Phenotypic variability observed in a Chinese patient cohort with biallelic variants in the CLN genes

Subjects

This retrospective study was approved by the Beijing Tongren Hospital Joint Committee on Clinical Investigation, and all procedures followed the tenets of the Declaration of Helsinki. Written informed consent was obtained from all patients or their guardians. A total of 14 patients (11 males and three females) from 13 unrelated families were recruited from the Genetics Laboratory of the Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, from 2012 to 2022. The inclusion criteria were that patients were clinically diagnosed with RD and harbored biallelic pathogenic or likely pathogenic variants in one of the CLN genes. The diagnosis of NCLs was established when patients suffered any other NCL-related neurologic manifestation.

All patients underwent thorough ocular examinations, including best-corrected visual acuity (BCVA), slit-lamp biomicroscopy, and fundus examinations. Most patients were subjected to multimodal imaging evaluations, including color fundus (CF) photography, infrared imaging (IRI), retinal spectral domain optical coherence tomography (OCT, Heidelberg, Germany), and short-wavelength fundus autofluorescence (FAF, Heidelberg, Germany). Ten patients underwent full-field electroretinography (ERG), consistent with the International Society for Clinical Electrophysiology of Vision standards.

For all patients, we documented the general medical records and systemic examination results, which involved motor function, language, intellect, and epilepsy. Some patients underwent electroencephalography (EEG) examinations and brain magnetic resonance imaging (MRI) scans. In this cohort, eight patients underwent an ophthalmic evaluation more than once, and all patients but one (067,020) participated in telephone surveys about their neurologic symptoms or signs.

Targeted exome sequencing (TES) and bioinformatics analysis

After informed consent was obtained, peripheral blood samples were collected from all participants and their available family members for genetic analysis. Genomic DNA was then extracted using a genomic DNA extraction kit (Vigorous, Beijing, China) according to the manufacturer’s protocol. We performed TES in all the probands using a capture panel comprising 533 known inherited retinal disease (IRD) genes (Supplementary Table S1, including the CLN1-8, 10, 11, and 13 genes). The Illumina library preparation and the capture experiment were conducted as previously described [16]. The HGMD database and ClinVar database were used to search for any reported pathogenic variants.

The pathogenicity of the variants was predicted by three in silico programs: Mutation Taster, SIFT, and PolyPhen-2. Three algorithms, NetGene2 Server (version 2.42), NNSplice (version 0.9), and Human Splicing Finder (HSF, version 2.05), were used to analyze any variant influences on splicing. The copy number variation (CNV) kit software (https://github.com/etal/cnvkit) was utilized to detect CNVs from variations in the read depth, and real-time, quantitative polymerase chain reaction (q-PCR) was subsequently performed to confirm the existence of the presumed CNVs. Segregation analysis was performed by Sanger sequencing in all the pedigrees.

RNA extraction and analysis

A novel variant c.963–13A>G, which was identified in patient 0,191,609 and her mother and located at a non-canonical splice site (NCSS), was strongly suspected of causing aberrant splicing (Figure 1A). Fresh peripheral blood samples were obtained from patient 0,191,609 and her parents and utilized to isolate peripheral blood lymphocytes using a peripheral blood lymphocyte isolation kit (Suzhou Melian Biotechnology Co., Ltd.). Total RNA was extracted from the lymphocytes using the RNAprep Pure Cell/Bacteria Kit (Tiangong, Beijing, China). Reverse transcription PCR (RT–PCR) was performed with a pair of specific primers (forward 5′-agcccctcataagaaccgag-3′ and reverse 5′-agacgaggtagatgcttggc-3′) using the Fasting One-Step RT–PCR Kit (Tiangong Biotech, China). The products were separated by electrophoresis on 2% agarose gels, excised, and sequenced. Homology modeling of a three-dimensional (3D) structure of the CLN3 protein was generated using Swiss-Model server (https://swissmodel.expasy.org/interactive). PyMOL software was then selected to analyze the spatial conformation and stability of the predicted wild-type (WT) and mutant proteins and the hydrogen bond formation between amino acid residues.

NCL genes and variants

We detected 21 distinct variants of three CLN genes (PPT1, CLN3, and MFSD8) in the 13 probands. The three NCL genes were identified in the following incidences: CLN3 in six probands, MFSD8 in four probands, and PPT1 in three probands. Among the 21 variants, 13 were identified in the current study. All of the novel variants were either not recorded in any public database or our in-house database or were present at a very low frequency (between 0.003% and 0.01%). All were described as pathogenic or likely pathogenic, based on the American College of Medical Genetics and Genomics (ACMG) guidelines and standards; one exception was a missense variant c.280A>C (p.T94P) in PPT1, which was defined as “uncertain of significance” (Table 1). This variant, which is absent from all public databases, was detected in trans with a pathogenic variant c.490C>T (p.R164X), and the retinal phenotype of proband (067,020) was similar to that of patients with PPT1 variants. Of the 21 variants, only p.(E336K) in MFSD8 was detected three times; the remaining 20 variants were identified only once (17/21) or twice (3/21).

Transcript analysis

The novel variant c.963–13A>G in CLN3 was predicted to cause an abnormal splicing of exon 14 by NetGene2, NSSP, and HSF software. We then conducted further cDNA analysis to investigate the influence of this variant on CLN3 pre-mRNA splicing. The RT–PCR results from the lymphocytes demonstrated that this novel NCSS variant caused a 12-nt retention of intron 13 (Figure 1B, C), resulting in an in-frame indel p.(W321delinsCPNLR) by triggering a cryptic acceptor splice site. The 3D structural modeling demonstrated that the mutant changed one segment of an α-helical secondary structure to a loop structure (Figure 1D). The bioinformatic analysis predicted that the mutant altered the lengths and angles of several hydrogen bonds between the related residues (Figure 1D).

Clinical findings

In the current cohort, six probands, four probands, and four patients from three unrelated pedigrees harbored compound heterozygous variants in CLN3, MFSD8, and PPT1, respectively. All the variants were verified by familial segregation analyses (Supplement Figure 1). Of the 14 patients, eight were diagnosed with NCLs and six were diagnosed with isolated RD. All the patients experienced night blindness or different extents of vision defect and were initially diagnosed with retinitis pigmentosa (RP, seven patients) or cone-rod dystrophy (CORD, seven patients; Table 2).

Of the six probands carrying biallelic CLN3 disease-causing variants, five were diagnosed with NCL and one was diagnosed with isolated RP. The five patients with NCL presented typical clinical features of CLN3 disease, and their mean age at last examination was 9.8 years (range: 7.0–15.3 years). These patients suffered a rapid visual defect, mostly combined with nyctalopia, followed by a neurologic manifestation, which included intellect and language function decline, motor ability defect, and epileptic seizure. The mean onset age for visual loss was 6.4 years (range: 6–8 years), while the mean onset age for neurologic symptoms was 8.8 years (range: 7–12 years). In contrast, patient 019,309 with isolated RP, whose last exam age was 31.4 years, had a late onset age (16 years) for his ocular symptoms. He had not developed any neurologic symptoms at his latest telephone follow-up at the age of 42.

The four probands with biallelic MFSD8 variants were clinically diagnosed with CORD, and their mean onset age for ocular symptoms was 8.1 years (range: 5–13.5 years). In addition to the ocular symptoms, proband 010,421 simultaneously suffered mental retardation and epileptic seizure at the age of five, followed by the development of speech impairment at age eight and then paralysis and bedridden in what resembled a vegetative state at the age of 12 at his most recent telephone follow-up. His brain MRI at age 12 revealed obvious cerebellar atrophy.

Of the four patients carrying biallelic PPT1 variants, two had a neurologic manifestation and two had only RD at their last examination (3.5 years and 15.8 years, respectively). All four patients suffered visual defect with night blindness, and their mean onset age was 8.9 years (range: 0.5–15 years). None of the four patients presented any typical clinical features of CLN1 disease. Proband 010,707 (11.5-year-old male) had suffered only mild mental retardation at the age of eight, while proband 019,953 (25.2-year-old female) had developed hypomnesia, mild dysarthria, emotional instability, and motor ability deterioration at the age of 24. Her brain MRI displayed whole-brain atrophy at her last examination. Her younger brother 019,953-3 (15.8 years old) only presented with rapid visual defects with night blindness.

Ophthalmic features

The 14 patients presented profound visual loss with a mean logMAR BCVA of 1.4 (range: 0.4–2.7), and eight of them experienced a rapid visual decline over 3–6 months. Nystagmus was observed in one patient (010,421). Among the 10 patients whose ERG recordings were available, four displayed severe dysfunction in rod and cone responses, three showed extinguished rod responses and severe dysfunction in cone responses, and three exhibited an extinguished recording (Table 2).

Retinal features

Fundus examinations revealed that all 14 patients exhibited macular alteration ranging from bull’s eye maculopathy to total macular atrophy, arteriolar attenuation, and tapetoretinal degeneration (TD) in the mid-peripheral retina (Figure 2, Figure 3, and Figure 4). A distinctive radial macular striation was observed in 12 of 14 patients (86%; Figure 2; Figure 3A–B, D; and Figure 4A, C). The two exceptions were patient 010,777 (Figure 3C) and patient 019,953-3 (Figure 4B), who carried variants in MFSD8 and PPT1, respectively. The linear striations were easily observed by IR imaging (Figure 2A,B). In addition, optic disc pallor was detected in 11 patients (Figure 2, Figure 3, and Figure 4). FAF examination showed an enlarged area of hypo-autofluorescence (AF) in the foveal region surrounded by an area of hyper-AF in the posterior pole (Figure 2A,B; Figure 3C,D; and Figure 4). Several cases displayed patchy hypo-AF in the midperiphery and along the arcades (Figure 4A,B). Retinal OCT scans were available for analysis in 10 patients. OCT images of all the cases showed macular atrophy with almost total loss of the ellipsoid zone (EZ) in the macula and paramacular regions (Figure 2A–D; Figure 3B–D; Figure 4A,C). The one exception was patient 019,953-3, who showed bilateral macular edema (Figure 4B). Numerous hyperreflective substances were observed at the level of the retinal pigment epithelium (RPE) in all eyes. With the exception of patient 019,953-3 and patient 010,777, all the cases showed fine rippling at the level of the internal limiting membrane, which correspond to the macular striation observed by CF and IR.

Longitudinal observation of eight follow-up patients

We followed eight patients for their ocular longitudinal changes for a mean follow-up time of 42 months (range: 3–96 months). During follow-up, seven patients presented a severely decreased BCVA, as their mean logMAR BCVA at the first exam time was 1.32 (range: 0.4–2.7), while the mean logMAR BCVA at the last exam time was 2.1 (range: 1.3–3.0). The one remaining patient 0,191,730 showed a stable BCVA but had only a three-month follow-up. Our fundus evaluations revealed different extents of progression of retinal degeneration in all the patients, with features that included enlargement of macular atrophy, expanded retinal and RPE atrophy, and the occurrence of optic pallor and vascular attenuation with aging (Figure 2, Figure 3, and Figure 4). The OCT scans disclosed obviously increasing hyperreflective dots at the RPE level even during a short three-month follow-up (Figure 2A). In addition, the normal retinal lamination vanished, and the fine macular striation decreased with aging (Figure 2D and Figure 4A).