BCL11B mutations in patients affected by a neurodevelopmental disorder with reduced type 2 innate lymphoid cells

Abstract

The transcription factor BCL11B is essential for development of the nervous and the immune system, and Bcl11b deficiency results in structural brain defects, reduced learning capacity, and impaired immune cell development in mice. However, the precise role of BCL11B in humans is largely unexplored, except for a single patient with a BCL11B missense mutation, affected by multisystem anomalies and profound immune deficiency. Using massively parallel sequencing we identified 13 patients bearing heterozygous germline alterations in BCL11B. Notably, all of them are affected by global developmental delay with speech impairment and intellectual disability; however, none displayed overt clinical signs of immune deficiency. Six frameshift mutations, two nonsense mutations, one missense mutation, and two chromosomal rearrangements resulting in diminished BCL11B expression, arose de novo. A further frameshift mutation was transmitted from a similarly affected mother. Interestingly, the most severely affected patient harbours a missense mutation within a zinc-finger domain of BCL11B, probably affecting the DNA-binding structural interface, similar to the recently published patient. Furthermore, the most C-terminally located premature termination codon mutation fails to rescue the progenitor cell proliferation defect in hippocampal slice cultures from Bcl11b-deficient mice. Concerning the role of BCL11B in the immune system, extensive immune phenotyping of our patients revealed alterations in the T cell compartment and lack of peripheral type 2 innate lymphoid cells (ILC2s), consistent with the findings described in Bcl11b-deficient mice. Unsupervised analysis of 102 T lymphocyte subpopulations showed that the patients clearly cluster apart from healthy children, further supporting the common aetiology of the disorder. Taken together, we show here that mutations leading either to BCL11B haploinsufficiency or to a truncated BCL11B protein clinically cause a non-syndromic neurodevelopmental delay. In addition, we suggest that missense mutations affecting specific sites within zinc-finger domains might result in distinct and more severe clinical outcomes.

Figure 1

Images of patients with BCL11B associated disorder. Facial images of Patient A:II-3 at age of 2 11/12 and 3 11/12 years (A); Patient B:II-2 at the age of 8 11/12 and 15 1/12 years (B); Patient C:II-2 at the age of 1 8/12 years (C); Patient D:II-1 at the age of 1 6/12 years (D); Patient E:II-1 at the age of 1 month and 2 1/12 years (E); Patient F:II-2 at the age of 5, 7 and 17 years (F); Patient H:II-1 at 11 years (G); and Patient J:II-1 at 6 2/12 years (H).

Figure 2

Genetic data of patients with BCL11B associated disorder. (A) Schematic protein structure of BCL11B: the position of the mutations identified in this study are marked with vertical arrows and shown in red, and the recently identified missense mutation (Punwani et al., 2016) is shown in black. C = C terminus; N = N terminus; ZnF = zinc-finger C2H2 domain. (B) Genomic context of the translocation breakpoints in Patients H:II-1 and I:II-2. Top: BCL11B genomic regulatory block (GRB) model. Middle: The position of the breakpoints in chromosome 14 is shown in blue. The yellow bar shows the position of the T cell-specific enhancer (Li et al., 2013). Bottom (in pink): Location of permissive enhancers according to the FANTOM5 algorithm. The legend at the bottom describes the symbols used.

Figure 3

Functional analysis of the human p.Gly820Alafs*27 BCL11B mutation. (A–R) Immunohistological analysis of Bcl11b^flox/flox;Emx1-Cre hippocampal slice cultures after in vitro Day 11 electroporation. Animals were electroporated with pIRES-EGFP (A, D, G, J, M and P), pIRES2-EGFP Bcl11b-hu wt (wild-type human BCL11B cDNA) (B, E, H, K, N and Q) as well as pIRES2-EGFP Bcl11b-hu dup (human BCL11B cDNA containing the c.2449_2456dupAGCCACAC, p.Gly820Alafs*27) (C, F, I, L, O and R). DAPI (blue) as morphological marker, GFP (green) and BrdU (red) as marker for cell proliferation as well as Bcl11b (red). Images were taken at 63× magnification, 2× zoom. (S) Statistical analysis of BrdU-positive cells in Bcl11b^flox/flox;Emx1-Cre and control hippocampal slice cultures. [t-test, *P < 0.0005; control n = 4, mutant (Bcl11b^flox/flox; Emx1-Cre n = 3)]. Scale bar = 10 µm (A).

Figure 4

Impaired T cell development in patients with BCL11B associated disorder. (A) Two-dimensional t-distributed stochastic neighbour embedding (t-SNE) plot showing the clustering of seven BCL11B patients (red circles) and 14 healthy controls aged under 16 (grey circles) according to 102 T cell traits. (B) Frequency of recent thymic emigrants (RTE) (CD4+ CD45RA+ CD31+) in relation to age in control donors (black circles) and BCL11B patients (red circles). (C) Shannon Diversity Index of the TCRαβ-repertoire in CD4 and CD8 T cells in unaffected donors aged 15 (HD) and in individuals with BCL11B mutations (mut). (D) Percentage of T-γδ cells and usage of TCRδ chains in unaffected donors aged under 16 (HD) and in BCL11B patients (mut). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 5

Drastic reduction in peripheral ILC2 in patients with BCL11B associated disorder. (A) Dot plots corresponding to the gating of ILC2s in an unaffected individual (left) and in Patient C:II-2 (right). The plots show events in the CD45+ lineage− HLA-DR− gate (left plots) and after selection of the CD127+ cells (right plots). (B) Frequency and absolute numbers of ILCs in unaffected individuals and in individuals with BCL11B mutations. *P < 0.05, **P < 0.01. Significant outliers (Grubb’s test P < 0.05) are displayed in parentheses, but were excluded from statistical analysis.