Mutation Analysis of Autosomal-Dominant Polycystic Kidney Disease Patients

Autosomal-dominant polycystic kidney disease (ADPKD) is characterized by bilateral kidney cysts that ultimately lead to end-stage kidney disease. While the major causative genes of ADPKD are PKD1 and PKD2, other genes are also thought to be involved. Fifty ADPKD patients were analyzed by exome sequencing or multiplex ligation-dependent probe amplification (MLPA), followed by long polymerase chain reaction and Sanger sequencing. Variants in PKD1 or PKD2 or GANAB were detected in 35 patients (70%). Exome sequencing identified 24, 7, and 1 variants in PKD1, PKD2, and GANAB, respectively, in 30 patients. MLPA analyses identified large deletions in PKD1 in three patients and PKD2 in two patients. We searched 90 cyst-associated genes in 15 patients who were negative by exome sequencing and MLPA analyses, and identified 17 rare variants. Four of them were considered “likely pathogenic” or “pathogenic” variants according to the American College of Medical Genetics and Genomics guidelines. Of the 11 patients without a family history, four, two, and four variants were found in PKD1, PKD2, and other genes, respectively, while no causative gene was identified in one patient. While the pathogenicity of each variant in these genes should be carefully assessed, a comprehensive genetic analysis may be useful in cases of atypical ADPKD.

While the main causative genes of ADPKD are PKD1 and PKD2, GANAB is also known as PKD3 [6]. The genetic analysis of PKD1 is complicated in comparison to PKD2 or GANAB because there are six PKD1 pseudogenes, from PKD1P1 to PKD1P6, as well as PKD1 on chromosome 16 [7]. Long polymerase chain reaction (PCR) was used in previous genetic studies to differentiate true PKD1 from the six PKD1 pseudogenes [8][9][10][11][12][13].
Exome sequencing is a powerful tool to identify causative genes for chronic kidney disease (CKD); 31% of the genetically diagnosed cases are due to PKD1 and PKD2 mutations [14]. Although the causative genes of ADPKD are mainly PKD1 and PKD2, many genes are associated with cystic kidney diseases, including nephronophthisis or autosomaldominant tubulointerstitial kidney disease (ADTKD). Therefore, exome sequencing is superior to Sanger sequencing for performing a comprehensive analysis. A multiplex ligation-dependent probe amplification (MLPA) analysis of PKD1 or PKD2 is also useful for detecting large deletions or insertions [15,16]. Therefore, we have analyzed 50 ADPKD cases by exome sequencing or multiplex ligation-dependent probe amplification (MLPA), followed by long PCR and Sanger sequencing.

Sanger Sequencing
Long polymerase chain reaction (PCR) was performed with KOD FX Neo (Toyobo, Tokyo, Japan) according to the manufacturer's instructions. The reaction volume was 25 µL, which included 12.5 µL of 2×PCR buffer for KOD FX Neo, 5 µL of 2 mM dNTP mix, 2 µL (30-50 ng) of template DNA, 0.75 µL of 10 µM forward primer, 0.75 µL of 10 µM reverse primer, 0.5 µL of KOD FX Neo, and 3.5 µL of PCR-grade water. The PCR conditions were as follows: pre-denaturation at 94 • C for 2 min, 5 cycles of denaturation at 98 • C for 10 s and extension at 74 • C for 30 s per kilobase (kb), 5 cycles of denaturation at 98 • C for 10 s and extension at 72 • C for 30 s per kb, 5 cycles of denaturation at 98 • C for 10 s and extension at 70 • C for 30 s per kb, 25 cycles of denaturation at 98 • C for 10 s and extension at 68 • C for 30 s per kb, and extension at 68 • C for 7 min. Long PCR primers and sequencing primers for PKD1 are shown in Table S1 [29]. Normal PCR was performed with HotStarTaq DNA Polymerase (Qiagen, Hilden, Germany). The PCR conditions were as follows: 96 • C for 15 min, 35 cycles of denaturation at 96 • C for 45 s, annealing at 57 • C for 45 s, and elongation at 72 • C for 1 min and 72 • C for 15 min. PCR primers for PKD2 and GANAB are shown in Table S2. Sanger sequencing was performed with an Applied Biosystems 3130 Genetic Analyzer (Applied Biosystems, Waltham, MA, USA). The sequence results were analyzed with the BioEdit software program and compared using the Ensembl database.

Pathogenicity Evaluation
The pathogenicity of the identified variants was evaluated according to the American College of Medical Genetics and Genomics (ACMG) guidelines [30]. Previous reports of each variant were examined in the ClinVar database [31] and Leiden Open Variation Database (LOVD) [32]. The minor allele frequency (MAF) of each variant was searched in the Genome Aggregation Database (gnomAD) and 3.5KJPNv2 database, and rare minor allele frequency (MAF) was defined as <1% [33]. The pathogenicity of variants was assessed in silico with software programs such as Polymorphism Phenotyping v2 (PolyPhen-2) and Sorting Intolerant From Tolerant (SIFT).

Background Data
Fifty patients were analyzed ( Table 1). The mean age of the 50 patients was 56 ± 13 years, and 18 patients (36%) were male. A family history of ADPKD was observed in 39 patients (78%), and the number of other affected members is shown in Table 1. Microhematuria was observed in nine patients (18%). The median protein/creatinine ratio was 0.13 (0.07-0.39) g/g·Cr, and the protein/creatinine ratio was >0.5 g/g·Cr in nine patients (18%). The mean estimated glomerular filtration rate (eGFR) was 48.7 ± 27.3 mL/min/1.73 m 2 . The median TKV was 1266 (877-1716) ml. Twenty-six patients (52%) received tolvaptan at an average dose of 57 ± 31 mg. The Mayo classifications of the patients were as follows: class IA (n = 6), class IB (n = 6), class IC (n = 11), class ID (n = 4), and class II (n = 18) [34]. Five patients were unclassified because four patients were diagnosed with stage 5 CKD and one had an eGFR of >100 mL/min/1.73 m 2 .

Mutation Analyses of Patients with Autosomal-Dominant Polycystic Kidney Disease Exome Sequencing
Exome sequencing was performed in 50 patients, and variants in PKD1, PKD2, or GANAB were detected in 30 patients (60%) ( Table 2). Among the 30 patients, there were 24 variants in PKD1, 7 variants in PKD2, and 1 variant in GANAB; patients 1 and 10 had two variants in PKD1. Nineteen of the thirty-two variants were considered to be "likely pathogenic" or "pathogenic" according to the ACMG guidelines (Table S3). Twenty-two of the thirty-two variants were unreported in the ClinVar or LOVD databases and were considered to be novel. To confirm variants in PKD1, long PCR followed by sequencing was performed (except for PKD1 c.12671C>A in patient 1, c.12386T>A in patient 5, and c.11441_11459dup in patient 32, which were confirmed by Sanger sequencing). The 24 variants in PKD1 were classified as follows: missense (n = 10), insertion frameshift (n = 4), deletion (n = 3 (2 were frameshifts)), near splice-site (n = 3), and nonsense (n = 4). Sanger sequencing was performed to confirm variants in PKD2 or GANAB. The seven variants in PKD2 were classified as follows: splice-site (n = 3), deletion frameshift (n = 2), missense (n = 1), and nonsense (n = 1). The variant in GANAB was a missense variant.

The Exome Analysis of Cyst-Associated Genes
Rare variants with MAF <1% in the 90 genes were examined in 15 patients who showed negative results in exome sequencing and MLPA analyses. The results were confirmed by Sanger sequencing and are summarized in Table 3. The PCR primers are shown in Table S4. In all, 19 variants were identified in 13 patients. While 17 were heterozygous, OFD1 c.1215A>C in patient 7 was hemizygous, and PKHD1 c.9629C>G in patient 28 was homozygous. Four of the nineteen variants were considered to be "pathogenic" or "likely pathogenic" according to the ACMG guidelines (Table S5). Seven of the nineteen variants were unreported in the ClinVar or LOVD databases and were considered to be novel. Rare pathogenic variants in other genes were examined in 15 patients (Table S6). There were no rare pathogenic variants in other genes in patients 28, 38, or 39.

Comorbidity Evaluation
The prevalences of liver cysts, VHD, and brain aneurysm are summarized in Table 4. Liver cysts were observed in 43 patients (86%), among whom 7 had severe liver cysts. Six of the seven patients with severe liver cysts were female. The causative genes of severe liver cysts were five variants in PKD1, one variant in GANAB, and one variant in PKHD1. Hypertension was observed in 40 patients (80%). The median value of brain natriuretic peptide (BNP) was 23.8 [13.1-38.2] pg/mL, and 30 patients (60%) had a value of >18.4 pg/mL. The mean ejection fraction (EF) in an ultrasound cardiography examination was 68.9 ± 6.6%, and 48 patients (96%) had a value of >50%. Mild regurgitation was observed at the aortic valve (n = 7), mitral valve (n = 23), tricuspid valve (n = 28), and pulmonary valve (n = 20). Moderate regurgitation was observed in the aortic valve (n = 5) and tricuspid valve (n = 1). Severe tricuspid valve regurgitation was observed in one patient. The mean E/A and E/e' ratios were 1.0 ± 0.4 and 9.1 ± 2.9, respectively. Septal e' < 7 cm/s, septal E/e' > 15, tricuspid regurgitant velocity (TRV) > 2.8 m/s, and left atrial volume index (LAVI) >34 mL/m 2 were observed in 21, 3, 0, and 2 patients, respectively. Patients 19 and 35 fulfilled two of the four criteria for heart failure with preserved ejection fraction (HFpEF), with increased BNP levels [35]. The causative genes of VHD were as follows: PKD1 (n = 19), PKD2 (n = 8), GANAB (n = 1), and other genes (n = 9); the causative gene was unknown in two patients. Brain aneurysms were observed in eight patients (16%), all of whom were female. The causative genes of brain aneurysm were three variants in PKD1, two variants in PKD2, and three variants in other genes. Thirty-seven patients had a low (score 0-3) Predicting Renal Outcome in Polycystic Kidney Disease (PROPKD) score, while thirteen had an intermediate (score 4-6) PROPKD score (Tables 4 and S8) [36].       Eighteen patients had a type II Mayo classification (Table 1); among these patients eight had "likely pathogenic" or "pathogenic" variants, including four PKD1/PKD2 deletions and four other variants such as DNAJB11, NEK8, PKHD1, or WDR19. Transverse and coronal CT images of the eight patients are shown in Figure 2. The kidney images of patients 8, 14, 29, and 44 were similar (Figure 2A). There were no apparent liver cysts in patients 35 and 46 ( Figure 2B). The kidney cysts in patient 35 were mostly seen in the kidney medulla. The kidney cysts and TKV in patient 38 were small. The kidney cysts in patient 41 had calcification. Each of the kidney cysts in patient 46 was relatively small.

Change in eGFR and TKV without and with Tolvaptan
Because tolvaptan was administered to 26 patients (52%), the change in eGFR and TKV was examined in patients who were followed up for more than one year (Table 5). Sixteen patients were managed without tolvaptan, and twenty-four were managed with tolvaptan. The mean eGFR changes per year in patients managed without and with tolvaptan were −2.6 ± 1.6 and −3.1 ± 2.6 mL/min/1.73 m 2 , respectively. The mean initial TKV values in patients managed without and with tolvaptan were 1059 ± 615 and 1429 ± 631 mL, respectively. The mean TKV changes per year in patients managed without and with tolvaptan were 3.8 ± 7.0% and 5.0 ± 5.5%, respectively.

Discussion
We demonstrated 32 variants in PKD1, PKD2, or GANAB in 30 of 50 patients (60%) in the exome analyses; 19 of these variants (59%) were considered to be "likely pathogenic" or "pathogenic" according to the ACMG guidelines, and 22 variants (69%) were considered to be novel. We also identified five novel large deletions in 5 of 20 patients (25%) in MLPA analyses. Additional exome analyses of 90 cyst-associated genes were conducted for 15 patients who showed negative results in exome sequencing and MLPA analyses; these identified 19 variants, of which 4 (21%) were considered to be "pathogenic" or "likely pathogenic" according to the ACMG guidelines, and 7 variants (37%) were considered to be novel. Overall, "likely pathogenic" or "pathogenic" variants were identified in 28 patients; these were identified by PKD1/PKD2 exome analyses in 19 patients; MLPA analyses in 5 patients; and exome sequencing of 90 genes in 4 patients. Other patients had variants of uncertain significance, which did not indicate a definite diagnosis until further proof was obtained.
Of the 50 patients in the present study, the causative genes were PKD1 in 25 patients (50%), PKD2 in 9 patients (18%), and GANAB in 1 patient (2%). Variants in other cystassociated genes that were identified in 13 of 50 patients (26%) were variable, including BBS12, CEP164, DNAJB11, GLIS2, IFT140, KIF7, NEK8, OFD1, SLC41A1, TMEM67, WDPCP, WDR19, and ZNF423 in 1 patient, NPHP3 in 2 patients, and PKHD1 in 4 patients. No causative genes were identified in 2 of the 50 patients (4%). While rare pathogenic variants in other genes were examined in 15 patients who showed negative results in exome sequencing or MLPA analyses, their functional roles in ADPKD were undetermined. Rare pathogenic variants in other genes were also examined in 11 patients with variants of uncertain significance in PKD1, PKD2, or GANAB in a similar way, and while the functional roles in ADPKD were undetermined for most of these variants, the roles of CPLANE1 in patient 18 and AHI1 in patient 47, both of which were cilia-associated genes, might be involved in the development of ADPKD to some degree.
Tolvaptan is frequently used in the treatment of ADPKD due to its effectiveness [40]. Patients with a TKV of >750 mL and with a growth rate of >5% are suitable for tolvaptan treatment. The goal of tolvaptan treatment is to slow the progression of kidney cysts, which can slow the progression of CKD in this disease. The present study showed a decline in eGFR of −3.1 mL/min/1.73 m 2 per year and a 5.0% TKV change per year, which might be acceptable according to previous research [40]. Careful follow-up of the patients receiving tolvaptan treatment should be continued.
The present study was associated with the following limitations. In the present study, no variants were identified in patient 25 or 39. Patient 25 had stage 5 CKD, and his kidney cysts may have been related to hemodialysis treatment. Patient 39 had a relatively small TKV for her age, and the causative gene may be unknown. While two variants in PKD1 were found in patient 1 and patient 10, respectively, we could not examine whether these acted as in cis or in trans variants, as blood samples could not be obtained from parents or siblings. While DNAJB11 is a known disease-causing gene of ADPKD [25], and kidney cysts can develop in patients who are heterozygous for a pathogenic variant in PKHD1, more evidence is required to determine whether heterozygous NEK8 and WDR19 pathogenic variants can mimic the ADPKD phenotype. Rare pathogenic variants in other genes were not validated in Sanger sequencing. We did not examine variants in the introns of 93 genes in the present study.

Conclusions
While the pathogenicity of each variant in cyst-associated genes should be carefully assessed, a comprehensive genetic analysis may be useful for atypical ADPKD cases.