or younger siblings and littermates, at least for first 9 months after birth. We constantly collected their longitudinal T2-weighted and diffusion weighted images (two b0 volumes and 30 volumes with b=1000s/mm2) from 1 to 18 months, total 114 time points, using 7 Tesla Biospec 70/16 MRI scanner (Bruker Biospin GmbH; Ettlingen, Germany) Receiver coil; Bespoke 4ch phased array coil (Takashima Seisakusho Co., Ltd.,Tokyo Japan) with actively shielded gradients at a maximum strength of 700mT/m. To estimate the regional volumes, the T2-weighted images were nonlinearly registered to the reference T2-weighted image of marmoset brain, which has its relative hippocampus labelmap cropped from the published common marmoset brain parcellation map1. Then using the registration information, the regional labels were inverted to the relative native spaces and modified manually if necessary. To assess the changes over all cell integrity in the orbitofrontal-limbic structures, we acquired diffusion tensor imaging (DTI) metrics, calculated from diffusion weighted MRI. The obtained volumes and DTI metrics were then statistically analyzed using mixed effect regression analysis, which generally used in longitudinal data, to investigate the age-related changes of the regions.

Results and Discussions

Volume growth followed a non-linear, biexponential curve for all brain regions, regardless of sex (Fig. 1). The highest volumetric growth rate occurred during infancy period (1 to 3 months of age) with a corresponding marked decrease in the apparent diffusivities in all the brain regions. It suggests that an increase in the number and size of cells, dendrites, and spines would have occurred, resulting in volume increase during this period. Such developmental neural growth in this period was also qualitatively observed in the image contrast changes (Fig. 2). The volumes were all steadily increased until adolescence (7 to 13 months of age). The diffusivity values of the DTI kept decreasing until approximately 9 months of age. Nevertheless, according to the previous study2, synaptogenesis and enlargement of neuropil reached the peak at 3 months old in common marmosets. Thus, the volmetric growth during this period may be driven by, at least in part, progression of axonal myelination rather than increase of the cell or synapse. Thereafter, the volume of the orbitofrontal and cingulate regions decreased considerably, while there was little change in hippocampus and amygdala. A possible cause of the reduction in grey matter volume involves tissue compression following maturation of adjacent white matter. Indeed, the heatmap derived from fiber density of tractography passing through the regions indicated the increase of fiber bundle density by age (Fig 3). No significant sex difference was found in the development of any measured regions. Rather familial factors seemed to put weight on the developmental patterns of all the measured regions (Fig 4). Marmosets are often multizygotic, sharing placenta circulation during pregnancy, and exposure of testosterone seems to occur soon after birth in males but not before birth3. In other words, maternal status including age, nutrient, and stress levels during pregnancy as well as postnatal rearing environment may be important drivers of brain growth.

Conclusion

We investigated the typical developmental patterns of non-human primate brain regions important for social skills, including the amygdala, hippocampus, cingulate cortex, and orbitofrontal cortex. We found a non-linear structural development in orbitofrontal-limbic regions, with robust growth in early life and prolonged white matter changes until adolescence. Further, familial factors, rather than sex, were important contributors to structural development in the orbitofrontal-limbic regions.

Acknowledgements

This research is partially supported by the program for Brain Mapping by Integrated Neurotechnologies for Disease Studies (Brain/MINDS) from Japan Agency for Medical Research and development, AMED and Grant-in-Aid for Japan Society for the Promotion of Science(JSPS) Research Fellow.