Novel Mutations in NPHS2 Detected in Both Familial and Sporadic Steroid-Resistant Nephrotic Syndrome

Patients

We analyzed 27 multiplex families (53 affected individuals were included) and 25 patients with “sporadic” SRINS originating mainly from Germany but also from other European countries. Parental consanguinity was not reported. For clinical evaluation, we used a standard questionnaire as previously described (7). The characteristic features defining the clinical diagnosis of SRINS included familial occurrence, age at onset in early childhood, resistance to steroid therapy, progression to end-stage renal disease within a few years, and absence of recurrence after renal transplantation (1).

Haplotype Analysis

Genomic DNA was extracted from leukocytes according to standard laboratory protocols. PCR was performed with five polymorphic markers spanning the critical region of NPHS2 on chromosome 1q25-q31 (1,2,7). The respective order from the centromeric to the telomeric border was: cen, D1S416, D1S1640, D1S2791, NPHS2, D1S215, D1S2883, tel (flanking markers are underlined). In the multiplex families, we first performed haplotype analysis. In case of consistency with linkage to NPHS2, mutational analysis was performed.

Mutational Analysis

We performed mutational analysis by single-strand conformation polymorphism (SSCP) and direct sequencing of both strands as previously described (8) in 17 SRINS families compatible with linkage to the NPHS2 gene locus on chromosome 1q21 and 25 patients with sporadic SRINS. The PCR of exons 3, 4, and 8 was carried out as previously described (2). For the remaining exons, the following primers and conditions were applied: exon 1: 5′-GCAGCGACTCCACAGGGACT-3′ and 5′-TCCACCTTATCTGACGCCCC-3′; exon 2: 5′-AGAATTGGACCAACAGATGC-3′ and 5′-AAGTGAGAATGGGCATGGTG-3′; exon 5: 5′-AAAGGAGCCCAAGAATCAAG-3′ and 5′-AAATATTTCAGCATATTGGCC-3′; exon 6: 5′-GTTTAGGCATGCTC TCCTC-3′ and 5′-GATATGGCTATAGTACTCAGTG-3′; exon 7: 5′-GTCTGTGTGAAAGCTTTGGC-3′ and 5′-GCAAAGGGGAAATGTTCTCC-3′; at annealing temperatures of 52°C (exon 5) and 60°C (exons 1, 2, 6, and 7). Because of the high CG content of exon 1, PCR was performed with Taq polymerase and Q solution according to manufacturer’s instructions (Qiagen, Hilden, Germany). To rule out polymorphism in 100 chromosomes of healthy individuals, we used SSCP for R138Q, SSCP and direct sequencing for V290M, digestion with HhaI (Amersham Pharmacia Biotech, Freiburg, Germany) for A284V and with ClaI (MBI Fermentas GmbH, St. Leon-Rot, Germany) for R229Q, polyacrylamide gel electrophoresis for 460-467insT (8), and allele-specific PCR for IVS-1 G→T (5′-TCCAAACTTTTTTCTGCCTAT-3′ and 5′-AAATATTTCAGCATATTGGCC-3′ [59°C]) and A196P (5′-GGAGATAGATGCCATTTGCTACTACCC-3′ and 5′-TAAGTACCTTTGCATCTTGGGCGATGC-3′ [62°C]). SSCP electrophoresis was performed for 6 h at 150 V with Power Pack P25 (Biometra, Göttingen, Germany). Direct sequencing of both strands was carried out as described previously (8).

Haplotype Analysis

To determine consistency with linkage to NPHS2 in the multiplex SRINS families, we first performed haplotype analysis followed by mutational analysis. All 12 families reported here were consistent with linkage to this chromosomal region (data not shown). To further evaluate potential consanguinity in the seven patients with apparently sporadic SRINS, which would give rise to homozygosity by descent (9) in the NPHS2 gene, we carried out haplotype analysis in these patients too (Figure 1). Six of the seven tested patients showed homozygosity (INS 72, 73, 74, 76, 83, and 90) and one heterozygosity (INS 86) in the critical NPHS2 interval. The haplotypes of the respective parents, analyzed in four patients (INS 72, 74, 86, and 90), were congruent with these findings (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Pedigrees of seven patients with steroid-resistant nephrotic syndrome. Haplotypes were generated by using five consecutive microsatellite markers spanning the critical genetic NPHS2 region on chromosome 1q25-q31. Affected individuals are indicated as black symbols. Male family members are shown as squares and female family members as circles. The differently shaded bars indicate heterozygous (white) and homozygous (black) haplotypes. The respective order from top to bottom was as follows: cen, D1S416, D1S1640, D1S2791, NPHS2, D1S215, D1S2883, tel (flanking markers are underlined). Note that in all families, haplotype data were compatible with homozygosity by descent, with the exception of INS 86. The haplotypes of the respective parents, analyzed in four patients (INS 72, 74, 86, and 90), were congruent with these findings (data not shown).

Mutational Analysis

A total of 51 people, 31 of whom were affected, were examined for mutational analysis. We detected 10 different missense, splice-site, and frameshift mutations. Of these, five mutations are novel (Table 2). Remarkably, we detected NPHS2 mutations not only in familial cases but also in individual patients with SRINS. None of the mutations were found in at least 100 control chromosomes. In eight families and all seven patients, potential loss-of-function mutations in NPHS2 were detected on both alleles; in four families, only one mutation was found (INS 18, 28, 29, and 46).

Missense Mutations

We identified three novel missense mutations affecting the carboxy-terminal cytoplasmic tail of podocin (Table 2) (Figure 2). The first, a C→T transition at position 851 leading to an alanine to valine substitution A284V, was found heterozygously in families INS 29 and 46, and homozygously in patient INS 76. The other two novel missense mutations were both identified in patients of family INS 86 in a heterozygous state: the 587G→C transversion affects the highly conserved arginine at position 196 (R196P), and the 868G→A transition replaces valine at position 290 (V290M).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2. Detection of five novel NPHS2 mutations by direct sequencing. The corresponding nucleotide sequence with the respective amino acids are shown at the bottom of each chromatogram. The left column exemplifies the respective chromatograms of healthy control individuals (1A to 5A). The right column shows the chromatograms of patients with steroid-resistant nephrotic syndrome with missense mutations (1B to 3B), a splice site mutation (4B), and a frameshift mutation (5B). In patient INS 76, we found a homozygous C→T substitution at position 851 leading to A284V (1B). Patient INS 86 showed two missense mutations in the heterozygous state. The first G→C substitution at position 587 leads to R196P (2B), whereas the G→A substitution at position 868 leads to V290M (3B). A homozygous splice site mutation IVS4-1G→T was detected in patient INS 74 (4B). A heterozygous frameshift mutation 460-467insT leading to a premature stop codon T181X was found in patient INS 92 (5B). The mutations cosegregated with the disease within the families, and none of them were found in at least 100 control chromosomes.

We identified the frequently reported R138Q mutation homozygously in six families (INS 11, 16, 37, 41, 43, and 67) and two patients (INS 73 and 90); in three families (INS 28, 50, and 92) only the maternal allele was affected. Furthermore, we identified a homozygous V180M exchange in patient INS 72 and a heterozygous R291W substitution in family INS 18 segregating from the maternal side (Table 2).

Splice Site and Frameshift Mutations

We identified the first splice site mutation in NPHS2. IVS4-1G→T involves the 3′ acceptor splice site of intron 4 and was detected in a homozygous state in patient INS 74 (Figure 2). Another novel mutation results in a 1-bp insertion at position 460 to 467 with a consecutive frameshift and premature stop codon T181X. This is the second potential loss-of-function mutation in family INS 92. In addition, we detected two further frameshift mutations due to deletions that have previously been described (2). In family INS 50, we found the paternal 419delG in both affected siblings in a heterozygous state. 855–856delAA was identified in the patient INS 83 in a homozygous state.

Polymorphisms

Aside from the potential loss-of-function mutations described, we detected a number of nucleotide exchanges that are probably without any influence on NPHS2 function (Table 2). Some were silent without an amino acid exchange: 954T→C (A318A), 102G→A (G34G), 288C→T (S96S), and 1038A→G (L346L). In fact, one led to an amino acid exchange but was also detected in 3% of all chromosomes in 100 healthy controls: 686G→A (R229Q). Interestingly, the R229Q exchange was found in a heterozygous state in three of four families with only one mutation (INS 18, 29, and 46) whereby the respective missense mutations were on the opposite chromosome. Furthermore, none of the healthy control individuals carrying R229Q was homozygous for this exchange. A functional role of R229Q is unlikely; however, it cannot completely been ruled out. The polymorphisms 288C→T, 954T→C, and 1038A→G have recently been published by Wu et al. (10,11).