Introduction
Nephrotic syndrome (NS) is defined by significant proteinuria, hypoalbuminemia, and edema. NS can be classified by the response to steroid to steroid-sensitive NS, steroid-dependent NS, and steroid-resistant NS (SRNS). SRNS is defined by an inability to achieve complete remission after a minimum of eight weeks of corticosteroid treatment [
1]. Moreover, SRNS accounts for 40% of adult-onset NS [
2]. NS responds differently to steroids, and this therapeutic response is associated with the prognosis. Focal segmental glomerulosclerosis (FSGS), which is histologically characterized by focal and segmental glomerular sclerosis and foot-process effacement, is clinically characterized by proteinuria and progressive renal failure. FSGS accounts for nearly 20% of the NS in children and adults [
3], and 24.3% and 3.3%–16% of primary glomerulonephritis (GN) in the US and China, respectively [
4]. FSGS etiology is largely unknown. Moreover, it is considered as a podocytopathy, and several monogenic FSGS subtypes have been reported in genetic studies that were primarily focused on familial FSGS. However, mutations in these genes account for only a small portion of the familial FSGS patients. A lower mutation rate of these genes was reported in an Asian population, suggesting that additional genes might be involved in FSGS susceptibility.
Known genes causing adult-onset FSGS/nephrotic syndrome
Most of these pathogenic genes are involved in the maintenance of podocyte structure and function, including protein assembly of glomerular basement membrane (GBM), signal transduction, podocyte skeleton, transcription factor, mitochondrion, and lysosome (Table 1, Fig. 1). Most of these mutations are inherited by autosomal dominant and recessive patterns. Mutations of type IV collagen α3-5 (COL4A3–5), inverted formin 2 (INF2), transient receptor potential cation channel 6 (TRPC6), α-actinin-4 (ACTN4), CD2 associated protein (CD2AP), andARHGAP24 can contribute to FSGS development under dominant forms. However, NPHS1,NPHS2, phospholipase C epsilon 1(PLCE1), myosin 1E (MYOE1), and Nei endonuclease VIII-like 1(NEIL1) can cause NS following autosomal dominant inheritance. Only some mutations of genes can be inherited by X-linked recessive models, such as nuclear RNA export factor 5 (NXF5). The clinical manifestations are various with different inherence patterns. The recessive form of FSGS presents early proteinuria and rapidly progresses, whereas the dominant form of FSGS is usually late onset with non-NS proteinuria.
Type IV collagen is a component of GBM.
COL4A3–5 encodes α chain of collagen IV in GBM; moreover,
COL4A3–5 mutations cause Alport syndrome, thin basement membrane nephropathy (TBMN), and benign familial hematuria. Eighty percent of Alport syndrome cases are inherited by X-linked patterns, 15% are recessive, and 5% by dominant patterns [
15]. Some studies showed that mutations of
COL4A3–5 can be detected in patients clinically diagnosed with Alport syndrome with histological FSGS features. With the development of next-generation sequencing (NGS), our team identified five
COL4A3 gene mutations (p.Cys1616Tyr, p.Gly801Arg, p.Gly118Arg, p.Gly997Glu, and p.Leu737Hi) in five familial FSGS, which have been excluded because of Alport syndrome, and these cases showed segmental thinning in GBM electron micrographs [
16]. At the same time, Gbadegesin
et al.[
5] presented seven families with mutations in
COL4A3 or
COL4A4 from a cohort of 70 families diagnosed with FSGS with nephrotic-range proteinuria. In one family, electron microscopy (EM) showed thin GBM. Gast
et al.[
6] screened 81 patients from 76 FSGS/SRNS families and identified six
COL4A3–5 mutations (eight patients) in six FSGS families and two in sporadic FSGS patients, accounting for 38% of familial FSGS and 3% of sporadic FSGS. Moreover, 26 glomerular genes in 50 patients with FSGS/SRNS were sequenced, and heterozygous mutations in
COL4A3 were identified in three patients, thereby indicating that some
COL4A3 mutations can present aggravating effects on FSGS [
17]. To date, no data has demonstrated the difference of the mutation domain between Alport syndrome and FSGS. Compared with the FSGS patients with mutation of other genes, patients with
COL4A3–5 mutations are likely to be young and manifest increased hematuria and GBM abnormalities. Segmental GBM thinly occurred in most of these FSGS patients; however, no typical EM features of classical Alport syndrome were observed, and only few patients displayed classical manifestation of Alport syndrome, such as hearing loss and lesions in the eyes. Mice with heterozygous
COL4A3 knockout defects exhibited TBMN phenotype or FSGS [
18]. Thus, studying the different
COL4A3–5 mutations that cause different clinical manifestations is necessary.
INF2 is a member of the formin family, which can regulate the actin polymerization and depolymerization.
INF2 mutations account for approximately 3.6%–17% [
7,
8] familial FSGS but not sporadic FSGS.
INF2 mutation in FSGS was first reported in 2010 [
19]. Brown
et al. detected nine missense mutations of
INF2 by linkage analysis and found that
INF2 R218Q and S186P were associated with FSGS; then, it was confirmed in European FSGS families [
8,
20–
22]. Our team identified an
INF2-S85W mutation in a Chinese FSGS family with one of the affected members with IgA nephropathy by genome-wide linkage analysis and subsequent sequencing [
7]. To date, most of the mutations are in diaphanous inhibitory domain of
INF2, leading to the loss of the inhibitory function of INF2 on actin and destroy the interaction with Rho GTPases and mDia. Moreover,
INF2 knockout in zebrafish [
23] showed increased proteinuria, which confirmed the significant role of
INF2 in the kidney. However,
INF2 R218Q knock-in mice showed no significant renal pathology or proteinuria at baseline. In mice with
INF2 R218Q mutation, heparin sulfate perfusion cannot rescue foot process effacement induced by protamine sulfate [
24]. These results indicated that normal
INF2 is required for response and/or recovery from kidney injury. Further studies are needed to support this theory.
TRPC6 is expressed in podocytes and is a component of the glomerular slit diaphragm (SD). TRPC6 is a member of transient receptor potential superfamily, which functions to increase the intracellular Ca
2+. Reiser
et al. [
25] screened 71 pedigrees (49% of the families were of Western European ancestry; 5% of African ancestry, and 27% of Hispanic) with familial FSGS for
TRPC6 by DNA sequence analysis and identified five autosomal dominant FSGS families (7%) with mutations in the
TRPC6 gene. Simultaneously, Rosenberg
et al. discovered a missense mutation (C335A) of
TRPC6 by haplotype analyses [
26]. Our team reported a novel
TRPC6 Q889K mutation in a Chinese late-onset FSGS family [
27].
TRPC6 mutation always caused “gain of function.”
In vitro functional research showed that
TRPC6 Q889K mutation caused increased intracellular Ca
2+ level and activated calcineurin/nuclear factor of activated T-cell pathway [
28], which decreased SD-associated proteins, such as nephrin. On the other side,
TRPC6 interacts with podocin and nephrin.
TRPC6 mutation breaks these connections and damage podocyte cytoskeleton [
25].
NPHS2 encodes protein podocin, which is a 383-amino acid lipid-raft-associated protein localized at the SD. Podocin plays an important role in the maintenance of the structural organization and regulation of GBM by interacting with nephrin, CD2AP, TRPC6, and NEPH1. It was first identified in early onset of familial SRNS 10 years ago, and most of them presented FSGS in pathology [
29]. Tsukaguchi
et al. reported that the rate (R229Q) in late-onset FSGS and sporadic adult-onset FSGS was 23% and 2%, respectively [
30]. Functional studies showed that R229Q decreases nephrin binding to mutant R229Q-podocin.
ACTN4 encodes protein a-actinin-4, which is an actin-filament crosslinking protein. The role of a-actinin-4 has been widely studied, and a-actinin-4 plays an important role in the kidney. In addition to binding to F-actin, a-actinin-4 can interact with multiple proteins, such as cell junction and cell adhesion proteins, activate phosphatidylinositol 4,5-bisphosphate signaling, and act as transcription activator of estrogen receptor-a, retinoic acid receptor, and NF-κB [
31]. Mice deficient of α-actinin-4 exhibited severe glomerular disease [
32]. Late-onset FSGS caused by mutation of
ACTN4 was first reported by the Martin group in three autosomal dominant forms [
33]. Weins
et al. screened
ACTN4 in 141 FSGS families and found five mutations (W59R, I149del, K255E, T259I, and S262I), accounting for 3.5% [
11]. Our team screened this gene in sporadic FSGS patients and found that the mutation in promoter in
ACTN4 caused FSGS among Chinese patients [
9,
34,
35]. The mutation rate was 2.4%. Moreover, a germline mosaicism for the p.Ser262Phe was detected in a patient’s father [
36]. All of the known mutations are located in the N-terminal actin binding domain. Study
in vivo showed that podocyte-specific expression of mutant α-actinin-4 caused proteinuria [
37].
CD2AP has been known as a SD component. Mice with
CD2AP haploinsufficiency displayed glomerular abnormality and had increased susceptibility to glomerular injury similar to FSGS [
38]. Moreover, in podocytes, CD2AP binded to nephrin and podocin and regulated podocyte cytoskeleton. Kim
et al. screened 45 sporadic FSGS patients and found 10 variants. One variant alters
CD2AP expression. Löwik
et al. [
39] found a stop-gain variant in one FSGS patient. This mutation caused decreased F-actin binding function. Gigante
et al. [
40] screened 80 Italian patients with idiopathic NS, and found three missense mutations (c.904A>T, c.1120A>G, and c.1573delAGA).
With the development of the technology of genetic testing methods, mutations of genes, such as
TTC21B,
ANLN,
PAX2,
LMX1B, and
NXF5 causing FSGS, were found. Esposito
et al.[
41] identified
NXF5-R113W mutation in a large Australian FSGS pedigree with progressive heart block (apparent X-linked recessive inheritance) by combination of linkage, haplotype analysis, and whole exome sequencing (WES) approach. Cong
et al.[
42] identified a homozygous missense mutation (p.P209L) in the
TTC21B gene in seven families with FSGS by WES combined with homozygosity mapping. Gbadegesin
et al.[
13] identified a missense mutation R431C in anillin (
ANLN) as a cause of FSGS by linkage analysis and WES approach. They found another variant (G618C) after screening 250 additional families. Barua
et al. [
12] found
PAX2 mutation in a dominant FSGS by exome sequencing.
LMX1B mutation caused familial FSGS without extrarenal manifestations by linkage analysis and exome sequencing [
43]. However, these genes were only reported by few researchers and account for a small part of FSGS.
Gene mutation, histological change, and treatment
Determining a mutation by histological change is impossible; however, some gene mutations display specific clinical characteristics. NPHS1 mutations always show diffuse foot process effacement. LAMB2, WT1,PLCE1,ARHGDIA, and TRPC6 mutation show diffuse mesangial sclerosis. Segmental thin of GBM (EM) can be seen in patients with mutation in COL4A3–5. Moreover, mutation of ACTN4 and ITGB4 leads to expression reduction of α-actinin-4 and integrin b4.
For the different molecular and pathogenesis bases of genetically associated FSGS and SRNS, the manifestation and prognosis are different. Conlon
et al. analyzed 31 patients among eight families with familial FSGS and found that FSGS that occurs in a familial pattern displays a poor prognosis [
44]. Tomero
et al.[
45] studied the members of three generations of a family and found the occurrence of end-stage renal disease. Chen
et al.[
46] enrolled 124 FSGS patients and 124 sporadic FSGS patients and found that familial FSGS patients experience severe pathological changes, low response to treatment, and worse renal prognosis.
Most FSGS/SRNS patients with genetic factors do not respond to steroid treatment; however, potential therapies are available for some gene mutation, such as
TRPC6,
ARHGDIA,
PLCE1, and
ADCK4. Treatment with coenzyme Q10 might be effective to patients with mutation of
COQ2 [
47],
COQ6 [
48],
ADCK4 [
49], or
PDSS2 [
50]. Hinkes
et al. [
51] found that two patients with
PLCE1 mutation responded to treatment with steroids or cyclosporine A.
TRPC6 mutations might potentially respond to the treatment with calcineurin inhibitors [
28]. Furthermore,
ARHGDIA mutation might be responsive to the eplerenone treatment [
52].
Gene detection in diagnosis of genetically associated FSGS/nephrotic syndrome
Mutation screening is significant for the adult-onset SRNS and FSGS with genetic factors. As mentioned previously, these patients always do not respond to common treatment and show poor prognosis. Identification of the causative mutation helps in defining the phenotype–genotype and in choosing an improved therapy.
With the development of sequencing technology, the methods we identified in mutations have changed from single gene Sanger sequencing to high-throughput sequencing. Moreover, high-throughput sequencing saves patients’ money and time. However, for clinical application, considering which way is good for patients is still necessary.
Sanger sequencing is a kind of chain-termination method of DNA sequencing widely used for a long time. Although time and money consuming, it is still one of the most reliable methods to detect the mutation.
NGS, including whole genome sequencing (WGS) and WES, use massively parallel DNA enrichment technology and has been widely used in recent 10 years. WGS can detect almost the whole genome of a patient. WES can detect 330 000 exons of human genome. When the deep was 10×, 78.6% of heterozygosis mutation can be detected, and when it was 30×, almost 100% was found [
53]. Most of the familial FSGS or NS are monogenic diseases, and mutations are located in exons. WES was first used in Miller syndrome and identified
DHODH mutations. WES is more advantageous in identifying mutations compared with WGS. However, when WES was applied, more than one candidate variants can be detected. Methods, such as application of highly stringent genetic criteria, can be used to decrease the variants. Moreover, elaborating and confirming this variant to be a disease-causing mutation is still challenging.
Gene panels can detect almost 30 genes and is a cost-effective method to determine disease mutation. They have been used widely in research. Bullich
et al. simultaneously sequenced 26 glomerular genes in 50 SRNS and/or FSGS patients [
17]. Korkmaz
et al. screened more than 500 prospective SRNS cases by panel sequencing and suggested that ADCK4 mutation might be the third most common hereditary cause of adolescence-onset SRNS [
54]. Besides, Gast
et al. screened 81 familial FSGS by targeted NGS panel covering 39 and found that
COL4A mutations were the most frequent in adult-onset FSGS [
6].
Conclusions
Over the past decades of studies, many causative genes have been found in adult-onset FSGS and NS, and most of these genes are expressed in podocytes. COL4A3–5 and INF2 mutations explain almost half of these patients. Further study still needs to be conducted to discover the other causative genes and illustrate the mechanism of the genes, such as COL4A3–5 and INF2. Considering the poor response to therapy and prognosis, screening these genes in familial FSGS/NS by utilizing and integrating of multiple methods is necessary.
Higher Education Press and Springer-Verlag Berlin Heidelberg