A splicing mutation in aryl hydrocarbon receptor associated with retinitis pigmentosa

Abstract

Retinitis pigmentosa (RP) refers to a group of retinal degenerative diseases, which often lead to vision loss. Although 70 genes have been identified in RP patients, the genetic cause of approximately 30% of RP cases remains unknown. We aimed to identify the cause of the disease in a cohort of RP families by whole exome sequencing. A rare homozygous splicing variant, c.1160 + 1G>A, which introduced skipping of exon 9 of the aryl hydrocarbon receptor (AHR), was identified in family RD-134. This variant is very rare in several exome databases and leads to skipping of exon 9 in the transcript. AHR is expressed in the human retina and is a ligand-activated transcription factor with multiple functions. Mutant AHR failed to promote 2, 3, 7, 8-tetrachlorodibenzo-p-dioxin (TCDD)-induced xenobiotic responsive element (XRE) luciferase activity. In parallel, mutation in AHR abolished activation of its downstream target gene, such as CYP1A1 and CYP1A2. To investigate the in vivo roles of Ahr in the retina, we generated a retina-specific conditional knockout mouse model of Ahr. Comparing with wild-type mouse, Ahr knockout mice exhibited reduced electroretinogram responses at 9 months of age. Retinal histology revealed retinal histology showed the degeneration of photoreceptors with a thinner outer nuclear layer. Thus, our data demonstrate that AHR is associated with RP.

Introduction

Retinitis pigmentosa (RP) is a group of hereditary degenerative diseases of the retina characterized by night blindness, progressive loss of peripheral vision in the early stage, and complete loss of vision in the late stages (1). Affected individuals generally experience early-onset night blindness followed by progressive loss of peripheral vision; decreased electroretinograph (ERG) findings; fundus changes, including arterial attenuation, disc pallor, retinal pigment epithelium (RPE) atrophy and the presence of intraretinal pigmentary deposits; and eventual loss of central vision or complete blindness. The prevalence of RP varies in different populations worldwide, ranging from approximately 1 in 4000 to 1 in 1000 (2–4), and affects more than 1 million individuals (3,5–7).

In the recent years, whole exome sequencing (WES) has been shown to be an effective method for identification of disease-associated genes and genetic variants (8–14). According to the RetNet database (https://sph.uth.edu/retnet/; date last accessed October 30, 2017), over 70 genes have been associated with non-syndromic RP (15). Most of the RP disease-associated genes are expressed preferentially in either the photoreceptors or the RPE and encode diverse proteins involved in the phototransduction cascade, the retinoid cycle, the photoreceptor structure, or transcription (RetNet). However, these disease genes can only explain only approximately 50–60% of the existing RP patients.

In the present study, we identified a homozygous splicing mutation in the aryl hydrocarbon receptor (AHR) gene in one consanguineous Indian family by WES. We applied WES to explore the disease-associated genes in this RP family. We found a novel homozygous splicing mutation in AHR, c.1160+1G>A, in an autosomal recessive RP (arRP) family from south India. cDNA analysis demonstrated skipping of exon 9 in the patient’s transcript. The transcriptional activity of AHR was abolished by the AHR c.1160+1G>A mutation. We generated knockout mouse models to investigate the in vivo roles of Ahr in the retina. Ahr mutant mice displayed late-onset retinal degeneration. Our study demonstrated that AHR was associated with RP.

Results

WES identified a homozygous variant in AHR as a candidate mutation in an arRP consanguineous family

Two patients affected with RP from one consanguineous Indian family (RP-134) were recruited (Fig. 1A). RD-ICP-78 was a 10-year-old boy, and RD-ICP-79 was an 8-year-old boy. These two RP patients presented with an early-onset increasing difficulty to adapt in dim light and gradually decreased vision acuity in both eyes. Typical RP features were observed in both patients. The fundus picture of patient RD-ICP-78 shown in Figure 1A and B reveals pigment deposits and a pale retina. The fundus examination of patient RD-ICP-79 is shown in Figure 1C and D. OCT examinations of patient RD-ICP-79 showed disorganized photoreceptor layer (Fig. 1E). Full-field ERG tests of patient RD-ICP-79 revealed severely reduced scotopic, photopic responses and oscillatory potentials in both eyes (Supplementary Material, Fig. S1).

Figure 1.

Clinical images of affected patients with RP in family RP-134. (A, B) Fundus photographs of patient RD-ICP-78 showed pigment deposits in the retina. (C, D) Fundus photographs of patient RD-ICP-79 showed a pale colour and pigment deposits in the retina. (E) OCT examination revealed xx in the photoreceptor layer.

To identify the causative mutations for this arRP family from India, we performed WES on genomic DNA samples from the proband patient RD-ICP-78 (VII: 1) and the unaffected relative V: 8. We generated 4.3 billion and 5.6 billion bases of sequence with an average throughput depth of 86.2× and 111.5× for the target regions, respectively. Variants were called using Atlas2 and noted by ANNOVAR. We identified 9011 SNPs and 417 indels that may affect amino acid sequences in patient RD-ICP-78 and 9301 SNPs and 429 indels that may affect amino acid sequences in sample V: 8.

We first compared the identified variants in patient RD-ICP-78 (VII: 1) with reported retinal degeneration genes (https://sph.uth.edu/Retnet/). The exons of these RP genes were well covered (Supplementary Material, Table S1). No causal mutations were found in the known RP genes. Variant frequency data were obtained from the dbSNP database, 1000 Genomes, NHLBI Exome Sequencing Project (ESP), the ExAC Browser (Beta), gnomAD browser beta and our in-house controls of 1700 WES data. After screening though the dbSNP138 and 1000 Genome Project databases (Supplementary Material, Table S2), we excluded the variants with high frequencies using the criteria described in the Materials and Methods section and identified 614 variants in patient RD-ICP-78 (Supplementary Material, Table S3). After excluding the heterozygous variants and synonymous mutations and considering effect of variant and gene expression data, we found 8 variants in patient RD-ICP-78 (Supplementary Material, Table S4). Subject V: 8 was unaffected individual, which would not have the homozygous or compound heterozygous mutations VII: 1 (RD-ICP-78). After these steps, these eight variants were tested in nine other family members by Sanger sequencing (Supplementary Material, Table S5). Only a homozygous variant in AHR, NM_001621.4 c.1160+1G>A, segregated with the disease in the family (Fig. 2), while all other seven variants did not segregate with the disease (Supplementary Material, Table S5).

Figure 2.

Identification of the AHR mutation in family RP-134. (A) Pedigrees of the RP-134 Indian families. Arrows indicate the two patients RD-ICP-78 (VII: 1) and RD-ICP-79 (VI: 6). VII: 1 and V: 8 were subjected to WES analysis. (B, C) Sanger sequencing analysis of AHR in the RD-134 family showing that the homozygous splicing mutation c.1160 + 1G>A co-segregated with the phenotype. Both patients RD-ICP-78 (VII: 1) and RD-ICP-79 (VI: 6) harboured the homozygous splicing mutation c.1160 + 1G>A in the AhR gene. V: 8, the mother of patient VI: 6, was a heterozygous carrier. (D) Genotypes of all members of RP-134. Note VII: 2 (103) does not carry the mutant allele.

This AHR c.1160+1G>A variant was absent in 1000 ethnicity-matched control samples screened by direct sequencing. Furthermore, this variant was not present in 228426 alleles from the Genome Aggregation Database (http://gnomad.broadinstitute.org). The c.1160+1G>A variant was also investigated in the human gene mutation database (http://www.hgmd.org/), and the results showed that this variant was novel. We used homozygosity mapping on WES data (VCF files) using HomozygosityMapper. There are several homozygous regions in RD-ICP-78 (Supplementary Material, Fig. S2). The AHR mutation identified in this study was located in the homozygous region on chromosome 7 of RD-ICP-78, indicated with a blue arrow (Supplementary Material, Fig. S2). Haplotype analysis using SNP surrounding the AHR locus in this family revealed a homozygous region in both RD-ICP-78 and RD-ICP-79 (Supplementary Material, Fig. S3). In addition, association analysis of RPGR locus with the RP disease in this family showed that RPGR is not associated with the disease (Supplementary Material, Fig. S4). These data demonstrated that the homozygous splicing mutation c.1160+1G>A in AHR was a good candidate for retinal degeneration in this Indian RP family.

The splicing mutation in AHR introduced a 142 bp deletion in the cDNA

cDNA samples from the two patients were prepared to assess the effect of the c.1160+1G>A mutation on splicing (Fig. 3A). This splicing mutation led to deletion of exon 9 and resulted in frameshift change at the 399th amino acid residue, Arginine, and a termination codon occurred after seven amino acid (p.R339fs7*) (Fig. 3B). The AHR protein contains distinct functional domains (Fig. 3B). The bHLH motif is located in the N-terminal of the protein and is commonly found in a variety of transcription factors. Two PAS domains, PAS-A and PAS-B, are stretches of 200–350 amino acids. The ligand-binding site of AHR is contained within the PAS-B domain and has several conserved residues critical for ligand binding. The ninth exon is located at the glutamine-rich (Q-rich) domain, which is located in the C-terminal region of the protein and is involved in co-activator recruitment and transactivation (Fig. 3B).

Figure 3.

Effect of the AHR splicing mutation c.1160 + 1G>A on protein function. (A) The schematic diagram shows the location and effect of the c.1160 + 1G>A mutation. (B) Functional domains of AHR WT and mutant protein. The basic helix-loop-helix (bHLH) motif is located in the N-terminal of the protein and is a common entity in a variety of transcription factors. AHR also are two PAS domains, PAS-A and PAS-B, which are stretches of 200–350 amino acids. The ligand-binding site of AHR is contained within the PAS-B domain and contains several conserved residues critical for ligand binding. A glutamine-rich (Q-rich) domain is located in the C-terminal region of the protein and is involved in co-activator recruitment and transactivation. Mutant protein (p.R339fs7*) only retains the first 339 amino acids and seven additional amino acids. The glutamine-rich transcriptional activation binding domain is lost. (C) A 142 bp deletion of exon 9 was identified in the cDNA samples from blood leukocyte cells of the two patients RD-ICP-78 (VII: 1) and RD-ICP-79 (VI: 6) by RT-PCR. (D) Sanger sequencing analysis confirmed the deletion of exon 9 in the cDNA sample of RD-ICP-78 (VII: 1). (E–G) The AHR c.1160 + 1G>A, p.R339fs7* mutation abolished the transcriptional activity of AHR. (E) pCMV-AHR-WT, pCMV-AHR-Mut (exon 9 deletion) and mock plasmids were cotransfected with the promoter reporter plasmid pGL3-XRE-luciferase. After 48 h, the XRE luciferase activity induced by the pCMV-AHR-WT expression plasmid was significantly higher than that of the normal control, while the XRE luciferase activity induced by the pCMV-AHR-Mut expression plasmid was dramatically decreased. (F, G) The pCMV-AHR-WT expression plasmid and pCMV-AHR-Mut expression plasmid were transfected into D407 cells. TCDD was used to activate AHR. After 48 h, AHR WT significantly induced the mRNA expression of CYP1A1 and CYP1A2, but AHR MUT could not activate these two genes (E, F). N = 4 for each groups in (E)–(G). ***P < 0.001. In all cases the P-values refer to comparison of AHR Mut KO vs. WT, by two-tailed t-test. (H) Detection of AHR mutant protein by western blotting analysis. The pCMV-AHR-WT expression plasmid and pCMV-AHR-Mut expression plasmid were transfected into D407 cells.

To assess the effect of AHR mutation on the transcript, we designed primers for AHR to amplify the cDNA fragment between exon 8 and exon 10. As shown in Figure 3C, the PCR products from the patients were 142 bp shorter than those of the normal controls. Sanger sequencing of the PCR products from the patients confirmed that there was a deletion of exon 9 (142 bp) of the AHR cDNA (Fig. 3D). Thus, the c.1160+G>A splicing mutation in the AHR gene introduced a frameshift deletion mutation (Fig. 3B).

The mutant AHR protein failed to activate downstream target genes

AHR is a transcription factor that is activated by xenobiotic substances such as dioxin. After activation, it binds to the XRE element of DNA, thereby inducing transcription of a variety of downstream target genes, such as xenobiotic-metabolizing enzymes (16,17). To investigate the effect of AHR c.1160+1G>A on its transcriptional activity, we constructed the pCMV-AHR-WT and pCMV-AHR-Mut plasmids (exon 9 deletion) and cotransfected them with the promoter report plasmid pGL3-XRE-luciferase. As shown in Figure 3E, the XRE luciferase activity induced by the pCMV-AHR-WT expression plasmid was significantly higher than that of the vector control. However, the XRE luciferase activity introduced by the pCMV-AHR-Mut expression plasmid was similar to that of the vector control (Fig. 3E). These results indicated that the c.1160+1G>A mutation abolished the transcriptional activity of AHR. To further identify the effect of the AHR mutation on downstream target gene transcription, we transfected the pCMV-AHR-WT expression plasmid and pCMV-AHR-Mut expression plasmid into D407 cells. After 24 h, TCDD was added to activate AHR, which was translocated to the nucleus and promoted the transcription of downstream genes. As shown in Figure 3F and G, WT AHR observably induced the expression of CYP1A1 and CYP1A2, while mutant AHR failed to active these two target genes. Taken together, these data suggested that the truncated AHR protein encoded by the c.1160+1G>A splicing mutation identified in our RP patients lost its transcriptional activity and failed to activate downstream target genes (Fig. 3H).

Ahr mutant mice exhibited a retinal degeneration phenotype

In addition to its expression in the RPE, Ahr is expressed in the mouse retina and photoreceptors (18). AHR expression in the human retina was investigated by immunohistochemistry and western blot. As shown in Supplementary Material, Figure S5, AHR was expressed in human retinal photoreceptors. Ahr has been reported to play roles in the RPE and retina (18,19). To investigate the potential roles of Ahr in the retina, we generated retina-specific knockout mouse model by deleting Ahr in neuron progenitor cells using the Six3-Cre line (20). Deletion of Ahr in the neuron progenitor cells (Ahr^flox/flox, Six3-Cre, named Ahr KO) was assessed by real-time PCR. Ahr expression levels in Ahr KO retinas were reduced to approximately 25% of that of controls (Fig. 4A), probably due to the fact that Six-Cre was expressed in neuronal progenitor cells. Immunohistochemistry staining of retinal cryosections revealed that no Ahr expression in Ahr KO sections (Fig. 4B) At 12 months of age, Ahr KO mice exhibited decreased scotopic responses (Fig. 4C), compared with WT mice. The mean b-wave amplitude was reduced by approximately 30% in Ahr KO mice at a light intensity of 0 and 1 log cd·s/m². A-wave was reduced about 15% in Ahr KO mice.

Figure 4.

Ahr knockout mice exhibited retinal degeneration phenotypes. (A) Reduced expression of Ahr in retinal knockout of Ahr mice. Real-time PCR analysis revealed reduced expression of Ahr in knock out mice. Quantification of expression revealed that expression of Ahr was decreased to 26% of that of control. In the brain, Ahr expression did not change. (B) Reduced expression of AHR in Ahr KO retina sections. AHR antibody was used at 1: 100 dilution. No AHR expression was observed in the photoreceptor layer. (C) Ahr mutant mice showed reduced scotopic electroretinogram (ERG) b-wave amplitude responses. Data were from an average of four samples per group. (D) H&E staining of retinal sections showed a thinner outer nuclear layer and thinner outer segment in Ahr KO retinas at 18 months of age compared to WT retinas. No defect was observed in heterozygous mice with the genotype of Ahr ^loxp/+, Six3-Tg. N = 4 for each groups in (A)–(C). *P < 0.05; **P < 0.01; ***P < 0.001. ns, no significant difference. In (A)–(C), the P-values refer to comparison of Ahr KO (retina-specific knockout) vs. WT mice, by two-tailed t-test. (E) Quantification of outer nuclear layer (ONL) nuclei revealed that the number of photoreceptor cell per row in Ahr mutant was reduced (left panel). There was no change in the number of cells per row in the inner nuclear layer (INL) in Ahr mutant (middle panel). ONL length was reduced in Ahr mutant retina (right panel). Note no defect was observed in both photoreceptor cell number per row or OS length quantification in Cre carrying Ahr heterozygous mice. Nuclei were counted every 200 μm to the optic nerve. OS length was also measured every 200 μm to the optic nerve. Sample size N = 4. *P < 0.05; **P < 0.01; ***P < 0.001. Scale bar: 20 μm.

To investigate the pathological changes underlying the abnormal ERG test results, we applied H&E staining and immunohistochemistry to study retinal sections of WT and Ahr KO mice. We used H&E staining to assess whether the retinal ONL thickness was changed in Ahr KO retinas. The results showed that Ahr KO mice exhibited retinal degeneration phenotype. Although the overall morphology of the retina appeared normal, the number of photoreceptor cells per row in 18-month-old Ahr KO mice was reduced compare to that of the WT mice (Fig. 4D and E). No difference in inner nuclear layer (INL) was observed (Fig. 4D and F). The outer segment of retinas from 18-month-old Ahr KO mice was also thinner than that of the WT mice (Fig. 4D and G). Cre carrying heterozygous mice with the genotype of Ahr^loxp/+, Tg did not exhibit any abnormality (Fig. 4D–G).

Consistent with the ERG results, immunostaining revealed increased expression of the proapoptotic protein CCAAT/-enhancer-binding protein homologous protein (CHOP) in the Ahr KO retinas (Fig. 5A), indicating cellular stress in the mutant retina. TUNEL analysis demonstrated increased cell death in the Ahr KO retinas (Fig. 5B).

Figure 5.

Analysis of CHOP expression and TUNEL in the retina of Tmem30a inducible knockout mutant mice by immunofluorescence microscopy. (A). Immunofluorescence labelling of retina cryosections from control (WT) and mutant (KO) littermates at 11 months of age using CHOP antibody (green) and DAPI (blue). CHOP expression was increased in Ahr KO mice. (B). Immunofluorescence labelling of retina cryosections from 11 months of age control and mutant littermates with the TUNEL labelling kit. TUNEL staining is shown in green and DAPI in blue. TUNEL positive cells were observed in Ahr mutant retina sections. Scale bar: 10 μm.

Discussion

RP is an inherited disease caused by a progressive decrease in rod and cone photoreceptor function. Although more than 70 known genes responsible for RP have been identified, there are still many unknown RP disease-associated genes. NGS-based methods have been shown to be a powerful strategy for mapping new loci and subsequent identification of new genes for all inherited forms of RP (8,21). In this study, we found a novel homozygous splicing mutation, c.1160+1G>A, in the AHR gene from an Indian arRP family using WES technology. Homozygosity analysis showed that this AHR mutation identified was in a homozygous region shared by these two patients (Supplementary Material, Figs S2 and S3).

AHR is a nuclear receptor that belongs to the basic-helix-loop-helix-Per-Arnt-Sim (bHLH/PAS) family of transcription factors and has been implicated in the regulation of developmental pathways and biological responses to xeno- and phytobiotics (16,17). AHR can bind to natural and synthetic ligands, polycyclic aromatic hydrocarbons such as those found in cigarette smoke, and polychlorinated dioxins such as TCDD (22). AHR can act on many types of cells and mediate a series of biological processes, including cell division, apoptosis, cell differentiation (23), oestrogen and androgen functions (24), immune system homeostasis (25) and reproduction (26,27). AHR has been reported to be expressed in the retina and in some fibroblast-like cells in the choroid (28), and the loss of AHR signalling increases the susceptibility of the retina to environmental stress, such as intense light (18), indicating critical roles for the AHR in regulating photoreceptors and RPE biological processes.

In our study, a novel homozygous splicing mutation, c.1160+1G>A, in the AHR gene was identified from an Indian arRP family. Further analysis by Sanger sequencing showed that the mother (V: 8) of the affected subject was an unaffected carrier with the heterozygous splicing mutation in the AHR gene. This homozygous splicing mutation co-segregated with the phenotype in this family and was very rare in several public databases. Therefore, our results indicated the role of this AHR mutation in RP pathogenesis.

As a member of the bHLH/PAS family of transcription factors, the AHR protein contains several domains critical for various in vivo functions (29–34) (Fig. 3B). The Q-rich domain is located in the C-terminal region of the protein and is involved in co-activator recruitment and transactivation (35). The human AHR gene G1661A polymorphism may affect AHR transactivation domain interactions, especially those with the TATA-binding protein, which influences AHR target gene expression and causes distinct outcomes in different individuals and tissues (36). The AHR rs2066853 gene polymorphism was associated with infertile oligoasthenoteratozoospermic men and seminal oxidative stress (37). In this report, the novel homozygous splicing mutation c.1160+1G>A in the AHR gene resulted in complete deletion of exon 9 and truncation of the C-terminal half of the AHR protein, which is essential for transcriptional activation (35). In our in vitro assays, the mutant AHR protein lost its transcriptional activation function and failed to activate downstream target genes (Fig. 3E–G).

Although there were different phenotypes reported in different Ahr knockout mouse lines, the common phenotypes of retinal degeneration in these models and our retina-specific knockout model validated the functional relevance of AHR in maintaining the integrity of the retina during stress conditions (18,19). These previous results emphasize the important role of AHR in maintaining photoreceptor and RPE homeostasis. In the present study, the homozygous splicing mutation c.1160+1G>A resulted in the deletion of exon 9 in AHR cDNA in RP patients, causing the RP phenotype. AHR is expressed broadly in human and mice including RPE and retina. Loss of AHR activity in our RP patients only resulted in retinal degeneration, not syndromic RP. One explanation is that AHR activity is essential for the highly metabolic active retina to remove toxic substrates generated during phototransduction.

Both Hu et al. (19) and Kim et al. (18) reported mild retinal abnormalities including subretinal deposits and choroid atrophy in Ahr KO mice. Our patients exhibited typical RP phenotypes: pigment deposits on the retina and reduced scotopic and photopic responses (Fig. 1). To assess the roles of Ahr in the neuronal retinal, we generated a neuronal specific knockout model by deleting Ahr using Six3-cre. Our conditional knockout model exhibited late-onset loss of photoreceptor cells and impaired ERG responses. Of course, there were some differences of our study and the two previous reported work. This discrepancy in phenotypes may reflect that the Ahr mouse models were independently generated. Differences in phenotyping time points and histological processing protocols. Further investigation is necessary to delineate the role of AHR in different cell populations in the eye.

The discrepancy between the phenotypes of patients AHR with mutation and that of knockout mice is not uncommon. One such example is the difference between patients harboring CRB1 mutations and rd8 mice phenotypes. Mutations in CRB1 led to LCA8, RP 12 or childhood- and juvenile-onset cone-rod dystrophy (38,39). Most patients exhibited severe vision loss. rd8 mice with a frameshift mutation in Crb1 only exhibited mild retinal degeneration (40). In addition, the severity of retinal pathology is dependent on genetic backgrounds (41,42).

Our study is the first report showing that an AHR mutation is linked to human RP. Taken together, our data demonstrated that AHR is a candidate gene for the pathogenesis of RP.

Materials and Methods

Patients

Samples from two patients and seven unaffected family members from one consanguineous Indian family and 1000 control individuals with no history of retinal degeneration were collected from the Aravind Eye Hospital of India. This research was carried out in accordance with the tenets of the Declaration of Helsinki and was approved by the Aravind Medical Research Foundation and the Sichuan Provincial People’s Hospital. Written informed consent was obtained from patients who participated in this study or from their legal guardians for minors.

DNA extraction and WES

All genomic DNA samples were extracted from peripheral blood leukocytes of the two RP patients, unaffected relatives and control individuals using a blood DNA extraction kit according to the protocol provided by the manufacturer (TianGen, Beijing, China). Exome sequencing was performed on the DNA sample of the index patient. The DNA Illumina TruSeq Exome Capture System (62 Mb) was used to collect the protein coding regions of human genomic DNA. The data covered 20 794 genes and 20 1121 exons in the Consensus Coding Sequence Region database (http://www.illumina.com/applications/sequencing/targetedresequencing.ilmn).

Data analysis and mutation validation

The high-quality sequencing reads were aligned to the human reference genome (NCBI build 37.1/hg19) with SOAPaligner (soap2.21). Based on the SOAP alignment results, SOAPsnp v1.05 was used to assemble the consensus sequence and call genotypes in target regions. Data were provided as lists of sequence variants (SNPs and short Indels). Variants were called using Atlas2 and noted by ANNOVAR. Variant frequency data were obtained from the dbSNP database, 1000 Genomes (http://browser.1000genomes.org/index.html), NHLBI Exome Sequencing Project (ESP) (http://evs.gs.washington.edu/EVS/, ESP6500), the ExAC Browser (Beta) (http://exac.broadinstitute.org/), gnomAD browser beta (http://gnomad.broadinstitute.org/). The following criteria were used for prioritization and determination of pathogenicity variants were: (1) Variants with total read depth >5× with the snp quality score > 50, (2) Minor allele frequency <0.5% in all the five variant databases for recessive genes, (3) Variants were SNVs (stoploss, stopgain, non-synonymous, non-frameshift_substitution, non-frameshift_insertion, non-frameshift_deletion, frameshift_insertion, frameshift_deletion) or splice site variants (splicing within 7-bp of a splicing junction), (4) Variants were consistent with the pattern of inheritance models (homozygous/compound heterozygous), (5) The gene is expressed in the human retina as determined by human retinal RNA-seq data (44), (6) Mutations were assessed for the potential deleteriousness as determined by 12 in silico prediction scores included in dbNSFP. Validation of the mutations in patients and their relatives was performed by Sanger sequencing on an ABI 3730XL Genetic Analyzer using the following primers: AHR-c.1160+1G>A-F, 5′-TGGCATGATAGTTTTCCGGC-3′; AHR-c.1160+1G>A-R, 5′-AGCTCTGCACA TGTAGCTCT-3′. Sequencing data were used to determine whether the remaining mutations co-segregated with the disease in these families and whether the splicing mutation was absent in the 1000 ethnicity-matched controls.

Cell line and cell culture

HEK293T, D407 and COS7 cells were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal calf serum, 100 mg/ml penicillin and 100 mg/ml streptomycin and cultured at 37°C in a 5% CO₂ incubator. TCDD (2, 3, 7, 8-tetrachlorodibenzo-p-dioxin) was dissolved in DMSO and administered at 10 nm for 18 h. Control plates received the same amount of DMSO. Final DMSO concentrations did not exceed 0.1% (v/v).

RNA extraction and PCR

Total RNA from the leukocytes of the two patients (RD-ICP-78, RD-ICP-79) and the transfected cells was extracted using TRIzol (Invitrogen, Austin, TX, USA) according to the manufacturer’s protocol. cDNA was synthesized as described previously. AhR splicing products from the two patients were amplified by an Eppendorf Mastercycler personal PCR system (Eppendorf, Germany) and were evaluated by 1% agarose gel electrophoresis. Equal amounts of cDNA from cells were submitted to PCR using the SYBR PCR Master Mix kit (Applied Biosystems, Foster City, CA) and the 7500 Fast real-time PCR detection machine (Applied Biosystems, Foster City, CA). Specific primers were as follows: AHR-splicing-F, 5′-TGGTTGTGATGCCAAAGGAAG-3′, AHR-splicing-R, 5′-CCTTGCTTAGAGTGGATGTGG-3′; CYP1A1-F, 5′-ACATGC TGACCCTGGGAAAG-3′, CYP1A1-R, 5′-GGTGTGGAGCCAATTCGGAT-3′, and CYP1A2-F, 5′-ATGCTCAGCCTCGTGAAGAAC-3′, CYP1A2-R, 5′-GTTAGGC AGGTAGCGAAGGAT-3′. The housekeeping gene GAPDH was used as an internal control.

Plasmids

The full-length AHR wild-type (WT) expression plasmid pCMV6-AHR-Myc/DDK was purchased from OriGene Technologies, Inc. (OriGene, Rockville, Maryland, USA). The AHR exon 9 deletion plasmid pCMV6-AHR-Mut-Myc/DDK was prepared using the QuikChange XL site-directed mutagenesis kit (Stratagene, La Jolla, CA, USA) with AHR mutant primers: AHR-Mut-F, 5′-TATTGTGCCGAGTCC CATATCCGAAAGATGAGGAAGGAACAGAGCATTTA-3′, AHR-Mut-R, 5′-TAAATGCTCTGTTCCTTCCTCATCTTTCGGATATGGGACTCGGCACAATA-3′. The pGL3-XRE-luciferase plasmid and pRL-TK plasmid were purchased from Life Technologies.

Transfection and luciferase activity

The pGL3-XRE-luciferase plasmid with the pCMV6-AHR-WT or pCMV6-AHR-MUT plasmid was co-transfected with the pRL-TK plasmid into subconfluent (80–90%) monolayer cells using Lipofectamine 2000 (Invitrogen Co.). Luciferase activity was detected with the Dual Luciferase Reporter Assay System (Promega, Madison, WI, USA) according to the manufacturer’s instructions. Transfection efficiency was determined as a percentage of control GFP-expressing cells.

Ahr knockout mouse model

All animal studies adhered to the ARVO statement for the use of animals in ophthalmic and vision research and were approved by the Animal Care and Use Committee of the Sichuan Provincial People’s Hospital.

We used Ahr^loxp/loxp mice (43) (Ahr^tm3.1Bra, http://jaxmice.jax.org/strain/006203.html) to generate tissue-specific deletion of exon 2 of Ahr on a C57BL/6J genetic background. Briefly, Ahr^loxp/loxp mice were crossed with Six-Cre mice (20) to generate the retina-specific Ahr^loxp/loxp Six3-Cre (named Ahr KO) knockout mice. Age-matched WT Ahr^loxp/loxp mice (Jackson Laboratory, Bar Harbor, Maine, USA) or heterozygous Ahr^loxp/+, Six3-Cre mice were used as controls. All WT and Ahr mutant mice were negative for the rd8 mutant allele as determined by Sanger sequencing.

Electroretinogram (ERG)

ERG was performed using an Espion Visual Electrophysiology System (Diagnosys, Lowell, MA, USA). Control and mutant mice were dark-adapted overnight. Animals were anaesthetized with a combination of ketamine (16 mg/kg body weight) and xylazine (80 mg/kg body weight) in normal saline. Eyes of the anaesthetized mice were dilated with a drop of tropicamide and phenylephrine, and tetracaine (0.5%) was applied before ERG. Body temperature was maintained at 37°C with a heating platform. Dark-adapted ERGs were recorded using gold wire loops with responses to flashes with intensities ranging from 0.001 to 10 cd·s/m². Cone-mediated ERGs were recorded with white flashes after 20 min of complete light adaptation.

Retinal histology and measurement of the outer nuclear layer

For haematoxylin and eosin (H&E) staining, eyes from WT and KO mice were enucleated, marked on the nasal side for orientation, fixed overnight in 1.22% glutaraldehyde and 0.8% paraformaldehyde in 0.08 m phosphate buffer, embedded in paraffin, and cut into 5 µm sections. Sections were collected at five locations near the optic nerve and stained with H&E. H&E-stained sections were used to count the rows of photoreceptors in the outer nuclear layer. Three measurements of the outer nuclear layer were taken every 200 μm from the optic nerve and averaged. The optic nerve was designated 0 µm.

Immunohistochemistry

For immunohistochemistry, enucleated eyes were removed, marked on the nasal side for orientation, fixed for 1 h in 4% paraformaldehyde in 100 mm phosphate buffer (pH 7.4) and then cryoprotected in 30% sucrose. Tissues were embedded in optimal cutting temperature (OCT) solution and frozen on dry ice for sectioning. Sections were blocked and permeabilized with 10% normal goat serum and 0.2% Triton X-100 in phosphate buffer for 30 min. Labelling with various antibodies was performed as previously described (44).

Websites

GenBank, http://www.ncbi.nlm.nih.gov/genbank/ 1000 Genomes, http://www.internationalgenome.org/ ExAC Browser, http://exac.broadinstitute.org/ GeneMatcher, https://genematcher.org/ gnomAD Browser, http://gnomad.broadinstitute.org/ HGMD, http://www.hgmd.cf.ac.uk/ Mouse Genome Informatics, http://www.informatics.jax.org/ NHLBI Exome Sequencing Project Exome Variant Server, http://evs.gs.washington.edu/EVS/ OMIM, http://www.omim.org/ Optimized CRISPR Design, http://crispr.mit.edu/ RetNet - Retinal Information Network, https://sph.uth.edu/retnet/ Protein Databank, http://www.rcsb.org/pdb/

Supplementary Material

Supplementary Material is available at HMG online.

Acknowledgements

The authors wish to thank all patients who participated in this study. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript. The authors alone are responsible for the content and the writing of the paper.

Conflict of Interest statement.None declared.

Funding

This research was supported by grants from the National Key Scientific Research Programme (2016YFC0905200 [Y.Z.], 2015CB554100 [X.J.Z.])), the Natural Science Foundation of China (81790643, 81430008 [Y.Z.], 81770950, 81470668, 21561142003 [X.J.Z.], 81400437 [Y.Z.]), the Department of Science and Technology of Sichuan Province (2016JZ0008 [Y.Z.], 2016TD0009, 2014JQ0023, 2017TJPT0010 [X.J.Z.], 2014SZ0169, 2015SZ0052 [Y.Z.]), and the Department of Sichuan Provincial Health [130176 (Y.Z.)].

Author notes

Yu Zhou and Shujin Li authors contributed equally to this work.