Mutational screening of inverted formin 2 in adult-onset focal segmental glomerulosclerosis or minimal change patients from the Czech Republic

Podocytes are highly specialized cells with a unique and highly differentiated cytoarchitecture. Their main functions include: support of the glomerular capillaries, synthesis of glomerular basement membrane (GBM), and regulation of permselectivity. Any damage to the podocytes is usually characterized by nephrotic syndrome and (if persistent) it may progress into end stage renal disease (ESRD). Minimal change disease (MCD) and focal segmental glomerulosclerosis (FSGS) are common causes of nephrotic syndrome.

The major pathological feature of both MCD and FSGS is increased permeability of the glomerular capillary filter. Under light microscopy the glomeruli of patients with MCD have a normal size and cellularity and show no evidence of segmental sclerosis; whereas electron microscopy demonstrates the same podocyte foot processes effacement as in glomeruli affected by FSGS. Whereas the outcome of MCD in terms of long-term preservation of renal function is good; by comparison, FSGS often leads to ESRD. We still have only limited insight into the pathogenesis of both MCD and FSGS. Moreover, both the etiology and histology of FSGS as well as of MCD may be very variable, including: viral infections, toxic agents, adaptive structural-functional responses, and mutations in the genes encoding proteins specific for podocytes (or highly expressed in them) [1–3], such as: ACTN4 [4], INF2 [5], and TRPC6 [6].

Inverted formin 2, encoded by INF2, is a member of the formin family of proteins, which regulates the dynamics of the actin cytoskeleton. It not only accelerates actin polymerisation, but also filament depolymerisation. INF2 is expressed in many tissues such as: kidney, liver, heart, skeletal muscles, and placenta. In renal podocytes, it plays a key role in the function and structure of the actin cytoskeleton. The main regions of INF2 are: the N-terminal diaphanous inhibitory domain (DID), the formin homology FH1 and FH2 domains, and the terminal diaphanous autoregulatory domain (DAD) [7, 8]. Most of the reported mutations are localized in the DID domain, which probably functions as a regulatory domain of the polymerisation and depolymerisation of the actin filaments. The previous experiments demonstrated that substitutions in INF2 caused an abnormal distribution of the slit diaphragm proteins nephrin and podocin, and dysregulation of the podocyte cytoskeleton; suggesting its involvement in the pathogenesis of the autosomal dominant form of FSGS [5, 9]. Mutations in INF2 are not only connected with FSGS, but also with Charcot-Marie-Tooth neuropathy [10].

In this study, we present a mutational analysis of INF2 (whose mutations are frequent causes of FSGS in adults) in patients with FSGS or MCD as well as in a group of patients with a positive family history of ESRD of unknown etiology, using a high resolution melting method (HRM) and Sanger sequencing.

Two mutations in INF2 (p.Arg218Gln and p.Arg214His) already known to be responsible for the familial form of FSGS [5] were detected in two different families with a positive family history for FSGS, or ESRD of unknown etiology. Both of the mutations found are localized in exon 4, which is involved in the highly conserved DID region, and that is situated at the N-terminus of the INF2 protein.

The most frequent substitution in INF2 (p.P35P) located in exon 2, had been found in all patients, mostly in the homozygous state (allele frequency = 97.4%). Exon 21 of INF2 seemed to be a very variable region because of the high number of changes identified. However, these substitutions probably did not influence the phenotype of the patients included in our study.

Family no. 133 (INF2, p.Arg218Gln)

This family (see Fig. 1a), with a positive family history of ESRD of unknown etiology, and with mild proteinuria (affecting two brothers, their father, aunt, and cousin), had already been identified as one described by the causal missense mutation p.Arg218Gln (c.653 G > A) in INF2, with a damaging effect on podocyte function [5]. The substitution was first detected in the two brothers who had already developed ESRD. Afterwards, the same mutation was found in their father and cousin. The ages of the brothers at ESRD were 27 and 31 years, respectively; while their father developed ESRD at age 57. For that reason, in the case of these two patients, we presume the co-action of more factors, such as other substitutions in INF2 or some influence from the environment. The younger affected brother as well as his healthy mother also had the SNP p.Pro528Ser (c.1582 C > T) in INF2. Additionally, we found other different intronic and exonic same-sense substitutions; however, there is not any hypothesis of their effects on the phenotype.

Family no. 50 (INF2, p.Arg214His)

Also identified was the known mutation p.Arg214His (c.641 G > A), with a proven damaging effect [5], in a 51 year old woman with FSGS (from a renal biopsy) and with a positive family history (her father had ESRD at age 33, and an uncle at age 30) (see Fig. 1b). Her two daughters were healthy. At her age, this patient has only mild chronic renal insufficiency with moderate proteinuria (1-3 g/24 h), although she suffers from diabetes mellitus, hypertension, and obesity (BMI = 35.58). The biopsy sample showed a perihilar variant of FSGS, with a mild increase of the mesangial matrix (see Fig. 2). This type of injury represents a common part of the so-called secondary FSGS, which is mediated by adaptive structural-functional responses, and frequently is associated with obesity of all of grades with a BMI ≥ 30.0. Even though the renal biopsy looked like secondary FSGS, a mutational analysis was done because of the positive family history of ESRD of unknown etiology.

Several genes highly expressed in podocytes (e.g., ACTN4, INF2, and TRPC6) had previously been reported to be a cause of familial and sporadic autosomal dominant forms of FSGS. Herein, we present the molecular genetic analysis of INF2 (using the HRM method and Sanger sequencing) performed on a large cohort of patients with FSGS/MCD (FSGS = 77 patients, MCD = 32 patients), as well as in a group of patients (6) characterized by a positive family history for ESRD in combination with advanced CKD/ESRD at the time of diagnosis. These types of analyses can lead to the identification of new causal mutations; therefore, they are very important in clinical practice for the diagnosis of patients.

The INF2 protein, belonging to the formin family, plays a key role in the influence of the actin cytoskeleton dependent processes in podocytes because of its capability to accelerate both the polymerisation and depolymerisation of actin. The protein consists of three main parts including: the DID domain, two FH domains, and the DAD domain. Previous studies have demonstrated the abnormal distribution of the detected mutations; all of them having been identified in exons encoding the DID domain, suggesting its crucial role for the protein function. The DID domain appears to be important for actin cytoskeleton regulation [7, 8].

All identified substitutions in the exons were determined to be disease causing candidates if they: (a) segregated with the disease (in the case that other relatives were also affected, and we had samples of their DNA); (b) were predicted to be damaging by the Mutational Taster program [12], PolyPhen-2 [13], PROVEAN [14], and SIFT [15]; (c) were not carried by both healthy family members (whose DNA samples we had) as well as the healthy controls; and (d) were not present in control chromosomes in the db SNP and 1000 Genomes project.

We identified the mutation p.Arg218Gln in INF2 with a damaging effect on the podocyte function in two brothers from a family with the positive history for ESRD. This mutation was first described by Brown et al. in a large family with the familial form of FSGS [5]. Both brothers quickly progressed to ESRD, aged 27 and 31, respectively.

The other detected casual mutation in INF2 was p.Arg214His, also first identified by Brown et al. [5]. The mutation was found in a woman with a perihilar variant of FSGS and simultaneously suffering from diabetes mellitus. Although the family history of the proband was positive for renal diseases (father and uncle), her two healthy daughters had refused the opportunity for genetic testing. The incidences of this variant in our analysis were 0.9% (the whole cohort of patients), 1.3% (only the cohort of FSGS patients), and 5.6% (the cohort of FSGS patients with a positive family history), respectively. Both identified casual mutations are localized in the DID domain of INF2, which is necessary for normal protein function because it is involved in actin cytoskeleton regulation.

The mutational analysis of INF2 also identified other different exonic changes and intronic substitutions that seemed to have no effect on the phenotype of the patients included in this study.

The most frequent substitution in INF2 was p.P35P within exon 2, and with an allele frequency of more than 97%. This incidence corresponds with the results available in the ExAC Browser (http://exac.broadinstitute.org/) and gnomAD Browser (http://gnomad.broadinstitute.org/) databases, confirming the sensitivity of HRM. Other highly frequent changes in INF2 were p.D1022D and p.D880D, with allele frequencies of 68% and 60%, respectively.

The HRM analysis was a very effective method for mutational screening in such a large cohort of patients. It proved to be of sufficient sensitivity in the detection of single nucleotide substitutions.

According to previous studies focused on INF2, it appears that this gene is responsible for a high percentage of patients with familial FSGS [5, 17, 18]. Even though the proportion of mutations in sporadic FSGS is significantly lower [9, 18], we included both familiar and idiopathic patients with FSGS. Although the incidence of pathogenic mutations was lower than our assumptions at the beginning of the study, we established the method for the molecular analysis of INF2 and analyzed a large cohort of patients with FSGS/MCD; confirming that MCD cases are less likely to harbor deleterious genetic variants in those genes implicated in FSGS [4–6, 19]. This was the first molecular genetic analysis focused on INF2 in the Czech Republic. A mutational analysis of INF2 should be performed in all patients with a positive family history of FSGS or unknown ESRD with an autosomal dominant inheritance. Proteinuria is frequently mild or moderate, the patients are asymptomatic, and advanced renal insufficiency is already present, so a renal biopsy is not indicated in many cases.

In conclusion, pathogenic mutations of INF2 are not a frequent cause of FSGS nor of MCD in the Czech Republic. The INF2 mutations must be taken into account; not only in patients with FSGS and a positive family history, but also in patients with ESRD of unknown etiology and a positive family history of ESRD in other generations. The HRM method seems to be a reliable and cheaper method than is direct sequencing for an analysis of the concrete gene in large cohorts of patients.