Children are not simply small adults  

It is clear when observing the behaviours and thought processes of children that they are not merely small adults. Less easily discerned, however, are the underling molecular and cellular mechanisms distinguishing the paediatric and adult human brain.  In the Hockman lab we are interested in exploring the process of how the human brain matures from birth through to adulthood. Specifically, we want to understand what genes are responsible for driving this process and how this differs between different types of cells in the human brain. The brain is an incredibly complex and sophisticated structure made up of many different types of cells that each perform different functions¹. This is made possible by each cell type having a unique gene expression profile. Instead of being born with fully developed brains, our brains are a work-in-progress and they continue to develop into adulthood, with dynamic changes in gene expression underpinning larger structural and functional changes². For example, changes in the expression of synaptic genes enable neuronal connections to be refined and strengthened postnatally while increases in the expression of myelin-related proteins enhance rates of neurotransmission².

Although changes in postnatal  gene expression trajectories have previously been studied across many cell types together^2–4, this is one of the first studies to examine it within individual cell types – allowing us to gain greater insight into the complexities of the developing brain. This research is not only exciting but also has important medical implications. For example, disrupted gene expression during the period from birth to adulthood may contribute to the onset of neurological conditions including neuropsychiatric conditions such as schizophrenia and major depressive disorder⁵.  Additionally, there are various brain conditions that show differential manifestations or responses to treatment between children and adults, such as epilepsy⁶ and brain tumours⁷. Towards addressing these challenges, our study shines light on the differences between the paediatric and adult human brain within different cell types. This may contribute to the development of therapeutic strategies for treating neurological diseases that take into account the age of the patient and target individual brain cell types, thereby enhancing efficacy whilst reducing side effects.

Notably,  our study includes data from individuals of African ancestry which is especially important considering that, by 2050, 37% of the world’s children will grow up in Africa⁸. We hope this study will contribute towards understanding conditions relevant to the local paediatric population such as tuberculous meningitis and HIV^9,10.

Image 1: The human brain is a work-in-progress and continues to develop from birth into early adulthood. Multiple cell types and cell subtypes exist in the brain and are involved in a variety of processes such as synaptic pruning and myelination. We are interested in comparing gene expression profiles between paediatric and adult brains within individual cell types instead of looking across multiple cell types. This close-up examination will enhance our understanding of mechanisms of brain maturation that are specific to certain cell types and shared across different cell types. This may contribute towards developing targeted therapies for treating neurological diseases that also account for the age of the patient. Adapted from Silbereis et al.(2016).

What did we do?

Over the past five years, we have been building a brain bank through a collaboration with neurosurgeons at local hospitals in Cape Town, South Africa. This gave us the opportunity to receive live human brain tissue samples from paediatric and adult patients undergoing surgeries to treat epilepsy. This tissue would otherwise have been discarded.

To explore gene expression profiles of our samples, we processed the tissue using single nucleus RNA sequencing technology which allowed us to obtain gene expression information from individual brain cells instead of from a whole mass of brain cells. To increase the sample size for the study, we combined our datasets with several comparable datasets from epilepsy surgeries which had already been published¹¹. We identified a diverse array of cell types and cell subtypes in our combined datasets and used a machine learning method to identify the top genes defining each cell subtype. We also examined the location of the various cell types within sections of brain tissue from the same patients. This was achieved using spatial transcriptomics which can approximate the location of different cell types based on their gene expression profiles. Subsequently, we compared the gene expression profiles of our paediatric and adult samples to find genes that were changing with age within individual cell subtypes.

Image 2: A snap shot of brain tissue sections from a 15-year-old and 31-year-old showing an array of cell types indicated by different colours. We see a very similar composition of cell types despite the tissue coming from individuals who are more than 15 years apart. Although the architecture of the paediatric and adult brain appears similar, we were interested in whether there are underlying differences in gene expression between paediatric and adult cell types. Images were obtained using Visium spatial transcriptomics.

What did we find?

Firstly, we identified the same number of cell types in paediatric and adult datasets. When examining the marker genes defining different cell types, we found a large overlap between paediatric and adult markers, suggesting that the fundamental identity of paediatric and adult cell types are the same.  Our spatial transcriptomics data validated this by showing comparable distributions of cell types within paediatric and adult tissue sections.

Interestingly, we found that most genes which changed during brain maturation were unique to certain cell subtypes. This highlights the importance of examining gene expression dynamics at the high resolution of individual cell types instead of looking across multiple cell types. For example, we found that the level of a gene known to shape synaptic connections, TENM1, was higher in paediatric samples compared to adults within a particular excitatory neuron subtype only. We also built on previous studies identifying developmentally regulated genes, such as LAMC3, SOX11, FNBP1L, and NOTCH2, by determining the specific cell subtypes in which their expression levels changed. These genes are important for normal brain maturation and disruption to their expression may prevent cells from becoming functionally distinct, in turn leading to neurological complications.

Most of the changes in gene expression trajectories appeared to be in neuronal versus non-neuronal cells suggesting that neurons form the basis for the features distinguishing paediatric and adult brain. Additionally, there were fewer shared marker genes between paediatric and adult cell states for neuronal cell types compared to non-neuronal cell types. However, we examined the non-neuronal cells at a more granular resolution than the non-neuronal cells which may explain this difference. It is possible that more differences in non-neuronal cells would be observed if we were to examine them at a finer resolution.

When examining functional differences between the paediatric and adult cell types, we found that genes highly expressed in paediatric brains were involved in processes related to cellular respiration, neurotransmitter transport, and synaptic plasticity. Moreover,  many of the genes changing during human brain maturation have previously been implicated in cognitive ability including genes associated with educational attainment and with human evolution. The expression signature of these genes can now be attributed to specific brain cell subtypes. This may help refine our understanding of the susceptibility of different cell types to adverse events as the brain matures and how disrupted gene expression during childhood can increase risk for neurological disease.

Potential impact of the study

We hope the datasets generated in this study will serve as a valuable resource to the neuroscience community that  can be explored further to identify medically-relevant differences between the paediatric and adult human brain within individual cell types. This may contribute to developing therapies for neurological conditions that target specific cell types which may be more effective than current strategies and have fewer side effects. We are grateful to the individuals who kindly donated their brain tissue for the purpose of this study, without whom this research would not be possible.

References

1. Silbereis, J. C., Pochareddy, S., Zhu, Y., Li, M. & Sestan, N. The Cellular and Molecular Landscapes of the Developing Human Central Nervous System. Neuron 89, 248–268 (2016).
2. Kang, H. J. et al. Spatio-temporal transcriptome of the human brain. Nature 478, 483–489 (2011).
3. Colantuoni, C. et al. Temporal dynamics and genetic control of transcription in the human prefrontal cortex. Nature 478, 519–523 (2011).
4. Werling, D. M. et al. Whole-Genome and RNA Sequencing Reveal Variation and Transcriptomic Coordination in the Developing Human Prefrontal Cortex. Cell Rep 31, 107489 (2020).
5. Paus, T., Keshavan, M. & Giedd, J. N. Why do many psychiatric disorders emerge during adolescence? Nat Rev Neurosci 9, 947–957 (2008).
6. Baram, T. Z. The Brain, Seizures and Epilepsy Throughout Life: Understanding a Moving Target. Epilepsy Curr 12, 7–12 (2012).
7. Merchant, T. E., Pollack, I. F. & Loeffler, J. S. Brain Tumors Across the Age Spectrum: Biology, Therapy, and Late Effects. Semin Radiat Oncol 20, 58–66 (2010).
8. O’Malley, J., Wardlaw, T., You, D., Hug, L. & Anthony, D. Africa’s child demographics and the world’s future. The Lancet 384, 730–732 (2014).
9. Schutte, C. M. Analysis of HIV-related mortality data in a tertiary South African neurology unit, 2006- 2012. South Afr J HIV Med 14, 121–124 (2013).
10. Rohlwink, U. K. et al. Clinical characteristics and neurodevelopmental outcomes of children with tuberculous meningitis and hydrocephalus. Dev Med Child Neurol 58, 461–468 (2016).
11. Thrupp, N. et al. Single-Nucleus RNA-Seq Is Not Suitable for Detection of Microglial Activation Genes in Humans. Cell Rep 32, 108189 (2020).