FOXP4 Variants Are Associated With Plateau Iris and Angle Closure Glaucoma

Abstract

Purpose: Angle closure glaucoma (ACG) is a common cause of adult-onset vision loss that often presents with iris abnormalities and short axial lengths. Although it is heritable, little is known about the genetic risk factors underlying this condition. We thus conducted a disease gene discovery study in a family exhibiting an autosomal dominant triad of ACG, plateau iris, and short axial lengths.

Methods: Pooled exome sequencing was performed to identify coding variants contributing to disease. The spatiotemporal expression pattern of candidate gene FOXP4 was evaluated via immunostaining in embryonic mouse eyes. YFP-tagged mutant and wild-type FOXP4 proteins were expressed in HEK-293T and ARPE-19 cells to evaluate nuclear localization, and an SRPX2-Luciferase reporter was used to ascertain variant effects on transcriptional regulation. We also reviewed more than 20,000 patients (primarily from the UK Biobank) diagnosed with glaucoma and/or disorders of the iris and ciliary body for additional FOXP4 variants and functionally validated them as described.

Results: We identified a single likely pathogenic variant in transcription factor FOXP4: c.1433A>G (p.Q478R). FOXP4 is highly expressed in multiple structures relevant to the drainage angle, such as the periocular mesenchyme, iris, ciliary body, and cornea. The p.Q478R variant appears to be a hypomorphic allele that retains its transcriptional activity, but often mislocalizes to cytosolic aggregates. Comparable variants, including one found in another glaucoma patient, show similar mislocalization that may indicate protein instability.

Conclusions: These data suggest that FOXP4 is important for anterior segment development and that variants therein are rare risk factors for ACG.

Figure 1.

Pedigree showing the segregation of the LCP1 (p.K534T) and FOXP4 (p.Q478R) variants, along with disease status. II-3 and II-4 affected per family report.

Figure 2.

Structure of the FOXP4 protein, along with the tolerance landscape of the forkhead domain (amino acids 456–542)^, and the evolutionary conservation of the Arginine 478 residue. FOX, forkhead domain, LZ, leucine zipper; Q-rich, glutamine-rich; ZF, zinc finger.

Figure 3.

Spatiotemporal expression pattern of FOXP4 in developing mouse eyes from embryonic day 14.5 to 8 weeks (A). This is complemented by single-cell RNA sequencing from mature human (B) and mouse (C) eyes.^, C, cornea; CB, ciliary body; GCL, ganglion cell layer; I, iris; INL, inner nuclear layer; L, lens; ONL, outer nuclear layer; POM, periocular mesenchyme; Ret, retina. Scale bar, 25 µm.

Figure 4.

Representative images showing the localization of FOXP4 WT, Q478R, and H517N at an original magnification of ×40 in HEK-293T cells (A). In the same cell line, mislocalization of WT and variant proteins was quantified over six replicates (B) and luciferase assays conducted (C). Representative images showing the localization of FOXP4 WT, Q478R, and H517N at an original magnification of ×10 in ARPE-19 cells (D). Green, YFP-FOXP4; blue, DAPI; Normalized LUX, Firefly/Renilla luciferase expression ratio normalized to the empty vector control; EV, empty vector; ns, not significant;. **P < 0.01; ****P < 0.0001.

Figure 5.

Predicted loss-of-function (pLOF) and forkhead variants found in the Genebass cohort of patients with glaucoma or disorders of the iris/ciliary body plotted along the FOXP4 tolerance landscape. Localization of the missense substitutions is visualized at an original magnification of ×20 and dn/ds scores for the residues at which they occur are noted (A). PROST scores were used to predict the stability of the mislocalizing forkhead variants as compared to likely benign variants and other forkhead variants that showed WT localization (B). Red dots, cytosolic mislocalization/aggregation; Blue dots, granular nuclear localization; Green dots, pathogenic variants that do not mislocalize. ***P < 0.001; ns, not significant; FH, forkhead.