De novo variants in NPTN cause a neurodevelopmental disorder with autism and neuroplastin-PMCA hypofunction

Present Address:

YiLiang1

RodrigoOrmazabal-Toledo2✉Emailrodrigo.herrera@ubo.cl

HariniSrinivasan3

AyseMalci4

WaldoAcevedo5

UlrichThomas6

JulieS.Cohen7,8

NilsRahner9

JohannesLuppe10

GabriellaVera11

FrancoisLecoquierre11

EdenKroin12

BradAngle12

HongCui13

MariaJ.Guillen Sacoto13

BertB.A.

Vries14

RolphPfundt14

GillianPrinzing15

KimberlyWiltrout15

YakiraBegun16

ElaineM.Pereira17

KonradPlatzer10✉,22Emailkonrad.platzer@medizin.uni-leipzig.de

DirkMontag1✉Emaildirk.montag@lin-magdeburg.deEmaildirk.montag@neurogenetic.de

RodrigoHerrera-Molina18✉,19,20,21Emailrherreramolina@email.gwu.edu

1Neurogenetics LaboratoryLeibniz Institute for NeurobiologyBrenneckestrasse 639118MagdeburgGermany

2Departamento de Química Orgánica y Fisicoquímica, Facultad de Ciencias Químicas y FarmacéuticasUniversidad de ChileIndependencia, SantiagoChile

3Department of Genetics and Molecular Neurobiology, Institute of BiologyOtto-von- Guericke UniversityMagdeburgGermany

4Center for Neuroscience ResearchChildren’s National Medical CenterWashingtonDCUSA

5Instituto de Química, Facultad de CienciasPontificia Universidad Católica de ValparaísoValparaísoChile

6Department Cellular NeurosciencesLeibniz Institute for NeurobiologyMagdeburgGermany

7Department of Neurology and Developmental MedicineKennedy Krieger Institute21205BaltimoreMDUSA

8Department of NeurologyJohns Hopkins University School of Medicine21287BaltimoreMDUSA

9MVZ Institute for Clinical Genetics and Tumor GeneticsBonnGermany

10Institute of Human GeneticsUniversity of Leipzig Medical Center04103LeipzigGermany

11Department of Genetics and Reference Center for Developmental DisordersUniversity Rouen Normandie, Inserm U1245 and CHU RouenF-76000RouenFrance

Division of GeneticsAdvocate Children’s HospitalPark Ridge60068ILUSA

13GeneDx, LLC20877GaithersburgMDUSA

14Department of Human GeneticsRadboud University Medical CenterNijmegenthe Netherlands

15Department of NeurologyBoston Children’s HospitalBostonMassachusettsUSA

16Department of Pediatrics, Division of Clinical GeneticsColumbia University Irving Medical Center10032NY, NYUSA

17Department of Pediatrics, Division of Clinical GeneticsColumbia University Irving Medical Center and NewYork Presbyterian10032New YorkNYUSA

18Department of Pharmacology & Physiology, School of Medicine & Health SciencesGeorge Washington UniversityWashingtonDCUSA

19Centro Integrativo de Biología y Química AplicadaUniversidad Bernardo O’HigginsSantiagoChile

20Department of Pharmacology & Physiology, School of Medicine & Health SciencesGeorge Washington UniversityUniversity, WashingtonDCUSA

21Centro Integrativo de Biología y Química AplicadaUniversidad Bernardo O’Higgins1702, 8320000General Gana, Santiago, SantiagoChile

22Clinical Genomics, Institute of Human GeneticsUniversity of Leipzig Medical CenterPhilipp-Rosenthal-Str. 5504103LeipzigGermany

Yi Liang¹, Rodrigo Ormazabal-Toledo², Harini Srinivasan³, Ayse Malci⁴, Waldo Acevedo⁵, Ulrich Thomas⁶, Julie S. Cohen^7,8, Nils Rahner⁹, Johannes Luppe¹⁰, Gabriella Vera¹¹, Francois Lecoquierre¹¹, Eden Kroin¹², Brad Angle¹², Hong Cui¹³, Maria J. Guillen Sacoto¹³, Bert B.A. de Vries¹⁴, Rolph Pfundt¹⁴, Gillian Prinzing¹⁵, Kimberly Wiltrout¹⁵, Yakira Begun¹⁶, Elaine M. Pereira¹⁷, Konrad Platzer^10#, Dirk Montag^1#, Rodrigo Herrera-Molina^18,19#

^#Corresponding authors

¹Neurogenetics Laboratory, Leibniz Institute for Neurobiology, Magdeburg, Germany

²Departamento de Química Orgánica y Fisicoquímica, Facultad de Ciencias Químicas y Farmacéuticas, Universidad de Chile, Independencia, Santiago, Chile

³Department of Genetics and Molecular Neurobiology, Institute of Biology, Otto-von-Guericke University, Magdeburg, Germany

⁴Center for Neuroscience Research, Children's National Medical Center, Washington, DC, USA

⁵Instituto de Química, Facultad de Ciencias, Pontificia Universidad Católica de Valparaíso, Valparaíso, Chile

⁶Department Cellular Neurosciences, Leibniz Institute for Neurobiology, Magdeburg, Germany

⁷Department of Neurology and Developmental Medicine, Kennedy Krieger Institute, Baltimore, MD 21205, USA

⁸Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD 21287, USA

⁹MVZ Institute for Clinical Genetics and Tumor Genetics, Bonn, Germany

¹⁰Institute of Human Genetics, University of Leipzig Medical Center, 04103 Leipzig, Germany

¹¹University Rouen Normandie, Inserm U1245 and CHU Rouen, Department of Genetics and Reference Center for Developmental Disorders, F-76000, Rouen, France

¹²Division of Genetics, Advocate Children's Hospital, Park Ridge, IL 60068, USA

¹³GeneDx, LLC, Gaithersburg, MD 20877, USA

¹⁴Department of Human Genetics, Radboud University Medical Center, Nijmegen, the Netherlands

¹⁵Department of Neurology, Boston Children's Hospital, Boston, Massachusetts, USA

¹⁶Department of Pediatrics, Division of Clinical Genetics, Columbia University Irving Medical Center, NY, NY 10032, USA

¹⁷Department of Pediatrics, Division of Clinical Genetics, Columbia University Irving Medical Center and NewYork Presbyterian, New York, NY 10032, USA

¹⁸Department of Pharmacology & Physiology, George Washington University, School of Medicine & Health Sciences, Washington, DC, USA

¹⁹Centro Integrativo de Biología y Química Aplicada, Universidad Bernardo O’Higgins, Santiago, Chile

Correspondence to:

Rodrigo Herrera-Molina

Department of Pharmacology & Physiology, George Washington University, University, School of Medicine & Health Sciences, Washington, DC, USA

Email: rherreramolina@email.gwu.edu

Centro Integrativo de Biología y Química Aplicada, Universidad Bernardo O’Higgins, General Gana 1702, 8320000 Santiago, Santiago, Chile

Email: rodrigo.herrera@ubo.cl

Dirk Montag

Neurogenetics Laboratory, Leibniz Institute for Neurobiology, Brenneckestrasse 6, 39118 Magdeburg, Germany

Email: dirk.montag@lin-magdeburg.de; dirk.montag@neurogenetic.de

Konrad Platzer

Clinical Genomics, Institute of Human Genetics, University of Leipzig Medical Center, Philipp-Rosenthal-Str. 55, 04103 Leipzig, Germany

Email: konrad.platzer@medizin.uni-leipzig.de

Running title: NPTN-related DD, ID, and ASD

Keywords:

autism

epilepsy

speech delay

neuroplastin

PMCA

calcium homeostasis

Abstract

NPTN encodes human neuroplastin (hNp), a subunit of plasma membrane Ca²⁺-ATPases (PMCA). The critical importance of hNp and its associations with PMCA are unknown for the human brain. Here, we describe de novo NPTN variants in seven individuals with autism and mild-to-severe DD/ID and evaluate them using animal models and in silico, molecular and cellular approaches. We identified NPTN variants with dominant-negative (missense) or loss-of-function (nonsense/ frameshift) effect on hNp-PMCA expression and function. The missense variants caused structural and thermodynamic molecular abnormalities and lower expression of hNp in HEK cells. In neurons, hNp missense variants affected PMCA levels and cytosolic Ca²⁺ regulation. In Drosophila, a missense mutation with affected PMCA interaction failed to prevent a lethal phenotype caused by hNp ortholog elimination. In Nptn^+/− mice, levels of Np and PMCA were reduced and insufficient for normal social behavior. Therefore, we show that de novo variants in NPTN cause a neurodevelopmental disorder with intellectual disability and autism, likely linked to PMCA dysfunction.

Introduction

Developmental delay (DD), often linked to intellectual disability (ID) and autism spectrum disorder (ASD)¹, has a prevalence of 1–3% which may vary from large international cohorts to local populations^1–3. Due to its heterogeneity, the etiology of DD is not always diagnosed, as it remains uncertain for 60% of the affected children^1,3. Exome and genome sequencing are being used to identify genetic variants as the plausible cause of DD, revealing that 40–60% of individuals with undiagnosed DD and 30–39% of the ASD cases may carry pathogenic de novo variants^3–5. Discovering genetic variants associated with ID and ASD will facilitate early diagnosis and open avenues for their treatment.

The gene NPTN encodes two isoforms of the type I transmembrane glycoprotein neuroplastin expressed in the brain, human neuroplastin55 (hNp55) and neuron-specific human neuroplastin65 (hNp65)⁶. High levels of neuroplastin mRNAs are detected at 19–24 post-coital weeks (PCW) in neurons across fetal brain regions⁷, with peak levels in the prefrontal cortex of 18-year-old individuals⁶. The isoforms hNp55 and hNp65 are enriched in human synapses⁶. NPTN is often deleted or duplicated in individuals with 15q24 microdeletion syndrome^8–10. Also, a single nucleotide polymorphism in the NPTN promoter is associated with thinner frontal and temporal lobes in the left hemisphere of the brain, correlating with verbal and non-verbal abilities in adolescents¹¹. Studies in Nptn^−/− mutant mice provide direct evidence for the necessity of normal Np55/65 expression for multiple cognitive functions^6,12.

Neuroplastin55/65 are obligatory binding partners in protein complexes for more than 95% of the four plasma membrane Ca²⁺-ATPases (PMCA1-4)¹³. The abundance and importance of these associations remain unexplored for human brain. PMCA1-4 are ATP-fed Ca²⁺-H⁺ co-transporters that pump Ca²⁺ towards the extracellular space to reinstate resting cytosolic levels and regulate intracellular Ca²⁺ signaling while setting the perisynaptic alkaline pH necessary for the activation of ionotropic glutamate receptors of the NMDA type (iGluNRs)^14–18. Mutations in the PMCA1-4 genes (ATP2B1-4) in individuals with DD, ASD, and other neurodevelopmental disorders decrease the expression and/or activity of these Ca²⁺ pumps, leading to defective restoration of cytosolic Ca²⁺ levels^19–30. Importantly, the expression, stabilization, and activity of PMCA1-4 are strongly dependent on Np55/65 binding^{6,12,13,31–33} and constitutive and inducible Nptn-deficient mice display massive PMCA1-4 loss associated with cognitive impairments and deficits in social and affective behaviours^6,12.

Here, we report mild to severe DD/ID and autism in seven individuals with de novo missense, nonsense or frameshift in NPTN. The missense variants cause structural abnormalities in hNp as evaluated in silico through protein modelling and molecular dynamics. We confirmed that, in a human cell line, cultured primary rodent neurons, and in vivo in Drosophila melanogaster, the missense variants are inefficiently expressed and exert dominant negative effects on PMCA levels resulting in failed cytosolic Ca²⁺ regulation. Nptn^+/− mice express reduced amounts of both Np and PMCA. In a social behavior test, Nptn^+/− mice display loss of preference for a novel mouse representing an endophenotype analog to social deficits that characterize autism. We conclude that these de novo variants in NPTN cause a neurodevelopmental disorder likely through direct Np-PMCA hypofunction and Ca²⁺-deregulation in central neurons.

Results

Clinical and genetic analysis

We describe a cohort of seven individuals with heterozygous variants in NPTN, six of which are of de novo origin. An overview of the clinical evaluation of all individuals is presented in table 1. Additional descriptions are provided as case reports in the supplementary data and

in supplementary table 3 and 4. All seven individuals presented with developmental delay (DD) and/or intellectual disability (ID) ranging from mild to severe. Four individuals displayed severe, two moderate, and one mild DD/ID, respectively. All individuals were diagnosed with autism spectrum disorder. Other behavioral findings included poor social interactions, high pain threshold, automutilation or repetitive behaviors. Individual 1 presented seizures starting at age seven months with epileptic spasms. That same individual developed focal impaired awareness seizures at eight years of age and was not seizure-free at age 17. Five of seven individuals received a cranial MRI. Of note, individual 3 presented with a vermal dysgenesis. Individual 4 was reported to have a period of significant regression of speech skills. Growth was found normal in all individuals. Subtle dysmorphic facial features were reported in five out of seven individuals and upslanting palpebral fissures and a prominent forehead were recurrently observed in individuals 1, 2, and 6 (Fig. 1).

Fig. 1

NPTN individuals. Photographs of three individuals carrying de novo missense (individuals 1 and 2) or frameshift (individual 6) mutations. These individuals are characterized along other individuals in Table 1 and described in the main text.

Genetic results

Trio exome sequencing revealed de novo variants in NPTN in individuals 1–4, 6, and 7. A single exome test was carried out in individuals 3 and 5. The nonsense variant in NPTN in individual 3 was segregated in the parents with Sanger sequencing. The biological parents of individual 5 were not available for testing. All variants are not recurrent and are absent from gnomAD (v4 dataset). Two distinct de novo missense variants were identified in addition to five predicted loss-of-function (pLoF) variants. Multiple in silico tools predict a deleterious effect of the two NPTN missense variants (Supplementary table 2 and 4). Missense variants as well as pLoF variants are highly depleted from the gnomAD database. This indicates a selective constraint on both types of variants in a general population that lacks severe, early-onset phenotypes such as DD and ID (LOEUF = 0.25; pLI = 0.99; o/e for missense variants = 0.52; z-score = 2.68).

Characterization of the NPTN variants

We located in the genomic sequence of the NPTN gene the position of each of the variants identified (Fig. 2A). NPTN consists of eight coding exons, one non-encoding 3'UTR exon, and exon 2 that can be spliced out by alternative splicing of the gene transcripts. From this process, two main glycoprotein isoforms are produced, hNp55 and hNp65. Additionally, they may display an alternative aminoacidic insert (DDEP) encoded by exon 7. The function of the DDEP insert is unknown but it is dispensable in Np for regulation of PMCA levels^31,33. hNp65 is 394 or 398 amino acids long, contains a signal peptide sequence, three extracellular Ig-like domains (Ig-like I-III) encoded by exons 1–6, a single transmembrane domain (TM) encoded by exon 6, and a 34 or 38 amino acids long intracellular domain encoded by exons 6–8. Np55 is 278 or 282 amino acids long and distinctively displays the Ig-like domains II-III. NPTN missense variants affected conserved amino acids located within conserved amino acid sequences in different species (Fig. 2B). The missense variant c.403T > A (p.W135R) is located at the hNp65-specific exon 2, and other variants affect exons 1, 2, 4, and 6 which are common for both hNp isoforms (Fig. 2A and C). Interestingly, the missense variant c.1025C > A, p.(Pro342Leu) (hNp65: p.P342L; hNp55: p.P226L) replaces a key transmembrane amino acid in Np that interacts directly with PMCA³⁰. The nonsense variant c.14C > A, p.(Ser5*) stops the transcription of the signal peptide in the mutated allele. The nonsense variants c.284C > G, p.(Ser95*) and c.342C > G, p.(Tyr114*) stop mRNA translation during the very early synthesis of the hNp65-specific Ig-like domain, probably resulting in non-functional peptides likely to be degraded (Supplementary table 1). The frameshift in the variant c.902del, p.(Asn301Thrfs*3) in exon 6 results in a stop in the mutated allele. Analysis of mRNA stability indicates that truncated mRNA from the frameshift variant c.902del (p.N301Tfs*3) is degraded and thus, not producing truncated Np65 protein (Supplementary Fig. 1).

Fig. 2

NPTN monoallelic variants. (A) Diagram of the NPTN gene and localization of missense (black font) and frameshift and nonsense (blue font) variants identified in the individuals. The NPTN gene contains eight exons with obligatory (black lines) and alternative (blue lines) splicing points and yields two hNp isoforms (hNp55 or hNp65). In exon 1, the gray band represents the 5’ untranslated region and the light green band the signal peptide sequence. The splicing of exon 1 to exon 3 results in the elimination of exon2 encoding the hNp65-specific Ig-like domain I and leads to the synthesis of hNp55. Exons 3–5 encode Ig-like domains II and III and exon 6 encodes a single transmembrane domain (light red band) common to all hNp isoforms. The small exon 7 can be removed from hNp55 and hNp65 mRNAs producing proteins with cytoplasmic tails lacking a four amino acidic DDEP insert. (B) Missense variants substitute conserved amino acids located in highly conserved amino acidic sequences across different species. (C) The localization of the missense variants is displayed in the protein structure of hNp65. Ig: immunoglobulin.

Computational analysis of the NPTN missense variants

We performed a customized structural and thermodynamic analysis of NPTN missense variants using molecular dynamics and protein-protein docking modeling and calculating binding energies (∆G binding) (Fig. 3 and SMaterial). Based on the resolved structure of the Np65-specific Ig-like domain I³⁴, our computational procedures were robust to recreate the hNp65^wt-hNp65^wt trans-homophilic binding (Fig. 3A, top-down view). We found that W135, N130, and I133 have an important participation in the thermodynamically spontaneous attraction between hNp65^wt Ig-like domain I F-G loops (Fig. 3A, upper frame in lateral view). In the variant hNp65^p.W135 resulting from c.403T > A, the F-G loop structure is altered and the N130 and I133 are far off from reaching effective trans-interaction positions (Fig. 3A, middle frame in lateral view). An increased ∆G confirms the reduced binding efficiency of the hNp65^p.W135R F-G loop to form the pair hNp65^wt-hNp65^p.W135R. The interaction of the pair hNp65^p.W135R-hNp65^p.W135R was worsened by the appearance of an abnormal P122-P122 interaction with an even higher ∆G for their binding (Fig. 3A, middle frame in lateral view).

Fig. 3

Structure of the proteins resulting from NPTN missense variants from individual 1 (A, p.W135R) and individual 2 (B, p.P342L). (A) Views and ΔG binding of each of the trans-homophilic interactions were extracted from our molecular docking simulations and based on previous kinetic and crystallography studies. hNp65^wt (red) and hNp65^p.W135R variant (green). The sequential replacement of hNp65^wt by hNp65^p.W135R results in severe conformational changes with energetically less favorable and more unstable dimerization. Arrows indicate distances between the key amino acids. (B) Interactions resulting from molecular docking simulations between the transmembrane domain 10 (T10) of hPMCA (blue) with the transmembrane domain of hNp^wt (red) or with hNp^p.P342L (green) are based on the crystallography of the Np-PMCA complex. PM: plasma membrane. Arrows indicate that the distance necessary for the interacting amino acids P342 in hNp^wt and W1043 in hPMCA is drastically reduced for the pair L342 in hNp^p.P342L and W1043 in hPMCA.

In agreement with the reported crystallographic structure of the hNp-hPMCA complex³¹, we localized P342 at the hNp transmembrane domain (TMD) facing W1043 at the hPMCA TM10 and found it participates in the stable interaction of the proteins within the cell plasma membrane (lateral view in Fig. 3B). When the hNp65^p.P342L variant resulting from 1025C > A was projected onto the Np-PMCA interaction surface, we observed that the larger mutant residue L342 violates the effective interaction distance with hPMCA TM10 W1043, creating a thermodynamic constraint that would interfere with the intermolecular interaction (lower frame in top-down view in Fig. 3B). Having studied other missense mutations affecting hNp Ig-like domain II structure³⁵, we also analyzed the missense variant hNp65^p.A210T from c.1025C > A located at Ig-like domain II which was previously identified but not characterized³⁶ (Supplementary Fig. 2). Briefly, the switch from A210 to T210 resulted in replacement of stabilizing intramolecular interactions by weak CH-CH interactions with highly variable interaction distances. Furthermore, T210 adds a mutant polar OH group to a normally apolar environment that causes extra steric congestion of the domain and stabilization constraints to the Ig-like domain II structure (Supplementary Fig. 2).

Expression of NPTN missense variants and effects on PMCA levels

Based on our previous studies^6,33,35, we tested the expression levels of hNp65^wt, hNp55^wt, hNp65^p.W135R, hNp65^p.P342L, and hNp55^p.P226L in transfected HEK293T cells (Supplementary Fig. 3). Other missense mutations identified³⁶ were also analyzed (Supplementary Fig. 3) and are described in Supplementary material. As expected^6,33,35, hNp65^wt and hNp55^wt were efficiently detected by Western blot analysis (Supplementary Fig. 3A). Importantly, decreased expression was found for hNp65^p.W135R (p < 0.001 vs hNp65^wt) and hNp65^p.P342L (p = 0.033 vs hNp65^wt) whereas hNp65^p.A210T expression was only slightly decreased compared to control (p = 0.077 vs hNp65^wt, SFig. 3A,B). Levels of hNp55^p.P226L and hNp55^p.T78P were similar to hNp55^wt (SFig. 3A,B). Neuroplastin is an obligatory binding partner and post-transcriptionally promotes the expression of PMCA^6,12,13. Therefore, we examined the effect of the missense variants on the capacity of Np to increase hPMCA2 levels^6,35. hPMCA2 levels in hNp65^WT- and hNp55^WT-expressing cells were higher than the ones in non-transfected control cells (Supplementary Fig. 3B,C). Compared to PMCA2-expressing single transfected cells and to PMCA2/hNp65^WT- or PMCA2/hNp55^WT-expressing double transfected cells, all missense variants promoted hPMCA2 but some of them inefficiently. Indeed, hPMCA2 was less in cells expressing hNp65^p.P342L (p = 0.004 vs hNp65^wt), hNp65^p.W135R (p = 0.011 vs hNp65^wt) or Np55^p.P226L (p = 0.057 vs hNp55^WT) (Supplementary Fig. 3B,C). Compared to controls, the ability to increase hPMCA2 (hPMCA2/hNp ratio) was reduced for hNp65^p.P342L and hNp55^p.P226L but not for hNp55^p.T78P, hNp65^p.A210T or hNp65^p.W135R (Supplementary Fig. 3D) pointing to a specific necessity of this mutated proline residue in both hNP isoforms for normal levels of hPMCA2 expression in human cells.

The impact of P226L on hNp55 functionality was evaluated in Drosophila melanogaster, a classical system to study neurodevelopment and synaptic mechanism^37,38 (Fig. 4). In contrast to the three mammalian paralogs NPTN, BSG, and EMB, only a single ortholog gene encoding dBsg exists in Drosophila. DBsg shows an overall 25% amino acid sequence identity and a transmembrane domain homology of 69% (including adjacent intra- and extracellular amino acid residues) with hNp55 (Fig. 4A, B). Deletion of dBsg expression in muscle is known to be lethal at the late embryonic to early larval (L1) stage³⁹. Thus, we tested whether the co-expression of hNp55^wt or hNp55^p.P226L rescues the lethal phenotype triggered by dBsg knockdown due to mef2-Gal4-induced expression of dsRNA (Fig. 4C,D). As expected, dBsg knockdown caused a highly penetrant lethality around the L1 stage. The larval lethality was fully rescued by hNp55^wt as virtually all progeny developed into viable adult flies, indicating tolerance to the differences between dBsg and hNp55^wt. Strikingly, hNp55^p.P226L displayed only minimal if any rescue capacity, as all progeny died before the L2 stage (Fig. 4D).

Fig. 4

Lethality due to dBsg-KD during early muscle development is rescued by hNp55 but not hNp55^P226L. (A,B) Structural homology between hNp55 and dBsg. (C) Crossing scheme for the assessment of the rescue capacity of hNp55 and hNpP226L during muscle development. Flies homozygous for the early-onset muscle Gal4 driver mef-Gal4 were crossed to effector lines homozygous for either UAS-hNp55 or UAS-hNp55P226L, and heterozygous for UAS-bsgdsRNA over the balancer chromosome TM6B carrying the dominant markers Humeral (Hu, visible on adults) and Tubby (Tb, visible on L3 larvae and pupae). Control crosses lacked the UAS-hNp55 or UAS-hNpP226L effectors. (D) Pupal progeny from the crosses in (C) on the wall of a culture vial. Note that in crosses with dbsg-KD alone or with dbsg-KD plus hNp55^P226L only round-shaped Tubby pupae are detectable, whereas crosses with dbsg-KD plus hNp55 give rise to both Tubby and normally shaped pupae (stars). This clear discrepancy is confirmed by counts of adult progeny with TM6B or without TM6B.

Effect of NPTN missense variants on cytosolic Ca²⁺ regulation

Neuroplastin-PMCA complexes are crucial for cytosolic Ca²⁺ extrusion and shaping of Ca²⁺ signaling in brain neurons^{6,12,13,31–33,35}. To evaluate the functional effect of NPTN missense variants on Ca²⁺ regulation, we investigate electrically-evoked cytosolic Ca²⁺ transients using Ca²⁺ imaging^31,34 in 14–16 days-old GCaMP5G-expressing primary hippocampal neurons (referred to as GCaMP5G-neurons) (Fig. 5A). We quantified peak amplitude, half-width, and decay time of the evoked Ca²⁺ transients, as these parameters reflect how Ca²⁺ transients are shaped by the levels and activity of Np-PMCA complexes in synapses and dendrites (Fig. 5A). In line with the previous reports showing that Np^wt over-expression adds on endogenous Np to promote PMCA levels and function (gain-of-function)^{6,12,31–33,35,40,41}, GCaMP5G-neurons co-expressing hNp65^wt or hNp55^wt displayed smaller evoked Ca²⁺ transients with faster restoration of basal Ca²⁺ levels compared to control GCaMP5G-neurons (Fig. 5B-D). In contrast, hNp65^p.W135R, hNp65^p.P342L and hNp55^p.P226L caused abnormal or incomplete Ca²⁺ transients. Indeed, whereas peak amplitudes were similarly reduced in GCaMP5G-neurons co-expressing either hNp65^p.W135R or hNp55^p.P226L compared with their wild-type expressing controls, both mutants displayed an increased half-width and decay time indicating longer Ca²⁺ transients with slower recovery to baseline (Fig. 5B-C). hNp65^p.P342L accelerated the restoration of basal Ca²⁺ levels but did not sufficiently reduce the Ca²⁺ peak amplitude vs hNp65^wt (Fig. 5D). Thus, these data indicate that these missense variants act dominant negatives rather than as a complete loss-of-function.

Fig. 5

Effect of NPTN missense variants from individual 1 and 2 on cytosolic Ca²⁺ transients. (A) Summarized Ca²⁺ imaging sequence (1pic/400ms) of a representative GCaMP5G-expressing secondary dendrite (100µm section) depolarized with a field electrical stimulation of 10 pulses at 20 Hz (red asterisk at t₁). From these image sequences, the parameters peak amplitude (P.A.), half-width (H.W.), and 90% decay time (D.T.) were calculated to describe the evoked cytosolic Ca²⁺ transients³¹. (B-D) Parameters were obtained from GCaMP5G-expressing dendrites (open circles) transfected or not with one of the wt isoforms or one of the NPTN variants. For data presentation clarity mean ± S.E.M. are displayed, but additionally S.D. is given for each condition. (B) For P.A.: control (no label) n = 81; mean = 1931 ± S.D.=1333; hNp65^wt n = 71; 1408 ± 1006, hNp65^p.W135R n = 57; 1335 ± 798. **p < 0.01 unpaired t-test vs. control; ns: no significant difference; unpaired t-test vs. hNp65^wt. H.W.: control 1349 ± 265; hNp65^wt 1108 ± 139; hNp65^p.W135R 1320 ± 306. ****p < 0.0001 or ns vs. control. ###p < 0.001 vs. hNp65^wt. D.T.: control mean = 1724 ± 520; hNp65^wt 1359 ± 448; hNp65^p.W135R 1775 ± 570. ****p < 0.0001 or ns vs. control. ####p < 0.0001 vs. hNp65^wt. (C) For P.A.: control (no label) n = 64, mean = 2116 ± 1237; hNp55^wt n = 60, 1121 ± 754; hNp55^p.P226L n = 82, 909 ± 510. ***p < 0.001 unpaired t-test vs. control. ##p < 0.01 or ns unpaired t-test vs. hNp55^wt. &&p < 0.01 one-way ANOVA. H.W.: control 1332 ± 467; hNp55^wt 976 ± 199; hNp55^p.P226L 1079 ± 231. **p < 0.01 or ***p < 0.001 vs. control. #p < 0.05 vs. hNp55^wt. D.T.: control 2034 ± 866; hNp55^wt 1123 ± 403; hNp55^p.P226L 1422 ± 600. **p < 0.01 or ****p < 0.0001 vs. control. ##p < 0.01 vs. hNp55^wt. &&p < 0.01 one-way ANOVA. (D) For P.A.: control (no label) n = 64, 2597 ± 1598; hNp65^wt n = 67, 1102 ± 772; hNp65^p.P342L n = 72, 1564 ± 1133. ***p < 0.001 unpaired t-test vs. control. ##p < 0.01 unpaired t-test vs. hN65^wt. &&p < 0.01 one-way ANOVA. H.W.: control 1574 ± 496; hNp65^wt 1100 ± 462; hNp65^p.P342L 1158 ± 552. ***p < 0.001 vs. control. D.T.: control 1831 ± 728; hNp65^wt 1137 ± 643; hNp65^p.P342L 1312 ± 830. ***p < 0.001 or ****p < 0.0001 vs. control.

Nptn heterozygosity affects PMCA brain levels and social behavior in mice

A reduction of Np expression by ~ 50% using a pan siRNA against Np mRNA resulted simultaneously in a partial reduction of PMCA1-4 in 14–16 days-old hippocampal neurons (Supplementary Fig. 4A-C) and increased half-width and decay time of electrically-evoked Ca²⁺ transients, as visualized using GCaMP7s-based Ca²⁺ imaging (Supplementary Fig. 4D). As anticipated, immunohistochemical evaluation in the hippocampus of 14 days-old Nptn^+/− pups demonstrated that Nptn heterozygosity, resulting in ~ 50% reduction in Np levels, causes ~ 45% loss of PMCA1-4 in developing neurons (Fig. 6A,B). Therefore, heterozygous levels of Np^wt are insufficient to maintain normal endogenous PMCA levels in the rodent hippocampus and cortex.

Fig. 6

Np-PMCA hypoexpression and social behavior in Nptn^+/− mice. (A,B) Immunoreactivity of anti-Np55/65 (red) and anti-PMCA1-4 (green) antibodies to brain slices from Nptn^+/+ and Nptn^+/− mice was evaluated using confocal microscopy. Nuclei were labelled with DAPI. Maximal projections and fluorescence signal quantification in CA1 hippocampal area. Nptn^+/+ (n = 10, green) and Nptn^+/− (n = 12, blue) mice were measured after habituation (phase 1) in the 3-chamber test. (C) The pictures show the experimental phases of the three-chamber test as initiated by the habituation of either Nptn^+/− or Nptn^+/+ mice (Phase 1). (D) The graph shows the time spent by the tested mouse in the compartments with an unfamiliar mouse (novel 1) or with an empty cup (empty) (Phase 2). The novel mouse is preferred over an empty cup. D. Nptn^+/+ mice prefer the novel 2 mouse over the familiar mouse, whereas Nptn^+/− mice displayed no significant preference for an unfamiliar mouse (novel 2) versus the familiar mouse (Phase 3). Data are presented as mean ± S.E.M. (One-way ANOVA with Scheffe post-hoc; **p < 0.01; ***p < 0.001; ****p < 0.0001; ns: no significant difference).

Autism is a common trait shared by the NPTN individuals (Table 1). While general stereotypic behaviors were not observed in the heterozygous Nptn^+/− mice and specifically not detected in the marble burying test (not shown; Supplementary Table 5), these mice displayed reduced anxiety in the O-maze test (not shown; Supplementary Table 5). Further, we assessed autism-like behavioral features related to social interactions using the three-chamber social interaction assay (Fig. 6C). Both Nptn^+/− and Nptn^+/+ mice displayed a similar preference for a novel mouse compared to an empty cup (Fig. 6C). However, in contrast to Nptn^+/+, Nptn^+/− mice did not prefer a stranger over an already familiarized mouse (Fig. 6D). Such altered social interactions observed in Nptn^+/− mice have also been reported in other mouse models of genetically caused autism⁴².

Discussion

We describe seven individuals with an overlapping, albeit nonspecific, phenotype. All individuals displayed de novo variants in NPTN and were diagnosed with autism and neurodevelopmental disorders, pointing to the existence of pathological mechanisms related to alterations in NPTN that affect neurodevelopment. The NPTN variants disrupted critical functions of Np on the regulation of PMCA levels and cytosolic Ca²⁺ dynamics. In particular, the missense variants exhibit dominant negative features, whereas the nonsense or frameshift variants resulted in loss of function, affecting Np expression, PMCA levels, and Ca²⁺ signal regulation. Furthermore, monoallelic Nptn condition proved insufficient to support normal PMCA function and led to autism-like social behavior in transgenic mice. This report not only confirms prior assumptions regarding the relevance of Np in human neurodevelopment derived from machine learning approaches⁴³ and correlative genetic studies¹¹, but also provides a wider perspective to prospective pathological mechanism relevant to other affected individuals, such as those with de novo mutations in PMCA1–4-encoding genes (ATP2B1–4)^19–30.

It is confirmed that Np binds, stabilizes into protein complexes, and promotes the catalytic function of PMCA1-4. It is also shown that a significant decrease in Np levels, due to Np gene deletion or mutation or Np mRNA-interference, results in PMCA1-4 loss in neurons. Notably, in our assays, missense variants of the NPTN individuals 1 and 2 resulted in decreased production of hNp levels carrying different dysfunctional residues that replaced key conserved amino acids. While the variant of individual 1 substitutes a critical tryptophan within the amino acid sequence at the extracellular and isoform-specific Ig-I of hNp65, individual 2‘s variant replaces a proline located at the common transmembrane segment of hNp55 and hNp65 which is embedded into the plasma membrane. The identified binding instability and conformational alterations in these human missense variants indicate that they are either not effectively produced or avidly degraded, as is the case for other Np variants affecting hearing in mice³⁵.

Our results in HEK293T cells and Drosophila melanogaster further support that the NPTN variant of patient 2, yielding hNp55^p.P226 and hNp65^p.P342L, results in impaired biosynthesis/ degradation and is insufficient to support normal PMCA function. This impairment seems to be most detrimental during early development of the flies and is compatible with a dominant negative effect on PMCA function. Interestingly, although human variants from individuals 1 and 2 were expressed at lower levels compared to wt Np isoforms, the NPTN Ig-I variant of patient 1 hNp65^p.W135R was able to promote PMCA, but the NPTN transmembrane variant of patient 2 was inefficient in this property. Stop codon-producing nonsense or frameshift variants in individuals 3-to-7 do not produce truncated proteins and, thus, their effects may result from a diminished haploid production of hNp55/65 insufficient to maintain normal PMCA levels. Indeed, we demonstrated that either titrated Np RNAi-targeted knockdown in isolated cultured neurons or monoallelic condition in Nptn^+/− mice yielded appr. 50% expression levels of Np with a significant reduction in PMCA levels. Ca²⁺ imaging confirmed that a progressive decrease on Np-PMCA content is mirrored by a progressive increase in the Ca^+ 2 signal parameters peak amplitude, half width, and decay time. Therefore, we conclude that missense variants display loss-of-function with dominant negative features and that nonsense or frameshift variants result in hypofunctional hNp levels with diminished hNp-PMCA function.

Insufficient PMCA function may not be the only mechanism underlying the phenotypes of the NPTN individuals in this cohort. The missense variant in individual 1 is located at an important extracellular module of hNp65 that mediates homophilic Np65-Np65 trans-interactions³³. In fact, competition of this motif with high-affinity peptides or antibodies destabilizes glutamatergic synaptic contacts and impairs synaptic transmission⁴⁴. As our thermodynamic calculations indicate that this Np65-specific variant displays weakened binding, it is possible to hypothesize the occurrence of an insufficient structural stabilization of excitatory synapses during synaptogenesis in individual 1. Additionally, as reduced synapse formation triggered by Np55 and Np65 occurs in Nptn^−/− neurons due to failed TRAF6 signaling^33,44, it is plausible that production of hNp variants in individuals 1 and 2, or haploid wild-type hNp production in individuals 3-to-7, would not trigger Np-TRAF6-dependent synapse formation sufficiently. On the other hand, insufficient hNp production could also alter the excitatory-inhibitory balance affecting development and maturation of neuronal circuits leading to epilepsy or ataxia in the NPTN individuals 1-to-3. This idea is based on the requirement of Np for the correct synaptic localization and function of AMPA receptors⁴⁵ and α1/2 subunit-containing GABA type A receptors^44,46. Interestingly, recent studies suggest that reduced synaptic transmission involving binding of Np to GABA type A receptors sensitizes rodents to pentylenetetrazole-induced epilepsy^40,47. Therefore, hypofunctions and/or malfunctions of other binding partners of hNp such as AMPA and/ or GABA type A receptors may play contributing role in the phenotype of the NPTN individuals. These possibilities remain to be evaluated in the future.

Our results in the Nptn^+/− mice shown here, in combination with our previous reports demonstrating behavioral abnormalities and cognitive deficiencies in constitutive Nptn^−/− mice and inducible Np-deficient mice¹², provide strong evidence for the necessity of Np during brain development. Indeed, induced Nptn elimination after normal development demonstrates that Np is acutely required for associative learning and memory in adult mutant mice, while it is not required for other behaviors tested in open field, O-maze, light/dark avoidance learning, light/dark avoidance memory, and startle responses¹². In contrast, constitutive Nptn^−/− mice fail to perform in all these behavioral tests¹². Here, we show that Nptn^+/− mice with incomplete Np levels display altered social interaction behavior. This confirms our previous results showing impaired social interaction in Nptn^−/−, but not in inducible Np-deficient mice, and highlights a strict requirement for normal Np function for successful cognitive development. Furthermore, the archetypical autism-model mouse behavior of reduced preference for a novel vs. a familiar mouse in the social interaction test diplayed by Nptn^+/− mice strongly supports the causality of the NPTN mutations for the autism diagnozed in all NPTN individuals.

The facts that all PMCA are obligatory binding partners of Np and that de novo variants in ATP2B1 and ATP2B2 are identified as causing neurodevelopmental disorders including DD/ID and autism^19–25 further strengthens the association of NPTN-ATP2B1-2 and neurodevelopmental disorders. Neuroplastin peripheral malfunctions identified in homozygous Np-deficient mutant mice are not evident in the NPTN cohort yet, and, thus they need to be confirmed. Both Nptn- and Atp2b2-deficient mice and ATP2B2 individuals are deaf, whereas the NPTN cohort described here do not show hearing deficits. This may be attributed to the young age at diagnosis where degeneration of hair cells due to loss of Np-PMCA may not have progressed yet dramatically. Supporting this idea, outer and inner hair cells in the Nptn^−/− cochlea initially develop normally before undergoing degeneration that leads to hearing loss⁴¹. Alternatively, the residual expression level of Np in our individuals may be sufficient to support hearing and eventually may delay hearing loss to later ages. While neurons express very high levels of Np55 and Np65, most peripheral cell types synthesize only Np55. Altered inflammatory responses by immune cells⁴⁸ or impaired pancreatic beta cells function⁴⁹ in mice associated to Np55-PMCA deficiency have not been tested nor clinically manifested in the NPTN cohort.

We have provided the first evidences in mice connecting Nptn expression with normal cognitive capabilities and PMCA expression^6,12. Consistently Np-PMCA molecular interaction and its functions in intracellular Ca²⁺ regulation have been confirmed and further detailed^{31–33,35,40,41}. Although Desriveres et al. using large-scale gene association identified NPTN playing a potential role for intellectual deficits in humans¹¹, and Dhindsa et al. using a machine learning approach based on gene constraint, expression, and other gene-level annotations, predicted NPTN as preferably causing an autosomal dominant neurodevelopmental condition (percentiles of 90.8 for DD, 94.0 for ASD, and 97.9 for developmental and epileptic encephalopathy)⁴³, more direct evidence for a critical relevance of NPTN for brain development was missing. To our knowledge, this is the first report linking NPTN to human neurodevelopment. Our clinical, animal model, cellular, and molecular data indicate that the de novo variants in NPTN are pathogenic due to dominant negative (missense variants) or as loss-of-function (nonsense/ frameshift variants) effects on PMCA function. Both types of variants lead to a clinically non-specific neurodevelopmental disorder with varying severity, albeit based on a relatively small cohort. During the writing of this report, additional individuals with variants NPTN and neurodevelopmental disorders and autism have been identified. We have, therefore, adopted NPTN at the Human Disease Genes series (www.humandiseasegenes.nl) to promote future clinical research revealing a genotype–phenotype correlation.

In summary, we establish that de novo variants in NPTN as causative for a neurodevelopmental disorder and autism. Based on several lines of evidence shown here and also in the literature, we proposed that impaired or insufficient function of the Np-PMCA complexes may contribute to the NPTN disorder.

Methods

Recruitment of affected individuals and animal experimentation

This study was approved by the ethics committee of the University of Leipzig (402/16-ek). Written informed consent for molecular genetic testing and permission for publication of the data were obtained from all individuals and/or their legal representatives by the referring physicians according to the guidelines of the ethics committees and institutional review boards of the respective institutes. The compilation of the cohort was supported by international collaboration and online matchmaking via GeneMatcher⁵⁰. Phenotypic and genotypic information was obtained from the referring collaborators using a standardized questionnaire. Heterozygous neuroplastin-deficient mice Nptn^+/− were described¹². Mice were kept with a 12 h light/dark cycle and food and water ad libitum.

Animal husbandry, behavioral tests, and tissue collection were conducted in accordance with German (Tierschutzgesetz TierSchG) and European legislations (European Communities Council Directive (2010/63/EU) for the care of laboratory animals) and with the respective legal and ethical approval by the legal authorities (Landesverwaltungsamt Halle, Sachsen-Anhalt, Germany).

Identification of NPTN variants

Trio exome sequencing was performed for all affected individuals and their parents, except for individuals 3 and 5 (singleton exome only). All individuals were analyzed in the context of local diagnostic protocols. As there was no causative variant identified in a known rare disease gene, and thus all individuals lacked a definite diagnosis, research evaluation of the sequencing data was conducted to potentially identify causative variants in candidate genes. The gnomAD v4 dataset served as the control population⁵¹. There were no significant findings, other than the described variants in NPTN, which could explain the phenotype of the respective individuals. All variants described were aligned to hg38, mapped to the NPTN MANE Select transcript NM_012428.4 and classified according to the ACMG criteria^52,53 (Table S1). NPTN gene sequences used for comparison are Homo sapiens, NP_001154835; Macaca fascicularis, XP_005560051; Mus musculus, NP_001392991; Rattus norvegicus, NP_001400276; Equus caballus, XP_023509800; Danio rerio, NP_991268.

In silico prediction

In silico predictions of the missense variants were assessed using CADD-v1.6n⁵², REVEL⁵³, MutPred2⁵⁶, VEST4⁵⁷, and BayesDel⁵⁸ using cutoffs for deleterious predictions from Pejaver et al.⁵⁹ (Supplementary table 4).

Molecular dynamics and docking

Molecular dynamics simulations were performed with Gromacs 2020.3⁶⁰ and the OPLS-AA force field⁶¹. The models were solvated in a cubic box with the SPCE water model, and NaCl counterions were added to neutralize the system⁶². System energy was minimized with the steepest descent algorithm up to a convergence criterion of 1000 kJ mol^–1nm^–1. To equilibrate temperature and pressure, an equilibration step of 200 ps in the NVT ensemble and an equilibration of 200 ps in the NPT ensemble were performed. After that, a 100ns production run in the NPT ensemble was performed to obtain geometrical information on all variants considered. The LINCS algorithm was used to restrain hydrogen bonds during equilibration and production runs⁶³, while the PME method was used for treating long-range electrostatic interactions with a cutoff of 10Å⁶⁴. Temperature was kept constant at 300 K with the V-rescale thermostats, while pressure was kept constant at 1 bar with the Parrinello-Rahman barostat^65,66. The atomic coordinates of hNp65^wt were taken from our previous work³⁴ and variants were obtained by replacing the corresponding amino acid using Pymol software. For the analysis of hNp65^p.W135R, we used the last frame obtained from the molecular dynamics simulation to compare Np65 dimers as described elsewhere³⁵. Crystallographic information described elsewhere³¹ was used for the docking of hNp65^p.P342L to hPMCA, which was performed directly from the structure obtained with Pymol, because this amino acid change occurs within the intermembrane space. The binding interfaces with low-energy conformations were identified using the HADDOCK docking protein-protein web server^{67, 68}.

Cell cultures

Human embryonic kidney cells (HEK293T) were prepared in Dulbecco’s Modified Eagle Medium (DMEM, Gibco) with 1% penicillin/streptomycin (Gibco), 1% L-glutamine (Gibco) and 10% fetal bovine serum (Gibco) at 37°C, 5% CO₂⁶. Primary hippocampal neurons were prepared from E16-18 rat or mouse embryos following our published protocols^6,32,44. In short, neurons were dissociated with trypsin-EGTA for 15 min at 37°C. 50,000 neurons were plated on poly-D-lysine-coated glass coverslips for 12-well plates. After one hour, when the cells were well-attached to the slips, the plating medium DMEM 1% penicillin/streptomycin, 1% L-glutamine, and 10% horse serum was replaced with 1 ml of Neurobasal medium with 2% B27 supplement (Thermo Fisher Scientific Inc.), 1% penicillin/streptomycin, and 1% L-glutamine. At day 7 in vitro, 100 µl of fresh culture medium was added.

Constructs, plasmids, siRNA, and transfections

Constructs of human Np wild-type (hNpwt) and human PMCA2 (hPMCA2; Addgene, ID: 47584, Cambridge, MA) have been characterized⁶. TagRFPT-GCaMP5G plasmid has been described³². PGP-CMV-jGCaMP7f plasmid was obtained from Addgene (ID: 104483). We used control (scrambled) siRNA duplexes and siRNA targeting all premature Np RNA (Santa Cruz Biotechnology, Inc., ID: sc-149938). NPTN missense variants were generated by PCR amplification (see Supplementary Material). DNA-fragments were inserted into linearized vector FUGW (Addgene, ID: 14883) using cold-fusion cloning (System Biosciences, Palo Alto, CA). Transfections were performed with Lipofectamine 2000 following the company’s protocol (Invitrogen, Darmstadt, Germany). Hippocampal neurons were transfected with plasmids (1 µg/well) with or without control (scrambled) siRNA or Np siRNA (0.1-1 µg) at days 10–11 in vitro. 300000 HEK293T cells were transfected with plasmids (1.5 µg/well) 24 hours after seeding in 6-well plates and harvested 24 hours later.

Western blot

HEK293T cells were harvested with lysis buffer (1% Triton X-100 in 50 mM Tris/HCl, pH 8.0 protease inhibitors), homogenized using an ultrasonic homogenizer, and spun down at 12.000g for 20 min. The supernatant was collected and mixed uniformly with 2X SDS loading buffer and boiled at 95°C for 5 min. Proteins were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) on 10% gels and electro-transferred to a nitrocellulose membrane (Cytiva, AmershamTM ProtranTM 0.45µm NC). After blocking with 5% nonfat milk in TBS-T solution, TBS containing 0.1% Tween 20 for 1 h, the membrane was incubated with primary antibodies (see list in SMaterial) overnight at 4°C. Afterwards, the membrane was washed three times with TBS-T and incubated with secondary antibodies (see list in SMaterial) for 1 h at room temperature. After washing, the membrane was incubated with Immobilon Western Chemiluminescent HRP Substrate (EMD Millipore) and chemiluminescence was detected using Intas Chemocam ECL Imaging system.

Immunocytochemistry

Cultured neurons were fixed and stained according to established protocols^6,32,44.

Briefly, cultures were fixed with 4% paraformaldehyde (PFA) containing 2% sucrose in 1X PBS for 8–10 min, washed carefully with 1X PBS and incubated with blocking solution (10% horse serum and 0.1% Triton X-100 in PBS) two times, 20 min each. Then they were incubated for one hour with primary antibodies (SMaterial) mixed with the blocking solution. After incubation, the coverslips were washed 3 times with 1X PBS for 10 min each. Secondary antibodies (SMaterial) diluted in the blocking solution were added to the coverslips and incubated for 1 h. The coverslips were then washed 3 times with 1X PBS and mounted on microscopy slides with Mowiol.

Confocal microscopy and image quantification

Images of neurons were acquired using an oil immersion objective HCXAPO 63X/1.40 NA coupled to an upright confocal microscope TCS SP5 (Leica), with a pinhole value of 0.75 AU, under sequential scanning mode (200 Hz), with 1-, 3- or 6-fold digital magnification for the whole cell, cell soma, or dendrites, respectively, and digitized in a 1054 X 512 format.

Calcium imaging

Ca²⁺ imaging was performed in transiently transfected primary hippocampal neurons at 14–16 days in vitro following our published procedures³². Glass coverslips were carefully placed into an imaging magnetic chamber equipped with two silver wires for field electrical stimulation (Warner Instruments, Hamden, CT) and then filled with 1 ml of Tyrode’s solution. Stimulation was 10 or 20 biphasic pulses (1 ms duration each) generated with a S48 stimulator (Astro-Med, Inc., West Warwick, RI, USA). Evoked Ca²⁺ transients were recorded using an inverted microscope Observer D1 (Zeiss, Jena, Germany) with a 63X/1.20 NA objective and an EMCCD camera Evolve 512 (Delar Photometrics, Tucson, AZ) under the control of VisiView software (Visitron Systems GmbH, Puchheim, Germany). Fluorescence intensity changes were quantified using Fiji/ImageJ software and parameters were extracted using pCLAMP 10 (Molecular Devices, San Francisco, CA).

Drosophila melanogaster studies

UAS-hNp55 transgenic constructs were established in the vector pJFRC12 (addgene clone 26222) and used for PhiC31-mediated germline transformation into the attP40 target site on the second chromosome (cytological position 25C6) of the recipient strain (y1 w67c23; P[CarryP]attP40)^53–55. This procedure was performed by BestGene (Chino Hills, CA). Transgenic flies were identified based on orange eye color and established as stocks carrying the CyO^GFP balancer chromosome. For further details on the transgenic lines see Supplementary material.

Social interactions of Nptn-deficient mice

Social interactions of Nptn^+/− and Nptn^+/+ adult male matched littermate mice were analyzed by the three-chamber test as described⁶⁹. Briefly, during 3 test phases (10 min each), the mouse could explore all compartments. In phase 1, the mouse was alone for habituation. In phase 2, an unfamiliar C57BL/6NCrl wild-type mouse (male, novel 1) was placed in one of the wire cups. In phase 3, another unfamiliar C57BL/6NCrl mouse (same sex, novel 2) was added to the other cup. Time spent in each compartment, in contact with strangers, and transitions between compartments were recorded.

Statistical analysis

Statistical analysis of data was performed using Prism9 software (GraphPad), with outliers from raw data screened out with the Grubbs' test (Q = 1). For Western blots and Immunocytochemistry, statistical analysis of optical density measurement data used Student’s t-test. For Ca²⁺ imaging, a Mann-Whitney U test was used for group comparisons for non-parametric data with sample sizes > 20, and a Wilcoxon matched-pairs test for the inner group comparison. For the analysis of behavior, Statview (SAS Institute Inc., Cary, NC) was used for Analysis of Variance (repeated measures ANOVA) and post hoc analysis (Scheffe's test). P-value smaller than 0.05 (P < 0.05) was considered significant.

Acknowledgements

We are extremely grateful to the families that participated in this study. We thank Kathrin Pohlmann for her assistance during the cell culture preparations and Karla Sowa for helping with the mouse experiments. We also thank Dr. Marnie Phillips for her helpful comments on the manuscript. A.M. and H.S. worked at and R.H-M. headed the Laboratory of Neuronal and Synaptic Signals at the Leibniz Institute for Neurobiology.

Author Contribution

Y.L., H.S., A.M., and R.H.M. produced and analyzed cellular and molecular data. U.T. and D.M., produced and analyzed animal model data. W.A. and R.O.T. produced and analyzed in silico data. J.S.C., N.R., J.L., G.V., F.L., E.K., B.A., H.C., M.J.G.S., B.B.A.deV., R.P., G.P., K.W., Y.B., E.M.P., K.P. evaluated patients and wrote clinical reports. All authors wrote text and agreed with final manuscript version publication. K.P., D.M., and R.H.M. prepared manuscript final version.

Funding

This work was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – 526008379 (R.H-M. and D.M.), the China Scholarship Council (Y.L.). Work by A.M. and H.S. was partially financed by the Center for Behavioral Brain Sciences (CBBS) grant to R.H-M. R.H-M. was supported by NIH Awards NS106244 and EY022730 (to Matthew T. Colonnese). R.O.T. thanks Powered@NLHPC: This research was partially supported by the supercomputing infrastructure of the NLHPC (ECM-02). W.A. was partially supported by HPC OCÉANO (FONDEQUIP Nº EQM170214). JSC receives support from NIH National Institute of Child Health and Human Development (P50 HD103538), NIH National Institute of Neurologic Disorders and Stroke (1U24NS131172-01), and U.S. Department of Health and Human Services, Health Resources and Services Administration (MCH T7317245). B.B.A.d.V was supported by the Dutch Organisation for Health Research and Development for ZON-MW grant no. 912-12-109. Multiple authors of this publication are members of the European Reference Network on Rare Congenital Malformations and Rare Intellectual Disability ERN-ITHACA.

Competing interests

HC and MJGS are employees of and may hold stock in GeneDx, LLC. The other authors report no competing interests.

Data availability

Identified variants in NPTN have been uploaded to ClinVar https://www.ncbi.nlm.nih.gov/clinvar/submitters/506086/.

References

Khan I, Leventhal BL Developmental Delay. StatPearls [Internet] 2024. Treasure Island (FL): StatPearls Publishing. https://www.ncbi.nlm.nih.gov/books/NBK562231/

Zeidan J, Fombonne E, Scorah J, Ibrahim A, Durkin MS, Saxena S, Yusuf A, Shih A, Elsabbagh M (2022) Global prevalence of autism: A systematic review update. Autism Res 15(5):778–790. 10.1002/aur.2696

Jansen S, Vissers LELM, de Vries BBA (2023) The Genetics of Intellectual Disability. Brain Sci. ;13(2):231. 10.3390/brainsci13020231. PMID: 36831774

Bowling KM, Thompson ML, Amaral MD, Finnila CR, Hiatt SM, Engel KL, Cochran JN, Brothers KB, East KM, Gray DE, Kelley WV, Lamb NE, Lose EJ, Rich CA, Simmons S, Whittle JS, Weaver BT, Nesmith AS, Myers RM, Barsh GS, Bebin EM, Cooper GM (2017) Genomic diagnosis for children with intellectual disability and/or developmental delay. Genome Med 9(1):43. 10.1186/s13073-017-0433-1

Husson T, Lecoquierre F, Cassinari K, Charbonnier C, Quenez O, Goldenberg A, Guerrot AM, Richard AC, Drouin-Garraud V, Brehin AC, Soleimani M, Taton R, Rotharmel M, Rosier A, Chambon P, Le Meur N, Joly-Helas G, Saugier-Veber P, Boland A, Deleuze JF, Olaso R, Frebourg T, Nicolas G, Guillin O, Campion D (2020) Rare genetic susceptibility variants assessment in autism spectrum disorder: detection rate and practical use. Transl Psychiatry 10(1):77. 10.1038/s41398-020-0760-7

Herrera-Molina R, Mlinac-Jerkovic K, Ilic K, Stöber F, Vemula SK, Sandoval M, Milosevic NJ, Simic G, Smalla KH, Goldschmidt J, Bognar SK, Montag D (2017) Neuroplastin deletion in glutamatergic neurons impairs selective brain functions and calcium regulation: implication for cognitive deterioration. Sci Rep 7(1):7273. 10.1038/s41598-017-07839-9

Ilic K, Mlinac-Jerkovic K, Sedmak G, Rosenzweig I, Kalanj-Bognar S (2021) Neuroplastin in human cognition: review of literature and future perspectives. Transl Psychiatry 11(1):394. 10.1038/s41398-021-01509-1

Smith M, Spence MA, Flodman P (2009) Nuclear and mitochondrial genome defects in autisms. Ann N Y Acad Sci 1151:102–132

Liu Y, Zhang Y, Zarrei M, Dong R, Yang X, Zhao D, Scherer SW, Gai Z (2020) Refining critical regions in 15q24 microdeletion syndrome pertaining to autism. Am J Med Genet B Neuropsychiatr Genet 183:217–226

10.

Sharp AJ, Selzer RR, Veltman JA, Gimelli S, Gimelli G, Striano P, Coppola A, Regan R, Price SM, Knoers NV, Eis PS, Brunner HG, Hennekam RC, Knight SJ, de Vries BB, Zuffardi O, Eichler EE (2007) Characterization of a recurrent 15q24 microdeletion syndrome. Hum Mol Genet 16(5):567–572. 10.1093/hmg/ddm016Epub 2007 Mar 14. PMID: 17360722

11.

Desrivières S, Lourdusamy A, Tao C, Toro R, Jia T, Loth E, Medina LM, Kepa A, Fernandes A, Ruggeri B, Carvalho FM, Cocks G, Banaschewski T, Barker GJ, Bokde AL, Büchel C, Conrod PJ, Flor H, Heinz A, Gallinat J, Garavan H, Gowland P, Brühl R, Lawrence C, Mann K, Martinot ML, Nees F, Lathrop M, Poline JB, Rietschel M, Thompson P, Fauth-Bühler M, Smolka MN, Pausova Z, Paus T, Feng J, Schumann G, IMAGEN Consortium (2015) Single nucleotide polymorphism in the neuroplastin locus associates with cortical thickness and intellectual ability in adolescents. Mol Psychiatry 20(2):263–274. 10.1038/mp.2013.197Epub 2014 Feb 11

12.

Bhattacharya S, Herrera-Molina R, Sabanov V, Ahmed T, Iscru E, Stöber F, Richter K, Fischer KD, Angenstein F, Goldschmidt J, Beesley PW, Balschun D, Smalla KH, Gundelfinger ED, Montag D (2017) Genetically Induced Retrograde Amnesia of Associative Memories After Neuroplastin Ablation. Biol Psychiatry 81(2):124–135. 10.1016/j.biopsych.2016.03.2107

13.

Schmidt N, Kollewe A, Constantin CE, Henrich S, Ritzau-Jost A, Bildl W, Saalbach A, Hallermann S, Kulik A, Fakler B, Schulte U (2017) Neuroplastin and Basigin Are Essential Auxiliary Subunits of Plasma Membrane Ca2+-ATPases and Key Regulators of Ca2 + Clearance. Neuron 96(4):827–838e9. 10.1016/j.neuron.2017.09.038

14.

cheuss V, Yasuda R, Sobczyk A, Svoboda K (2006) Nonlinear [Ca2+] signaling in dendrites and spines caused by activity-dependent depression of Ca2 + extrusion. J Neurosci 26(31):8183–8194

15.

Alves VS, Oliveira FA (2024) Plasma membrane calcium ATPase powered by glycolysis is the main mechanism for calcium clearance in the hippocampal pyramidal neuron. Life Sci 344:122554. 10.1016/j.lfs.2024.122554

16.

Chen HY, Chesler M (2015) Autocrine boost of NMDAR current in hippocampal CA1 pyramidal neurons by a PMCA-dependent, perisynaptic, extracellular pH shift. J Neurosci 35(3):873–877

17.

Stawarski M et al (2020) Neuronal glutamatergic synaptic clefts alkalinize rather than acidify during neurotransmission. J Neurosci 40(8):1611–1624

18.

Sibarov DA, Antonov SM (2018) Calcium-dependent desensitization of NMDA receptors. Biochem (Mosc) 83(10):1173–1183

19.

Rahimi MJ, Urban N, Wegler M, Sticht H, Schaefer M, Popp B, Gaunitz F, Morleo M, Nigro V, Maitz S, Mancini GMS, Ruivenkamp C, Suk EK, Bartolomaeus T, Merkenschlager A, Koboldt D, Bartholomew D, Stegmann APA, Sinnema M, Duynisveld I, Salvarinova R, Race S, de Vries BBA, Trimouille A, Naudion S, Marom D, Hamiel U, Henig N, Demurger F, Rahner N, Bartels E, Hamm JA, Putnam AM, Person R, Abou Jamra R, Oppermann H (2022) De novo variants in ATP2B1 lead to neurodevelopmental delay. Am J Hum Genet 109(5):944–952. 10.1016/j.ajhg.2022.03.009

20.

Stehr AM, Lenberg J, Friedman J, Dobbelaere D, Imbard A, Levy J, Donoghue S, Derive N, Stoeva R, Gueguen P, Zech M (2024) Consolidating the Role of Mutated ATP2B2 in Neurodevelopmental and Cerebellar Pathologies. Clin Genet

21.

Yap P, Riley LG, Kakadia PM, Bohlander SK, Curran B, Rahimi MJ, Alburaiky S, Hayes I, Oppermann H, Print C, Cooper ST, Le Quesne Stabej P (2024) Biallelic ATP2B1 variants as a likely cause of a novel neurodevelopmental malformation syndrome with primary hypoparathyroidism. Eur J Hum Genet 32(1):125–129. 10.1038/s41431-023-01484-9

22.

Poggio E, Barazzuol L, Salmaso A, Milani C, Deligiannopoulou A, Cazorla ÁG, Jang SS, Juliá-Palacios N, Keren B, Kopajtich R, Lynch SA, Mignot C, Moorwood C, Neuhofer C, Nigro V, Oostra A, Prokisch H, Saillour V, Schuermans N, Torella A, Verloo P, Yazbeck E, Zollino M, Jech R, Winkelmann J, Necpal J, Calì T, Brini M, Zech M (2023) ATP2B2 de novo variants as a cause of variable neurodevelopmental disorders that feature dystonia, ataxia, intellectual disability, behavioral symptoms, and seizures. Genet Med 25(12):100971. 10.1016/j.gim.2023.100971

23.

Vicario M, Zanni G, Vallese F, Santorelli F, Grinzato A, Cieri D, Berto P, Frizzarin M, Lopreiato R, Zonta F, Ferro S, Sandre M, Marin O, Ruzzene M, Bertini E, Zanotti G, Brini M, Calì T, Carafoli E (2018) A V1143F mutation in the neuronal-enriched isoform 2 of the PMCA pump is linked with ataxia. Neurobiol Dis. ;115:157–166. 10.1016/j.nbd.2018.04.009. Epub 2018 Apr 12. PMID: 29655659

24.

Bortolozzi M, Mammano F (2018) PMCA2 pump mutations and hereditary deafness. Neurosci Lett. ;663:18–24. 10.1016/j.neulet.2017.09.059. PMID: 29452611

25.

Smits JJ, Oostrik J, Beynon AJ, Kant SG, de Koning Gans PAM, Rotteveel LJC, Klein Wassink-Ruiter JS, Free RH, Maas SM, van de Kamp J, Merkus P, DOOFNL Consortium, Koole W, Feenstra I, Admiraal RJC, Lanting CP, Schraders M, Yntema HG, Pennings RJE, Kremer H (2019) De novo and inherited loss-of-function variants of ATP2B2 are associated with rapidly progressive hearing impairment. Hum Genet 138(1):61–72. 10.1007/s00439-018-1965-1Epub 2018 Dec 8. PMID: 30535804; PMCID: PMC6514080

26.

Vallese F, Maso L, Giamogante F, Poggio E, Barazzuol L, Salmaso A, Lopreiato R, Cendron L, Navazio L, Zanni G, Weber Y, Kovacevic-Preradovic T, Keren B, Torraco A, Carrozzo R, Peretto F, Peggion C, Ferro S, Marin O, Zanotti G, Calì T, Brini M, Carafoli E (2022) The ataxia-linked E1081Q mutation affects the sub-plasma membrane Ca2+-microdomains by tuning PMCA3 activity. Cell Death Dis 13(10):855. 10.1038/s41419-022-05300-yPMID: 36207321; PMCID: PMC9546857

27.

Calì T, Brini M, Carafoli E (2018) The PMCA pumps in genetically determined neuronal pathologies. Neurosci Lett 663:2–11. 10.1016/j.neulet.2017.11.005Epub 2017 Nov 17. PMID: 29155350

28.

Vicario M, Calì T, Cieri D, Vallese F, Bortolotto R, Lopreiato R, Zonta F, Nardella M, Micalizzi A, Lefeber DJ, Valente EM, Bertini E, Zanotti G, Zanni G, Brini M, Carafoli E (2017) A novel PMCA3 mutation in an ataxic patient with hypomorphic phosphomannomutase 2 (PMM2) heterozygote mutations: Biochemical characterization of the pump defect. Biochim Biophys Acta Mol Basis Dis 1863(12):3303–3312 Epub 2017 Aug 12. PMID: 28807751

29.

Ho PW, Pang SY, Li M, Tse ZH, Kung MH, Sham PC, Ho SL (2015) PMCA4 (ATP2B4) mutation in familial spastic paraplegia causes delay in intracellular calcium extrusion. Brain Behav 5(4):e00321. 10.1002/brb3.321Epub 2015 Feb 16. PMID: 25798335; PMCID: PMC4356846

30.

Li M, Ho PW, Pang SY, Tse ZH, Kung MH, Sham PC, Ho SL (2014) PMCA4 (ATP2B4) mutation in familial spastic paraplegia. PLoS ONE 9(8):e104790. 10.1371/journal.pone.0104790PMID: 25119969; PMCID: PMC4132067

31.

Gong D, Chi X, Ren K, Huang G, Zhou G, Yan N, Lei J, Zhou Q (2018) Structure of the human plasma membrane Ca2+-ATPase 1 in complex with its obligatory subunit neuroplastin. Nat Commun 9(1):3623. 10.1038/s41467-018-06075-7

32.

Malci A, Lin X, Sandoval R, Gundelfinger ED, Naumann M, Seidenbecher CI, Herrera-Molina R (2022) Ca2 + signaling in postsynaptic neurons: Neuroplastin-65 regulates the interplay between plasma membrane Ca2 + ATPases and ionotropic glutamate receptors. Cell Calcium 106:102623. 10.1016/j.ceca.2022.102623

33.

Vemula SK, Malci A, Junge L, Lehmann AC, Rama R, Hradsky J, Matute RA, Weber A, Prigge M, Naumann M, Kreutz MR, Seidenbecher CI, Gundelfinger ED, Herrera-Molina R (2020) The Interaction of TRAF6 With Neuroplastin Promotes Spinogenesis During Early Neuronal Development. Front Cell Dev Biol 8:579513. 10.3389/fcell.2020.579513

34.

Owczarek S, Soroka V, Kiryushko D, Larsen MH, Yuan Q, Sandi C, Berezin V, Bock E (2011) Neuroplastin-65 and a mimetic peptide derived from its homophilic binding site modulate neuritogenesis and neuronal plasticity. J Neurochem 117(6):984–994. 10.1111/j.1471-4159.2011.07269.x

35.

Liang Y, Ormazabal-Toledo R, Yao S, Shi YS, Herrera-Molina R, Montag D, Lin X (2024) Deafness causing neuroplastin missense variants fail to promote plasma membrane Ca2+-ATPase levels and Ca2 + transient regulation in brain neurons. J Biol Chem 300(7):107474. 10.1016/j.jbc.2024.107474

36.

Kaplanis J, Samocha KE, Wiel L, Zhang Z, Arvai KJ, Eberhardt RY, Gallone G, Lelieveld SH, Martin HC, McRae JF, Short PJ, Torene RI, de Boer E, Danecek P, Gardner EJ, Huang N, Lord J, Martincorena I, Pfundt R, Reijnders MRF, Yeung A, Yntema HG, Deciphering Developmental Disorders Study, Vissers LELM, Juusola J, Wright CF, Brunner HG, Firth HV, FitzPatrick DR, Barrett JC, Hurles ME, Gilissen C, Retterer K (2020) Evidence for 28 genetic disorders discovered by combining healthcare and research data. Nature 586(7831):757–762. 10.1038/s41586-020-2832-5Epub 2020 Oct 14. PMID: 33057194; PMCID: PMC7116826

37.

Bateman JR, Lee AM, Wu CT (2006) Site-specific transformation of Drosophila via phiC31 integrase-mediated cassette exchange. Genetics 173(2):769–777. 10.1534/genetics.106.056945

38.

Chou VT, Johnson SA, Van Vactor D (2020) Synapse development and maturation at the drosophila neuromuscular junction. Neural Dev 15(1):11. 10.1186/s13064-020-00147-5

39.

Schnorrer F, Schönbauer C, Langer CC, Dietzl G, Novatchkova M, Schernhuber K, Fellner M, Azaryan A, Radolf M, Stark A, Keleman K, Dickson BJ (2010) Systematic genetic analysis of muscle morphogenesis and function in Drosophila. Nature 464(7286):287–291. 10.1038/nature08799

40.

Doğanyiğit Z, Okan A, Yılmaz S, Uğuz AC, Akyüz E (2024) Gender-related variation expressions of neuroplastin TRAF6, GluA1, GABA(A) receptor, and PMCA in cortex, hippocampus, and brainstem in an experimental epilepsy model. Synapse. ;78(2):e22289. 10.1002/syn.22289. PMID: 38436644

41.

Lin X, Brunk MGK, Yuanxiang P, Curran AW, Zhang E, Stöber F, Goldschmidt J, Gundelfinger ED, Vollmer M, Happel MFK, Herrera-Molina R, Montag D (2021) Neuroplastin expression is essential for hearing and hair cell PMCA expression. Brain Struct Funct 226(5):1533–1551. 10.1007/s00429-021-02269-w

42.

Kazdoba TM, Leach PT, Crawley JN (2016) Behavioral phenotypes of genetic mouse models of autism. Genes Brain Behav 15(1):7–26. 10.1111/gbb.12256Epub 2015 Oct 22. PMID: 26403076; PMCID: PMC4775274

43.

Dhindsa RS, Weido BA, Dhindsa JS, Shetty AJ, Sands CF, Petrovski S, Vitsios D, Zoghbi AW (2025) Genome-wide prediction of dominant and recessive neurodevelopmental disorder-associated genes. Am J Hum Genet 112(3):693–708. 10.1016/j.ajhg.2025.02.001Epub 2025 Feb 26. PMID: 40015282; PMCID: PMC11947176

44.

Herrera-Molina R, Sarto-Jackson I, Montenegro-Venegas C, Heine M, Smalla KH, Seidenbecher CI, Beesley PW, Gundelfinger ED, Montag D (2014) Structure of excitatory synapses and GABAA receptor localization at inhibitory synapses are regulated by neuroplastin-65. J Biol Chem 289(13):8973–8988. 10.1074/jbc.M113.514992

45.

Jiang CH, Wei M, Zhang C, Shi YS (2021) The amino-terminal domain of GluA1 mediates LTP maintenance via interaction with neuroplastin-65. Proc Natl Acad Sci U S A 118(9):e2019194118. 10.1073/pnas.2019194118

46.

Sarto-Jackson I, Milenkovic I, Smalla KH, Gundelfinger ED, Kaehne T, Herrera-Molina R, Thomas S, Kiebler MA, Sieghart W (2012) The cell adhesion molecule neuroplastin-65 is a novel interaction partner of γ-aminobutyric acid type A receptors. J Biol Chem 287(17):14201–14214. 10.1074/jbc.M111.293175Epub 2012 Mar 2

47.

Li S, Wei X, Huang H, Ye L, Ma M, Sun L, Lu Y, Wu Y (2023) Neuroplastin exerts antiepileptic effects through binding to the α1 subunit of GABA type A receptors to inhibit the internalization of the receptors. J Transl Med 21(1):707. 10.1186/s12967-023-04596-4PMID: 37814294; PMCID: PMC10563248

48.

Korthals M, Langnaese K, Smalla KH, Kähne T, Herrera-Molina R, Handschuh J, Lehmann AC, Mamula D, Naumann M, Seidenbecher C, Zuschratter W, Tedford K, Gundelfinger ED, Montag D, Fischer KD, Thomas U (2017) A complex of Neuroplastin and Plasma Membrane Ca2 + ATPase controls T cell activation. Sci Rep 7(1):8358. 10.1038/s41598-017-08519-4PMID: 28827723; PMCID: PMC5566957

49.

Kitamura RA, Hummel D, Ustione A, Piston DW, Urano F (2024) Dual role of neuroplastin in pancreatic β cells: Regulating insulin secretion and promoting islet inflammation. Proc Natl Acad Sci U S A 121(33):e2411234121. 10.1073/pnas.2411234121Epub 2024 Aug 6. PMID: 39666939; PMCID: PMC11331099

50.

Sobreira N, Schiettecatte F, Valle D, Hamosh A (2015) GeneMatcher: a matching tool for connecting investigators with an interest in the same gene. Hum Mutat 36(10):928–930. 10.1002/humu.22844

51.

Chen S, Francioli LC, Goodrich JK, Collins RL, Kanai M, Wang Q, Alföldi J, Watts NA, Vittal C, Gauthier LD, Poterba T, Wilson MW, Tarasova Y, Phu W, Grant R, Yohannes MT, Koenig Z, Farjoun Y, Banks E, Donnelly S, Gabriel S, Gupta N, Ferriera S, Tolonen C, Novod S, Bergelson L, Roazen D, Ruano-Rubio V, Covarrubias M, Llanwarne C, Petrillo N, Wade G, Jeandet T, Munshi R, Tibbetts K, Donnell-Luria A, Solomonson M, Seed C, Martin AR, Talkowski ME, Rehm HL, Daly MJ, Tiao G, Neale BM, MacArthur DG, Karczewski KJ (2024) A genomic mutational constraint map using variation in 76,156 human genomes. Nature 625(7993):92–100 Genome Aggregation Database Consortium; O'. 10.1038/s41586-023-06045-0

52.

Morales J, Pujar S, Loveland JE, Astashyn A, Bennett R, Berry A, Cox E, Davidson C, Ermolaeva O, Farrell CM, Fatima R, Gil L, Goldfarb T, Gonzalez JM, Haddad D, Hardy M, Hunt T, Jackson J, Joardar VS, Kay M, Kodali VK, McGarvey KM, McMahon A, Mudge JM, Murphy DN, Murphy MR, Rajput B, Rangwala SH, Riddick LD, Thibaud-Nissen F, Threadgold G, Vatsan AR, Wallin C, Webb D, Flicek P, Birney E, Pruitt KD, Frankish A, Cunningham F, Murphy TD (2022) A joint NCBI and EMBL-EBI transcript set for clinical genomics and research. Nature 604(7905):310–315. 10.1038/s41586-022-04558-8

53.

Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, Grody WW, Hegde M, Lyon E, Spector E, Voelkerding K, Rehm HL, ACMG Laboratory Quality Assurance Committee (2015) Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med 17(5):405–424. 10.1038/gim.2015.30

54.

Rentzsch P, Schubach M, Shendure J, Kircher M (2021) CADD-Splice-improving genome-wide variant effect prediction using deep learning-derived splice scores. Genome Med 13(1):31. 10.1186/s13073-021-00835-9

55.

Ioannidis NM, Rothstein JH, Pejaver V, Middha S, McDonnell SK, Baheti S, Musolf A, Li Q, Holzinger E, Karyadi D, Cannon-Albright LA, Teerlink CC, Stanford JL, Isaacs WB, Xu J, Cooney KA, Lange EM, Schleutker J, Carpten JD, Powell IJ, Cussenot O, Cancel-Tassin G, Giles GG, MacInnis RJ, Maier C, Hsieh CL, Wiklund F, Catalona WJ, Foulkes WD, Mandal D, Eeles RA, Kote-Jarai Z, Bustamante CD, Schaid DJ, Hastie T, Ostrander EA, Bailey-Wilson JE, Radivojac P, Thibodeau SN, Whittemore AS, Sieh W (2016) REVEL: An Ensemble Method for Predicting the Pathogenicity of Rare Missense Variants. Am J Hum Genet 99(4):877–885. 10.1016/j.ajhg.2016.08.016

56.

Pejaver V, Urresti J, Lugo-Martinez J, Pagel KA, Lin GN, Nam HJ, Mort M, Cooper DN, Sebat J, Iakoucheva LM, Mooney SD, Radivojac P (2020) Inferring the molecular and phenotypic impact of amino acid variants with MutPred2. Nat Commun 11(1):5918. 10.1038/s41467-020-19669-x

57.

Carter H, Douville C, Stenson PD, Cooper DN, Karchin R (2013) Identifying Mendelian disease genes with the variant effect scoring tool. BMC Genomics 14(Suppl 3):S3. 10.1186/1471-2164-14-S3-S3Epub 2013 May 28

58.

Feng BJ (2017) PERCH: A Unified Framework for Disease Gene Prioritization. Hum Mutat 38(3):243–251. 10.1002/humu.23158

59.

Pejaver V, Byrne AB, Feng BJ, Pagel KA, Mooney SD, Karchin R, O'Donnell-Luria A, Harrison SM, Tavtigian SV, Greenblatt MS, Biesecker LG, Radivojac P, Brenner SE, ClinGen Sequence Variant Interpretation Working Group (2022) Calibration of computational tools for missense variant pathogenicity classification and ClinGen recommendations for PP3/BP4 criteria. Am J Hum Genet 109(12):2163–2177. 10.1016/j.ajhg.2022.10.013

60.

Abraham MJ, Murtola T, Schulz R, Páll S, Smith JC, Hess B, Lindahl E (2015) GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1–2:19–25

61.

Jorgensen WL, Tirado-Rives J (1988) The OPLS [optimized potentials for liquid simulations] potential functions for proteins, energy minimizations for crystals of cyclic peptides and crambin. J Am Chem Soc 110:1657–1666

62.

Berendsen HJC, Grigera JR, Straatsma TP (1987) The missing term ineffective pair potentials. J Phys Chem 91:6269–6271

63.

Ryckaert J-P, Ciccotti G, Berendsen HJC (1977) Numerical integration of the cartesian equations of motion of a system with constraints: molecular dynamics of n-alkanes. J Comput Phys 23:327–341

64.

Essmann U, Perera L, Berkowitz ML, Darden T, Lee H, Pedersen LG (1995) A smooth particle mesh Ewald method. J Chem Phys 103:8577–8593

65.

Bussi G, Donadio D, Parrinello M (2007) Canonical sampling through velocity rescaling. J Chem Phys 126:014101

66.

Parrinello M, Rahman A (1981) Polymorphic transitions in single crystals: A new molecular dynamics method. J Appl Phys 52:7182–7190

67.

Van Zundert GCP, Rodrigues JPGLM, Trellet M, Schmitz C, Kastritis PL, Karaca E, Melquiond ASJ, Van Dijk M, De Vries SJ, Bonvin AMJJ (2016) The HADDOCK2.2 Web Server: User-Friendly Integrative Modeling of BiomolecularbComplexes. J Mol Biol 428(4):720–725. https://doi.org/10.1016/j.jmb.2015.09.014

68.

Xue LC, Rodrigues JP, Kastritis PL, Bonvin AM, Vangone A (2016) PRODIGY: a web server for predicting the binding affinity of protein–protein complexes. Bioinformatics 32(23):3676–3678. https://doi.org/10.1093/bioinformatics/btw514

69.

Schmeisser MJ, Ey E, Wegener S, Bockmann J, Stempel AV, Kuebler A, Janssen AL, Udvardi PT, Shiban E, Spilker C, Balschun D, Skryabin BV, Dieck St, Smalla KH, Montag D, Leblond CS, Faure P, Torquet N, Le Sourd AM, Toro R, Grabrucker AM, Shoichet SA, Schmitz D, Kreutz MR, Bourgeron T, Gundelfinger ED, Boeckers TM (2012) Autistic-like behaviours and hyperactivity in mice lacking ProSAP1/Shank2. Nature 486(7402):256–260

Figures and Figure legends

Yes