Novel Mutations in NPHP4 in a Consanguineous Family With Histological Findings of Focal Segmental Glomerulosclerosis

Recent Advances in Molecular Genetics: Tools in the Diagnosis of Hereditary Kidney Disease 

Family history is a critical component of the evaluation of a patient with anatomic or functional kidney abnormalities. In many cases, careful history taking can make the correct diagnosis readily apparent and obviate the need for exhaustive workups. However, the absence of a clear family history does not necessarily imply absence of the phenotype in other family members, as certain manifestations of disease, eg, proteinuria, will not always manifest symptomatically, but require specific testing or measurement for detection. This information will help guide the clinician in obtaining relevant investigations to establish the diagnosis and assist in determining the mode of inheritance. Although not exclusively the case, in general, families with a single generation of affected individuals most likely will have a recessive mode of inheritance, whereas involvement of multiple generations implies a dominant inheritance pattern. When only male members of families are affected, this suggests an X-linked inheritance pattern. There are instances when the mode of inheritance may not be entirely clear from the pedigree. Figure 1 outlines a general stepwise approach to the diagnosis of hereditary kidney disease.

View Large Image Download to PowerPoint

Figure 1. An approach to the diagnosis of hereditary kidney disease. Abbreviation: OMIM, Online Mendelian Inheritance in Man.

In many cases, a specific phenotype will be established after careful history, examination, and appropriate investigations, including urinalysis, laboratory and imaging studies, and, if appropriate, renal biopsy, in all affected individuals. In patients with suspected familial focal segmental glomerulosclerosis (FSGS), it is important that such secondary causes of FSGS such as obstruction, vesicoureteral reflux, and infection be ruled out before the diagnosis of familial FSGS is made.

If there is suspicion of mutations in a specific candidate gene based on the clinical information, mutational screening can be used to confirm the diagnosis, eg, sequencing of the NPHS1 gene in an infant with nephrotic syndrome, or sequencing the NPHP1 gene in a patient with suspected nephronophthisis (NPHP).

There will be instances in which there is a clear family history of kidney disease, suggesting Mendelian inheritance, for which the majority of disease-causing genes are unknown, eg, familial FSGS. In these circumstances, whole-genome scanning or genotyping may be helpful to establish a likely genetic region segregating with the disease.

Recent years have witnessed remarkable advances in genetic molecular biology, providing scientists with powerful analytic tools to study genetic disease. Mutational analysis of novel candidate genes is guided by our rapidly growing knowledge of cell biology. Although identification of proteins interacting with known disease-associated genes, proteins integral to biochemical and signaling pathways, and structural proteins provide insight into cell biology, little is known about the role of the vast majority of proteins in human disease. Thus, sequencing specific candidate genes in case-control studies, together with biological evidence of the effect of variants on protein structure, localization, and function, can provide compelling evidence of association with disease.

With the recent development of ever-denser single-nucleotide polymorphism (SNP) chips, the entire human genome can be more easily scanned to discover loci associated with disease. Whole-genome SNP analysis can be used for linkage studies and homozygosity mapping. This is a high-throughput method, and results can be available within days, making it extremely efficient. The most popular technologies for these analyses are the high-density SNP chips made by the companies Illumina (San Diego, CA) and Affymetrix (Santa Clara, CA). This technology is rapidly evolving. The 10K, 100K, and, most recently, 500K SNP chips made by Affymetrix, for example, genotype the respective number of SNPs throughout the genome. For the 10K SNP chip, this implies an intermarker distance of approximately 300 kilobases (kb) compared with every 6 kb for the 500K chip.

Before the advent of SNP chips, microsatellite repeats were the most widely used marker type for linkage studies. The resolution of microsatellite maps typically used provides coverage of the genome at approximately every 7 megabases. The development of new SNP chips has enabled much denser coverage of the genome. However, microsatellite markers can provide information for multiple (as opposed to two) alleles at each locus, which has significant advantages in some genetic studies.

Case Vignette 

We describe a consanguineous Filipino family with a history of proteinuria, end-stage renal disease (ESRD), and histological diagnosis of FSGS. Patient numbers correspond to the pedigree numbers in Fig 2.

View Large Image Download to PowerPoint

Figure 2. Family pedigree. The 10K single-nucleotide polymorphism (SNP) genotype calls identified a 5.8865-Mb homozygous region (5.0450 to 10.9315 Mb) on chromosome 1p in affected individuals. Sequencing the NPHP4 gene, located within this region, showed 2 novel homozygous missense sequence variants in exons 18 and 21 that segregated with disease in this family. At position c.2314, in exon 18, the reference sequence for the 2 alleles is C/C. The homozygous c.2314C→T variant leads to a p.R772C amino-acid substitution. At position c.3037, in exon 21, the reference sequence for the 2 alleles is G/G, the homozygous c.3037G→A variant leads to a p.E1013K amino-acid substitution. Black box, affected; white box, unaffected; hatched box, unknown.

Patient 11 

Patient 11 is a 31-year-old man who presented with chronic kidney disease (serum creatinine level, 5 mg/dL [442 μmol/L]) at age 25 years. He rapidly progressed to ESRD within 1 month. He has a history of proteinuria (24-hour urine protein excretion measured in 2000 was 1,200 mg) and microscopic hematuria of unknown duration. He recalls admission to the hospital during childhood for poor growth. He denies a history of polydipsia, polyuria, enuresis, recurrent urinary tract infections, or vesicoureteral reflux.

In addition, patient 11 reports jaundice of unclear cause since infancy. Serum total bilirubin level was increased at 5.1 mg/dL (87.2 μmol/L) at the time of our evaluation in 2000, with normal aspartate aminotransferase, alanine aminotransferase, and alkaline phosphatase levels. An ophthalmological examination performed as workup for poor vision showed a diagnosis of retinitis pigmentosa. Imaging studies at the time of presentation included a renal ultrasound, which showed bilateral small 9.3-cm echogenic kidneys with three 0.5- to 3-cm simple cysts. The left upper pole collecting system was dilated. No comment was made on bladder appearance or size.

A renal biopsy was performed soon after presentation. Light microscopy showed 13 glomeruli, of which 8 were globally sclerosed and 2 showed segmental sclerosis (Fig 3). There was moderate tubulointerstitial fibrosis and tubular atrophy. Immunofluorescence microscopy results were unremarkable. Electron microscopy showed focal podocyte foot-process effacement, and there were no deposits (Fig 4).

View Large Image Download to PowerPoint

Figure 3. Renal biopsy tissue from patient 11. Masson trichrome stain shows global (arrow) and segmental (arrowhead) glomerulosclerosis. There is moderate interstitial fibrosis and tubular atrophy. No cysts are noted, and tubular basement membranes are unremarkable. (Bar = 200 μm.)

View Large Image Download to PowerPoint

Figure 4. Electron microscopy in patient 11 shows ultrastructural features of preserved podocyte foot processes (arrowheads) in some areas of the glomerular tuft and foot-process effacement (arrows) in other areas. This finding is suggestive of podocyte injury from a secondary cause, but also may occur in some hereditary forms of focal segmental glomerulosclerosis. (Bar = 2 μm.)

Patient 11 has 5 siblings who have ESRD, 2 of whom died, 1 at age 55 years (patient 19) and the other at age 40 years (patient 20).

Application of Molecular Genetics Techniques to Determine the Diagnosis 

We describe a consanguineous family with an autosomal recessive mode of inheritance of ESRD. The initial limited history, imaging, and histological evaluation (late presentation with ESRD made diagnosis difficult and histological findings correlated with FSGS) suggested a diagnosis of familial FSGS. To identify a disease-associated locus in this family, we obtained a whole-genome scan by using the 10K SNP chip (the only density available at the time) for homozygosity mapping, ie, identifying chromosomal regions that are homozygous in affected individuals, but not in those unaffected.

Following whole-genome scan using Affymetrix 10K SNP chips in 4 affected siblings, 3 unaffected siblings, 1 sibling of unknown status (person 13) with mild proteinuria (urine albumin, 208 mg/g of creatinine), and 2 controls, a 5.8865-Mb homozygous region (5.0450 to 10.9315 Mb) was identified on chromosome 1p in affected individuals (Fig 2). We were surprised that the NPHP4 locus lies within this region (chromosome 1: 5,857,136 to 5,986,797). On further probing of clinical history and the finding of this family’s extrarenal manifestations of hepatic and ocular involvement, we were suspicious that the correct clinical and pathological diagnosis may be NPHP. We therefore sequenced all 29 coding exons of the NPHP4 gene in all family members for whom DNA was available and 12 controls (4 whites, 1 Chinese, 1 Korean, and 6 Africans).

We detected 2 novel homozygous missense sequence variants in exons 18 and 21 that segregated with disease in this family (Fig 2). These are not previously reported polymorphisms. One individual of unknown status (person 13) and the mother (person 10) were heterozygous at both these loci, as were 4 unaffected individuals. All 12 controls had the normal expected sequence.

The mother does not have a history of kidney disease. She is 77 years old. Because she is heterozygous for the abnormal variant at both loci in question, it follows that 1 disease-causing allele was inherited from her. She was married to her first cousin, who died at age 75 years. He had ESRD of unclear cause. However, given the late age of onset of ESRD and history of diabetes and hypertension, we suspect that he had diabetic or hypertensive nephropathy, rather than NPHP, which typically results in ESRD by early adulthood. We were unable to obtain a DNA sample from him. However, because the affected children are all homozygous for the variant, 1 unaffected child (person 12) is homozygous wild type, and his partner was heterozygous, it follows that he must be heterozygous for the disease-associated mutation at this locus.

Without biological or functional studies, it is difficult to determine which of these 2 homozygous missense mutations is disease causing or whether both together have a cumulative effect. Based on the nature of the amino-acid substitution, ie, the substitution of a hydrophilic, charged arginine to hydrophobic cysteine, and the very high degree of conservation of this sequence between species, it is likely that the R772C variant in exon 18 is the more significant by way of affecting protein conformation and function. The sequence variant in exon 21 leads to a substitution of lysine for glutamate, which are both hydrophilic and charged; therefore, this change is less likely to have an effect on protein structure or function. This region is also less conserved.

We performed TaqMan SNP Genotyping Assays (Applied Biosystems, Foster City, CA) for both variants on 73 control samples (55 whites, 14 African Americans, 3 Hispanics, and 1 Korean). None of the controls had the abnormal allele, indicating that these variants are not common in the general population. Additionally, because these variants are located in NPHP4, a known disease-causing gene, this provides significant supportive evidence for association with disease and the diagnosis of NPHP in this family.

Nephronophthisis 

ESRD is a significant cause of morbidity in the United States and worldwide. According to the US Renal Data System 2006 Annual Data Report, between May 1995 and June 2004, almost 7,600 children began treatment for ESRD in the United States. Of these, 3,400 had cystic kidney disease as the cause.1 Although NPHP is a rare kidney disease with an estimated incidence of 1:1,000,000 inhabitants in the United States, it is the most frequent genetic cause of ESRD in children and young adults, accounting for 6% to 10% of all ESRD and 15% of all kidney transplantations in children.2, 3, 4

NPHP is a form of tubulointerstitial disease characterized histologically by renal tubular cysts, interstitial fibrosis, and tubular basement membrane disruption.5 This is a genetically and clinically heterogeneous disorder. To date, disease-causing mutations in 6 genes, NPHP1 through 6, have been identified by using positional cloning.6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16

The clinical phenotype consists of renal manifestations with or without extrarenal involvement. There is impaired urinary-concentrating ability and salt loss with progression to ESRD, usually within the first 2 to 3 decades of life. Extrarenal involvement includes ocular abnormalities, hepatic fibrosis, situs inversus, cerebral involvement, and skeletal abnormalities. There appears to be little genotype-phenotype correlation in that mutations in a single gene manifest with varying extrarenal phenotypes, and mutations in several different genes may lead to a similar extrarenal phenotype (Table 1).

Table 1. NPHP Genes and Associated Phenotypes

	NPHP Gene	Gene Map Locus	Protein	Clinical Form of NPHP	Extrarenal Phenotype	References
	NPHP1	2q13	Nephrocystin	Juvenile NPHP	Retinitis pigmentosa, oculomotor apraxia, Joubert syndrome	4, 14, 17, 18, 19
	INVS/NPHP2	9q31	Inversin	Infantile NPHP	Hepatic involvement, situs inversus	8, 20
	NPHP3	3q22	Nephrocystin 3	Adolescent NPHP	Retinitis pigmentosa, hepatic fibrosis	9, 21
	NPHP4	1p36	Nephrocystin 4	Juvenile NPHP	Retinitis pigmentosa, Leber congenital amaurosis	12, 13, 22
	IQCB1/NPHP5	3q21.1	Nephrocystin 5	Variable	Retinitis pigmentosa	7
	CEP290/NPHP6	12q21.3	Nephrocystin 6	Variable	Retinitis pigmentosa, Joubert syndrome 5	16, 23

Abbreviation: NPHP, nephronophthisis.

The diagnosis of NPHP usually can be made by means of clinical features of polyuria, polydipsia, and poor growth; radiological findings of renal cystic disease; and classic histological features of tubulointerstitial fibrosis, inflammation, tubular cysts, and basement membrane changes. However, these findings may not always be present, especially when patients present with established ESRD.

The initial symptoms of polyuria and polydipsia are nonspecific, and patients may not recognize these as abnormal. Growth retardation may be the only indication of an underlying pathological process. Signs that typically alert one to renal disease, such as hypertension, active urinary sediment (hematuria with presence of cellular casts), proteinuria, and abnormal serum electrolyte or creatinine levels, are absent.

Later symptoms usually are related to the development of chronic renal failure and associated uremia, anemia, and metabolic derangements. NPHP almost uniformly progresses to ESRD by the teenage years or early adulthood, with the rate of progression determined by the genetic defect. Extrarenal manifestations may occur, although the mechanisms are unclear.

With late presentation after onset of ESRD, as in the reported family, the clinical, radiological, and histological features may be misleading. Renal ultrasonography is too insensitive to appreciate small cysts (which may not always be present). The common finding of increased renal echogenicity is a nonspecific finding in patients with a multitude of acute and chronic injuries, and kidney size may vary from normal to small.24, 25, 26 Proteinuria, sometimes heavy, is nonspecific and may be seen with the development of secondary forms of glomerulosclerosis. Likewise, histological evidence of NPHP may be masked by the development of chronic tubulointerstitial changes and secondary glomerulosclerosis.

Pathogenesis of NPHP 

To date, mutations in 6 genes have been described as causing NPHP (Table 1). Identification of these genes by means of positional cloning continues to greatly advance our understanding of the molecular and biochemical processes involved in the pathogenesis of cystic kidney disease.

The majority of patients with NPHP (∼85%) are accounted for by mutations in the NPHP1 gene.14, 27 Mutations in IQCB1 appear to be the most common cause of Senior-Loken syndrome.7 In patients with NPHP, Senior-Loken syndrome, or ocular motor apraxia, 35 sequence variants in the NPHP4 gene were described to date, with variable inheritance patterns and extrarenal phenotype.3 Although some of these variants may be associated with disease, the role of other variants, especially heterozygous sequence variants, is less clear.

The mechanisms by which alterations in the respective encoded proteins of NPHP genes cause disease and why some patients develop extrarenal manifestations while others do not is incompletely understood. Cilia, which in the kidney act as mechanosensory flow detectors, clearly have an important role in the pathogenesis of cystic kidney disease, evidenced because all the genes mutated in polycystic kidney disease and NPHP are expressed in cilia of renal tubular epithelial cells.8, 9, 13, 28 Cilia have been shown to be important in the development of left-right orientation, and defective function leads to situs inversus.

NPHP most commonly is caused by large homozygous deletions in the NPHP1 gene, which manifests as juvenile NPHP with or without extrarenal manifestations of retinitis pigmentosa, oculomotor apraxia, and Joubert syndrome.4, 14, 29 NPHP1 encodes a ciliary protein nephrocystin, which interacts with 3 other NPHP proteins, inversin, nephrocystin 3, and nephrocystin 4, to form a large complex. It also interacts with such proteins as filamin, tensin, and tubulin, involved in cell-cell and cell-matrix signaling, suggesting that nephrocystins have a role in the integrity and architecture of renal tubular epithelial cells.8, 9, 13, 28, 30

Nephrocystins localize to the cytoplasm of renal tubular epithelial cells at the base of the cell cilium (basal body). Microtubules in the cilium originate from the basal body, which also functions as the spindle-organizing center in mitosis.31 CEP290 localizes to the centrosome.16

Thus, to date, data suggest that nephrocystins are important in several cellular processes, including ciliary function, cell interactions with extracellular matrix and adjacent cells, cytoskeletal organization, and cell-cycle regulation. Exactly how these various functions cause cystic disease or interstitial fibrosis is unclear.

The relationship between NPHP and retinal abnormalities may be explained in part by the interaction of nephrocystin 4 and retinitis pigmentosa guanosine triphosphatase regulator–interacting protein 1 (RPGRIP1). Mutations in RPGRIP1 cause Leber congenital amaurosis. Both proteins colocalize in the retina, and mutations in either NPHP4 or RPGRIP1 disrupt their interaction.32

Pathological Characteristics of FSGS: Primary Versus Secondary 

Almost any chronic kidney injury, irrespective of cause, can lead to secondary glomerular scarring (glomerulosclerosis), which is identified histologically on light microscopy by segmental or global fibrosis of a number of glomeruli. This injury, when secondary, is caused in part by hyperfiltration in surviving glomeruli when the functional nephron mass is reduced. When significant chronic kidney disease has developed, it may not be possible to determine the underlying cause from the biopsy findings. In some cases, the histological finding of FSGS may be misinterpreted as representing glomerulosclerosis resulting from primary podocyte injury. This may have occurred in one of the cases described in this report. This emphasizes the importance of ultrastructural examination by using electron microscopy in discerning the role of podocyte injury in the glomerular scarring process.

Ultrastructural examination of biopsy material is essential to distinguish between processes that have podocyte injury as a root cause (eg, primary FSGS, minimal change disease, and familial FSGS) and processes in which podocyte injury and glomerulosclerosis appear subsequent to damage in other compartments of the kidney (ie, secondary glomerulosclerosis). In patients with primary FSGS or minimal change disease, ultrastructural examination typically shows diffuse podocyte foot-process effacement, along with other signs of podocyte injury, including cytoplasmic vacuolization and microvillous change. When glomerulosclerosis is secondary to other forms of kidney damage, signs of podocyte injury are focal and segmental and most likely to be seen in glomeruli undergoing senescence. However, the degree of foot-process effacement can be variable and thus cannot be used reliably to distinguish primary from secondary FSGS.33

The familial forms of FSGS also appear to develop as a result of podocyte dysfunction and show ultrastructural features of podocyte injury on biopsy, similar to those seen in patients with primary FSGS.34 However, in some of these hereditary diseases, particularly the autosomal dominant and later-onset forms of familial FSGS, like that caused by mutations in ACTN4, the podocyte injury may be less severe and often shows a focal and segmental distribution, thus making the distinction from secondary FSGS difficult. Ultrastructurally, some familial forms of FSGS may show unique findings that could aid in their recognition and distinguish them from secondary FSGS. Examples include the loss of slit diaphragms associated with homozygous NPHS1 mutations and the occasional observation of cytoplasmic electron-dense aggregates in heterozygous ACTN4 mutations.35, 36 In some hereditary diseases, the glomerular lesion has not been fully characterized because of the limited availability of kidney biopsy material from these patients. As studies of genotype/phenotype correlation continue, distinguishing histological and ultrastructural features will likely become more apparent.

Summary 

This report emphasizes the importance of genetic testing, when possible, in establishing a diagnosis for the purpose of providing appropriate genetic counseling. We discuss an approach to the use of current tools in molecular genetics to assist in making a diagnosis or narrowing down a chromosomal region of interest in Mendelian kidney disease.

On initial evaluation, the only diagnostic clue to the cause of progressive kidney disease in the reported family was mild to moderate proteinuria and the histological findings of FSGS. We performed homozygosity mapping to identify the disease-associated locus and identified a region of interest in chromosome 1 that we noted harbored the NPHP4 gene. Therefore, before additional analysis of specific genes, we obtained more complete clinical information to determine whether there were features suggestive of NPHP. We found this to be the case. These features included a history of enuresis, poor growth, and extrarenal manifestations of retinitis pigmentosa and hepatic disease.

Sequencing the coding exons of the NPHP4 gene showed 2 novel missense homozygous mutations that segregated with disease and were not found in controls. Given both the genetic linkage to the NPHP4 gene and the existence of rare nonconservative variants, we believe the diagnosis of NPHP is correct.

Mutations in NPHP4 typically are associated with extrarenal ocular manifestations of retinitis pigmentosa and Leber congenital amaurosis, as well as hearing loss.3 This family also has a history of hepatic involvement, not described previously in patients with mutations in NPHP4.

In the case of proteinuric kidney disease, especially with associated renal failure, clinicians tend to rely heavily on pathological features for diagnosis. We must remember that FSGS is a pattern of injury and not a clinical disease. Any patient with this phenotype warrants careful history, physical examination, and biochemical and radiological assessment to evaluate for causes of secondary disease. When there is a family history suggestive of Mendelian inheritance, diagnoses other than familial FSGS must be considered, especially in the presence of extrarenal manifestations, which are unusual in patients with familial FSGS and nephrotic syndromes. If possible, appropriate genetic studies should be performed to establish the diagnosis, thus enabling prognostication, avoidance of potentially toxic medications, and genetic counseling to the affected family.

Additionally, this report shows how clinicians could use genetic techniques, made more and more accessible and available by the brisk pace of technological advances, for clinical testing, thus increasing the armamentarium of investigational tools at their disposal. These techniques clearly are blurring the lines between clinical testing and research. DNA sequencing technology is becoming less expensive, and scientists already are working toward the goal of sequencing the entire human genome for $1,000 or less.37 When this goal is realized, the implications are momentous and will propel the pace of genetic research by placing vast amounts of data at the disposal of researchers. Not only will this provide insight into human health and disease, but it will also very likely lead to individual tailoring of drug therapy based on one’s genetic makeup and susceptibility to response.