The protocol for this study was reviewed and approved by the Ethics Committee of The Affiliated Hospital of Qingdao University before its performance. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. We obtained written informed consent from the patient for publication of identifying information/images in an online, open-access publication.

The patient was diagnosed with FH at The Affiliated Hospital of Qingdao University. We used the Dutch Lipid Clinic Network diagnostic criteria for the diagnosis of FH (Catapano et al., 2016). Briefly, the Dutch Lipid Clinic Network criteria evaluate the possibility for the diagnosis of FH based on scores for the following four aspects: (1) family history of premature vascular disease, LDL-C level above the 95th percentile, presentation of tendinous xanthoma and/or arcus cornealis in a first-degree relative, or LDL-C level above the 95th percentile in a child < 18 years of age; (2) clinical presentation of the patient with premature vascular disease or LDL-C level above the 95th percentile; (3) physical examination of the patient showing tendinous xanthomata and/or arcus cornealis; (4) degree of LDL-C elevation of at least 4.0 mmol/L; and (5) DNA analyses showing functional genomic variants in the LDLR, APOB, or PCSK9 gene. The diagnosis of FH is considered definite if the total score is over 8.

We obtained peripheral venous blood samples from the proband and her first-degree relatives in a fasting condition for the measurement of TC, LDL-C, triglycerides, and high-density lipoprotein cholesterol. Measurements of the lipid parameters mentioned earlier were performed with a chemiluminescence method with an automated biochemistry analyzer (Beckman AU 4500, United States) in The Affiliated Hospital of Qingdao University.

Genomic DNA of the proband, as well as the family members, was extracted from the peripheral venous blood sample with the phenol–chloroform centrifugation method as previously described (Schiebelhut et al., 2017). After treatment with ethylenediaminetetraacetic acid, genomic DNA samples of the proband were sent for customized target exome capture-based next-generation sequencing of the LDLR, PCSK9, and APOB genes at the Beijing Genomic Institution. The samples were sequenced simultaneously and analyzed for genomic variants of LDLR on the Illumina Platform, and a single-base variant within exon 16 was detected. Then, polymerase chain reaction (PCR) was performed targeting the mutated exome (exon 16) for the proband, her father, and paternal aunt and uncle (Table 1). Briefly, the primers were designed with Prime 5 as 5′-CCTTCCTTTAGACCTGGGCCT-3′ and 5′-CATAGCGGGAGGCTGTGACC-3′. We used 100 ng of genomic DNA as the template, which was mixed to obtain a total 50-μl reaction system using the HotStarTaq Plus DNA Polymerase Kit (Qiagen, Germany). The reactive conditions were as follows: Taq polymerase activation at 94±C for 5 min, 35 cycles of denaturing at 94±C for 30 s, annealing at 61–64±C for 30 s, extension at 72±C for 45 s, and final extension at 72±C for 5 min. We used agarose gel to separate and confirm the amplified products by electrophoresis. The PCR products were further purified using the DNA fragment purification kit (TaKaRa, Japan). Sequencing of the products was performed at the Beijing Genomic Institution with the Sanger sequencing method.

Total RNA from the peripheral venous blood samples obtained from the proband and an age- and sex-matched health control was extracted using the Trizol method as previously described (Kim et al., 2017). Then, reverse transcription PCR was performed to evaluate the transcription of exon 16 with the primers designed by Primer 5 software, forward: 5′-TGAA CTGGTGTGAGAGGACC-3′, and reverse 5′-CTGGTTGTGG CAAATGTGGA-3′. After electrophoresis with agarose gel, the targeted bands were confirmed and separated, and the purified PCR products were sequenced with the Sanger sequencing method at the Beijing Genomic Institution.

The construction of lentiviral vectors containing the coding regions of the mutant LDLR mentioned earlier at exon 16 and the wild-type (WT) LDLR genes, both with a merged FLAG protein on the C-terminus, was performed by the GeneChem Biotech Company (Shanghai, China). A GV416 plasmid with BamHI enzymatic sites was used when preparing the vector, and the vectors were labeled with a green fluorescent protein. 293T cells were used to measure the supernatant virus titer.

HepG2 cells were purchased from YRgene Biotech Company (Shanghai, China) and cultured with Roswell Park Memorial Institute 1640 medium (Gibco, United States) supplemented with 10% fetal bovine serum, 100 units/ml penicillin, and 100 μg/ml streptomycin at 37±C in a humidified atmosphere at 5% carbon dioxide. For vector transfection, HepG2 cells were plated in six-well culture plates and infected with the lentiviruses carrying the WT LDLR or mutant LDLR with Polybrene Reagent (Invitrogen Life Technologies, Spain) according to the manufacturer’s instructions. Infected cells were maintained in culture for 72 h to achieve the maximal LDLR expression, and then the cells were cultured in a medium with Puromycin (1 μg/ml) to exclude the non-transfected cells.

The membrane proteins of HepG2 cells were biotinylated and extracted as previously described (Whitaker et al., 2007). Briefly, HepG2 cells were washed with ice-cold phosphate-buffered saline (PBS) and incubated with buffer A [1 mg/ml Sulfo-NHS-(LC)-biotin in PBS, containing 1-mM MgCl₂ and 1.3-mM CaCl₂, pH 8.0] for 45 min at 4±C. Then, the cells were washed with cold PBS three times, scraped down, and centrifuged. After the cells were cultured in 300-μl buffer B (1% Triton X-100, 4-mM egtazic acid, 10-mM Tris, pH 8.0), they were lysed with ultrasonication, and the lysates were clarified by centrifugation for 15 min at 20,000 × g at 4±C. Then, 500-μg total protein was taken for each sample, and buffer B was added to a total volume of 280 μl before the protein sample was quantified. After the addition of 30 μl of a 50% slurry of High Capacity NeutrAvidin Agarose Resin (Pierce, United States) followed by mixing for 1 h at room temperature, the biotinylated proteins were precipitated from 280 μl of lysate. Then, the biotinylated proteins binding to the agarose were washed three times with ice-cold PBS and eluted from NeutrAvidin agarose after the addition of 20 μl of 2 × sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis sample buffer. Then, 20 μl of 50-mM dithiothreitol in SDS sample buffer was added in each sample and centrifuged at room temperature, and Bromophenol Blue was added and mixed before centrifugation for 1 min at 10,000 × g at 4±C. Then, the samples were used for Western blotting, with 500 μg of total protein loaded for each sample on 10% SDS-polyacrylamide gel electrophoresis gels, transferred to polyvinylidene difluoride membranes, and immunoblotted for FLAG proteins. All experiments for each sample were repeated at least three times.

Confocal laser scanning microscopy (CLSM) was used to evaluate the colocalization of LDLR-FLAG fusion proteins in the HepG2 cells. Briefly, the transfected cells with WT or mutant LDLR were seeded on coverslips, fixed with 4% paraformaldehyde for 15 min, and washed three times with ice-cold PBS. Then, cells were permeabilized with 1% Triton X-100 for 30 min at room temperature. The samples were washed and blocked with PBS-5% bovine serum albumin for 1 h. After washing three times, the samples were incubated with tetramethylrhodamine-conjugated concanavalin A (1:100, Molecular Probes, United States) at room temperature for 1 h. Then, the cells were washed three times with PBS and fixed with 4% paraformaldehyde for 15 min at room temperature for observation of the intracellular fluorescent dye by CLSM (Olympus IX 81, Tokyo, Japan). The intracellular fluorescent dye was observed with sequential excitation and capture image acquisition with a digital camera (Axiocam NRc5; Zeiss, Jena, Germany). We used the Fluoview v50 software (Olympus, Miami, FL, United States) to obtain the images, and quantitative analyses of the fluorescence intensities were performed with ImageJ software (NIH, Bethesda, MD, United States). Finally, the intracellular fluorescent dye was observed by CLSM.

To evaluate the LDL uptake ability of the cells, the HepG2 cells transfected with the mutant and WT LDLR were incubated with 20 mg/ml fluorescent 1,19-dioctadecyl-3,3,3939-tetramethylindocarbocyanine perchlorate (Dil)-conjugated LDL (Molecular Probes) in serum-free media and kept at 37±C for 4 h. The Golgi apparatus was labeled with anti-syntaxin 6. Then, the medium was removed, and the cells were washed with PBS three times. After fixation with 4% paraformaldehyde for 15 min at room temperature, the intracellular fluorescent dye, which is reflective of the uptake ability of LDL, was observed by CLSM.

We used the GraphPad Prism 6 (GraphPad Software, San Diego, CA) and SPSS 16.0 software programs (SPSS, Inc., Chicago, IL, United States) for statistical analyses. Data are presented as the mean ± standard error of the mean. The Student’s t-test was used for comparisons between two groups. A P < 0.05 was considered to be statistically significant.

The proband was a 19-year-old female who presented at our hospital with high levels of TC and LDL-C and multiple xanthomas on the buttocks that had been present she was 9 years old. Physical examination showed multiple xanthomas on the buttocks, tendon, and popliteal space of the proband (Figure 1A). As for the proband’s family history, her father had died at 48 years of age due to myocardial infarction, and he also had hypercholesterolemia and multiple tendon xanthomas. Moreover, one paternal aunt and one paternal uncle of the proband also had hypercholesterolemia and multiple tendon xanthomas. None of the family members mentioned earlier had arcus cornea. The paternal grandmother of the proband died at the age of 41 years from myocardial infarction, whereas the mother, brother, and grandfather of the proband had no hypercholesterolemia. The family tree of the proband is shown in Figure 1B, and the results of lipid analyses before treatment are shown in Table 1. Based on the proband’s family history, physical examination, lipids results, and subsequent DNA analyses, the proband was diagnosed with FH. According to the statement of the proband, statins were occasionally used before the diagnosis. After the diagnosis, atorvastatin (20 mg per night) and ezetimibe (10 mg per night) were prescribed, and the proband was followed in the clinic each month for 6 months. The serum TC and LDL-C varied between 6–7 and 4–5 mmol/L during follow-up.

Given that functional genomic variants in LDLR, PCSK9, and APOB are the main causes of FH (Henderson et al., 2016; Sharifi et al., 2017), we performed target exome capture-based next-generation sequencing of these genes to narrow down the potential genomic variants in this case of FH. Our genomic DNA sequencing found no genomic variants in the PCSK9 and APOB genes. However, there was a single-base substitution (G < A) at position 2389 in the 16th exon of the LDLR gene as shown in Figure 2A, which would lead to a change from valine (V) to methionine (M) at the protein sequence of 797 (V797M), if the protein is translated successfully. However, the genomic variants (c. 2389G > A) was located in the last base of exon 16, which led to the hypothesis that the variant in this case of FH may have a potential influence on mRNA cleavage progression. To test this hypothesis, we further performed quantitative reverse transcription PCR of peripheral venous blood samples from the proband and a healthy control to identify differences in the LDLR mRNA. With the primers for the 16th exon, the results showed that the proband had an additional PCR product of a different length from the one PCR product obtained for a healthy control subject (535 bp, Figure 2B). The sequencing results showed that the shorter PCR product exclusively observed in the proband had a deletion of the 16th exon of the mutant LDLR (Figure 2C), indicating that c. 2389G > A in LDLR disrupted normal mRNA transcription via disturbing mRNA cleavage. Genomic DNA sequencing by Sanger sequencing showed a similar variant (c. 2389G > A) in the family members with FH (father, paternal uncle, and aunt, Table 1) but not in the family members without FH (mother and brother).

To further characterize the localization of the mutant LDLR protein, we constructed viral vectors of LDLR with exon 16 deletion, and WT LDLR with FLAG merged on the C-terminus and infected the conventional cell line of human liver cells. We then used Western blot analyses to evaluate the amount of LDLR protein on the membrane and among total cellular proteins. As shown in Figure 3A, opposite to the considerable membrane expression of LDLR as detected by Western blotting on HepG2 cells transfected with WT LDLR, LDLR protein was barely detected on the membrane cells transfected with the mutant LDLR with exon 16 deletion. However, a considerable amount of LDLR at 120 kDa was detected when the total protein was used for Western blot analysis in the mutant cells, indicating the accumulation of an LDLR precursor-like protein in the plasma of the HepG2 cells with mutant LDLR. To further determine the cellular localization of the LDLR-FLAG fusion proteins, a CLSM examination was used. To our surprise, WT LDLR was mainly distributed at the cell surface. However, the mutant LDLR with exon 16 deletion was retained primarily in the Golgi apparatus (Figure 3B). Taken together, these results indicate that LDLR exon 16 deletion led to LDLR protein re-localization in the Golgi apparatus rather than at the cell surface.

To evaluate the functional consequence of mutant LDLR with exon 16 deletion, we used DiI-LDL to evaluate the uptake ability of the LDLR in HepG2 cells. We observed that the internalization capacity of HepG2 cells with mutant LDLR was substantially lower than that of cells with WT LDLR (13.6 vs. 19.5%, P < 0.05; Figure 4). From their results taken together, this variant was classified as likely pathogenic with autosomal dominant inheritance by the American College of Medical Genetics and Genomics. Because this variant has been reported in individuals with FH in various studies¹, this variant has been found in ≥ 10 unrelated FH cases (PS4). Moreover, our functional analysis fulfilled PS3_Moderate (1) and (3) criteria.