You are here

Article Text

Article menu
Download PDFPDF
Genotype-phenotype correlations
Original article
Different mutations in PDE4D associated with developmental disorders with mirror phenotypes
  1. Anna Lindstrand1,2,3,
  2. Giedre Grigelioniene1,2,3,
  3. Daniel Nilsson1,2,3,4,
  4. Maria Pettersson1,2,
  5. Wolfgang Hofmeister1,2,
  6. Britt-Marie Anderlid1,2,3,
  7. Sarina G. Kant5,
  8. Claudia A L Ruivenkamp5,
  9. Peter Gustavsson1,2,3,
  10. Helena Valta6,
  11. Stefan Geiberger7,
  12. Alexandra Topa8,
  13. Kristina Lagerstedt-Robinson1,2,3,
  14. Fulya Taylan1,2,4,
  15. Josephine Wincent1,2,3,
  16. Tobias Laurell1,2,9,
  17. Minna Pekkinen10,
  18. Magnus Nordenskjöld1,2,3,
  19. Outi Mäkitie1,2,3,6,10,
  20. Ann Nordgren1,2,3
  1. 1Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden
  2. 2Center for Molecular Medicine, Karolinska Institutet, Stockholm, Sweden
  3. 3Department of Clinical Genetics, Karolinska University Hospital, Stockholm, Sweden
  4. 4Science for Life Laboratory, Karolinska Institutet Science Park, Solna, Sweden
  5. 5Department of Clinical Genetics, Leiden University Medical Center, Leiden, The Netherlands
  6. 6Children's Hospital, Helsinki University Central Hospital, University of Helsinki, Helsinki, Finland
  7. 7Department of Pediatric Radiology, Karolinska University Hospital Solna, Stockholm, Sweden
  8. 8Department of Clinical Genetics, Sahlgrenska University Hospital, Gothenburg, Sweden
  9. 9Department of Clinical Science and Education, Södersjukhuset, Karolinska Institutet, Stockholm, Sweden
  10. 10Folkhälsan Institute of Genetics, Helsinki, Finland
  1. Correspondence to Dr Anna Lindstrand, and Ann Nordgren Department of Molecular Medicine and Surgery, Karolinska Institutet, Clinical Genetics Unit, Karolinska Institutet and Karolinska University Hospital Solna, Stockholm S-171 76, Sweden; Anna.Lindstrand{at}ki.se

Abstract

Background Point mutations in PDE4D have been recently linked to acrodysostosis, an autosomal dominant disorder with skeletal dysplasia, severe brachydactyly, midfacial hypoplasia and intellectual disability. The purpose of the present study was to investigate clinical and cellular implications of different types of mutations in the PDE4D gene.

Methods We studied five acrodysostosis patients and three patients with gene dose imbalances involving PDE4D clinically and by whole exome sequencing, Sanger sequencing and array comparative hybridisation. To evaluate the functional consequences of the PDE4D changes, we used overexpression of mutated human PDE4D message and morpholino-based suppression of pde4d in zebrafish.

Results We identified three novel and two previously described PDE4D point mutations in the acrodysostosis patients and two deletions and one duplication involving PDE4D in three patients suffering from an intellectual disability syndrome with low body mass index, long fingers, toes and arms, prominent nose and small chin. When comparing symptoms in patients with missense mutations and gene dose imbalances involving PDE4D, a mirror phenotype was observed. By comparing overexpression of human mutated transcripts with pde4d knockdown in zebrafish embryos, we could successfully assay the pathogenicity of the mutations.

Conclusions Our findings indicate that haploinsufficiency of PDE4D results in a novel intellectual disability syndrome, the 5q12.1-haploinsufficiency syndrome, with several opposing features compared with acrodysostosis that is caused by dominant negative mutations. In addition, our results expand the spectrum of PDE4D mutations underlying acrodysostosis and indicate that, in contrast to previous reports, patients with PDE4D mutations may have significant hormone resistance with consequent endocrine abnormalities.

  • Clinical Genetics
  • Copy-Number
  • Developmental
  • Other Neurology
  • Other Endocrinology

https://doi.org/10.1136/jmedgenet-2013-101937

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    cAMP-specific 3′,5′-cyclic phosphodiesterase 4D (PDE4D) is an enzyme that is encoded by the PDE4D gene in humans. PDE4D belongs to the phosphodiesterase subfamily and is one of the enzymes responsible for degrading cAMP in the cell. The protein shares large similarities with other PDE4 variants A, B and C. PDE4D (human locus 5q12.1) encodes at least nine functional transcript variants due to alternative splicing and multiple promoters. The longest transcript (NM_001104631) contains 15 exons that may be subdivided into three basic transcriptional units. All functional protein isoforms harbour a phosphodiesterase catalytic domain encoded by the 3′ terminus end of the gene and zero, one or two upstream conserved regions regulating the enzymatic activity.1 ,2 Recently, four separate publications reported point mutations in PDE4D in patients with acrodysostosis type 2 (Mendelian Inheritance in Man (MIM) 614613).3–6 This rare (<1:100 000) autosomal dominant syndrome is characterised by midfacial and nasal hypoplasia, skeletal dysplasia, small hands and feet, and variable intellectual disability (ID). Patients often have blue eyes and red or blond hair. Typical radiological manifestations include generalised shortening and broadening of metacarpals, metatarsals and phalanges, advanced skeletal maturation, first ray hyperplasia of the foot and stenosis of the lumbar spine.7 The other known genetic acrodysostosis locus is PRKAR1A responsible for acrodysostosis type 1 (MIM 101800). PRKAR1A is located upstream of PDE4D in the Gsα signalling pathway. Genotype–phenotype correlation studies have suggested that patients with PRKAR1A mutations are more likely to have short stature and hormone resistance while patients with PDE4D gene mutations have the characteristic facial features and ID, usually normal stature and lack hormone resistance. Here, we report eight additional unrelated patients with genetic variants affecting PDE4D: five individuals with acrodysostosis and three individuals with structural variants on chromosome 5q11.2-12.1. Our patients with 5q11.2-12.1 structural abnormalities do not have acrodysostosis but share a different distinct phenotype with ID, low body mass index (BMI) and characteristic facial features, including a prominent nose and a relatively small chin. Clinical comparison of the phenotype between these patients and a review of two previously published patients with structural variants affecting PDE4D raised the hypothesis that different types of genetic changes in PDE4D result in opposing phenotypes in humans.

    Materials and methods

    Subjects

    Five unrelated patients with a clinical diagnosis of acrodysostosis were analysed by whole exome (n=4) and/or direct sequencing (n=2) of the PDE4D gene. The samples were collected as a multicentre effort and patients originated from Sweden, Finland and Turkey. Two patients with a deletion of PDE4D were identified by array comparative genomic hybridisation (CGH) in a cohort of cases referred to the Clinical Genetic department’s clinical laboratory in Stockholm between 2008 and 2013 due to suspected genomic imbalances (n=3117). We then identified a third patient with a structural variant affecting PDE4D in the Database of Chromosome Imbalance and Phenotype in Humans using Ensembl Resources (DECIPHER).

    DNA was isolated from peripheral blood using standard methods. Informed consent was obtained from all participants, and the study was approved by the ethical committees at Karolinska University Hospital (ethical approval number: 2010/1930-32) and at Helsinki University Central Hospital (ethical approval number: 119/E7/2007). All patients were clinically evaluated, and their clinical data, biochemistry and radiological findings were reviewed.

    Array comparative genomic hybridisation

    For copy number variation detection, the following genome-wide array platforms were used: a 180K custom array from Oxford Gene Technology (Oxford, UK); a 244k catalogue array from Agilent Technologies (Santa Clara, USA); and 38K BAC array (Swegene, Lund, Sweden). All experiments were performed according to the manufacturer's recommendations with minor modifications. Scanning and computational procedures have been described previously.8–10

    Whole exome capture and resequencing

    DNA samples from four of the affected patients as well as from their healthy parents were submitted to the Science for Life Laboratory in Stockholm, which according to the manufacturer's instructions performed exom capture with Agilent SureSelect Human All Exon 50M (Agilent) followed by resequencing on Illumina HiSeq2000 (Illumina) at an average sequence depth of 87-111x. An in-house pipeline (http://github.com/dnil/etiologica) was used to process reads and determine candidate genes as has been described previously.11 Called variants were filtered for passable quality, a minor allele frequency in 1000 genomes less than 2%, absence from dbSNP132 and a set of 100 locally produced exomes using ANNOVAR.12 Only variants predicted to change an amino acid sequence were retained, and non-synonymous changes filtered to a PolyPhen2 score of above 0.85. The presence of suspected pathogenic variants was verified by Sanger sequencing.

    Sanger sequencing

    Sanger sequencing confirmed PDE4D point mutations detected by exome sequencing. In the remaining two acrodysostosis patients, all PDE4D exons were amplified by PCR and sequenced according to standard procedures (primers and PCR conditions are available on request).

    Expression of Zebrafish pde4d

    The temporal and spatial expression of pde4d was investigated by whole mount in situ hybridisation on 3 dpf embryos. A digoxin-labelled RNA antisense probe was generated targeting the coding region of pde4d. The in vitro transcription template was amplified by reverse transcriptase (RT)-PCR using the primers indicated in supplementary 2c (forward primer: TGC TTT AAC AAA TGT ACA ACA GGA A; reverse primer: CTT CTT AAC ACC GCT AAT CTG AGA C) on cDNA from 4 dpf zebrafish embryos. The T7 sequence (TAA TAC GAC TCA CTA TAG GG) was added to the reverse primer for the antisense PCR and to the forward primer to generate the sense template. The probe was then generated using the T7 RNA polymerase (Promega). Embryos were prepared, and the in situ reaction completed according to the protocol published by Thisse and Thisse.13

    Zebrafish embryo microinjection and manipulation

    Two morpholinos (MOs) targeting pde4d were designed by and obtained from Gene Tools, blocking translation (tb, CAG GCA TTT TCT TGC ATT GTT TGT C) or splicing (sb) by targeting the junction of exon 7/intron 7-8 (AAA CAC CTG TAC GCT CTT ACC TTG T). The longest wild type (WT) PDE4D open-reading frame (ORF) (NM_001104631.1) was obtained from GenScript (Piscataway, USA) in the pCS2+ plasmid. In addition, 16 mutated transcripts each harbouring one of the identified PDE4D acrodysostosis mutations identified to date (table 1) were ordered in the same vector. The ORFs were linearised with ApaI, and in vitro transcription was performed with the SP6 mMessage mMachine kit (Ambion).

    View this table:
    Table 1

    Detected and previously published PDE4D (NM_001104631.1) coding changes in patients with acrodysostosis

    About 1 nl of the indicated concentration was injected into WT TL (tupfel long fin) zebrafish embryos at the one-to-four cell stage (n=47–91 embryos/injection, repeated at least twice; with masked scoring). Embryo batches were kept at 28.5°C and scored live at 4 days postfertilisation (dpf). To test the sb-MO efficiency, we harvested whole embryos in Trizol (Invitrogen) and generated oligo-dT primed cDNA (Superscript III, Invitrogen) for a RT-PCR. For the overexpression experiments, 200 pg of mutated and 100 pg of WT PDE4D mRNA were injected. Live images of TL embryos were acquired on a Leica DFC 230 microscope using Leica application Suite V.4.1.0 (Leica Microsystems, Switzerland). Replicates were tested for homogeneity between samples using χ2 tests and then pooled. Subsequent comparisons between control and experimental groups were made using χ2 tests.

    Results

    PDE4D mutations in patients with acrodysostosis

    DNA from four patients with a clinical diagnosis of acrodysostosis and absence of PRKAR1A mutations was collected and screened for PDE4D (NM_001104631.1) mutations by either Sanger or exome sequencing. Out of the four acrodysostosis cases analysed with exome sequencing, three were found to have a heterozygous PDE4D mutation (c.676T>G, p.Phe226Val; c.907A>G, p.Met303Val; c.986T>C, p.Val329Ala). In the fourth case, the PDE4D locus was deemed normal by the massive parallel sequencing, but a heterozygous mutation was later detected by Sanger sequencing (c.677T>G, p.Phe226Cys). A fifth patient, with a typical acrodysostosis phenotype, was also investigated by Sanger sequencing and found to harbour a heterozygous point mutation in PDE4D (c.2033T>C, p.Ile678Thr). We then reviewed the literature for all previously reported patients with acrodysostosis and PDE4D mutations and found altogether 14 additional cases.3–6 Among these 19 unrelated acrodysostosis patients, 16 specific PDE4D point mutations have been identified (figure 1, table 1).

    Figure 1
    Figure 1

    The PDE4D gene and mutations. (A) A zoomed depiction of the PDE4D locus with a schematic illustration of exons (vertical black bars). The solid coloured bars represent regions coding for UCR1 (blue) UCR2 (green) and the catalytic domain (grey). The genetic position of all the 16 PDE4D point mutations associated with acrodysostosis is shown. (B) A schematic illustration of the three main functional units of the PDE4D protein upstream conserved region 1 (UCR1), UCR2 and the catalytic domain. Protein isoforms lacking one or both regulatory domains have a higher activity level.

    Structural variants affecting PDE4D

    Two simple microdeletions on chr5q11.2-12.1 encompassing the entire PDE4D gene were identified in the Karolinska University Hospital clinical array CGH cohort. They sized 10.2 Mb (patient 6; genomic position hg19 chr5:53169698-63350902) and 8.4 Mb (patient 7; genomic position hg19 chr5:52628315-61033057) and affected 48 and 44 transcripts, respectively. A third patient with a 784 kb intragenic duplication, including exon 2 to exon 6 of PDE4D, was identified through the DECIPHER database (patient 8; Decipher ID LEI250419; genomic position hg19 chr5:58330268-59114805, table 2, online supplementary figure 1). This type of copy number gain may be benign, and despite several attempts to clone the breakpoint region by PCR we did not succeed. Therefore, we were left to predict the functional effects of the duplication by indirect means. Segregation in the family revealed that the duplication occurred de novo, and in silico prediction showed that in both direct and inverted orientations the duplication, if occurrence in tandem, would render all but one PDE4D isoform out of frame. Taken together, these data suggest that the intragenic duplication results in a loss of gene function comparable to the patients with PDE4D deletions.

    View this table:
    Table 2

    Detected and previously published structural variants affecting the PDE4D gene

    Clinical features

    All five patients (4 males, 1 female) with acrodysostosis and PDE4D point mutations had a typical phenotype, including nasal, midface and maxillary hypoplasia, ID and skeletal abnormalities (table 3, figure 2). Spinal stenosis was detected in two out of three patients, examined with spine MRI. All patients with acrodysostosis were clinically disproportionate with short extremities and brachydactyly. Four had red hair, and all had blue eyes, regardless of the parents’ eye and hair colour. Several endocrine abnormalities and evidence for hormone resistance were observed. Four of the five patients with acrodysostosis had mild or transient parathyroid hormone (PTH) resistance; one of the four boys also had cryptorchidism, one had type 1 diabetes and lack of pubertal growth spurt and one had severe hypertension. Several additional features were observed and are summarised in table 3. Laboratory findings in patients with acrodysostosis are shown in online supplementary table S1.

    View this table:
    Table 3

    Clinical symptoms of patients with PDE4D missense mutations and 5q11.2-12.1 structural abnormalities

    Figure 2
    Figure 2

    Photographs and radiographs of the patients with acrodysostosis and with 5q11.2-12.1 structural rearrangements. Typical acrodysostosis features including red hair and blue eyes, nasal and maxillary hypoplasia (A, B, E, F, J, K), extremely short fingers and toes (C, D, G, H, L, M), and compact body composition with short arms (I). Patients with acrodysostosis: 1 (A, B, C, D), 2 (E, F, G, H, I) and 3 (J, K, L, M). Patients with 5q11.2-12.1 structural rearrangements were slim, showed down slanting palpebral fissures, bulbous nasal tip, prominent cheeks, posteriorly rotated ears, a relatively small chin and long arms, fingers and toes (N–W). Patients with 5q11.2-12.1 structural rearrangements: 6 (N, O, P, Q, R), 7 (S, T, U) and 8 (V, W).

    The three girls with PDE4D structural variants did not present with features of acrodysostosis but had a mild ID and low BMI. They also had long limbs, fingers and toes and characteristic faces, including a prominent nose, and a small chin. Low blood pressure was reported in both patients with deletions. We evaluated all the available clinical data and found several parameters that diverged between the acrodysostosis patients harbouring point mutations in PDE4D and the individuals with structural changes affecting the PDE4D locus. This mirror phenomena were apparent for the facial features, reflected in a small versus large nose, short versus long philtrum and a maxillary hyperplasia versus hypoplasia, as well as short versus long extremities and short versus long fingers. Finally, we noticed a tendency to high versus low BMI (2/3 Z score <−2 for haploinsufficiency patients; 2/4 Z score >+1.5 for acrodysostosis patients) and high versus low blood pressure (2/2 measured Z score <−0.5 for haploinsufficiency patients; 2/3 measured Z score >+1 for acrodysostosis patients). Clinical features of the patients are summarised in table 3. Phenotypic features are shown in figure 2.

    Pde4d depletion and overexpression of mutant variants in zebrafish cause developmental defects

    To further evaluate the PDE4D changes detected in our patients, we used overexpression of mutated human PDE4D messenger RNA (mRNA) and MO-based suppression of pde4d in zebrafish embryos. Using reciprocal BLAST, we identified a single zebrafish PDE4D orthologue, pde4d, with a high degree of similarity to the humans (92%), indicative of conservation of protein function. To characterise the expression pattern of zebrafish pde4d, we used whole mount RNA in situ hybridisation and found that zebrafish pde4d was expressed ubiquitously in the brain and head at 3 dpf (see online supplementary figure S2).

    To knockdown pde4d, two different MOs were designed to block either translation of the protein (tb) or to cause miss-splicing (sb) of the pre-mRNA. Aberrant splicing of the sb-MO was confirmed by RT-PCR at the highest dose tested (9 ng, online supplementary figure S2), confirming the efficacy of the sb-MO. Injection of either MO independently at the one-to-four cell stage resulted in similar dose-dependent phenotypes (see online supplementary figure S2). Knockdown embryos were subsequently classified into class I and class II morphants reflecting the degree of severity of the phenotypes (figure 3). Developmental defects observed at 4 dpf were variable and included shortened body length and heart oedema. To further confirm specificity of the observed phenotypes, knockdown embryos were coinjected with WT human PDE4D mRNA. Coinjection with WT human PDE4D mRNA significantly rescued the tb-MO (11% vs 32% abnormal, for 200 pg WT vs 12 ng of tb-MO alone; p=0.0123; n=50–61 embryos/injection; figure 3) and the sb-MO phenotype (20% vs 55% abnormal for 100 pg WT vs 9 ng of sb-MO alone; p=0.0007; n=42–59 embryos/injection; figure 3).

    Figure 3
    Figure 3

    Suppression of zebrafish pde4d and overexpression of mutated PDE4D mRNA produces specific defects. General dysmorphology of pde4d MO-injected and RNA-injected embryos. (A, B) Lateral (top) and dorsal (bottom) views of 96 hpf controls. (C, D, E, F, G, H, I and J) Time-matched embryos injected with 12 ng of pde4d tb-MO. (C, D) A class I embryo injected with pde4d tb-MO displaying bent body axis. (E, F) Pericardial oedema in a class I pde4d tb-MO-injected embryo, indicated by the arrow. (G, H) Short body with a curved tail, a large head and heart oedema (arrow) in a class II embryo injected with pde4d tb-MO (I, J) Generalised oedema (arrow) in a class II embryo injected with pde4d tb-MO. (K, L) normally looking embryo injected with 200 pg of human wild type (WT) PDE4D mRNA. (M, N, O, P, Q, R, S, T) Representative embryos injected with 200 pg of human mutated PDE4D mRNA. The specific mutation is indicated. (M, N) A class I embryo with an enlarged protruding jaw (arrow) and cyclopia. (O, P) A class I embryo with pericardial oedema and a sharply bent body. (Q, R) A class II embryo with generalised oedema (arrow), cyclopia and a curved body. (S, T) A class II embryo injected with generalised oedema and absence of the tail. (U) Quantification of live scoring of pde4d MO and human mRNA (co)injections. Embryos were injected with the indicated dose (9 ng sb-MO and/or 100 pg PDE4D mRNA; 12 ng tb-MO and/or 200 pg PDE4D mRNA) and scored live according to the criteria shown in panels A–L. n=42–61 embryos/injection, repeated at least twice; with masked scoring; NS, not significant; all comparisons were made using χ2 tests. (V) Quantification of live scoring of embryos injected with human mutated and WT PDE4D mRNA. Embryos were injected with 200 pg of mutated mRNA and scored live according to criteria shown in panels M–T, n=44–91 embryos/injection, repeated at least twice; with masked scoring; **p<0.005, ***p<0.0001, NS=not significant; all comparisons were made using χ2 tests.

    To further evaluate the pathogenicity of the identified PDE4D coding changes, we then proceeded with independent injections of the 16 identified PDE4D point mutations, including the five new variants detected in this study and all the mutations reported in previous publications (figure 1, table 1). Injection of WT alone did not result in any developmental defects. In contrast, injection of PDE4D mRNA harbouring the mutations identified in acrodysostosis patients resulted in consistent changes in 4 dpf embryos, confirming the pathogenicity of these mutations. We observed a shorter curved body with a fragile tail, microcephaly and heart oedema. Occasionally, we also observed cyclopia and an enlarged protruding jaw (figure 3). While these defects were reproducible, the percentage and severity of embryos showing a specific phenotype varied, with cyclopia and jaw defects being the rarest. The abnormal embryos were again subclassified into class I and class II (figure 3). For all 16 mutations, we observed aberrant embryos (between 9% and 41% abnormal; p<0.005 compared with uninjected control embryos for tested mutations; 44–96 embryos/injection; table 1; figure 3).

    Discussion

    Here, we report five new patients with acrodysostosis caused by heterozygous PDE4D missense mutations and three individuals with PDE4D structural variants predicted to cause PDE4D haploinsufficiency. We propose that this loss of PDE4D results in a previously unreported novel ID syndrome, with several mirror features compared with acrodysostosis. Our hypothesis is that the PDE4D point mutations resulting in acrodysostosis are dominant negative mutations in contrast to the haploinsufficiency caused by the structural variants of chromosome 5q11.2-12.1.

    The five acrodysostosis patients described here had a classical form of the disease, with nasal and maxillar hypoplasia, ID and skeletal abnormalities. Of the identified disease-causing mutations in the PDE4D gene, three are novel and two have been reported previously. To date, 16 mutations have been described in PDE4D in 19 unrelated patients with acrodysostosis.3–6 Coding changes are spread throughout the gene with a tendency to clustering in exons 3 and 6. The triplet encoding phenylalanine in amino acid position 226 seems to be a mutational hotspot as it has been mutated in three different ways, resulting in different amino acid changes (patient 1: c.676T>G, p.Phe226Val; patient 2: c.677T>G, p.Phe226Cys; Michot et al6: c.677T>C, p.Phe226Ser; table 1).

    In one of our acrodysostosis patients (patient 2), exome sequencing failed to detect the causal mutation. A clear heterozygous missense change was later identified by bidirectional Sanger sequencing. Subsequent manual inspection of the exome data identified one read from the mutant allele, but 160 from the WT. While theoretically possible, a statistical anomaly of this magnitude is hard to accept. However, detailed sequence inspection yielded no obvious explanation for the discrepancy. The coverage of the exon was good. Problems with specific sequences in massively parallel sequencing remain, as mutation detection is not always perfect.18

    Our results contribute to the spectrum of PDE4D mutations underlying acrodysostosis type 2. In contrast to previous reports, findings in our patients with PDE4D mutations indicate that even PDE4D mutations may lead to clinically significant endocrine abnormalities such as resistance to PTH and possibly gonadotropins and growth-hormone-releasing hormone. Some endocrine abnormalities were observed in all our patients with PDE4D mutations. Previous data suggest that patients with PRKAR1A mutations are more likely to have short stature and hormone resistance than patients with PDE4D gene mutations who have characteristic facial features and ID but usually normal stature and lack hormone resistance.4 Based on our observations, careful endocrine follow-up is necessary even for patients with PDE4D mutations. Regarding the reported patients with loss of function mutations, no endocrine abnormalities have been observed, except for delayed menarche in patient 6. Further studies are needed to determine whether these patients have a risk of developing endocrine abnormalities.

    Previous reports have shown that acrodysostosis patients have an increased risk of vascular stenosis and deep vein thrombosis.6 ,19 Metabolic and cardiovascular manifestations, including hypertension, have also been reported.20 In addition, genome-wide association studies have suggested PDE4D as a candidate gene in ischaemic stroke.21 One acrodysostosis patient in our study suffers from severe hypertension requiring medication, one has diabetes type 1 and another patient has been investigated for suspected deep vein thrombosis (table 3). Taken together, these data emphasise the importance of careful clinical follow-up, including weight and blood pressure monitoring in patients with acrodysostosis.

    Two patients with ID harboured a loss of the entire PDE4D coding region and one girl had a de novo intragenic duplication seemingly disrupting the transcripts. These patients suffer from a shared novel ID syndrome with several features mirroring acrodysostosis: long/short nose, arms and digits, and small/large chin. We also noted a tendency to opposing traits regarding underweight/overweight and low/high blood pressure. Additional studies are needed in order to understand the underlying mechanism of the metabolic and cardiovascular changes.

    Zebrafish have been increasingly used to successfully model human disorders.22 To further study the impact of PDE4D variants in early embryogenesis, we used MO knockdown (loss of function) and mRNA overexpression (dominant negative) in zebrafish embryos. By this approach, we could confirm that both loss of function and expression of mutant mRNA are detrimental to the developing embryo. In Drosophila, the pde4d (dunce) mutant has been associated with memory and learning defects (for a review, see Davis et al 199123); similarly previous knockout studies in mouse showed significant behavioural defects in associated learning.24 Additional defects such as progressive cardiomyopathies were also observed in knockout mouse studies,25 but there was no report of craniofacial defects. In contrast, we observed a tendency of the jaw to be underdeveloped in MO-treated embryos, although this is consistent with features observed in human patients with a haploinsufficiency in the PDE4D gene; it is also frequently associated with MO off-target effects26 and requires further validation. We confirmed that all the mutations tested are pathogenic in both human patients and zebrafish. While overexpression of WT mRNA was not detrimental to the embryo, various gross developmental defects were observed when an equivalent amount of mutant transcript was overexpressed. While the molecular basis for these defects still needs to be determined, this assay was able to validate the biological pathogenicity of these mutations. No paternal DNA sample was available for mutation carrier testing in patient 4, which precluded us from determining if the variant c.986T>C (p.Val329Ala) has occurred de novo. With support from the zebrafish assay, we conclude that the mutation is pathogenic.

    Observed phenotypic differences associated with different types of PDE4D mutations and different positions in the transcript may be explained by analysing the structure and function of the protein product. The PDE4D gene consists of 22 exons and its encoded protein, phosphodiesterase 4D, is involved in the degradation of cAMP, an important signal transducer in different cellular signalling pathways. Detailed reviews regarding PDE4D protein structure and function have been published by Housley and later by Conti and Beavo.1 ,2 Briefly all the PDE4 cAMP-specific phosphodiesterases (4A, 4B, 4C and 4D) are organised in a similar manner and are able to generate long, short and super-short isoforms. The proteins consist of two conserved upstream regions (UCR1 and UCR2) and an active catalytic unit encoded by the 3′ end of the gene (figure 1). UCR1 and UCR2 are able to interact with the catalytic unit and inhibit the cAMP reaction. Hence, the activity of the short and super-short isoforms lacking regulatory subunits is higher than that of the long form.1 ,2 The 16 PDE4D mutations detected to date are spread throughout the gene, affecting UCR1, UCR2 and the catalytic unit (figure 1). It is reasonable to predict that the common cellular consequence is higher cAMP hydrolysing activity. In contrast, patients with a PDE4D deletion or disruption will have a lower amount of protein and thus lower enzymatic activity (table 4).

    View this table:
    Table 4

    The type and location of the genetic mutation determines the phenotype of the patient

    This observation is supported by a previous publication by Jaillard et al27 that describes four cases with deletions encompassing this locus. Photographs are shown for two cases (patients 1 and 3). In support of our hypothesis, patient 3 in the report by Jaillard et al, lacking one copy of the entire PDE4D, has clinical features reminiscent of the new 5q12.1-haploinsufficiency syndrome described here. In contrast, the facial features of patient 1 in the same report are more similar to the ones observed in acrodysostosis with maxillar hypoplasia and a short nose. This observation reflects the composition of the PDE4D protein (figure 1). The deletion in patient 1 has its breakpoint within the PDE4D locus,27 and only the exons coding for the UCR1 inhibitory domain are affected (see online supplementary figure S1). The predicted cellular consequence is therefore an increased activity of the enzyme and lower cAMP levels as in acrodysostosis (table 4). A summary of the deletions reported by Jaillard et al and a comparison with our patients is shown in online supplementary figure S1.

    Mutations affecting the PRKAR1A gene also result in different disorders. Deactivating point mutations result in Carney complex (an autosomal dominant condition characterised by the presence of skin pigmentary lesions, endocrine abnormalities and cardiac myxomas) while milder loss-of-function mutations result in acrodysostosis.4 This is consistent with our hypothesis that acrodysostosis is caused by a reduction in cAMP signal as PRKAR1A acts downstream of cAMP.4

    The data presented here indicate that mutations affecting the activity of PDE4D result in two mirror phenotype syndromes; de novo dominant negative mutations leading to acrodysostosis type 2 and haploinsufficiency of the locus resulting in a different syndrome with opposing features, the 5q12.1-haploinsufficiency syndrome. After clinical comparisons and in vivo modelling in zebrafish embryos, we propose that the genetic type of PDE4D mutations explains the cellular mechanism underlying the phenotypic changes. Our work highlights the challenge of correctly interpreting the functional consequence of detected mutations to understand their role in disease pathogenesis.

    Acknowledgments

    We are grateful to the patients and their parents who participated in this study. We thank Aron Luthman and Nina Jäntti for technical assistance and Dr Lars Hagenäs for clinical advice.

    References

    1. ↵
      1. Houslay MD
      . PDE4 cAMP-specific phosphodiesterases. Prog Nucleic Acid Res Mol Biol 2001;69:249–315.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Conti M,
      2. Beavo J
      . Biochemistry and physiology of cyclic nucleotide phosphodiesterases: essential components in cyclic nucleotide signaling. Annu Rev Biochem 2007;76:481–511.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Lee H,
      2. Graham JM Jr.,
      3. Rimoin DL,
      4. Lachman RS,
      5. Krejci P,
      6. Tompson SW,
      7. Nelson SF,
      8. Krakow D,
      9. Cohn DH
      . Exome sequencing identifies PDE4D mutations in acrodysostosis. Am J Hum Genet 2012;90:746–51.
      OpenUrlCrossRefPubMed
    4. ↵
      1. Linglart A,
      2. Fryssira H,
      3. Hiort O,
      4. Holterhus PM,
      5. Perez de Nanclares G,
      6. Argente J,
      7. Heinrichs C,
      8. Kuechler A,
      9. Mantovani G,
      10. Leheup B,
      11. Wicart P,
      12. Chassot V,
      13. Schmidt D,
      14. Rubio-Cabezas O,
      15. Richter-Unruh A,
      16. Berrade S,
      17. Pereda A,
      18. Boros E,
      19. Munoz-Calvo MT,
      20. Castori M,
      21. Gunes Y,
      22. Bertrand G,
      23. Bougneres P,
      24. Clauser E,
      25. Silve C
      . PRKAR1A and PDE4D mutations cause acrodysostosis but two distinct syndromes with or without GPCR-signaling hormone resistance. J Clin Endocrinol Metab 2012;97:E2328–38.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Lynch DC,
      2. Dyment DA,
      3. Huang L,
      4. Nikkel SM,
      5. Lacombe D,
      6. Campeau PM,
      7. Lee B,
      8. Bacino CA,
      9. Michaud JL,
      10. Bernier FP,
      11. Parboosingh JS,
      12. Innes AM
      . Identification of novel mutations confirms Pde4d as a major gene causing acrodysostosis. Hum Mutat 2013;34:97–102.
      OpenUrlCrossRefPubMed
    6. ↵
      1. Michot C,
      2. Le Goff C,
      3. Goldenberg A,
      4. Abhyankar A,
      5. Klein C,
      6. Kinning E,
      7. Guerrot AM,
      8. Flahaut P,
      9. Duncombe A,
      10. Baujat G,
      11. Lyonnet S,
      12. Thalassinos C,
      13. Nitschke P,
      14. Casanova JL,
      15. Le Merrer M,
      16. Munnich A,
      17. Cormier-Daire V
      . Exome sequencing identifies PDE4D mutations as another cause of acrodysostosis. Am J Hum Genet 2012;90:740–5.
      OpenUrlCrossRefPubMed
    7. ↵
      1. Spranger JW,
      2. Brill P,
      3. Superti-Furga A,
      4. Unger S,
      5. Nishimura G
      . Bone dysplasias: an Atlas of genetic disorders of skeletal development. USA: Oxford University Press, 2012.
    8. ↵
      1. Winberg J,
      2. Lagerstedt Robinson K,
      3. Naess K,
      4. Lesko N,
      5. Wibom R,
      6. Lieden A,
      7. Anderlid BM,
      8. Graff C,
      9. Nordenskjold A,
      10. Nordgren A,
      11. Gustavsson P
      . Partial tetrasomy 14 associated with multiple malformations. Am J Med Genet A 2013;161:1284–90.
      OpenUrlCrossRefPubMed
      1. Bruno DL,
      2. Anderlid BM,
      3. Lindstrand A,
      4. van Ravenswaaij-Arts C,
      5. Ganesamoorthy D,
      6. Lundin J,
      7. Martin CL,
      8. Douglas J,
      9. Nowak C,
      10. Adam MP,
      11. Kooy RF,
      12. Van der Aa N,
      13. Reyniers E,
      14. Vandeweyer G,
      15. Stolte-Dijkstra I,
      16. Dijkhuizen T,
      17. Yeung A,
      18. Delatycki M,
      19. Borgstrom B,
      20. Thelin L,
      21. Cardoso C,
      22. van Bon B,
      23. Pfundt R,
      24. de Vries BB,
      25. Wallin A,
      26. Amor DJ,
      27. James PA,
      28. Slater HR,
      29. Schoumans J
      . Further molecular and clinical delineation of co-locating 17p13.3 microdeletions and microduplications that show distinctive phenotypes. J Med Genet 2010;47:299–311.
      OpenUrlAbstract/FREE Full Text
    10. ↵
      1. Lindstrand A,
      2. Malmgren H,
      3. Verri A,
      4. Benetti E,
      5. Eriksson M,
      6. Nordgren A,
      7. Anderlid BM,
      8. Golovleva I,
      9. Schoumans J,
      10. Blennow E
      . Molecular and clinical characterization of patients with overlapping 10p deletions. Am J Med Genet A 2010;152A:1233–43.
      OpenUrl
    11. ↵
      1. Kvarnung M,
      2. Nilsson D,
      3. Lindstrand A,
      4. Korenke GC,
      5. Chiang SC,
      6. Blennow E,
      7. Bergmann M,
      8. Stodberg T,
      9. Makitie O,
      10. Anderlid BM,
      11. Bryceson YT,
      12. Nordenskjold M,
      13. Nordgren A
      . A novel intellectual disability syndrome caused by GPI anchor deficiency due to homozygous mutations in PIGT. J Med Genet 2013;50:521–8.
    12. ↵
      1. Wang K,
      2. Li M,
      3. Hakonarson H
      . ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res 2010;38:e164.
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. Thisse C,
      2. Thisse B
      . High-resolution in situ hybridization to whole-mount zebrafish embryos. Nat Protoc 2008;3:59–69.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Cole TJ,
      2. Bellizzi MC,
      3. Flegal KM,
      4. Dietz WH
      . Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000;320:1240–3.
      OpenUrlAbstract/FREE Full Text
    15. ↵
      1. Karlberg J,
      2. Luo ZC,
      3. Albertsson-Wikland K
      . Body mass index reference values (mean and SD) for Swedish children. Acta Paediatr 2001;90:1427–34.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Wikland KA,
      2. Luo ZC,
      3. Niklasson A,
      4. Karlberg J
      . Swedish population-based longitudinal reference values from birth to 18 years of age for height, weight and head circumference. Acta Paediatr 2002;91:739–54.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      National High Blood Pressure Education Program Working Group on High Blood Pressure in C, Adolescents. The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents. Pediatrics 2004;114(2 Suppl 4th Report):555–76.
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Harismendy O,
      2. Ng PC,
      3. Strausberg RL,
      4. Wang X,
      5. Stockwell TB,
      6. Beeson KY,
      7. Schork NJ,
      8. Murray SS,
      9. Topol EJ,
      10. Levy S,
      11. Frazer KA
      . Evaluation of next generation sequencing platforms for population targeted sequencing studies. Genome Biol 2009;10:R32.
      OpenUrlCrossRefPubMed
    19. ↵
      1. Sezer N,
      2. Sutbeyaz ST,
      3. Koseoglu F,
      4. Aras M,
      5. Akin C
      . Adult case of acrodysostosis with severe neurologic involvement. J Back Musculoskelet Rehabil 2009;22:125–9.
      OpenUrlPubMed
    20. ↵
      1. Atabek ME,
      2. Pirgon O,
      3. Sert A
      . Metabolic syndrome manifestations in an adolescent with acrodysostosis. J Pediatr Endocrinol Metab 2007;20:739–41.
      OpenUrlPubMed
    21. ↵
      1. Lovkvist H,
      2. Olsson S,
      3. Hoglund P,
      4. Melander O,
      5. Jern C,
      6. Sjogren M,
      7. Engstrom G,
      8. Smith JG,
      9. Hedblad B,
      10. Andsberg G,
      11. Delavaran H,
      12. Jood K,
      13. Kristoffersson U,
      14. Luthman H,
      15. Norrving B,
      16. Lindgren A
      . A large-sample assessment of possible association between ischaemic stroke and rs12188950 in the PDE4D gene. Eur J Hum Genet 2012;20:783–9.
      OpenUrlCrossRefPubMed
    22. ↵
      1. Lieschke GJ,
      2. Currie PD
      . Animal models of human disease: zebrafish swim into view. Nat Rev Genet 2007;8:353–67.
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      1. Davis RL,
      2. Dauwalder B
      . The Drosophila dunce locus: learning and memory genes in the fly. Trends Genet 1991;7:224–9.
      OpenUrlPubMedWeb of Science
    24. ↵
      1. Rutten K,
      2. Misner DL,
      3. Works M,
      4. Blokland A,
      5. Novak TJ,
      6. Santarelli L,
      7. Wallace TL
      . Enhanced long-term potentiation and impaired learning in phosphodiesterase 4D-knockout (PDE4D) mice. Eur J Neurosci 2008;28:625–32.
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Lehnart SE,
      2. Wehrens XH,
      3. Reiken S,
      4. Warrier S,
      5. Belevych AE,
      6. Harvey RD,
      7. Richter W,
      8. Jin SL,
      9. Conti M,
      10. Marks AR
      . Phosphodiesterase 4D deficiency in the ryanodine-receptor complex promotes heart failure and arrhythmias. Cell 2005;123:25–35.
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      1. Robu ME,
      2. Larson JD,
      3. Nasevicius A,
      4. Beiraghi S,
      5. Brenner C,
      6. Farber SA,
      7. Ekker SC
      . p53 activation by knockdown technologies. PLoS Genet 2007;3:e78.
      OpenUrlCrossRefPubMed
    27. ↵
      1. Jaillard S,
      2. Andrieux J,
      3. Plessis G,
      4. Krepischi AC,
      5. Lucas J,
      6. David V,
      7. Le Brun M,
      8. Bertola DR,
      9. David A,
      10. Belaud-Rotureau MA,
      11. Mosser J,
      12. Lazaro L,
      13. Treguier C,
      14. Rosenberg C,
      15. Odent S,
      16. Dubourg C
      . 5q12.1 deletion: delineation of a phenotype including mental retardation and ocular defects. Am J Med Genet A 2011;155A:725–31.
      OpenUrlCrossRef

    Footnotes

    • AL and GG contributed equally.

    • Contributors All authors of this manuscript fulfil the criteria of authorship. There is no left who fulfils the criteria but has not been included as an author.

    • Funding This study was supported by grants from the Swedish Research Council, Kronprinsessan Lovisa, Karolinska Institutet, Frimurarna Barnhuset in Stockholm, Sällskapet Barnavård and Linnea and Joseph Carlsson foundation, Sweden; the Academy of Finland, Sigrid Juselius Foundation, Foundation for Pediatric Research, Folkhälsan Research Foundation, Finland, and Sabbatical Leave Program of the European Society for Pediatric Endocrinology through an educational grant from Eli Lilly International Corporation (OM). Computer resources were provided by SNIC through the Uppsala Multidisciplinary Center for Advanced Computational Science (UPPMAX) under project b2012085.

    • Competing interests None.

    • Patient consent Obtained.

    • Ethics approval Regional Ethics Committee, Stockholm, Research Ethics Committee, Helsinki University Central Hospital, Helsinki.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Web resources The URLs for data presented herein are as follows: Online Mendelian Inheritance in Man (OMIM), http://www.omim.org/; DECIPHER, http://decipher.sanger.ac.uk; 1000 Genomes Project, http://www.1000genomes.org/; dbSNP, http://www.ncbi.nlm.nih.gov/Genbank/; PolyPhen 2, http://genetics.bwh.harvard.edu/pph2/