GUCY2D Gene Loss-of-Function Mutations Responsible for Leber Congenital Amaurosis 1

Background: Leber congenital amaurosis (LCA) is a group of severe congenital neurodegenerative diseases. Variants in the guanylate cyclase 2D gene ( GUCY2D ), which encodes guanylate cyclase 1 (ROS-GC1), are associated with LCA1 and account for 6%– 21% of all LCA cases. Methods: In this study, one family with LCA1 was recruited from China. A combination of next generation sequencing and Sanger sequencing was used to screen for disease-causing mutations. Additionally, immunohistochemistry and HPLC-coupled tandem mass spectrometry (HPLC-MS/MS) were used to confirm the cellular location and catalytic activity of ROS-GC1 mutants, respectively. Results: We found three novel mutations (c.139_139delC, c.835G>A and c.2783G>A) in the GUCY2D gene. Mutation c.139_139delC results in a truncated protein. Mutations c.835G>A and c.2783G>A exert no effects on cellular location but reduce significantly the catalytic activity of ROS-GC1. Conclusions: Our findings highlight the clinical range of LCA. Moreover, HPLC-MS/MS was used to analyze the concentration of 3',5'-cyclic guanosine monophosphate (cGMP), suggesting that HPLC-MS/MS is an effective alternative method to evaluate the catalytic activity of wild-type and mutant ROS-GC1 c.835G>A (Asp279Asn) and c.2783G>A (Gly928Glu). This study probes the pathogenesis of LCA1 and addresses the effects of mutations on ROS-GC1 activity. The three novel mutations may be


Background
Leber congenital amaurosis (LCA), accounts for at least 5% of all retinal dystrophies, and is the earliest and most severe form of all inherited retinal dystrophies [1]. LCA is generally inherited in an autosomal recessive manner [2][3][4] and characterized by genetic and phenotypic heterogeneity. Currently, mutations in 20 genes are associated with LCA [5]. The distribution of pathogenic genes vary considerably among different populations; 4 however, guanylate cyclase 2D (GUCY2D) is one of the most prevalent LCA gene since GUCY2D mutations account for 6%-12% of LCA cases (LCA1) [4][5].
Guanylate cyclase 1 (ROS-GC1), encoded by GUCY2D, is expressed in rod and cone cells of vertebrae retina and catalyzes the synthesis of 3',5'-cyclic guanosine monophosphate (cGMP). ROS-GC1 is negatively controlled by a Ca 2+ feedback loop because a reduction in Ca 2+ concentration causes an increase in ROS-GC1 activity. However, the effects of the Ca 2+ concentration on ROS-GC1 activity are indirect because this feedback loop is regulated by small Ca 2+ -binding guanylate cyclase-activating proteins (GCAPs), which interact with overlapping binding sites within the juxtamembrane domain (JMD), kinase homology domain (KHD) and dimerization domain (DD) of ROS-GC1. In addition to these latter domains, ROS-GC1 also possesses a leader sequence (LS), an extracellular domain (ECD), a transmembrane domain (TM) and a cyclase catalytic domain (CCD) [6]. Currently, 127 GUCY2D point mutations in different domains of ROS-GC1 cause protein dysfunction and subsequently result in LCA1 [5]. Mutation Pro575Leu located at the JMD domain lacks sensitivity to GCAP1 regulation and induces dysregulation of the Ca 2+ -sensitive cyclase activation profile. Patients carrying the Pro575Leu mutation suffer from LCA1 by distortion of Ca 2+ -cGMP homeostasis, and these patients require longer periods in the dark to recover [7].
In the present study, targeted-next generation sequencing (NGS) was performed to screen 222 genes that are responsible for 24 kinds of ophthalmic genetic diseases in the proband. After confirming the results by Sanger sequencing, the identity of three novel variants of the GUCY2D gene were found, c.139_139delC (Ala49Profs*36), c.835G>A (Asp279Asn) and c.2783G>A (Gly928Glu). This study probes the pathogenesis of LCA1 and addresses the effects of mutations on ROS-GC1 activity. The three novel mutations may be suitable for LCA1 screening, and HPLC-MS/MS represents a convenient and effective method for cGMP quantification.

LCA patients
A family with a 5-year-old proband (III-2) and two suspected patients (III-1 and III-3) were recruited for this study. Written informed consent was obtained from each individual to participate in this study. The proband was diagnosed with LCA at Guangdong Zhongshan Hospital, Shanghai General Hospital and the Second Affiliated Hospital of the School of Medicine, Zhejiang University. The pedigree was constructed for the proband based on information provided by the guardians.

DNA isolation and qualification
Total genomic DNA was extracted using the Relax Gene Blood DNA System (Tiangen, Beijing, China) following the manufacturer's instructions. All DNA was dissolved in sterilized double-distilled water and kept at -20 °C until assayed.
One percent agarose gels were used to monitor DNA degradation and contamination. All DNA samples were examined for protein contamination (as indicated by the A 260 /A 280 ratio) and reagent contamination (indicated by the A 260 /A 230 ratio) with a NanoDrop ND 1000 spectrophotometer (NanoDrop, Wilmington, DE, USA).

Targeted-Next generation sequencing (NGS)
DNA samples obtained from the proband were sequenced using targeted-NGS. A customized Sequence Capture 2.1M Human Array from Roche NimbleGen (NimbleGen, Madison, USA) was designed to capture 3093 exons (including 100 bp regions that flanked 6 the exons) from 222 genes known to be associated with common genetic diseases, including retinitis pigmentosa, Waardenburg syndrome, X-linked juvenile retinoschisis, crystalline retinitis pigmentosa, albinism, LCA, Bardet Biedl syndrome and cone-rod dystrophy (Additional Table 1). The procedure for the preparation of the libraries was consistent with standard protocols published previously [8]. The data of targeted-NGS assay was analyzed by Illumina basecalling Software 1.7.

Mutation validation by Sanger sequencing
Sanger sequencing was used to validate candidate variants identified by NGS. Primers of the GUCY2D gene (NG_009092.1) used in Sanger sequencing were designed by Primer-BLAST (http://www.ncbi.nlm.nih.gov/tools/ primer-blast/) and synthesized by Sangon Biotech (Shanghai, China) ( Table 1). All amplifications were examined by electrophoresis using 2% agarose gels and sequenced by BioSune Biotechnology Co., Ltd. (Shanghai, China). The sequencing results were further compared and analyzed by Mutation Surveyor [9].

Construction of wild-type (wt) and mutant human ROS-GC1 recombinant plasmids
The cDNA of human ROS-GC1 was obtained from Gene Copoeia (EX-Z0715-M98). Primers F and R containing XhoI and AgeI restriction sites were used to amplify ROS-GC1 ( Table 2).
Site-directed mutagenesis PCR was used to construct the ROS-GC1 mutants. Primers used in site-directed mutagenesis PCR are shown in Table 2. Each mutant was achieved by twostep PCRs using pEGFP-GC1 as the template. For c.835G>A (Asp279Asn), two primer pairs 7 F and r-835 and R-BamHI and f-835, were used in the first PCR step. Primers F and R-BamHI were used in the second step. For c.2783G>A (Gly928Glu), primers F-BamHI and r-2783 and R and f-2783 were used in the first PCR step. Primers F-BamHI and R were used in the second step. For each mutation, amplification products in the first step were cleaned (Axygen, CA, USA), mixed and used as the template in the second PCR reaction.
All final PCR amplifications were ligated into the digested pEGFP-N1 vector using ClonExpress.
Recombinant plasmids pEGFP-GC1, pEGFP-Asp279Asn and pEGFP-Gly928Glu were transformed into Escherichia coli DH5a cells. DNA was prepared by using a plasmid DNA purification kit from Macherey-Nagel following the manufacturer's instructions. Sanger sequencing was used to verify sequences.

Western blotting
Proteins from transfected HeLa cells were electrophoresed on a 12% sodium dodecyl

Validation of cGMP quantitation by HPLC-MS/MS
HeLa cells were transfected with pEGFP-N1, pEGFP-GC1, pEGFP-Asp279Asn and pEGFP-Gly928Glu. After 36 h, cells were collected from 100 mm plates and washed three times with PBS. The supernatant was removed carefully and 300 mL of ice-cold extraction medium (acetonitrile/methanol/water, 2/2/1 v/v/v) was added to each tube. Twenty-five ng/mL Tenofovir (TNF) was added as the internal standard. After dissolving, the sample was frozen immediately in liquid nitrogen for 30 s to terminate cGMP metabolism, and this was followed by incubation in a 37 °C water bath for 60 s. After repeating 6 times, samples were heated at 98 °C for 20 min. Samples were cooled on ice and centrifuged at 20,000 × g at 4 °C for 10 min. The supernatant was transferred into a new tube and the non-dissolved residue was extracted two more times with 400-μL extraction medium. After evaporation, residues were collected and dissolved in water for further analysis.
The cGMP concentrations were analyzed via HPLC-MS/MS. The samples were applied to an HPLC utilizing an ACQUITY CSH-C18 column (1.7 μm, 2.1 × 100 mm column, Waters, Ireland). The binary pump system supplied two eluents for chromatographic analysis, eluent A (10 mM formic acid) and eluent B (acetonitrile). The flow rate was 0.3 mL/min.
Analyte detection was conducted on a sensitive triple quadrupole mass spectrometer (Waters TQ-XS, USA). Nitrogen was used as the collision gas. One-hundred ng/mL cGMP 9 (G7504, Sigma, Germany) was used as the standard. All processes were referenced to the method described previously [11]. cGMP concentrations presented as means ± SEM are based on three independent experiments. P-values were calculated by means of the independent sample T-test.

Clinical description of the proband
The proband (III-2) was a five-year-old girl who showed the disorder visually but could perceive light in a dark room. The proband displayed the classic signs of oculo-digital symptoms. Ophthalmological examination revealed pendular nystagmus, roving eye movements, macular "colobomas" and optic disc abnormalities. Additionally, dim retina and salt-and-pepper pigmentation were found in the fundus. The results of flash visual evoked potential (VEP) showed that the latent period of binocular P-waves was severely prolonged and the amplitudes were severely reduced. The electroretinogram (ERG) results showed that the latent period and the amplitude of binocular scotopia rod response bwave, mixed response a-wave and mixed response b-wave were flat. With an attenuated photopic 30-Hz flicker, the full-field ERG (ffERG) showed a worsening delay of the cone-dependent response. The latent period and amplitude of the binocular photopic cone response a-wave and b-wave were flat. Clinical examination of the parents confirmed that neither was affected.

GUCY2D mutation analysis
In order to discover the pathogenesis of the proband, targeted-NGS was performed using  (Fig. 1B1,2). III-3 was described as an LCA1 patient (Fig. 1A). Sanger sequencing was used to determine the GUCY2D gene sequences of the suspected sibling of the patient (III-4) and parents (II-7 and II-8). The results showed that II-7 is a c.2783G>A mutation carrier, and II-8 possesses another novel mutation c.835G>A (Fig. 1B3). The sibling (III-4) possessed no GUCY2D mutation. In addition, all heterozygous mutation carriers showed no clinical symptoms.
The high-resolution 3D structure of full-length ROS-GC1 remains unsolved. The theoretical model (PDB ID: 1AWL), which contains 158 amino acids from 871 to 1028 was used to study the impact of c.2783G>A (Gly928Glu) on the 3D structure of ROS-GC1. Although it is theoretical and not validated, 3D modeling analysis showed that Glu928 directly interacts with GTP (Fig. 2) and this change may reduce the size of the catalytic core, leading to a change in the relative position of GTP and the catalytic core. Furthermore, this mutation likely hampers ligand binding and reduces the catalytic activity of ROS-GC1.

Localization of wt and mutant ROS-GC1
ROS-GC1 is a member of the membrane guanylyl cyclase family, the correct localization of ROS-GC1 is critical for the synthesis of cGMP. In this study, immunofluorescence was used to confirm the localization of wt and mutant ROS-GC1. pEGFP-N1, pEGFP-GC1, pEGFP-Asp279Asn and pEGFP-Gly928Glu were transfected into HeLa cells after verification by Sanger sequencing (Additional Fig. 2). In Hela cells, ROS-GC1 was found to be present in cell membranes and cytoplasm (endoplasmic reticulum within cytoplasm [7]). The locations of Asp279Asn and Gly928Glu were the same as those found in the wt ROS-GC1. (Fig. 3, top panels). The anti-Na + /K + -ATPase antibody was used as a specific marker of the plasma membrane (Fig. 3, second panels).

Catalytic features of wt and mutant ROS-GC1
HPLC-MS/MS was used to analyze cGMP concentrations in HeLa cells. The cGMP concentration in cells transfected with pEGFP-N1 was undetectable. In contrast, the cGMP concentration in cells transfected with pEGFP-GC1 was about 3.631 pmol/mL. Compared with wt, cells transfected with pEGFP-Asp279Asn or pEGFP-Gly928Glu showed significantly lower concentrations of cGMP (Fig. 4A), implying that Asp279Asn and Gly928Glu significantly reduced the catalytic activity of ROS-GC1. In addition, Gly928Glu disrupted the catalytic activity of ROS-GC1 more severely than the other mutants, which is consistent with the bioinformatics analysis. Moreover, all missense mutations did not affect the expression level of ROS-GC1 (Fig. 4B).

Discussion
Twenty known genes associated with LCA have been identified. The large number of genes and exons and the lack of mutational hot spots make individual screening for diseasecausing mutations by Sanger sequencing difficult and expensive. However, NGS technologies provide more convenient and effective strategies to screen for pathogenic mutations in genes [12][13][14]. In this study, a family with one LCA proband was analyzed.
Targeted-NGS and Sanger sequencing were combined to identify three mutations (c.139_139delC, c.2783G>A and c.835G>A) in the GUCY2D gene in this pedigree, and all mutations have yet to be documented by the 1000 Genomes Project. According to the description, there are two suspected patients in this pedigree, III-1 and III-3. However, both of them died at a very young age, thereby we only verified the GUCY2D gene mutation in II-7 and II-8. Additionally, in this pedigree all the heterozygous mutation carriers showed no clinical symptoms.
The novel disease-causing mutation c.139_139delC (Ala49Profs*36) carried by II-1 and putatively by II-2 and III-3 generates a truncated protein (83 amino acids), containing only a part of the LS (Fig. 2C). As all domains of ROS-GC1 are presumably absent, there should be no enzymatic activity for this ROS-GC1 mutant and thus no cGMP production. Thus, the 13 absence of sufficient levels of cGMP results in photoreceptor cell polarization, which finally leads to vision loss.
Multiple sequence alignments indicated that aspartic acid and glycine at positions 279 and 928 are highly conserved across species, indicating that these residues are important for protein function. This is consistent with previous observations that mutations located in the ECD domain decrease cGMP production and result in LCA1 [15]. In this study, the endogenous concentration of cGMP in HeLa cells, which were transfected with pEGFP-N1, is too low to be measured. This phenomenon can be observed in other studies of GUCY2D mutations, such as HEK293 cells expressed mutant ROS-GC1 (A710V and P873R) showed undetectable cGMP concentrations, indicating that endogenous cGMP concentration exerted no contribution to cGMP levels. Although GUCY2D gene encodes a 14 retina-specific protein, use other cell lines to investigate the localization and enzymatic activity of ROS-GC1 is feasible, for example, HEK293 cells were widely used to study the cellular localization of ROS-GC1 [7,10]. Additionally, HEK293, COS-7 and HeLa cells had been applied in the ROS-GC1 activity assays [17,18].

Conclusions
In this study, we found three novel mutations (c.

Consent to publish
not applicable.

Availability of data and materials
The data used and analyzed in this study are available from the corresponding author.

Competing of interests
The authors report no conflict of interests.

Funding
The study was supported by grants from the National Natural Science Foundation of China

Supplementary Files
This is a list of supplementary files associated with the primary manuscript. Click to download.