Nanophthalmos-Associated MYRF Gene Mutation Causes Ciliary Zonule Defects in Mice

Abstract

Purpose: Patients with nanophthalmos who undergo intraocular surgery often present with abnormal ciliary zonules. In a previous study, we reported mutation in MYRF that is implicated in the pathogenesis of nanophthalmos. The aim of this study was to model the mutation in mice to explore the role of MYRF on zonule structure and its major molecular composition, including FBN1 and FBN2.

Methods: Human MYRF nanophthalmos frameshift mutation was generated in mouse using the CRISPR-Cas9 system. PCR and Sanger sequencing were used for genotype analysis of the mice model. Anterior chamber depth (ACD) was measured using hematoxylin and eosin–stained histology samples. Morphologic analysis of ciliary zonules was carried out using silver staining and immunofluorescence. Transcript and protein expression levels of MYRF, FBN1, and FBN2 in ciliary bodies were quantified using quantitative real-time PCR (qRT-PCR) and Western blot.

Results: A nanophthalmos frameshift mutation (c.789delC, p.N264fs) of MYRF in mice showed ocular phenotypes similar to those reported in patients with nanophthalmos. ACD was reduced in MYRF mutant mice (MYRF^mut/+) compared with that in littermate control mice (MYRF^+/+). In addition, the morphology of ciliary zonules showed reduced zonular fiber density and detectable structural dehiscence of zonular fibers. Furthermore, qRT-PCR analysis and Western blot showed a significant decrease in mRNA expression levels of MYRF, FBN1, and FBN2 in MYRF^mut/+ mice.

Conclusions: Changes in the structure and major molecular composition of ciliary zonules accompanied with shallowing anterior chamber were detected in MYRF^mut/+ mice. Therefore, MYRF mutant mice strain is a useful model for exploring pathogenesis of zonulopathy, which is almost elusive for basic researches due to lack of appropriate animal models.

Several eye disorders present with a reduced eye size, including microphthalmia, anterior microphthalmos, posterior microphthalmos, and nanophthalmos. Nanophthalmos is mainly characterized by a short axial length (AL; less than 20 mm) accompanied with small corneal diameter, shallow anterior chamber, anomalies of the vein plexus, and thick choroid and sclera, but it appears morphologically normal.¹ These anatomic characteristics present higher risk for secondary pathologies, including angle-closure glaucoma, ciliary zonule defects, and retinal folds.^2,3 Moreover, risk of complications, such as aqueous misdirection, retinal detachment, vitreous hemorrhage, and choroidal effusion, increases significantly during intraocular surgery or postoperatively in patients with nanophthalmos.^2,4 Therefore, pathologic mechanisms of nanophthalmos should be explored to provide scientific understanding of this complex condition and form a basis for improving current surgical strategies, thus achieving better prognosis for the patients. 

From a molecular perspective, the mechanisms of nanophthalmos corresponding to its anatomic features include abnormal regulation in AL; anterior chamber angle-–related structure configurations, including lenses and ciliary zonules; and anomaly in retina-choroid-sclera complex. Previous genetic studies report that most genes associated with nanophthalmos play a role in the development of these ocular structures. Transmembrane protein 98 (TMEM98, OMIM 615949),^3,5 crumbs homolog 1 (CRB1, OMIM 604210),⁶ and bestrophin 1 (BEST1, OMIM 607854)⁷ are implicated in the retina-choroid-sclera complex. Further, membrane-type frizzled-related protein (MFRP, OMIM 606227)⁸ and serine protease 56 (PRSS56, OMIM 613858)⁹ are implicated in the regulation of AL. However, currently no studies have explored the relationship of nanophthalmos-associated genes with anterior segment anatomy, specifically ciliary zonules, which is related to zonulopathy and narrowing of anterior chamber angle. Understanding the molecular mechanisms of zonulopathy and narrowing of anterior chamber angle in patients with nanophthalmos will provide information on primary angle-closure glaucoma (PACG) associated clinical phenotypes. In addition, zonulopathy is common in patients with nanophthalmos and is a significant risk factor for intraocular surgery complications. Therefore, the aim of this study was to explore the structure and main molecular components of zonules in nanophthalmos gene mutations. 

In a previous study, we have reported that the MYRF gene is implicated in the pathogenesis of nanophthalmos using a de novo mutation (DNM) strategy by trio-based whole-genome sequencing (WGS). Two DNMs of MYRF (c.789delC, p.S264fs and c.789dupC, p.S264fs) were detected in the probands of 11 trios.¹⁰ Myelin-gene regulatory factor (MYRF) is a key transcriptional regulator for oligodendrocyte maturation and central nervous system myelination.¹¹ Furthermore, haploinsufficiency of MYRF results in a genetic syndrome characterized by congenital heart defects, urogenital anomalies, congenital diaphragmatic hernia, and pulmonary hypoplasia.¹² In addition, Garnai et al.¹³ report that MYRF is important for development of the retinal pigment epithelium (RPE) and retina by the generation of MYRF conditional knockout mice. Moreover, truncation mutations in the C-terminal region of MYRF has been reported in high hyperopia families, and knockdown of MYRF results in small eye size in zebrafish.^13,14 However, the exact role of MYRF in eyes should be explored further. Therefore, in this study, we modeled human nanophthalmos frameshift mutations in mice to explore the role of MYRF in ciliary zonule structure and its major molecular composition, including fibrillin 1 (FBN1) and fibrillin 2 (FBN2), through immunofluorescent staining, quantitative real-time PCR (qRT-PCR) analysis and Western blot. Significantly, silver staining was used to evaluate the fine structure of zonules for the first time. 

Methods

Mice

C57BL/6J mice were purchased from Beijing Biocytogen Co., Ltd. (Beijing, China). All animal procedures were conducted according to protocols approved by the Institutional Animal Care and Use Committee in the Zhongshan Ophthalmic Center and strictly complied with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. All mice were housed in an assessment pathogen-free animal facility on a 12-hour light/dark cycle during the experiments. Animals were maintained under controlled temperature (18–29°C) and humidity (40%–70%) with food and water available ad libitum. 

Screening of Single-Guide RNAs

The primary RNA transcript of MYRF in ensemble is ENSMUST00000189897.1. For MYRF targeting, six single-guide RNAs (sgRNAs) were designed to target a region upstream or downstream of exon 6 (Supplementary Table S1). Accurate target sequence of CRISPR sgRNA was designed using the Optimized CRISPR Design tool (http://crispr.mit.edu). CRISPR sgRNAs were screened for on-target activity using the UCA kit (Beijing Biocytogen Co., Ltd.), and finally sgRNA 6 was chosen as it was the most active sgRNA. 

Production of Cas9 mRNA and sgRNAs

Cas9 mRNA and sgRNAs were transcribed with T7 RNA polymerase in vitro. T7 promoter sequence was added to the Cas9 and sgRNA templates by PCR amplification for Cas9 mRNA and sgRNA production. T7-Cas9 and T7-sgRNA PCR products were gel purified and used as the template for in vitro transcription with the mMESSAGE mMACHINE kit and MEGAshortscript T7 kit (Life Technologies, Carlsbad, CA, USA), respectively following the manufacturers’ protocol. Cas9 mRNA and sgRNA were purified using the MEGAclear kit and eluted with RNase-free water. The targeting oligo had one nucleotide deletion (c. 789del) and two homology arms of ∼60 bp, which were used as templates to repair the double-strand breaks generated by Cas9. 

Microinjection

C57BL/6 female mice (3–4 weeks old) were used as embryo donors and pseudopregnant foster mothers. Superovulated fertilized eggs were isolated from female C57BL/6J mice treated with 30 IU pregnant mare serum gonadotropin (Sigma-Aldrich, St. Louis, MO, USA) and 100 IU human chorionic gonadotropin (Sigma-Aldrich) at intervals of 50 hours. The mixtures containing Cas9 mRNA (50 ng/µL), sgRNA (100 ng/µL), and oligo donor were then co-microinjected into the cytoplasm of zygotes, and surviving zygotes were transplanted into the oviducts of pseudopregnant female mice to generate live animals. 

Genotyping of MYRF^mut/+ Mice

PCR and Sanger sequencing were used to analyze the genotype of MYRF^mut/+ mice. Total genomic DNA was extracted from mouse tails and then dissolved following the protocol of genomic DNA extraction kit (Tiangen, Beijing, China) for PCR and Sanger sequencing analysis. Forward primer and reverse primer sequences used for PCR are shown in Supplementary Table S2. 

Histology and Anterior Chamber Depth Measurement

Mice at 4, 8, and 20 weeks old were deep-anesthetized by carbon dioxide inhalation, followed by cervical dislocation. Eyes were harvested and immediately fixed in FAS fixed liquid (Wuhan Servicebio Technology Co., Ltd., Wuhan, China) for 24 hours at 4°C. Eyeballs were then transferred to ethanol for paraffin embedding after dehydration. After embedding, samples were cut into 5-µm-thick sections through the optic nerve head. Sections were stained with hematoxylin and eosin (HE) using standard protocols. Ocular anterior segments were then examined under a light microscope DMi8 (Laica, Wetzlar, Germany) and images obtained. Anterior chamber depth (ACD) was measured as the distance from the corneal endothelium to the anterior lenticular surface centered over the pupil. AL measurements were taken from the base of the optic nerve to the center of the cornea, whereas lens thickness (LT) was measured as the distance between anterior and posterior surface of lens. Simple crowding value (calculated as (LT − ACD)/AL) was used to evaluate the crowding condition of the eye.¹⁵ 

Silver Staining

Various tissue staining techniques, including HE staining and oxytalan staining (aldehyde-fuchsin staining), were repeatedly tested for visualizing and quantifying ciliary zonules. Silver staining was the most effective technique for quantifying major zonular fibrils between ciliary body and lens. Eyes in paraffin embedding were cut into 10-µm-thick sections through the optic nerve head. After dewaxing with xylene and dehydrating samples with gradient ethanol, sections were fixed using fixative solution (50% methanol, 10% acidic acid, 50 µL formaldehyde/100 mL solution) for 1 hour to overnight. Fixed sections were then treated for 1 minute with hypo solution (sodium thiosulfate solution 20 mg/100 mL) and washed with distillation-distillation H₂O three times (5 minutes each). Sections were stained with silver nitrite solution (200 mg/100 mL) for 30 minutes, developed in 100 mL developing solution (6 g sodium carbonate, 2 mL hypo solution, 50 µL formaldehyde) for 10 minutes and stopped with 5% acidic acid. Images focused on ciliary zonules were then obtained using LSM710 laser scanning confocal microscope (ZEISS, Jena, Germany). Five MYRF^mut/+ mice and 5 MYRF^+/+ mice were randomly selected and used for quantification of major zonular fibrils. One eye from each mouse was cut into paraffin sections through the optic nerve head and then silver staining performed. Major zonular fibrils in two fields (right and left fields) of each section were counted and the average major zonular fibrils number of two fields was recorded as the major zonular fibrils number of the section. 

Immunohistochemistry

After dewaxing with xylene and dehydrating sections with gradient ethanol, they were placed into citrate buffer (0.01 mol/L, pH 6.0) and heated in boiling water for 30 minutes to repair antigenic ability. Sections were then washed twice with PBS at room temperature for 10 minutes and incubated with blocking solution (5% normal goat serum in PBS) at room temperature for 1 hour. Further, sections were incubated with primary antibodies, including rabbit anti-MYRF (bs-11191R, 1:500; Biosynthesis Biotechnology Co., Ltd., Beijing, China), rabbit anti–fibrillin 1 (ab53076, 1:200; Abcam, Cambridge, MA, USA), and mouse anti–fibrillin 2 (sc-393968, 1:200; Santa Cruz Biotechnology, Santa Cruz, CA, USA), separately overnight at 4°C. Samples were then washed with PBS and incubated with Alexa Fluor 488–conjugated donkey anti-rabbit secondary antibody (A21206, 1:800; Invitrogen, Carlsbad, CA, USA) or Alexa Fluor 594–conjugated donkey anti-mouse secondary antibody (A11058, 1:800; Invitrogen) for 2 hours at room temperature. Samples were then counterstained with 4, 6-diamidino-2-phenylindole (Biofroxx GmbH, Germany) and washed with PBS. Immunofluorescence images focused on ciliary zonules were then obtained using the LSM710 laser-scanning confocal microscope (ZEISS). 

Tissue Collection, RNA Isolation, and Quantitative Real-Time PCR Analysis

Mice were euthanized and ciliary bodies were harvested directly under a stereoscopic microscope. A small hole was made using a syringe needle, and a small incision was made in the cornea. The posterior portion of the globe, including sclera, choroid, and retina, was dissected to the level of the pars plana and discarded. The iris was then isolated completely from the iris–ciliary body complex as much as possible, and finally the ciliary body was dissociated carefully from the sclera and lens using micro-forceps. Total RNA was extracted from the ciliary body with TRlzol (TaKaRa, Dalian, China) following the manufacturer’s instruction. Purified total RNA was quantified using a Nanodrop-1000 spectrophotometer (Thermo Fisher Scientific, Carlsbad, CA, USA). RNA was reverse-transcribed into cDNA using the First-Strand cDNA Synthesis Kit (Applied Biological Materials [ABM], Inc., Richmond, BC, Canada). qRT-PCR was performed using the SYBR Green Master Mix (Roche Life Science, Basel, Switzerland) in the Roche LightCycler 384. Primer sequences used are summarized in Supplementary Table S2. 

Western Blot Analysis

Ciliary bodies were immediately frozen in liquid nitrogen after dissecting them from eyes. Each sample was then lysed in 100 µL ice-cold RIPA buffer (Fude Biological Technology Co., Ltd., Hangzhou, China) supplemented with protease inhibitors and sonicated. Homogenates were centrifuged at 12,000 × g at 4°C for 5 minutes. Supernatants were collected and mixed with 6× loading buffer (Fude Biological Technology Co., Ltd.). After protein denaturation at 95°C for 5 minutes, samples were subjected to electrophoresis with 12% SDS-PAGE and electroblotted to polyvinylidene fluoride membranes (Millipore, Billerica, MA, USA). Membranes were blocked with 5% skim milk at room temperature for 2 hours and then incubated with primary antibodies: rabbit anti-MYRF (bs-11191R, 1:2000; Biosynthesis Biotechnology Co., Ltd.), rabbit anti–fibrillin 1 (ab53076, 1:2000; Abcam), and mouse anti–fibrillin 2 (sc-393968, 1:2000; Santa Cruz Biotechnology, Inc.) overnight at 4°C. After washing three times for 10 minutes with TBST, membranes were incubated with goat anti-mouse horseradish peroxidase (HRP)-conjugated secondary antibody (1:5000; Thermo Fisher Scientific, Carlsbad, CA, USA) or goat anti-rabbit HRP-conjugated secondary antibody (1:5000; Thermo) at room temperature for 2 hours, followed by washing with TBST for three times. Membranes were then scanned using the MP ChemiDoc Imaging System (Bio-Rad, Berkeley, California, USA). ImageJ software (National Institutes of Health, Bethesda, MD, USA) was used to analyze densitometric values of Western blot bands. 

Statistical Analysis

All data are presented as mean ± SEM. Comparisons of two individual experimental groups were performed using Student's t-test or the Mann-Whitney U-test based on the distribution of the data. SPSS 20.0 software (SPSS, Chicago, IL, USA) and GraphPad Prism 6 (version 6.01; GraphPad Software, La Jolla, CA, USA) were used for statistical analysis and generation of figures. P  <  0.05 was considered statistically significant. 

Results

Generation of MYRF^mut/+ Mice With CRISPR/Cas9 Genetic Editing System

To explore the effect of a human nanophthalmos frameshift mutation in MYRF on ciliary zonules, we successfully established MYRF^mut/+ mice in the C57BL/6J mouse strain. Exon 6 was scanned for sgRNA target sequence design to delete a cytosine deoxyribonucleotide causing frameshift mutation (c.789del, p.N264fs) (Fig. 1A). MYRF^mut/+ founders (F0) were generated by co-microinjection of the cocktail of Cas9 mRNA, sgRNA, and oligo donor into the cytoplasm of zygotes (Fig. 1B). Founder (F0) mice with targeted MYRF mutation were validated by PCR and Sanger sequencing analysis (Fig. 1D). In case of the rapid cleavage in the early embryonic stage, F0 mice would be mosaic. Therefore, three heterozygous mice (F1 mice) with heritable genotype were obtained after verifying that the targeted deoxyribonucleotide in exon 6 was mutated. Subsequently, F1 mice were intercrossed with C57BL/6J wild-type mice to generate more F2 heterozygous offspring for subsequent experiments (Fig. 1C). In addition, we have tried to generate homozygote offspring through hybridization between heterozygotes. However, no live MYRF homozygote mutation mice were obtained because homozygote MYRF gene mutation led to premature death during the embryonic period, and the offspring of heterozygotes had normal Mendelian segregation ratios (19 MYRF^mut/+ mice vs. 8 MYRF^+/+ mice, χ² P = 0.683). Heterozygous offspring with the same targeted MYRF mutation were verified by PCR and Sanger sequencing analysis. All the MYRF mutation mice used in our experiments had the same targeted mutation. Weight, growth, and behavior of MYRF^mut/+ mice were normal. 

Figure 1.

 

Generation of MYRF^mut/+ mice using the CRISPR/Cas9 system. (A) Schematic diagram of the MYRF gene encoding deoxyuridine 5′-triphosphate nucleotidohydrolase (dUTPase). Exons are indicated with blue rectangles and marked with E and numbers while introns are indicated with the short black line. Guide RNA (gRNA) target sites and protospacer adjacent motif (PAM) sequence are underlined in blue and red, respectively. (B) Embryonic stem (ES) cell was microinjected with sgRNA and Cas9 mRNA and matured as a blastocyst, which was subsequently transferred to pseudo-pregnant female mice. (C) The chimeras (F0) were crossbred with wild-type mice to generate littermate control mice (MYRF^+/+) and heterozygous (MYRF^mut/+) offspring (F1) containing the targeted locus through germline transmission. (D) Base sequence in MYRF^mut/+ and MYRF^+/+.

mRNA and Protein Expression Levels of MYRF Decreased in MYRF^mut/+ Mice Ciliary Bodies

To determine MYRF expression pattern in ocular tissues, immunohistochemistry was carried out on the whole mice eyeball. Relative high expression level of MYRF was observed in the ciliary body and retina, including RPE, inner nuclear layer (INL), and ganglion cell layer (GCL) (Fig. 2). In addition, a similar pattern of MYRF mRNA expression was observed in human ocular tissues. RPE showed the highest expression level of MYRF mRNA followed by the ciliary body and retina.¹³ To quantify expression of MYRF at the transcript and protein level in MYRF^mut/+ mice ciliary bodies, quantitative real-time PCR and Western blot of ciliary bodies were performed. A significant decrease in MYRF mRNA and protein expression was observed in MYRF^mut/+ mice ciliary bodies compared with expression levels in MYRF^+/+ mice (P < 0.01, Figs. 5, 6). 

Figure 2.

 

MYRF expression pattern in mice ocular tissues. (A) Eyeball panorama gram of MYRF immunofluorescence. Scale bar: 0.5 mm. (B) Partial enlarged detail of anterior chamber structures shows the high MYRF expression in the ciliary body. (C) Partial enlarged detail of retina shows the high MYRF expression in RPE, INL, and GCL.

Human Nanophthalmos Frameshift Mutation of MYRF in Mice Results in a Shallow Anterior Chamber Depth

A shallow anterior chamber is an important characteristic of nanophthalmos in humans. To investigate the effect of MYRF mutation on mice ocular anterior segments, HE staining was used to measure ACD (from the corneal endothelium to the anterior lenticular surface, centered over the pupil) and AL of total six littermate control mice (MYRF^+/+) and six mutant mice (MYRF^mut/+) (Fig. 3A). Mean ACDs of MYRF^+/+ mice and MYRF^mut/+ mice were 0.43 ± 0.04 and 0.37 ± 0.005 mm, respectively (P < 0.01, Fig. 3B). This finding implies that the ACD of MYRF^mut/+ mice was significantly reduced compared with the ACD of the control group. In addition, we analyzed ACD and AL ratio and simple crowding value of these mice. The mean ACD/AL was significantly higher in MYRF^+/+ mice (0.14) compared with the ACD/AL in MYRF^mut/+ mice (0.12) (P < 0.01). Simple crowding was significantly lower in MYRF^+/+ mice (0.44) compared with simple crowding in MYRF^mut/+ mice (0.48) (P < 0.001). Significantly higher simple crowding value in MYRF^mut/+ mice implies a crowding condition of the eye. However, the mean AL in MYRF^mut/+ mice was not significantly different compared with the mean AL of MYRF^+/+ mice (Supplementary Fig. S1). 

Figure 3.

 

Reduced anterior chamber depth in MYRF^mut/+ mice. (A) Representative HE staining images of midsagittal sections of mouse eyes showed a difference in ACD between MYRF^mut/+ mice and MYRF^+/+ mice. Scale bar: 50 µm. (B) The mean ACD, ACD/AL ratio, and simple crowding value in MYRF^mut/+ mice and MYRF^+/+ mice. (C) ACD measurements at 4, 8, and 20 weeks old in MYRF^mut/+ mice and MYRF^+/+ mice. Data represent mean ± SEM. n = 6. *P < 0.05, **P < 0.01, ***P < 0.001 versus MYRF^+/+ mice.

The patient with frameshift mutation (c.789delC, p.S264fs) was a 16-year-old girl with bilaterally short AL (OD: 18.09 mm; OS: 18.07 mm).¹⁰ Notably, she was diagnosed with secondary angle-closure glaucoma after a visit to our facility. Ultrasound Biomicroscopy (UBM) analysis showed that her ACD was 2.68 mm OD and 2.37 mm OS, which implies that the abnormal ciliary zonule and lens states as lens dislocation were characterized by binocular asymmetry of anterior chamber depth.¹⁶ Mice used for ACD and AL measurement were 8 weeks old (young adults), which matched with the disease onset in this patient. In addition, reduced ACD was reported from postnatal 4 weeks and did not progress significantly until postnatal 20 weeks (Fig. 3C). Therefore, 8-week-old mice (young adults) were used for all subsequent experiments. 

Ciliary Zonule Defects Were Present in MYRF^mut/+ Mice as Demonstrated by Silver Staining and Immunofluorescence

Ciliary zonules are made of fibrillin microfibril bundles and play an important role in anchoring the eye lens to the ciliary body.¹⁷ Clinically, zonule defects are usually secondary to nanophthalmos and are associated with shallow ACD and crowding condition of ocular anterior segments. Therefore, we focused on phenotypes related to ciliary zonule defects in MYRF^mut/+ mice. For the first time, we tested and adopted silver staining strategy to explore the delicate structure of ciliary zonules. The density of zonular fibers extending from the ciliary epithelium to the lens equator was reduced, and the zonule morphology was altered in MYRF^mut/+ mice (Fig. 4B). On the contrary, the fibrous structure of ciliary zonules remained intact in MYRF^+/+ mice (Fig. 4B). After performing silver staining, several zonular fibers were identified (shown with asterisks in Fig. 4B), which were defined as major zonular fibrils to quantify the change of zonular fibers. Further quantification showed that the mean number of major zonular fibrils was significantly decreased (P < 0.05, Fig. 4D), indicating that structural defects of ciliary zonules occurred in nanophthalmos-associated MYRF mutants. FBN1 and FBN2 are the most important molecular markers of fibrillin microfibrils in zonules.¹⁸ Therefore, an immunofluorescence technique was used to stain eye samples with respective antibodies. Immunofluorescent analysis showed that both major FBN1-stained and FBN2-stained fibers were reduced in MYRF^mut/+ zonules compared with those in MYRF^+/+ zonules (P < 0.01, Figs. 4C, 4E, 4F), which was consistent results obtained with silver staining. 

Figure 4.

 

Ciliary zonule defects in MYRF^mut/+ mice. (A) Pattern of anterior angle structures. (B) Silver staining demonstrated reduced zonular fibers density and zonular fibers dehiscence in MYRF^mut/+ mice compared to those in MYRF^+/+ mice. Major zonular fibers were labeled with asterisks. Scale bar: 25 µm. (C) Immunofluorescence analysis showed that fluorescence intensity of both FBN1 and FBN2 staining was significantly reduced in the zonules of MYRF^mut/+ mice compared with those in MYRF^+/+ controls. The deficiency of FBN1 and FBN2 staining was labeled with red arrows. Major zonular fibers were labeled with asterisks. Scale bar: 50 µm. (D) Quantification of the number of major zonule fibers per eye section field in MYRF^mut/+ and MYRF^+/+ mice. The mean value was 5.7 ± 0.6 in MYRF^+/+ mice and 4.5 ± 0.4 in MYRF^mut/+ mice. Data represent mean ± SEM. n = 5. *P < 0.05 versus MYRF^+/+ mice. (E, F) Quantification of the number of major FBN1- and FBN2-stained fibers per eye section field in MYRF^mut/+ and MYRF^+/+ mice. Data represent mean ± SEM. n = 3. **P < 0.01 versus MYRF^+/+ mice.

mRNA Expression of Both FBN1 and FBN2 was Decreased in MYRF^mut/+ Mice

Fibrillins, the major components of mouse zonules similar to those in humans, are mainly derived from nonpigmented ciliary epithelium (NPCE). FBN1 and FBN2 encoding abundant zonular proteins are expressed predominantly in the NPCE throughout development.¹⁸ Therefore, we dissected ciliary bodies to analyze expression levels of both FBN1 and FBN2 by qRT-PCR analysis and Western blot. The results showed a significant decrease in mRNA and protein expression levels of FBN1 and FBN2 in MYRF^mut/+ mice (P < 0.05, Figs. 5, 6). These results imply that production of fibrillins from NPCE was reduced in MYRF^mut/+ mutants, which is consistent with the phenotype of zonule defects. 

Figure 5.

 

Transcript expression in the ciliary body. A significant decrease of MYRF, FBN1, and FBN2 mRNA expression was shown in the MYRF^mut/+ mice ciliary body. Data represent mean ± SEM. n = 8. *P < 0.05, **P < 0.01 versus MYRF^+/+ mice.

Figure 6.

 

Protein expression in the ciliary body. Western blots showed a decrease of MYRF, FBN1, and FBN2 protein expression in the MYRF^mut/+ mice ciliary body. n = 3. *P < 0.05, **P < 0.01 versus MYRF^+/+ mice.

Discussion

In a previous study, we reported a mutation in MYRF (c.789delC, p.S264fs), which is a gene implicated in nanophthalmos through trio-based WGS analysis.¹⁰ Therefore, in this study, we modeled the mutation in mice to explore the role of MYRF in ciliary zonule structure and its major molecular composition. 

At first, we performed immunofluorescence analysis to evaluate the MYRF expression pattern in mice. Results showed significantly high expression levels of MYRF in ciliary bodies and retina, including RPE, INL, and GCL. A similar pattern of MYRF mRNA expression is reported in human ocular tissues, with the highest expression level recorded in RPE and relatively high expression level in ciliary bodies and retina.¹³ Therefore, further studies should explore the role of MYRF in RPE, retina, and ciliary bodies. However, significant abnormal nanophthalmos-related ocular phenotypes, including short AL, retinal folds, RPE defeats, high intraocular pressure, and thinner retinal nerve fiber layer or significant abnormal nonocular phenotypes found in MYRF Olig2 conditional knockout mice, including tremors, ataxia, and seizures,¹¹ were not observed (data not shown). Instead, ciliary zonule defects accompanied with a shallow ACD, decreased ACD/AL ratio, and increased simple crowding value were observed in MYRF^mut/+ mice. These findings imply that MYRF is important for integrity of ciliary zonules. Analysis of silver staining results showed that the morphology of ciliary zonules was altered with reduced zonular fiber density and significant structural dehiscence in zonular fibers in MYRF^mut/+ mice compared with the features in the control group. The principal structural component of the ciliary zonule is fibrillin, a cysteine-rich glycoprotein. Three fibrillin isoforms (FBN1–3) are present in humans, whereas the gene FBN3 is inactive in mouse.¹⁹ Therefore, the profile FBN1 and FBN2 in MYRF^mut/+ mice was evaluated to determine if MYRF deficiency affects the principal structural component of the ciliary zonule. Immunofluorescent analysis showed that FBN1 staining and FBN2 staining were reduced in MYRF^mut/+ zonules. Furthermore, mRNA and protein expression of MYRF, FBN1, and FBN2 in ciliary bodies was downregulated in mutant mice. 

Low expression level of MYRF at the transcript and protein level in MYRF^mut/+ mice can be attributed to two reasons. First, after the frameshift mutation, the termination codon appeared in advance at the amino acid position 271, resulting in a truncated protein (length of 270). The premature termination codon might trigger the nonsense-mediated mRNA decay (NMD) pathway, resulting in MYRF mRNA degradation. Second, the truncated protein might not trigger the NMD pathway, whereas the shorter MYRF fragment resulted in decreased transcript and protein expression of MYRF. Whatever the actual mechanism was, MYRF transcript and protein expression levels were lower in MYRF^mut/+ mice ciliary bodies, implying that MYRF plays an important role in zonular defects. On the other hand, we could not exclude the possibility that the MYRF truncated protein may act in a dominant-negative manner, resulting in the observed phenotype. The underlying mechanism of zonular defects caused by MYRF frameshift mutation should be further investigated. 

Several nanophthalmos-associated gene mutation mouse models have been used to investigate the human nanophthalmos phenotype. Neither Mfrp knockout mice nor Tmem98 mutation mice showed a statistically significant reduction in AL measurements.^3,20 Similarly, our findings on MYRF^mut/+ mice showed that AL was slightly reduced, but the decrease was not statistically significant. Although there is a measurable difference in ACD, AL may not change significantly due to the relatively minor proportion of ACD in AL. Additionally, mouse eyes are very small, which may require novel and more precise methods to achieve biologically meaningful accuracy. Use of the HE-stained section to measure AL does not accurately simulate the condition of a living eye, because intraocular pressure change after dehydration may result in measurement bias. The ACD structure is relatively more stable compared with the overall AL in an eyeball; therefore, measurement of ACD using HE staining is more accurate in demonstrating the phenotype of MYRF^mut/+ mice. Therefore, spectral-domain optical coherence tomography with higher imaging speed and higher resolution is a better choice to measure the dimension of mouse eyes in future research.²¹ On the other hand, the development of AL may be a consequence of polygenetic regulation; therefore, a single gene mutation may not be sufficient to cause significant change in AL development. 

Nanophthalmos²² is a rare ocular disorder, attributed to developmental arrest of eyes in the early stages of embryogenesis. Short AL, narrow iridocorneal angle, and shallow anterior chamber are associated with several clinical complications, with secondary angle-closure glaucoma being the most common complication.^2,23 Unlike primary angle-closure glaucoma, traditional laser peripheral iridotomy and trabeculectomy do not completely relieve high intraocular pressure in nanophthalmos, and the incidence of aqueous misdirection is significantly increased.²⁴ One of our hypotheses is that developmental ciliary zonule defects in nanophthalmos result in forward movement of a relatively larger-sized lens, following pupillary blockage and a shallow peripheral anterior chamber. The effect of these changes results in movement of aqueous seek backward into the vitreous cavity, thus pushing the lens-iris diaphragm forward, a trait associated with aqueous misdirection. Surgical operation in such a narrow and crowded anterior chamber with increased vitreous pressure may result in the occurrence of malignant glaucoma.²⁵ Our findings imply that ciliary zonule defects in MYRF mutation mice may be associated with developmental arrest of eyes in nanophthalmos. Experimental findings and clinical observations show that ophthalmologists should pay extra attention to changing surgical details, especially ciliary zonule defects in patients with angle-closure glaucoma secondary to nanophthalmos. 

Ciliary zonules are the circumferential suspensory ligaments that span the gap between the lens and adjacent ciliary bodies. The main role of zonules is to maintain the lens in the center of the visual pathway and transmit contraction forces from ciliary bodies to the lens for accommodation.^26,27 Zonules can be classified into three subgroups: anterior, equatorial, and posterior parts based on the attachment point of these fibers on the lens.²⁸ Reduction in tensile strength of ciliary zonules or rupture of any fibers in any subgroup can weaken traction force to the lens. In either case, the lens might become thicker and move forward or backward. This partially explains the phenotype of ciliary zonule defects in MYRF^mut/+ mice that present with a shallow anterior chamber compared with that in the control group with intact ciliary zonules. However, fiber groups that are defective were not identified due to currently available technical limits in distinguishing fine details among the three fiber subgroups as the mouse ciliary zonule fibers are too delicate. Furthermore, available techniques for studying ciliary zonules are limited. Each zonular fiber is composed of hundreds or thousands of microfibrils with long filamentous structures about 10 to 12 nm in diameter, which would not be visible with a light microscope.²⁹ Ciliary zonule is a type of oxytalan fiber,³⁰ and therefore, traditional aldehyde fuchsin and HE staining are difficult to make it clearly visible.³¹ Moreover, it is difficult to achieve good views under scanning electron microscopy due to artifacts associated with reagents for fixation and/or drying of fibers.³⁰ To our knowledge, this is the first time that analysis of the structure of ciliary zonules by silver staining is being reported. The findings show that ciliary zonules are distinguishable, and major zonular fibrils can be quantified. These findings provide a basis for further research on ciliary zonules, which is important but obstructed by previous lack of appropriate methodology. 

Fibrillin microfibrils are the main components of ciliary zonules, and proteomic analysis shows dozens of proteins in ciliary zonular fibers, with FBN1 and FBN2 as the main type of proteins.^17,18 Other important proteins include latent-transforming growth factor β-binding protein 2 (LTBP-2), microfibrillar-associated protein 2 (MFAP-2), disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS), and ADAMTS-like proteins (ADAMTSL).^32–34 In this study, FBN1 and FBN2 were used to perform immunofluorescent analysis as they are the highest expressed proteins in the ciliary zonule. Reduction level in FBN1 and FBN2 was further confirmed at the molecular level, indicating zonule abnormality in MYRF^mut/+ mice. The relative expression levels of FBN1 and FBN2 in NPCE, which play a role in the synthesis of the two zonular proteins,³⁵ were decreased. However, the underlying molecular mechanism of zonule defects associated with the MYRF mutation should be explored further. 

Mutations of FBN1 and FBN2 caused ciliary zonule defeats and lens dislocation,^29,36–38 demonstrating that FBN1 and FBN2 regulate structural integrity of the ciliary zonule. Other gene mutations, such as LTBP-2, ADAMTSL4, and CTRP5, have also been reported to cause ciliary zonule defects.^39–41 In contrast, the MYRF gene is not a directly related fibrillin microfibril gene. Therefore, ciliary zonule defects in MYRF^mut/+ mice may be due to the low FBN1 and FBN2 expression. Nonetheless, the relationship between MYRF, FBN1, and FBN2 needs to be further investigated. Given the lack of clinical measurements to evaluate the integrity of ciliary zonules, zonule abnormalities are often not identified until intraocular surgery, rendering patients susceptible to fatal intraoperative complications such as aqueous misdirection and posterior capsule rupture.⁴² By screening for gene mutations, the risk of intraocular surgery complications caused by unstable ciliary zonules can be predicted in a timely manner. 

Some limitations in this study should be considered. First, CRISPR editing might have a possibility of off-target effects, yet we have taken some measures to reduce it. We have designed six sgRNAs and finally chose the most efficient and specific one to avoid the off-targets. Additionally, Cas mRNA was used for microinjection, and the short half-life mRNA would decrease the risk of continuous cutting off the target. On the other hand, the mixtures containing Cas9 mRNA, sgRNA, and oligo donor were co-microinjected into the cytoplasm of zygotes. The targeting oligo carried one nucleotide deletion (c. 789del) and two homology arms of ∼60 bp, which were used as templates to repair the double-strand breaks generated by Cas9. Second, although we have identified that there was a significant decrease in the mRNA and protein expression of FBN1 and FBN2 in MYRF^mut/+ mice ciliary bodies, the underlying molecular mechanism of zonule defects associated with MYRF mutation has yet to be further investigated through RNA-seq or tandem mass tag (TMT)-based proteome profiling. 

In summary, our previous findings revealed that a DNM (c.789delC, p.S264fs) in MYRF is associated with human nanophthalmos.¹⁰ We found that the same frameshift mutation in MYRF caused a shallowing anterior chamber and ciliary zonule defects in mice, which suggests that MYRF plays an important role in zonule development. More important, we have utilized a silver staining strategy to detect the structure of ciliary zonules for the first time, which may contribute to progress in future investigations of ciliary zonules. Clinically, our findings indicate that ophthalmologists should pay attention to preoperative ciliary zonule assessment in patients with nanophthalmos during cataract surgery or antiglaucoma surgery to avoid severe complications. 

Acknowledgments

The authors thank Beijing Biocytogen Co., Ltd. (Beijing, China) for its technical support in the CRISPR design. This study was financially supported by Guangzhou Science and Technology Plan Project (Grant 201803040020, 201903010065; Guangzhou, China), Guangdong Natural Science Foundation (Grant 2020A151501168; Guangzhou, China), and Research Funds of the State Key Laboratory of Ophthalmology (Grant PT1001022). 

Disclosure: X. Yu, None; N. Sun, None; X. Yang, None; Z. Zhao, None; X. Su, None; J. Zhang, None; Y. He, None; Y. Lin, None; J. Ge, None; Z. Fan, None