A multi-level developmental approach to exploring individual differences in Down syndrome: genes, brain, behaviour, and environment

Abstract

In this article, we focus on the causes of individual differences in Down syndrome (DS), exemplifying the multi-level, multi-method, lifespan developmental approach advocated by Karmiloff-Smith (1998, 2009, 2012, 2016). We evaluate the possibility of linking variations in infant and child development with variations in the (elevated) risk for Alzheimer's disease (AD) in adults with DS. We review the theoretical basis for this argument, considering genetics, epigenetics, brain, behaviour and environment. In studies 1 and 2, we focus on variation in language development. We utilise data from the MacArthur-Bates Communicative Development Inventories (CDI; Fenson et al., 2007), and Mullen Scales of Early Learning (MSEL) receptive and productive language subscales (Mullen, 1995) from 84 infants and children with DS (mean age 2;3, range 0;7 to 5;3). As expected, there was developmental delay in both receptive and expressive vocabulary and wide individual differences. Study 1 examined the influence of an environmental measure (socio-economic status as measured by parental occupation) on the observed variability. SES did not predict a reliable amount of the variation. Study 2 examined the predictive power of a specific genetic measure (apolipoprotein APOE genotype) which modulates risk for AD in adulthood. There was no reliable effect of APOE genotype, though weak evidence that development was faster for the genotype conferring greater AD risk (ε4 carriers), consistent with recent observations in infant attention (D'Souza, Mason et al., 2020). Study 3 considered the concerted effect of the DS genotype on early brain development. We describe new magnetic resonance imaging methods for measuring prenatal and neonatal brain structure in DS (e.g., volumes of supratentorial brain, cortex, cerebellar volume; Patkee et al., 2019). We establish the methodological viability of linking differences in early brain structure to measures of infant cognitive development, measured by the MSEL, as a potential early marker of clinical relevance. Five case studies are presented as proof of concept, but these are as yet too few to discern a pattern.

Keywords: Alzheimer’s disease; Down syndrome; apolipoprotein APOE gene; brain imaging; genetics; individual differences; socio-economic status; vocabulary development.

Fig. 1

Parent ratings of (a) receptive and (b) expressive vocabulary sizes for children with DS, according to the CDI (Fenson et al., 2007). Children with DS are split by whether or not they are carriers of the APOE ε4 risk allele for dementia. Typically developing norms and standard deviations are also shown for 6-18 months of age (Fenson et al., 2007).

Fig. 2

Parent ratings of (a) receptive and (b) expressive vocabulary sizes for children with DS according to the CDI (Fenson et al., 2007) plotted against SES, estimated by parental occupation (1 = highest SES, 9 = lowest SES). Children with DS are split by whether they are carriers of the APOE ε4 risk allele for dementia (filled markers) or non-carriers (non-filled markers).

Fig. 3

Composite language measure for the children with DS, derived from CDI (Fenson et al., 2007) and MSEL receptive and expressive language measures (Mullen, 1995) plotted against age. Children with DS are split by whether or not they are carriers of the APOE ε4 risk allele for dementia.

Fig. 4

Structural MRI images of fetuses and a neonate with DS after motion correction and reconstruction (SVR) (left to right: sagittal, axial and coronal planes). (a) 30-week-old fetus also demonstrating unilateral ventriculomegaly, enlargement of the ventricles of the brain (shown by arrow); (b) 34 week old fetus; and (c) 43 week old neonate.

Fig. 5

(a) Supratentorial brain volume (cm³), (b) cortex volume (cm³), and (c) cerebellar volume (cm³) changes with age in a sample of fetuses and neonates with DS (age 22 weeks GA to 46 PMA). Labels are attached to those neonatal cases who were later assessed with the MSEL. TD cross-sectional growth trajectory and 95% CI shown in grey for comparison (DS and TD trajectories were generated from data in Patkee et al., 2019). The dotted vertical line shows usual full term (though babies may be born prematurely).

Fig. 6

MSEL age-equivalent (AE) score changes with age in a sample of infants and children with DS for the subtests of gross motor, fine motor, visual reception, receptive language and expressive language. Labels are attached to those cases who had an MRI scan in the perinatal period. TD cross-sectional growth trajectory and 95% CI shown in grey for comparison (DS and TD group data taken from D’Souza et al., 2020).