Two novel mutations affecting the same splice site of PKD1 correlate with different phenotypes in ADPKD

Abstract

Genetic heterogeneity is the main factor for significant variation in the course of autosomal dominant polycystic kidney disease (ADPKD). PKD1 patients have more severe renal outcomes compared with PKD2 patients. Co-inheritance of a mutation in both genes is associated with more severe phenotypes than that found with either mutation alone. However, the genotype–phenotype relationship is far from clear in ADPKD. Here, we observed two novel mutations, PKD1:c.12444G > A and PKD1:c.12444 + 1G > A, which alter the same splice donor site of intron 45, correlate with different renal outcomes. To explain the phenomenon, we analyzed the genic and allelic background of the patients, as well as the genetic modifiers, DKK3 and HNF-1β as suggested. Only PKD1 variants were found, which highlights the allelic influence of PKD1 gene to be the last candidate factor. Segregation analysis, online mutation prediction, and recurrence mutation searching were applied to sort the variants. However, none of variants was found to be damaging or associated with the disease except PKD1:c.12444G > A and PKD1:c.12444 + 1G > A. Cloning and sequencing of the mutated cDNA sequences had shown unexpected different splicing effects caused by the mutations. PKD1:c.12444 + 1G > A definitely destroyed the native splice site and created a novel donor site with truncating effect on PC1. In contrast, PKD1:c.12444G > A mainly weakened the site and decreased the expression of normal PC1. Since PC1 negatively regulates cell proliferation in the process of cyst formation and enlargement, our observation may explain this new genotype–phenotype correlation and help to improve genetic counseling and diagnosis of the disease.

• Genetic heterogeneity
• PKD genes
• ADPKD
• genotype
• phenotype

Introduction

Autosomal dominant polycystic kidney disease (ADPKD) is a hereditary renal disorder that is characterized by numerous gradually enlarged fluid-filled epithelial cysts in bilateral kidneys.1 Currently, two causative genes were identified: PKD1 (16p13.3) and PKD2 (4q21).2,3 PKD1 that encodes an approximately 14 Kb transcript with 46 exons is found mutated in 85% of ADPKD cases. Its entire 5′ region up to exon 33 has been reiterated approximately six times further on chromosome 16p. The presence of these highly homologous pseudogenes has made genetic analysis of PKD1 very difficult. Until recently a long-range and locus-specific amplification method has enabled complete mutation analysis of this complex gene.4,5 PKD2, a single-copy gene encoding a 3 Kb open reading frame with 15 exons is responsible for about 15% of ADPKD cases. There is an argument about the existence of PKD3, however, no more genes have been identified.6,7

The severity of renal disease in ADPKD is highly variable ranging from neonatal death, through typical presentations with end-stage renal disease (ESRD) in the sixth decade to adequate function into old age.8 The phenotypic variability is due to heterogeneity at the genic and allelic levels as accepted. Adjusted for age and gender, PKD1 patients have more severe renal disease with earlier diagnosis, higher incidence of hypertension and earlier onset of ESRD compared with PKD2 patients. Co-inheritance of a mutation in both genes is not lethal, but it seem to be associated with more severe disease than that found with either mutation alone.9 One correlation that has been described was the mutation position in PKD1, patients with 5′ mutations seem to have more severe renal disease and be more prone to develop a vascular phenotype.

Additionally, genetic modification and environmental factors may account for considerable intrafamilial variance in disease severity.10,11 Recent studies show that genetic variation of the Dickkopf 3 gene (DKK3) may modify severity of ADPKD resulting from PKD1 mutations.12 Hepatocyte nuclear factor 1β (HNF1β) directly regulates the transcription of PKD2 and PKHD, patients who carry, in addition to their expected familial germ-line defect, additional mutations in HNF1β gene likely aggravate the phenotype.13 Here, we report a study of multiple PKD genes in two Chinese families with different severity of renal disease. The genic and allelic heterogeneity of the disease, as well as genetic modification of PKD genes was analyzed to explain the inconsistent genotype–phenotype correlations in the two families.

Materials and methods

Study subjects and clinical manifestations

Patients in the two families were first diagnosed with ADPKD according to the ultrasound criteria.14 Family A (ID: 09051) has four participants: II1, II2, II4, and III1 (Figure 1A). The proband II2 and his affected brother II1 observed with bilateral renal cysts had only abdominal and back pain. Renal function tests showed the mild decreased epidermal growth factor receptor (eGFR) in II1 and II2. I1 and I2 who had already dead in their 70s were of unknown renal function. Five members were recruited in family B (ID: 09053) including II1, II2, III1, III2, and III3 (Figure 1B). The proband III2 suffering from back pain were found with hepatic cysts, bilateral renal cysts and moderately decreased renal function. His 55-year-old mother II2 suffering from hypertension, polycystic liver, and kidney disease, has ESRD and relies on dialysis to sustain her life. I2, grandmother of the proband, died of renal failure at 60s as claimed by families alive. Patients in family B have shown an early and severe polycystic kidney disease compared with patients in family A. The clinical characteristics of the families, particularly two primary renal outcomes eGFR and time to ESRD, are recorded in Table 1. The study was approved by the Institutional Ethical Review Boards, Chongqing Medical University, and the participants gave informed consent.

Figure 1. Pedigrees of the two Chinese families with ADPKD. (A) The pedigree of family A (ID: 09051). I1 and I2 dead in their 70s with unknown renal function. (B) The pedigree of family B (ID:09053). The probands are shown with the arrows.

Table 1. Clinical characteristics of the patients and classification of renal disease severity.

				Renal function
Patients	Gender	Age (years)	Weight (kg)	Scr^a (μmol/L)	BUN^b (mmol/L)	Ccr^c (mL/min)	eGFR^d (mL/min per 1.73 m^2)	CKD stages^e
Family A
II1	Male	41	75	134.0	4.2	68.0	66.7	2
II2	Male	38	70	127.6	5.0	68.7	71.8	2
Family B
II2	Female	55 (ESRD)	58	966.3	29.3	5.3	4.8	5
III2	Male	29	68	125.8	4.9	73.7	77.1	2
III3	Female	32	51	121.8	5.9	47.2	58.2	3

Notes: ^aScr, Serum Creatinine in μmol/L; ^bBUN, blood urea nitrogen in mmol/L; ^cCockcroft and Gault equation Ccr = (140 − Age) × Weight/(72 × Scr/88.4) × (if female 0.85); ^dGlomerular filtration rate (GFR) is estimated from serum creatinine measurements by using the modification of diet in renal disease (MDRD) study equation based on age, sex, race, and calibration for serum creatinine. MDRD eGFR = 186 × (Scr/88.4)^−1.154× Age^−0.203× (0.742 if female) × (1.233 if Chinese).15

^eCKD, chronic kidney disease, defined as GFR less than 60 mL/min/1.73 m² for three or more months with or without kidney damage. Stages 1 to 5 indicate patients with CKD. For stage 1, defined as kidney damage with normal or increased GFR ≥ 90 mL/min/1.73 m²; stage 2, defined as kidney damage with mild decreased GFR 60–89 mL/min/1.73 m²; stage 3, moderately decreased GFR 30–59 mL/min/1.73 m²; stage 4, severely decreased GFR 15–29 mL/min/1.73 m²; stage 5, kidney failure 15 mL/min/1.73 m² or ESRD.16

Mutation analysis of multiple PKD genes

Mutation analysis of PKD genes was performed by sequencing. For PKD1 and PKD2, a locus-specific amplification method was used.17,18 High-quality genomic DNA was extracted from the peripheral blood sample using a standard phenol–chloroform procedure. The entire PKD1 duplicated region was amplified by five long-range polymerase chain reaction (PCR). Then the long-range segments were diluted 1:10⁴ to avoid genomic DNA carryover and served as templates for 50 nested PCR reactions. Meanwhile, the unique region of PKD1 and the entire PKD2 gene including DKK3 and HNF1β were amplified from the genomic DNA covering exonic regions and the splice junctions. All PCR segments ranging from 150 to 450 bp were separated on 2.0% agarose gel to check the amplification efficiency, and sent for sequencing.

Mutation prediction and confirmation

All variations detected were first classified by analyzing recurrence as reported in the literature, ADPKD mutation database (PKDB) and Single Nucleotide Polymorphism database (dbSNP). Then novel variations were checked in their families and evaluated computationally using online mutation prediction programs: NNSplice and NetGene2. Splicing mutations were first confirmed by RT-PCR across the exon–exon junctional region of the cDNA followed by sequencing. Then, the fragments were cloned into Escherichia coli competent cells using pGEM-T Easy vector. Positive clones were selected and the DNA of the recombinant plasmid was extracted for sequencing in order to show clearly the novel splice sites.

TaqMan real-time PCR and the reference sequences

The differences in the expression of PKD1 were reported to cause the diversity of cell proliferation and disease severity. So the aberrant transcripts from the mutant alleles were analyzed using a relative quantification method, real-time quantification (ΔΔC_T), in order to find the potential causes of the phenotypic divergence. TaqMan real-time PCR was performed in an ABI 7500 real-time PCR system using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as the endogenous control. Amplification reaction contained 10 μL of 2.5× Real Master Mix, 1.0 μL each of 10 μM primers, 0.2 μL of 25 μM probe, 1.25 μL of 20× Probe Enhancer Solution and 2.0 μL of template cDNA and nuclease-free water to a final volume of 25 μL. The thermal profile consisted of 50 °C for 2 min, followed by 95 °C for 10 min and 40 cycles of 95 °C for 15 s and 60 °C for 40 s. Fluorescence was measured once per cycle at the end of the 60 °C segment.

The NCBI RefSeq sequences were used for PKD1 [GeneBank: NM_001009944.2, PKD2 [GeneBank: NM_000297.3], HNF1β [GeneBank: NM_000458.2], and DKK3 [GeneBank: NM_015881.5] in this study. The standard nomenclature recommended by the Human Genome Variation Society (HGVS) was used to number nucleotides and name variations.

Results

Mutation screening of all the exons and splice junctions of the PKD genes in the probands has identified five variants in PKD1 (Table 2); no PKD2 mutation was found. Mutation checking of PKD1:c.8444C > T and PKD1:c.9156C > A in other individuals of each family showed no co-segregation with the disease. Variation PKD1:c.4674G > A found to co-segregate with the disease was a well known polymorphism described in many previous studies. Novel changes PKD1:c.12444G > A and PKD1:c.12444 + 1G > A were also found to be associated with the disease, which observed in their affected relatives, but not found in the healthy relatives and 100 controls. Mutation analysis of two genetic modifiers, DKK3 and HNF1β, had been negative in the present study.

Table 2. Details of the variations detected in the families.

Families	Region	cDNA change	Amino acid change	Mutation type	Described or novel
Family A	PKD1
II1	EX45	c.12444G > A	E4148fs	Splice	Novel
II2	EX45	c.12444G > A	E4148fs	Splice	Novel
	EX23	c.8444C > T	A2815V	Substitution	18
II4	EX23	c.8444C > T	A2815V	Substitution	18
Family B	PKD1
II1	EX25	c.9156C > A	G3052G	Synonymous	19
II2	EX15	c.4674G > A	P1558P	Synonymous	20
	IVS45	c.12444 + 1G > A	E4148fs	Splice	Novel
III1	EX25	c.9156C > A	G3052G	Synonymous	19
III2	EX15	c.4674G > A	P1558P	Synonymous	20
	EX25	c.9156C > A	G3052G	Synonymous	19
	IVS45	c.12444 + 1G > A	E4148fs	Splice	Novel
III3	IVS45	c.12444 + 1G > A	E4148fs	Splice	Novel
	EX15	c.4674G > A	P1558P	Synonymous	20

Splice site prediction using online programs suggested that “synonymous” variant PKD1:c.12444G > A may abolish the native splice donor site, which shown a splicing effect similar to that of PKD1:c.12444 + 1G > A in family B (Supplement 1). RT-PCR using primers located in exon-exon junctional region of the cDNA (Supplement 2) has confirmed the prediction (Figure 2, Supplement 3). The novel splice sites were clearly shown by cloning and sequencing. Positive clones from the two probands were selected and the purified recombinant plasmid DNA was sent for sequencing. As was expected, only one aberrant transcript with a novel splice donor site and insertion sequence (–ATGG–) was observed in the proband of family B carrying PKD1:c.12441 + 1G > A (Figure 2-B3). Unexpectedly two different aberrant transcripts were observed in the proband of family A carrying PKD1:c.12444G > A (Figure 2): the transcript I with a novel splice donor site and insertion sequence (–GTGG–) (Figure 2-A3), the transcript II with only a substituted nucleotide and the native splice donor site (Figure 2-A4). The transcript II resulted from PKD1:c.12444G > A encodes unchanged amino acids comparing with the wild-type transcript (Figure 2C).

Figure 2. The splice site mutations detected by direct sequencing and confirmed by RT-PCR. (A) Diagram of mutant genomic sequence and cDNA sequences illustrating the effects of PKD1:c.12444G > A in family A. The mutation was observed by direct sequencing of the PCR product from genomic DNA (A1) then confirmed by RT-PCR (A2). Cloning and sequencing of the mutated cDNA found that PKD1:c.12444G > A resulted in two aberrant transcripts (A3; A4). Substituted nucleotide is underlined, exon–intron and exon–exon boundaries are indicated with vertical lines or arrows, respectively. (B) Mutated genomic sequence and cDNA sequences of PKD1:c.12444 + 1G > A illustrating the typical splicing effects in family B. The mutation was first observed by direct sequencing (B1), then confirmed by RT-PCR (B2). Cloning and sequencing shows an aberrant transcript (B3). (C) The wild-type cDNA sequence demonstrating the native splice donor site.

The allele specific primers (131204F/131204R) were designed according to the sequence of PKD1 to amplify the aberrant transcripts with the 4 bp insertion. The TaqMan probes (131204P, GAPDH-P) contained 6-carboxy-fluorescein (FAM) reporter dye at the 5′ end and 6-carboxytetramethylrhodamine (TAMRA) fluorescent quencher at the 3′ end (Supplement 4). In all TaqMan real-time PCR reactions, no fluorescence was observed for the negative blank controls, but successful detection was observed in the positive probands and the endogenous controls (Figure 3). The results of the quantification of the aberrant transcripts and the difference of the cycle threshold (C_t) values between PKD1:C.12444G > A (–GTGG–) and PKD1:C.12444 + 1G > A (–ATGG–) are shown in Table 3. The ΔΔC_T value and the rate of expression for PKD1:C.12444G > A (–GTGG–) relative to PKD1:C.12444 + 1G > A (–ATGG–) are also calculated (Table 3).

Figure 3. Relative quantification using the comparative C_T method. (A) The amplification plot of real-time quantitative PCR for the mutations and GAPDH. (B) Comparison of gene expression of PKD1:c.12444G > A (GTGG) relative to PKD1:c.12444 + 1G > A (ATGG).

Table 3. Analysis of the data from relative quantification using the comparative C_T method.

Mutations	Mutation Average CT	GAPDH Average CT	ΔCT Mutation-GAPDH^a	ΔΔCT ΔCT − ΔCT(–ATGG)^b	Rel. to (–ATGG–)^c
C.12444 + 1G > A (–ATGG–)	32.575 ± 0.110	25.819 ± 0.311	6.759 ± 0.189	0.000 ± 0.189	1.0 (0.88–1.14)
C.12444G > A (–GTGG–)	35.63 ± 0.275	28.335 ± 0.182	7.297 ± 0.129	0.538 ± 0.129	0.69 (0.63–0.75)

Notes: ^aThe ΔC_T value is determined by subtracting the average GAPDH C_T value from the average mutation C_T value. The standard deviation of the difference is calculated from the repeated mutation and GADPH C_T values.

^bThe calculation of ΔΔC_T involves subtraction by the ΔC_T calibrator value. This is subtraction of an arbitrary constant, so the standard deviation of ΔΔC_T is the same as the standard deviation of the ΔC_T value.

^cThe range given for C.12444 G > A (–GTGG–) relative to C.12444 + 1G > A (–ATGG–) is determined by evaluating the expression: with ΔΔC_T+ s and ΔΔC_T− s, where s = the standard deviation of the ΔΔC_T value.

Discussion

Family-based segregation analysis found that PKD1:c.12444G > A and PKD1:c.12441 + 1 G > A were co-segregated with the disease in family A and B, respectively. PKD1:c.12441 + 1 G > A was a typical splicing mutation with clear effect on pathogenesis, while PKD1:c.12444G > A was a “synonymous” variant that only altered the third position of the codon. Considering the fact that no clearly pathogenic mutation was found in family A and the mutated site of PKD1:c.12444G > A bordered upon a splice donor site, online programs were chosen to evaluate the splicing effect potential of this change. Finally, both online prediction and the following RT-PCR had confirmed the mutation. PKD1:c.12444G > A resulted in aberrant splicing products encoding truncated proteins similar to PKD1:c.12441 + 1 G > A. Since all exons and splice junctions of PKD1 and PKD2 were analyzed in the patients and no other mutation was found, PKD1:c.12444G > A and PKD1:c.12441 + 1 G > A are classified as the ADPKD causing mutations currently.

In our study, patients III2 and III3 in family B have an onset age of 30.5 years in average which is nine years earlier than that of family A (39.5 years). III3 and III2 in family B are predicted to be at-risk for ESRD when they are 50 s. While no ESRD patient described in family A indicates that patients II1 and II2 in this family may develop moderately decreased renal function into old age just like their deceased mother or father did. Classification of the CKD stages clearly shows the different severity of renal outcomes in the two families. To explain why these two mutations affecting the same splice site correlate with different phenotypes, we first analyzed the genetic heterogeneity including gene and allele influences. However, only unique PKD1 mutation was identified in each of the family without PKD2 mutation. Besides, no other pathogenic mutation potential which may aggravate the phenotype was identified in PKD1 except c.12444G > A and PKD1:c.12441 + 1 G > A. The possibility for genic or allelic heterogeneity causing the different renal disease severity in the two families was therefore excluded. Further mutation analysis of DKK3 and HNF1β had been negative in the study and no genetic modification influence was observed currently.

The PKD1 gene mutated in 85% ADPKD cases, encodes polycystin-1 (PC1), which is a receptor protein for cell–cell/matrix interactions in the regulation of cell proliferation and apoptosis.21 In vitro functional studies show that low expression of PKD1 contributes to increased cell proliferation, whereas its over expression slows cell growth, indicating that PC1 may negatively regulate cell proliferation.21–23 In the present study, two different aberrant transcripts were observed in the proband of family A carrying PKD1:c.12444G > A. One with only a substituted nucleotide should lead to the production of a normal protein and the other with a 4 bp insertion (–GTGG–) produced a premature translational-termination codon. Since a part of the non-mature RNA carrying the c.12444G > A mutation underwent normal splicing, it is possible that there was a lower amount of the aberrant transcript (–GTGG–) that led to milder clinical manifestations. Therefore, a relative quantification method, real-time quantification (ΔΔCt) was performed to analyze the aberrant transcripts from the mutant alleles in order to explain the phenotypic divergence (Figure 3). The ΔC_T (the subtraction of the average GAPDH C_T value from the average mutation C_T value) were calculated and then the rate of expression was calculated using . The C_T value for PKD1:c.12444G > A (–GTGG–) which is higher than PKD1:c.12444 + 1G > A (–ATGG–) indicates a lower expression of the aberrant transcript. In other words, mutation PKD1:c.12444G > A may not totally disrupt normal splicing and it resulted in a higher level of normal PC1 than PKD1:c.12444 + 1G > A in the development of renal cysts, finally negatively regulated cell proliferation and apoptosis. And this may be an interpretation for the phenotypic variability in the present study.

The phenotypic variability of ADPKD relates to multiple factors: genic and allelic background, genetic modification, and environmental influences. To understand the factors that underlie phenotypic variability is clearly of prognostic value and may also shed light on pathogenesis of the disease. Current study has identified two novel splicing mutations, of which PKD1:c.12444G > A is an atypical splicing mutation. Both the mutations altered the native splice donor site of intron 45 of PKD1 but correlated with different phenotypes. Current findings provide a novel genotype–phenotype correlation in ADPKD and highlight the importance of “synonymous” mutations in genetic diagnosis and counseling of the disease.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

This work was supported in part by grants from the National Natural Science Foundation of China (90919013, 30871103, 21305165), and Natural Science Foundation of Chongqing (2010BA5008).

Acknowledgments

We would like to thank all the patients and their families for their cooperation and interest in this study.

Supplementary material available online Supplements 1, 2, 3 and 4.