De Novo Mutations Affecting the Catalytic Cα Subunit of PP2A, PPP2CA, Cause Syndromic Intellectual Disability Resembling Other PP2A-Related Neurodevelopmental Disorders

Abstract

Type 2A protein phosphatases (PP2As) are highly expressed in the brain and regulate neuronal signaling by catalyzing phospho-Ser/Thr dephosphorylations in diverse substrates. PP2A holoenzymes comprise catalytic C-, scaffolding A-, and regulatory B-type subunits, which determine substrate specificity and physiological function. Interestingly, de novo mutations in genes encoding A- and B-type subunits have recently been implicated in intellectual disability (ID) and developmental delay (DD). We now report 16 individuals with mild to profound ID and DD and a de novo mutation in PPP2CA, encoding the catalytic Cα subunit. Other frequently observed features were severe language delay (71%), hypotonia (69%), epilepsy (63%), and brain abnormalities such as ventriculomegaly and a small corpus callosum (67%). Behavioral problems, including autism spectrum disorders, were reported in 47% of individuals, and three individuals had a congenital heart defect. PPP2CA de novo mutations included a partial gene deletion, a frameshift, three nonsense mutations, a single amino acid duplication, a recurrent mutation, and eight non-recurrent missense mutations. Functional studies showed complete PP2A dysfunction in four individuals with seemingly milder ID, hinting at haploinsufficiency. Ten other individuals showed mutation-specific biochemical distortions, including poor expression, altered binding to the A subunit and specific B-type subunits, and impaired phosphatase activity and C-terminal methylation. Four were suspected to have a dominant-negative mechanism, which correlated with severe ID. Two missense variants affecting the same residue largely behaved as wild-type in our functional assays. Overall, we found that pathogenic PPP2CA variants impair PP2A-B56(δ) functionality, suggesting that PP2A-related neurodevelopmental disorders constitute functionally converging ID syndromes.

Keywords: PP2A; PP2A-related neurodevelopmental disorders; PPP2CA; de novo mutation; epilepsy; intellectual disability; syndrome.

Figure 1

Schematic Overview of the PP2A Complex, Variants in the PP2A Catalytic Cα Subunit, and Photographs of Individuals (A) Overview of the PP2A complex based on the canonical subunits. Gene names encoding the subunit components are listed in each subunit. PPP2CA encodes the catalytic subunit alpha, highlighted in green. Subunits known for their involvement in ID and DD are annotated with an asterisk. (B) De novo mutations identified in individuals with ID and DD are shown with respect to their exonic and Cα protein localization; exon notation is indicated in the ribbon protein structure. The C-tail sequence involved in carboxymethylation and phosphorylation is enlarged for visualization purposes. Note that the partial deletion is not depicted. (C) Facial features of individuals with a de novo mutation in PPP2CA. Some individuals had mild dysmorphisms, but a consistent facial gestalt was not observed. Informed consent was provided by the parents.

Figure 2

Levels of Mutant PP2A Cα Subunits in HEK293T Cells HA-tagged WT and mutant Cα proteins were co-expressed with GFP (transfection control) in HEK293T cells. Protein levels were monitored in total protein lysates by immunoblotting with anti-HA and anti-GFP antibodies. Vinculin was used as a loading control. Abbreviated amino acid changes are as follows: H191R, p.His191Arg; R295^∗, p.Arg295^∗; F308dup, p.Phe308dup; R214^∗, p.Arg214^∗; F146fs, p.Phe146fs; Q125^∗, p.Gln125^∗; Y127C, p.Tyr127Cys; D131H, p.Asp131His; D223V, p.Asp223Val; Y265C, p.Tyr265Cys; G60V, p.Gly60Val; D88G, p.Asp88Gly; Q122H, p.Gln122His; and D223H, p.Asp223His.

Figure 3

Phosphatase Activity of Mutant PP2A Cα Subunits HA-tagged WT and mutant Cα proteins were purified from transfected HEK293T cells by HA pull-down, and absolute PP2A activity was determined on the K-R-pT-I-R-R phosphopeptide substrate with BIOMOL Green. Specific PP2A activities were calculated via correction of the measured activities for actual Cα inputs, determined by anti-HA immunoblotting. Results represent the average specific activity ± SEM for a given Cα mutant in relation to the specific activity of WT Cα (set at 100% in each experiment), as determined in at least three independent experiments (n ≥ 3). A one-sample t test (compare to 100%) was used for assessing statistical significance (^∗p ≤ 0.05, ^∗∗p ≤ 0.01, ^∗∗∗p ≤ 0.001, ^∗∗∗∗p ≤ 0.0001).

Figure 4

Carboxyterminal Methylation of Mutant PP2A Cα Subunits (A) HA-tagged WT and mutant Cα proteins were isolated from transfected HEK293T cells by HA pull-down. PP2A C subunit methylation was subsequently assessed by immunoblotting with a demethyl-specific anti-C monoclonal antibody. Immunoblot data from a single representative experiment are shown (n ≥ 4). (B) The average ratios of the quantified demethyl-specific signals over the quantified anti-HA signals ± SEM are displayed (n ≥ 4) in relation to those of WT Cα (set at 100%). A one-sample t test (compare to 100%) was used for assessing statistical significance (^∗p ≤ 0.05, ^∗∗p ≤ 0.01).

Figure 5

Binding of Mutant PP2A Cα Subunits to the A Subunit (A) A subunit binding. HA-tagged WT and mutant Cα proteins were purified from transfected HEK293T cells by HA pull-down, and the presence of endogenous A subunit in the complexes was determined by anti-A immunoblotting. Abbreviated amino acid changes are as follows: H191R, p.His191Arg; R295^∗, p.Arg295^∗; F308dup, p.Phe308dup; Y127C, p.Tyr127Cys; D131H, p.Asp131His; D223V, p.Asp223Val; Y265C, p.Tyr265Cys; D88G, p.Asp88Gly; Q122H, p.Gln122His; and D223H, p.Asp223His. (B) Same experimental setup as in (A), but in this case, HA-agarose-bound proteins were eluted from the beads with an excess of free HA-tagged peptide before anti-A immunoblotting of the eluates. This setup was necessary for enabling unambiguous visualization of HA-tagged p.Arg214^∗, which co-migrated with immunoglobulin light chains at approximately 25 kDa. (C) Quantified values of A subunit binding. Results represent the average value ± SEM of the ratios of the quantified anti-A signal to the quantified anti-HA signal for a given Cα mutant in relation to those of WT Cα (set at 100% in each experiment), as determined in at least four independent experiments (n ≥ 4). A one-sample t test (compare to 100%) was used for assessing statistical significance (^∗p ≤ 0.05, ^∗∗p ≤ 0.01, ^∗∗∗p ≤ 0.001, ^∗∗∗∗p ≤ 0.0001). (D) Alpha 4 subunit binding. HA-tagged WT and p.Arg214^∗ mutant Cα proteins were purified from transfected HEK293T cells by HA pull-down and eluted from the beads by excess HA peptide, and eluates were subjected to immunoblotting with anti-alpha 4 monoclonal antibodies for the detection of binding of endogenous alpha 4.

Figure 6

Binding of Mutant PP2A Cα Subunits to B-type Subunits (A) GST-tagged B-type subunits were co-expressed with HA-tagged WT or mutant p.Tyr127Cys Cα protein in HEK293T cells. The presence of the PP2A C variant was subsequently assessed in GST pull-downs by anti-HA immunoblotting. Representative blots of similar assays with other Cα variants can be found in Figure S1. (B) GFP-tagged B″′/STRN3 or GFP alone was co-expressed with HA-tagged WT or mutant Cα proteins in HEK293T cells. The presence of Cα variants was subsequently assessed in GFP traps by anti-HA immunoblotting. (C) Quantified values of C subunit binding. Results represent the average value ± SEM of the ratios of the quantified anti-HA signal to the quantified anti-GST or anti-GFP (in the case of STRN3) signal for a given Cα mutant in relation to those of WT Cα (set at 100% in each experiment, dotted line), as determined in at least three independent binding experiments (n ≥ 3). A one-sample t test (compare to 100%) was used for assessing statistical significance (^∗p ≤ 0.05, ^∗∗p ≤ 0.01, ^∗∗∗p ≤ 0.001, ^∗∗∗∗p ≤ 0.0001). Abbreviated amino acid changes are as follows: H191R, p.His191Arg; R295^∗, p.Arg295^∗; F308dup, p.Phe308dup; Y265C, p.Tyr265Cys; Y127C, p.Tyr127Cys; D131H, p.Asp131His; D223V, p.Asp223Val; D88G, p.Asp88Gly; Q122H, p.Gln122His; and D223H, p.Asp223His.