Identification of a Novel ACTN4 Gene Mutation Which Is Resistant to Primary Nephrotic Syndrome Therapy.

ACTN4, a gene which codes for the protein α-actinin-4, is critical for the maintenance of the renal filtration barrier. It is well known that ACTN4 mutations can lead to kidney dysfunction, such as familial focal segmental glomerulosclerosis (FSGS), a common cause of primary nephrotic syndrome (PNS). To elucidate whether other mutations of ACTN4 exist in PNS patients, we sequenced the ACTN4 gene in biopsies collected from 155 young PNS patients (≤16 years old). The patients were classified into five groups: FSGS, minimal change nephropathy, IgA nephropathy, membranous nephropathy, and those without renal puncture. Ninety-eight healthy people served as controls. Samples were subjected to Illumina's next generation sequencing protocols using FastTarget target gene capture method. We identified 5 ACTN4 mutations which occurred only in PNS patients: c.1516G > A (p.G506S) on exon 13 identified in two PNS patients, one with minimal change nephropathy and another without renal puncture; c.1442 + 10G > A at the splice site in a minimal change nephropathy patient; c.2191-4G > A at the cleavage site, identified from two FSGS patients; and c.1649A > G (p.D550G) on exon 14 together with c.2191-4G > A at the cleavage sites, identified from two FSGS patients. Among these, c.1649A > G (p.D550G) is a novel ACTN4 mutation. Patients bearing the last two mutations exhibited resistance to clinical therapies.

1. Introduction

PNS is a type of immune-mediated glomerular disease, resulting in a series of pathophysiological changes due to increased permeability of the glomerular filtration membranes, thus caused the loss of plasma proteins. Based on clinical characteristics, PNS is usually classified into 5 classes based on clinical features: FSGS, minimal change nephropathy, IgA nephropathy, membranous nephropathy, and without renal puncture. It is well known that this disease severely affects the growth and development of children [1, 2]. Genetic mutations play a critical role in the etiology of PNS; thus, it is important to clarify the relevant genetic mutations in PNS patients, which could be helpful for developing new diagnostic methods and genetic therapies [3].

α-actinin-4, a protein expressed widely throughout the body, but enriched in the kidney, is encoded by the ACTN4 gene. Several mutations of this gene have been found, and it usually correlates with abnormal serum levels and the function of α-actinin-4, especially in those patients suffering with FSGS [4]. The presence of the ACTN4 gene p.Lys255Glu mutation relates to abnormal affinity of the actin-binding domain (ABD), which is mainly due to conformational changes in the molecular structure of α-actinin-4 [5, 6]. An Y265H variant of the ACTN4 gene has also been detected in an adolescent patient with FSGS [7]. Some patients with the p.Ser262Phe mutation of the ACTN4 gene have shown full-blown rapidly progressing nephrotic syndrome in early childhood [8]. In addition, the p.G195D and c.465C > T mutations have also been discovered in FSGS patients [9, 10].

However, there is a possibility of the existence of more ACTN4 mutations in PNS patients. To clarify this, we used Illumina's next generation sequencing technology, in combination with the FastTarget target gene capture method, to screen and sequence peripheral blood samples collected from 155 young PNS patients (≤16 years of age) and compared these with 98 healthy controls. Based on bioinformatic analysis, we confirmed that ACTN4 mutations exist not only in FSGS patients but also in patients with minimal change nephropathy and those without renal puncture. More importantly, we discovered a new ACTN4 mutation, which could confer resistance to certain clinical therapies.

2. Results

2.1. Comparison of Serum α-Actinin-4

The serum levels of α-actinin-4 were 544.7 ± 108.11 and 241.20 ± 153.11 ng/mL in healthy controls and PNS patients, respectively. It is significantly different (p < 0.001).

2.2. Sequencing Depth and Coverage

The average sequencing depth in the target regions was 216.965–1118.708, 86.50–98.20% target area > 2 × coverage; 82.50–97.10% target area > 10 × coverage; 78.40–95.30% target area > 20 × coverage; and 76.00–94.70% target area > 30 × coverage.

2.3. ACTN4 Gene Mutations Identified from PNS Patients

Analysis of sequencing results revealed 5 ACTN4 mutations that only occurred in PNS patients (Table 1).

Table 1

ACTN4 mutations detected from PNS patients. Five types of ACTN4 mutations were detected in samples obtained from 155 PNS patients which were then classified into groups: FSGS, minimal change nephropathy, IgA nephropathy, membranous nephropathy, and without rental puncture. Mutation areas and related functions are listed and mutated patient's numbers are documented. represents a mutation detected from the same patients.

Gene region	Function	Predicted protein variants: NM_001322033 NM_004924	SNP ID	FSGS	Minimal change nephropathy	IgA nephropathy	Membranous nephropathy	Without renal puncture
Exonic	Nonsynonymous SNV	Exon13:c.1516 G > A (p.G506S)	rs753348354		1			1
Splicing		Exon12:c.1442 + 10G > A	rs772524653		1
Splicing		Exon18:c.2191-4G > A	rs371779934	2
Exonic	Nonsynonymous SNV	Exon14:c.1649 A > G (p.D550 G)		2∗
Exonic	Nonsynonymous SNV	Exon18:c.2315 C > T (p.A772 V)	rs760946329	2∗

2.4. Exon 13: c.1516G > A (p.G506S) Mutation

This mutation was detected in 2 patients. One was detected in a 5-year-old girl with no family history of nephrotic syndrome but was diagnosed with the disease just over 7 months before being admitted to hospital. Administration of prednisone acetate was not effective, and renal pathological examination showed the characteristics of minimal change nephropathy (Figure 1).

Figure 1

Minimal change of nephropathy was noted under renal pathological examination. (a) HE staining showed mild lesions of the glomeruli, pointed by the white arrow. Scale bar indicates 40 μm. (b) PAS staining showed obvious proliferation of mesangial cells and matrix, pointed by the white arrow. Scale bar indicates 25 μm. (c) PASM staining showed swollen epithelium, vascular degermation, diffused foot processes, and mesangial cell and stromal segmental hyperplasia, pointed by a white arrow. Scale bar indicates 25 μm. (d) Transmission electron microscope imaging showed podocytes diffuse fusion, pointed by a black arrow. Scale bar indicates 2 μm.

A 13-year-old girl with no family history of nephrotic syndrome was also diagnosed with PNS (without renal puncture) 9 months prior to admission. She was treated with prednisone acetate combined with cyclophosphamide for more than 3 months, and the urinary protein changed from negative to positive during the first month. However, the disease returned when the drug was reduced or when she had an infection. Pathological examination showed minimal renal puncture in this patient and she also had the exon 13: c.1516G > A (p.G506S) mutation.

2.5. Exon 12 at the Splice Site: c.1442 + 10G > A Mutation

This mutation was detected in a 15-year-old boy with no family history of nephropathy syndrome and 30 months after diagnosis. His symptoms were alleviated after receiving prednisone acetate and cyclophosphamide over a 16-month period. The renal pathological examination was similar to Figure 1 and consistent with minimal change in nephropathy.

2.6. c.2191-4G > A Mutation

This mutation was detected in two boys: one was 9 and another was 13 years old. Neither boys had a family history of renal diseases. The first one received irregular treatment with prednisone acetate due to repeated renal pathology. Under pathological examination, marked visible glomerular sclerosis was detected, which was consistent with focal stage glomerulosclerosis with acute tubular injury (Figure 2). The latter one showed alleviated symptoms after receiving hormone treatment combined with the immunosuppressant–cyclophosphamide. The renal pathological examination was similar in both patients.

Figure 2

Familial focal segmental glomerulosclerosis under renal pathological examination. (a) HE staining of Kidney biopsies showed tubular lumen expansion and the disappearance of the brush border, pointed by the white arrow. Scale bar indicates 100 μm. (b) Mesangial cell and stromal cell proliferation, pointed by the white arrow. Scale bar indicates 15 μm. (c) Capillary spasm opening and little or no significant thickening of the basement membrane, pointed by the black arrow. Scale bar indicates 25 μm.

2.7. Exon 14: c.1649A > G (p.D550G) and Exon 18: c.2315C > T (p.A772V) Mutations

The presence of these two mutations was detected in of two patients. One was an 8-year-old boy and another was a 11-year-old girl. Both were diagnosed with nephropathy syndrome. There was no prior family history of renal disease. Renal pathological examination showed mild glomerular lesions with characteristic mesangial cell and stromal cell proliferation, but no obvious change of the thickness of the basement of membrane. Photomicrographs were similar to that observed in Figure 2. However, both patients showed no alleviation of symptoms after receiving standardized treatment for PNS.

Importantly, c.1649A > G (p.D550G), a mutation of ACTN4 gene, had not been reported previously in any of the major databases including dbSNP, 1000Genomes, ESP6500, ExAC03, ExAC03_EAS, gnomAD, Hrcr1, Kaviar_20150923, and GENESKYDB_Freq (Table 2). There was a suggestion that the presence of this mutation could possibly result in the patient being resistant to clinical therapies.

Table 2

The probability of all low-frequency samples in the database. Five types of mutations were detected, checked, and compared through genome databases and/or other software, including Freq_Alt (1000 g), 1000 g_chbs, Exac03, ExAC03_EAS, esp6500, gnomAD, Hrcr1, Kaviar, GENESKYDBHITS_Freq, and GeneskyGenomeDB. The probability of each mutation is listed.

Polymorphisms: NM_001322033 NM_004924	Freq_Alt (1000 g)	1000 g_chbs	ExAC03	ExAC03_EAS	esp6500	gnomAD	Hrcr1	Kaviar	GENESKYDBHITS_Freq	GeneskyGenomeDB
exon13:c.1516 G > A (p.G506S)			0	0		0		6.5 × 106
exon12:c.1442 + 10G > A			8.446 × 106	0.0001		1.233 × 105		6.5 × 106
exon18:c.2191-4G > A			1.667 × 105	0	0.000077	5.29 × 105		1.29 × 105
exon14:c.1649 A > G (p.D550 G)
exon18:c.2315 C > T (p.A772 V)			8.284 × 106	0.0001		2.844 × 105		6.5 × 106

3. Discussion

Previous studies proved that ACTN4 mutations play an important role in the development of PNS [4, 11]. Detsika et al. found that the abnormal expression of α-actinin-4 and glomerular-associated proteins was related to the pathogenesis of FSGS [12]. Xie et al. found that the expression of α-actinin-4 in renal podocytes was decreased in FSGS and patients with IgA nephropathy [13]. Bartram et al. found that ACTN4 gene p.g195 d mutation in FSGS sporadic children and uroepithelial cells showed that the expression levels of α-actinin-4 was lower than those in healthy controls [9], while Wagrowska-Danilewicz et al. showed no change in α-actinin-4 expression in patients with minimal nephrotic patients, but lesions were observed in the glomeruli [14].

Luimula et al. found that the expression of α-actinin-4 protein in nephrotic rats did not change significantly, but abnormal distribution and increases of α-actinin-4 mRNA expression were seen. They suggested that the difference between the two experimental results may be related to the different experimental models used (puromycin rat/in vitro cultured glomerular podocytes) [15]. Suvanto et al. found that the renal podocytes slit diaphragm protein expression was reduced in Finnish type congenital nephrotic syndrome, and similar changes were not observed in minimal change nephrotic podocytes. However, NEPH1 FAT1, ACTN4, and CD2AP were found to be expressed normally in proteinuric and nonproteinuric kidneys of minimal change nephrotic patients [16]. Above all, this is a dynamic process which is associated with the course of PNS. Abnormal serum levels of α-actinin-4 could be a consequence of ACTN4 mutations.

In line with the aforementioned studies, kidney disease is consistent with decreased serum levels of α-actinin-4. In our study, the serum levels of ACTN4 in the healthy group were significantly higher than those of the PNS patients. In addition, we detected only 5 types of ACTN4 heterozygous mutations from PNS patients. They were c.1516G > A (p.G506S) mutation on exon 13, c.1442 + 10G > A mutation at the splice site, c.2191-4G > A mutation at the splice site, c.1649A > G (p.D550G) on exon 14, and c.2315C > T (p.A772V) on exon 18. Among these, the mutation c.1649A > G (p.D550G) on exon 14 is a newly described mutation, and its presence together with the c.2191-4G > A mutation at the cleavage site in two patients appeared to confer resistance to clinical therapies in their host.

In summary, the ACTN4 gene is a candidate gene involved in the development of PNS. It is anticipated that, in future its mutations could be helpful for the diagnosis and for the prediction of clinical therapies. Moreover, it could possibly serve as a novel therapeutic target.

4. Materials and Methods

4.1. Study Subjects

Ninety-eight healthy children were recruited from the Physical Examination Center of the Affiliated Hospital of Youjiang Medical College for Nationalities, and all those with nephrotic diseases were excluded after clinical tests. A total of 155 children with PNS (proteinuria ≥50 ng/kg/day and serum albumin <30 g/L, case group) were recruited from the Outpatient Department, Affiliated Hospital of Youjiang Medical College for Nationalities. All the study subjects in the case group were under 16 years of age and were excluded if they had secondary nephrotic syndrome. The 155 patients with PNS were classified into five groups based on pathological tests: FSGS (n = 47), minimal change nephropathy (n = 37), IgA nephropathy (n = 36), membranous nephropathy (n = 17), and without renal puncture (n = 18).

This research was approved by the Ethics Committee of the Affiliated Hospital of Youjiang Medical College for Nationalities. All the parents/guardians of the study subjects understood the research principles and agreed to participate in this study.

4.2. Extraction of DNA

2 mL of venous blood was collected in EDTA anticoagulant tubes from each subject, and DNA was extracted with a Dneasy Blood & Tissue DNA Extraction Kit (Qiagen, #60606) by following the manufacturer's instructions. The quality of the DNA was validated using an Invitrogen Qbit spectrophotometer.

4.3. FastTarget Target Gene Capture

Primers for sequencing the ACTN4 target region were designed based on a template of the standard human genome and selected/optimized to obtain a clear, single band. Thirty-four pairs of optimized primers (Table 3) were mixed thoroughly into multiplex PCR primer panels according to the protocol, and the standard human genome was used for quality control. Primers with index sequences were used to introduce specific tag sequences compatible with the Illumina platform to the ends of the library by PCR amplification. The reaction used an 11-cycle PCR program to minimize the propensity of products. Amplified products were mixed with equal amounts of buffer and were tapped to obtain the final FastTarget sequencing library. The length of the fragments was determined using an Agilent 2100 Bioanalyzer. After quantification of library molarity, high-throughput sequencing was performed through the Illumina Miseq platform in a 2 × 150 bp/2 × 250 bps double-end sequencing mode to obtain FastQ data.

Table 3

A list of gene locus primers and fragment lengths obtained. For the FastTarget Target gene capture method, 34 pairs of primers were used. Each pair of primers was able to achieve a single and clear PCR product.

Target gene	Forward primer	Reverse primer	Product (bps)
ACTN4_1_1	5′-CGCGGCCTTGGTGCCTTTTCT-3′	5′-ACTGGTTCGCCGCGTGGTAGTCC-3′	239
ACTN4_1_2	5′-AGCTGAGGCGGGAGCGGACA-3′	5′-CCCGGGCCCCCTCAGAAAAG-3′	264
ACTN4_2	5′-GCTGCGGTTCTCCTGAGGT-3′	5′-GCTGTGGCAGAGCACCTGT-3′	275
ACTN4_3	5′-TGCTTTTGGAGAACAGAGGAGACT-3′	5′-GTTGTGCTTCAGAGCCTAAAGGTC-3′	290
ACTN4_4	5′-GGAGGAGCCTCACTCTGGTTTTA-3′	5′-GTGAGTGACCCCAAGGAAACAG-3′	261
ACTN4_5	5′-TGGGCTGAGTTCTGAGGGTTTAT-3′	5′-TCTCACAGACCACGACAAAAACA-3′	248
ACTN4_6	5′-CAGACTGCAGTGAATGGGAATTAGT-3′	5′-CGGAGTTAGGGGGTCAGACAG-3′	249
ACTN4_7	5′-GGCTGAGAACTGCCTGAAGAAA-3′	5′-GAAGCACAGTGGTGGCTGAAC-3′	252
ACTN4_8	5′-CCCGTGGATCCCAGTGAGT-3′	5′-CCGTCTGCAAGAGAAATGAGGT-3′	258
ACTN4_9	5′-CCTCCCTGCGTCTTTCACTCT-3′	5′-CAGGGTCAGTCTGTGTGGTGTG-3′	280
ACTN4_10	5′-CTCCTTCCCCTCTGTGAGGAGT-3′	5′-CCTCTGGCTGAGGATAATGAGGT-3′	264
ACTN4_11_1	5′-TAGCAGGAATCGTGGAGAAGTTG	5′-CGGTAGTCGCGGAAGTCCT-3′	262
ACTN4_11_2	5′-GCCCCAAAAGACTATCCAGGAG-3′	5′-AAAGATTACGCTGGCCAAACTG-3′	272
ACTN4_12	5′-CCCTGGGTGCCTCCACTT-3′	5′-ATGCATGCCTGAGAGACAGGAG-3′	274
ACTN4_13	5′-GACAGCCCCTCCAGACTCCT-3′	5′-TGGTGAGAGCCAGGTGATGATA-3′	263
ACTN4_14	5′-GGGTCCAATCCATCTAGCCACT-3′	5′-GGAGCTCACAGGTCTGGACACTA-3′	260
ACTN4_15	5′-CCTCCTGCTCACATACTGACCTG-3′	5′-CACAGAGGCTCTTGGGAAGATG-3′	246
ACTN4_16_1	5′-ATCGTCCATACCATCGAGGAGA-3′	5′-TCCCACTTGGAGTTGATGATTTG-3′	285
ACTN4_16_2	5′-CTCAGCCCATGACCAGTTCAA-3′	5′-CTGCACCTGGCAGAGGAGAC-3′	272
ACTN4_17	5′-CAACTCCAAGTGGGAGAAGGTG-3′	5′-CTTGGAACCTTCTCAGCTCTGTG-3′	287
ACTN4_18_1	5′-CTCCTCCAGGTGGTCAGTGG-3′	5′-GCACCTCCATGGTATAGTTGGTGT-3′	251
ACTN4_18_2	5′-GAGCCACCTGAAGCAGTATGAAC-3′	5′-CCAAAGTGCTGGTCTCTTCAATAA-3′	275
ACTN4_19	5′-TGAACCACGGTGAGGACAGTT-3′	5′-CAGATGCAGAGACGAAGGTGTG-3′	286
ACTN4_20	5′-CTAACTCTGTGTTTCCCTCCCCTAC-3′	5′-GGCGAGGGGAGAAAGAGAGA-3′	282
ACTN4_21	5′-GGCCCCTCTTGCCTACTCTG-3′	5′-CTCGGGCGGAGGAGTGTC-3′	259
ACTN4_22	5′-CCCCTGCCCCACTAAATGTC-3′	5′-ACACACTGGCCCCCTCAG-3′	290
ACTN4_23_1	5′-GCATGGGGGCTGGCGAGAGG-3′	5′-GTCGGGGGTGTTGGGTCAGGTCTC-3′	287
ACTN4_23_2	5′-TGCCCGGTGCCCTCGACTACAA-3′	5′-AGGTTGGGGAGACTTGGGGCCA-3′	279
ACTN4_23_3	5′-CTCTGTATCTATGCAAAGCACTCTCTG-3′	5′-AGGGACCTCAGAGCAAAGGAAGA-3′	286
ACTN4_23_4	5′-GGGATGCCTCACCACACC-3′	5′-GGATGGGGTGCGGTTCAG-3′	287
ACTN4_23_5	5′-CACTTGCCATTGCCAGGAGA-3′	5′-ATCCGTAAGTTAATAAAGTAAATAGTAATTCTCTGA-3′	288
ACTN4_23_6	5′-TTGTCTGGCCTCACRTGTCT-3′	5′-GGGCAGAGAATCGGCTATGT-3′	257
ACTN4_23_7	5′-TAGCAACRTATCTCTGCCGTCTCTC-3′	5′-CCAGAGGGTGGTTTATCCAGAA-3′	245
ACTN4_23_8	5′-TGATGCTCCTCCGGGTCT-3′	5′-GCTCTGCCCTGGCTCTCCT-3′	281

4.4. Bioinformatic Analysis

By using Burrows-Wheeler Aligner (BWA) software (http://bio-bwa.sourceforge.net/) [8], the sequencing data were compared with the UCSC hg19 reference genome, and the GATK standard program was used to validate the preliminary comparison results obtained by the BWA software. The advanced comparison identified SNV/InDel was used to check the accuracy of the sequences. The SNV/InDel of each sample was determined by using the VarScan software and the GATK HaplotypeCaller software, respectively. The SNV/InDel determined by the abovementioned two detection schemes was compared, and all the samples were combined accordingly. SNV/InDel loci were compared with the dbSNP, thousands of human genomes, ESP6500, ExAC03, ExAC03_EAS, genomAD, and Hrcr1Kaviar_20150923 databases by ANNOVAR to evaluate the frequencies, functional characteristics, conservation of these sites, and pathogenicity and to confirm the most significant SNV/InDel sites for the data obtained.