PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of blackwellopenThis ArticleFor AuthorsLearn MoreSubmit
Human Mutation
 
Hum Mutat. 2017 August; 38(8): 942–946.
Published online 2017 June 6. doi:  10.1002/humu.23246
PMCID: PMC5518294

A recurrent de novo mutation in ACTG1 causes isolated ocular coloboma

Abstract

Ocular coloboma (OC) is a defect in optic fissure closure and is a common cause of severe congenital visual impairment. Bilateral OC is primarily genetically determined and shows marked locus heterogeneity. Whole‐exome sequencing (WES) was used to analyze 12 trios (child affected with OC and both unaffected parents). This identified de novo mutations in 10 different genes in eight probands. Three of these genes encoded proteins associated with actin cytoskeleton dynamics: ACTG1, TWF1, and LCP1. Proband‐only WES identified a second unrelated individual with isolated OC carrying the same ACTG1 allele, encoding p.(Pro70Leu). Both individuals have normal neurodevelopment with no extra‐ocular signs of Baraitser–Winter syndrome. We found this mutant protein to be incapable of incorporation into F‐actin. The LCP1 and TWF1 variants each resulted in only minor disturbance of actin interactions, and no further plausibly causative variants were identified in these genes on resequencing 380 unrelated individuals with OC.

Keywords: ACTG1, eye development, ocular coloboma, tissue fusion

Ocular coloboma (OC) is a closure defect affecting epithelial tissues in the embryonic optic fissure. OC accounts for ~11% of childhood blindness and has a birth prevalence of 0.0002–0.0005 (Morrison et al., 2002; Shah et al., 2012) and most commonly presents as the absence of iris and/or retina in the inferonasal quadrant of the eye (Nakamura, Diehl, & Mohney, 2011). OC often co‐occurs with ipsi‐ and/or contra‐lateral microphthalmia (small eye) or contralateral anophthalmia (absent eye), suggesting that these structural eye defects can represent a phenotypic continuum (Morrison et al., 2002). Many of the known OC loci encode transcription factors or signaling molecules that drive the growth of the optic cup (Chow & Lang, 2001; Zagozewski, Zhang, & Eisenstat, 2014), suggesting growth failure as a mechanism for OC occurrence, where fusion‐competent fissure margin cannot appose. Failure of epithelial fusion is another obvious mechanism, although our knowledge of the genes and pathways required for this process is limited (Brown et al., 2009). OC is proving to be genetically heterogeneous (Williamson & FitzPatrick, 2014) with no individual locus accounting for more than 3% of cases, and the majority of nonsyndromal OC‐affected individuals (>70%) have no identified genetic cause. Whole‐exome sequencing (WES) within the rare diseases component of the UK10K project (http://www.uk10k.org/) has been successful in identifying causative loci in families with OC in YAP1 (http://www.omim.org/, MIM# 606608) (Williamson et al., 2014), MAB21L2 (MIM# 604537) (Rainger et al., 2014), and FZD5 (MIM# 601723) (Liu et al., 2016).

Here, we report WES on 12 trios, each comprising one affected individual with isolated OC and both unaffected parents. In three of these trios, there was a wider family history of eye malformation that would be compatible with nonpenetrance in the intervening parent. DeNovoGear analysis (Ramu et al., 2013), together with a maximum minor allele frequency in ExAC of <10−4, was used to identify candidate de novo mutations (DNM) among all technically robust variant calls in the affected child. Following review and Sanger sequence validation, 10 heterozygous de novo, ultrarare, plausibly disruptive variants were confirmed in 10 different genes from eight of the trios surveyed (Fig. (Fig.1,1, and summarized in Supp. Table S1 and Supp. Fig. S1 and have been deposited in the DECIPHER database [http://decipher.sanger.ac.uk]). Each trio was also screened for plausibly pathogenic homozygous or compound heterozygous variants in known developmental disorder genes. In proband COL5103597, a causative homozygous loss‐of‐function mutation was identified in ADAMTS18, which has been reported separately (Chandra et al., 2014), and no DNM were identified in this trio. No other plausible autosomal‐recessive genotypes were identified in the other trios.

Figure 1
Whole‐exome de novo variant identification. Identification of 10 novel de novo variants in patients with isolated coloboma from family trios. Pedigree structures are shown with the gene variant detailed below each pedigree (chromatograms for each ...

The DNM identified in ACTG1 (NM_001199954.1; MIM# 102560) in COL5231458 (family 12; c.209C>T (p.(Pro70Leu), RNA not analyzed)) represented a strong candidate because DNM in the ubiquitous cytoplasmic actins (encoded by ACTB and ACTG1) cause Baraitser–Winter syndrome (BWS) with OC as a prominent feature (MIM# 243310 and 614583) (Di Donato et al., 2014; Rivière et al., 2012). Review of the WES data from the remaining UK10K OC probands identified another individual, COL5103624 (family 1,135) with the identical DNM (c.209 C>T (p.(Pro70Leu), RNA not analyzed)). Clinical review confirmed that both probands had bilateral irido‐chorioretinal OC, with no involvement of the optic nerve and no evidence of the neurological and neurodevelopmental impairments that typify BWS. Neither had hearing loss. Individual COL5103624 had mild bilateral ptosis. In family 1,135, the mother did not carry the mutation, and the father was unavailable for testing.

ATP binding to actin, and subsequent hydrolysis to ADP, is coupled to conformational changes in actin monomers that appear to be essential for filament dynamics (Otterbein, Graceffa, & Dominguez, 2001) and thus actin turnover during dynamic cytoskeletal remodeling. Actin monomers have four subdomains, structurally organized so that each presents a surface interface, and arranged such that they converge around the centrally bound nucleotide. The Pro70 residue in ACTG1 is positioned N‐terminal to the previously reported BWS mutations (Fig. (Fig.2A),2A), and is immediately adjacent to the nucleotide‐binding domain of the actin monomer, in a loop (Pro70‐Thr77) that connects the COOH‐terminal end of subdomain 2 to subdomain 1. This loop undergoes conformational reorganization upon ATP hydrolysis, where a 10o rotation of subdomain 2 occurs, which in turn induces changes to interdomain interactions that affect the orientation of subdomain 4. In combination, these are predicted to have major effects on the overall structure and stability of each actin‐ADP polymer (Otterbein et al., 2001). A severe destabilizing effect of the mutant Leu70 amino acid change to the protein was predicted using FoldX (Schymkowitz et al., 2005), with a difference in free energy of folding (ΔΔG) between mutant and wild type (WT), of 6.31 kcal/mol in the ADP‐bound actin crystal structure, and ΔΔG 7.4 kcal/mol in the ATP‐bound actin crystal structure (Fig. (Fig.2B).2B). Mouse embryonic fibroblasts (MEFs) were derived from F0 embryos following CRISPR/Cas9 gene editing to create homozygous p.Pro70Leu lines (Supp. Fig. S2). Immunofluorescence showed a marked reduction of mutant ACTG1 in filamentous F‐actin (Fig. 2C) compared with MEFs from WT littermate control embryos. We then created tetracyclin (TET)‐inducible HEK293 cell lines expressing either mutant (Leu70) or WT ACTG1 tagged with N‐terminal eGFP. Again, we observed reduced incorporation of Leu70 ACTG1 into F‐actin (Supp. Fig. S2b), whereas cosedimentation assays of these cells also showed a ~50% reduction of Leu70 ACTG1 in the G‐actin phase compared with WT (Fig. (Fig.2D).2D). In all cell types examined, endogenous ACTB and ACTG1 appeared to be unaffected (Fig. (Fig.2C2C and D; Supp. Fig. S2c). Western blotting and semiquantitative mass spectrometry on ACTG1–GFP immunoprecipitates from these HEK293 lines showed a marked reduction in the recovery of established actin‐binding partners for Leu70 ACTG1 (Fig. (Fig.2E;2E; Supp. Table S2).

Figure 2
Functional effects of de novo variants on the ACTG1 interactome. A: A scaled cartoon illustrating the positions of the de novo variants in each protein. Domains for each protein are indicated: EF‐Hand; CH, calponin homology; ADF‐H, actin ...

Individual COL5061742 (family 7) had a DNM in TWF1 (NM_001242397.1; MIM# 610932; c.1045 C>T; (p.(Pro349Ser), RNA not analyzed) that encodes Twinfillin‐1. However, the identification of a plausible DNM in this family is difficult to reconcile with the presence of an affected third‐degree maternal relative (Fig. (Fig.1),1), unless the microphthalmia in this individual is coincidental. The eye phenotype of this relative is detailed as one smaller eye and low vision, but unfortunately more detailed clinical information or DNA were not available. Twinfilins are highly conserved ubiquitous actin‐binding proteins that influence actin polymerization by forming 1:1 complexes with ADP–actin monomers to moderate F‐actin filament assembly (Palmgren, Vartiainen, & Lappalainen, 2002). Twf1 also influences the depolymerization and severing of actin filaments (Johnston, Collins, & Goode, 2015; Moseley et al., 2006). The Twinfilin‐1 p.(Pro349Ser) substitution is located at the C‐terminal “tail” region, out with the canonical actin interaction ADF‐H domains (Fig. (Fig.2A).2A). This region was implicated in determining binding affinity to F‐actin, and complete deletion of the C‐terminal tail significantly reduces barbed‐end capping activity (Paavilainen et al., 2007). In addition, the yeast Twinfilin tail region alone can bind F‐actin, and contributes to the binding of the full‐length protein, whereas loss of the tail region significantly affects F‐actin depolymerization in vitro (Johnston et al., 2015). TET‐inducible HEK293 cells expressing the WT and mutant Twinfilin‐1 failed to identify any significant protein–protein interaction differences using immunoprecipitation with mass spectrometry (Supp. Table S2), and cosedimentation assays showed no differences in F‐/G‐actin phase distributions. However, transient transfection of FLAG‐tagged, WT ACTG1 into these HEK293 cells showed slightly increased binding of variant Twinfilin‐1 to ACTG1, compared with WT TWF1 (Supp. Fig. S2D).

Individual COL5001071 (family 1) was found to have a DNM c.1823 A>G; (p.(Asn608Ser), RNA not analyzed) in LCP1 (NM_002298.4; MIM# 153430) that encodes LCP1 (or L‐plastin/plastin‐2), a conserved F‐actin filament cross‐linking protein of the plastin family found in structures involved in cell adhesion, motility, and invasion, such as focal adhesions, membrane ruffles, and cell projections (Janji et al., 2006). LCP1 contains two actin‐binding domains, and the p.(Asn608Ser) variant is located within the fourth of four calponin homology domains, in the second actin‐binding domain (Fig. (Fig.2A).2A). The variant was not predicted to affect LCP1 protein stability by in silico analysis but structural modeling predicts this residue to be on the surface of the protein (Supp. Fig. S2e). The variant was assessed using TET‐inducible HEK293 cells expressing WT and mutant versions of GFP‐tagged LCP1. We did not observe any reduction of protein stability (not shown), and both versions strongly associated with actin, but no clear differences in binding were observed by mass spectrometry. Cosedimentation analysis revealed these cells had increased signal in the F‐actin phase compared with WT (Supp. Fig. S2e). This may indicate abnormal LCP1–actin interactions or increased F‐actin stability in an LCP1‐Ser608 containing cellular environment.

Targeted resequencing of 380 affected individuals with OC revealed no further plausible causative variants in any of the DNM genes identified here. All alleles, with the exception of the variants in PPP1R15B (MIM# 613257) and STAB2 (MIM# 608561), had ExAC allele frequencies of zero (Supp. Table S1). Using established online tools (SIFT [http://sift.bii.a-star.edu.sg/index.html] and Polyphen‐2 [http://genetics.bwh.harvard.edu/pph2/index.shtml]) to assess the biochemical impact of the encoded amino acid substitutions, only the ABI3BP (MIM# 606279) DNM c.2009A>G; p.(Asp670Gly) was predicted to encode a possibly damaging substitution (Supp. Table S1). ABI3BP has two fibronectin type II domains, but a role for this protein in development has not been firmly established. A recent Abi3bp knock‐out mouse study revealed no overt phenotype (Yang et al., 2016); however, the ABI3BP locus has been associated with optic disk morphology in a meta‐analysis GWAS study (Springelkamp et al., 2015). None of the other genes have been implicated in developmental eye disorders, and no ocular phenotypes have been reported in mouse knock‐out models for their orthologues.

This work strongly supports a causative role for a recurrent de novo nonsynonymous variant affecting ACTG1 in human coloboma, although its precise role in eye development requires further investigation. It was remarkable that two other actin‐interacting proteins were implicated from the DNM screen; however, the clinical and biochemical evidence for causation is significantly less compelling, and these variants cannot currently be considered pathogenic. Trio‐based genome‐wide sequence analysis shows promise in identifying novel genetic causes and genetic mechanisms for OC, but a significantly larger number of families have to be analyzed to determine the true number of disease‐associated loci for this important eye malformation.

All datasets supporting the conclusions of this article are included within the article and the Supp. files: Table S1, Table S2, Figure S1, Figure S2, Figure S2 Legend, and Supp. Materials and Methods. The UK10K exome data are available via European Genome‐phenome Archive (https://www.ebi.ac.uk/ega/home) under the study accession number EGAS00001000127. Mass spectrometry data are available in EBI Pride repository as a ProteomeXchange submission http://www.ebi.ac.uk/pride/; (ID# PXD005090).

Supporting information

Supporting Information

Table S1: Summary of validated de novo variants identified in UK10K Trio analyses

Table S2: Summary of quantitative heavy vs light labelled mass spectrometry data from GFP‐TRAP Co‐immunoprecipitates

ACKNOWLEDGMENTS

We are grateful to staff from the Core Technical Services at The MRC IGMM for Sanger sequencing, and to staff at The IGMM Evans Building Transgenic Facility for CRISPR/Cas9 injections and animal husbandry.

Notes

Rainger J, Williamson KA, Soares DC, et al. A recurrent de novo mutation in ACTG1 causes isolated ocular coloboma. Human Mutation. 2017;38:942–946. https://doi.org/10.1002/humu.23246

Notes

Contact grant sponsors: Fight for Sight (UK) (Early Career Investigator Fellowship, grant number 1590/1591); MRC University Unit Grant to the University of Edinburgh; Wellcome Trust (WT091310).

Communicated by Daniel F. Schorderet

Contributor Information

Joe Rainger, ku.ca.de.nilsor@regniar.eoj.

David R FitzPatrick, ku.ca.de@kcirtapztif.divad.

REFERENCES

  • Brown J., Dutta S., Bharti K., Bonner R. F., Munson P. J., Dawid I. B., … Brooks B. P. (2009). Expression profiling during ocular development identifies 2 Nlz genes with a critical role in optic fissure closure. Proceedings of the National Academy of Sciences of the United States of America, 106, 1462–1467. [PubMed]
  • Chandra A., Arno G., Williamson K., Sergouniotis P. I., Preising M. N., Charteris D. G., ... Moore A. T. (2014) Expansion of ocular phenotypic features associated with mutations in ADAMTS18. JAMA Ophthalmology, 132(8), 996–1001. [PubMed]
  • Chow R. L., & Lang R. A. (2001). Early eye development in vertebrates. Annual Review of Cell and Developmental Biology, 17, 255–296. [PubMed]
  • Di Donato N., Rump A., Koenig R., Der Kaloustian V. M., Halal F., Sonntag K., ... Verloes A. (2014). Severe forms of Baraitser‐Winter syndrome are caused by ACTB mutations rather than ACTG1 mutations. European Journal of Human Genetics, 22, 179–183. [PubMed]
  • Janji B., Giganti A., De Corte V., Catillon M., Bruyneel E., Lentz D., ... Friederich E. (2006). Phosphorylation on Ser5 increases the F‐actin‐binding activity of L‐plastin and promotes its targeting to sites of actin assembly in cells. Journal of Cell Science, 119, 1947–1960. [PubMed]
  • Johnston A. B., Collins A., & Goode B. L. (2015). High‐speed depolymerization at actin filament ends jointly catalysed by Twinfilin and Srv2/CAP. Nature Cell Biology, 17, 1504–1511. [PubMed]
  • Liu C., Widen S. A., Williamson K. A., Ratnapriya R., Gerth‐Kahlert C., Rainger J., … Swaroop A. (2016). A secreted WNT‐ligand‐binding domain of FZD5 generated by a frameshift mutation causes autosomal dominant coloboma. Human Molecular Genetics, 25, 1382–1391. [PubMed]
  • Morrison D., FitzPatrick D., Hanson I., Williamson K., van Heyningen V., Fleck B., ... Campbell H. (2002). National study of microphthalmia, anophthalmia, and coloboma (MAC) in Scotland: Investigation of genetic aetiology. Journal of Medical Genetics, 39, 16–22. [PubMed]
  • Moseley J. B., Okada K., Balcer H. I., Kovar D. R., Pollard T. D., & Goode B. L. (2006). Twinfilin is an actin‐filament‐severing protein and promotes rapid turnover of actin structures in vivo. Journal of Cell Science, 119, 1547–1557. [PubMed]
  • Nakamura K. M., Diehl N. N., & Mohney B. G. (2011). Incidence, ocular findings, and systemic associations of ocular coloboma: A population‐based study. Archives of Ophthalmology, 129, 69–74. [PubMed]
  • Otterbein L. R., Graceffa P., & Dominguez R. (2001). The crystal structure of uncomplexed actin in the ADP state. Science, 293, 708–711. [PubMed]
  • Paavilainen V. O., Hellman M., Helfer E., Bovellan M., Annila A., Carlier M.‐F. F., … Lappalainen P. 2007. Structural basis and evolutionary origin of actin filament capping by twinfilin. Proceedings of the National Academy of Sciences of the United States of America 104, 3113–3118. [PubMed]
  • Palmgren S., Vartiainen M., & Lappalainen P. (2002). Twinfilin, a molecular mailman for actin monomers. Journal of Cell Science, 115, 881–886. [PubMed]
  • Rainger J., Pehlivan D., Johansson S., Bengani H., Sanchez‐Pulido L., Williamson K. A., … FitzPatrick D. R. (2014). Monoallelic and biallelic mutations in MAB21L2 cause a spectrum of major eye malformations. The American Journal of Human Genetics, 94, 915–923. [PubMed]
  • Ramu A., Noordam M. J., Schwartz R. S., Wuster A., Hurles M. E., Cartwright R., & Conrad D. F. (2013). DeNovoGear: De novo indel and point mutation discovery and phasing. Nature Methods, 10, 985–987. [PubMed]
  • Rivière J‐B., van Bon B. W. M., Hoischen A., Kholmanskikh S. S., O'Roak B. J., Gilissen C., … Dobyns W. B. (2012). De novo mutations in the actin genes ACTB and ACTG1 cause Baraitser‐Winter syndrome. Nature Genetics, 44, 440–444, S1–S2. [PubMed]
  • Schymkowitz J., Borg J., Stricher F., Nys R., Rousseau F., & Serrano L. (2005). The FoldX web server: An online force field. Nucleic Acids Research, 33(Web server issue), W382–W388. [PubMed]
  • Shah S. P., Taylor A. E., Sowden J. C., Ragge N., Russell‐Eggitt I., Rahi J. S., & Gilbert C. E. (2012). Anophthalmos, microphthalmos, and coloboma in the United Kingdom: Clinical features, results of investigations, and early management. Ophthalmology, 119, 362–368. [PubMed]
  • Springelkamp H., Mishra A., Hysi P. G., Gharahkhani P., Höhn R., Khor C. C., … Hammond C. J. (2015). Meta‐analysis of genome‐wide association studies identifies novel loci associated with optic disc morphology. Genetic Epidemiology, 39, 207–216. [PubMed]
  • Williamson K. A., & FitzPatrick D. R. (2014). The genetic architecture of microphthalmia, anophthalmia and coloboma. European Journal of Medical Genetics, 57, 369–380. [PubMed]
  • Williamson K. A., Rainger J., Floyd J. A. B., Ansari M., Meynert A., Aldridge K. V., … Fitzpatrick D. R. (2014). Heterozygous loss‐of‐function mutations in YAP1 cause both isolated and syndromic optic fissure closure defects. The American Journal of Human Genetics, 94, 295–302. [PubMed]
  • Yang H., Chen W., Shi J., Huang D., Li J., Hu B., ... Fei J. (2016). Knockout of Abi3bp in mice does not affect their olfactory function, mental state and NNK‐induced lung tumorigenesis. Acta Biochimica et Biophysica Sinica (Shanghai), 48, 820–826. [PubMed]
  • Zagozewski J. L., Zhang Q., & Eisenstat D. D. (2014). Genetic regulation of vertebrate eye development. Clinical Genetics, 86, 453–460. [PubMed]

Articles from Wiley-Blackwell Online Open are provided here courtesy of Wiley-Blackwell, John Wiley & Sons