Clinical and pathohistological characteristics of Alport spectrum disorder caused by COL4A4 mutation c.193-2A>C: a case series

Aim To present the pathohistological and clinical characteristics of five Croatian families with Alport spectrum disorders caused by splice acceptor pathogenic variant c.193-2A>C in COL4A4 at the genomic position chr2:227985866. Methods The study enrolled five probands with kidney biopsy analysis and five family members. Mutation screening was performed with Illumina MiSeq platform. The pathogenic variant was confirmed with standard dye-terminator sequencing. Results The only homozygous patient, aged two, had proteinuria and hematuria with preserved kidney function and no extrarenal manifestations. This patient had changes characteristic for Alport syndrome observed on electron microscopy of the kidney biopsy. In the heterozygous group, six patients had hematuria, four biopsied probands had proteinuria, and only one had moderately reduced kidney function. Heterozygous probands had variable kidney biopsy findings. Three patients had thin glomerular basement membrane nephropathy visible on electron microscopy and focal segmental glomerulosclerosis on light microscopy, two of them with focal lamellation on electron microscopy. One heterozygous patient had changes characteristic for Alport syndrome on electron microscopy without focal segmental glomerulosclerosis. Conclusion The homozygous patient had hematuria and proteinuria with preserved kidney function. The heterozygous patients presented with reasonably mild clinical phenotype and variable pathohistological findings.

Alport syndrome (AS) is a structural disorder of the glomerular basement membrane (GBM). Its genetic basis lies in the diverse mutations of COL4A3, COL4A4, and COL4A5 genes and it phenotypically manifests as a progressive nephropathy with hematuria, ultrastructural changes of the GBM, sensorineural hearing impairment, and eye abnormalities (1)(2)(3)(4)(5). The most frequent mutations (85%) are COL4A5 mutations, resulting in X-linked AS (6). Individuals with autosomal recessive AS (ARAS), caused by two mutations in COL4A3 and/or COL4A4, have similar clinical features to men with X-linked AS (7,8). The type of mutation affects disease phenotype and manifestation. The phenotype is usually severe both in men and women, with early onset of end-stage renal disease (ESRD) and frequent extrarenal disorders (9,10).
The spectrum of AS disorders has recently been expanded (11). Naming and describing individuals with heterozygous COL4A3 and COL4A4 mutations is still a matter of debate (10,12,13). A number of studies showed a correlation between thin glomerular basement membrane nephropathy (TBMN) with the heterozygosity for COL4A3 or COL4A4 mutation and benign familial hematuria (2,(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25). However, a variable proportion of COL4A3 or COL4A4 carriers progress to proteinuria, hypertension, and ESRD, which raises the question of the nomenclature of autosomal dominant AS (ADAS) (14,(25)(26)(27)(28)(29)(30)(31). Some scientists advocate the use of the term ADAS, others continue to use the term TBMN, while a Cyprus research group uses the term late-onset Alport nephropathy (10,12,13,32). The rationale behind ADAS nomenclature for heterozygous COL4A3 or COL4A4 patients lies in the presence of thin GBM in the kidney biopsy specimens of patients with X-linked AS and ARAS and the heterozygous carriers of COL4A3 or COL4A4 mutation (10). The authors suggest that this approach would improve clinical and diagnostic evaluation, with the possibility of ESRD rate reduction and treatment optimization (10). There are also rationales behind the use of the term TBMN. Savige et al (13) stated that most of heterozygous COL4A3 and COL4A4 carriers show either no decline in kidney function or show only mild decline with inconstant progression to ESRD and hearing impairment. The authors also argue that there is no unmistakable evidence that one mutation in COL4A3 or COL4A4 gene without disease modifying factors can be responsible for the characteristic ultrastructural signs of AS, hearing impairment, or eye abnormalities (13). Furthermore, in other genetic diseases autosomal dominant (AD) term is not used for the carriers of autosomal recessive (AR) disease because it can lead to the diagnosis of AD and AR disease in different members of the same fam-ily (13). However, there are emerging reports of autosomal dominant Alport spectrum disorders, especially in the cases that are hard to diagnose clinicopathologically (33). Here, we present the pathohistological and clinical characteristics of disorders caused by splice site mutation c.193-2A>C in COL4A4 at the genomic position chr2:227985866.

Patients
This study is a part of the research project Genotype-Phenotype Correlation in Alport's Syndrome and Thin Glomerular Basement Membrane Nephropathy funded by the Croatian Science Foundation. Five probands and their five family members were enrolled. The inclusion criterion for proband selection was kidney biopsy with glomerular changes on electron microscopy (EM) suggestive of AS or TBMN. The patients were selected by a retrospective review of the renal biopsy registry of the Department of Nephropathology and Electron Microscopy, Dubrava University Hospital Zagreb, covering the period from 2003 to 2019. All available clinical data, including the information about patients' and family medical history, onset of disease, kidney function (estimated glomerular filtration rate calculated by Chronic Kidney Disease Epidemiology Collaboration [CKD-EPI] equation, where values of 90 mL/ min/1.73m 2 or above were considered as preserved kidney function) and information about ocular abnormalities and sensorineural hearing loss was collected. An expert in clinical genetics conducted counselling with all probands, created family pedigrees, and identified family members at risk, who were later included in the study. All probands and family members gave a written consent for study participation. All procedures were performed in accordance with the ethical standards of the institutional research committee and the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The project Genotype-Phenotype Correlation in Alport's Syndrome and Thin Glomerular Basement Membrane Nephropathy was approved by the Ethics Committee of the University of Zagreb School of Medicine.

Methods
Probands and family members underwent mutation screening with Illumina MiSeq platform (Illumina, San Diego, CA, USA). Truseq Custom Amplicon Low Input kit was designed (Illumina) for re-sequencing of COL4A3, COL4A4, and COL4A5 genes. This custom-made panel includes 378 primer pairs that amplify the coding regions and flanking splice regions and generate amplicons ranging in size from 225 bp to 275 bp. Each sample was barcoded for multiplexing. The coverage for all exons was 99.09%. The quality of the libraries was assessed with the Agilent Bioanalyzer HS DNA Kit (Agilent Technologies, Santa Clara, CA, USA), showing correct size and concentration of the samples. The amplified libraries were pooled and sequenced on MiSeq Nano Flow Cell (Illumina). FastQ files generated by sequencing were subsequently submitted for analysis. The mean coverage depth of all amplicons was 270 × . For bioinformatical analysis, Illumina VariantStudio software was used. All variants were assigned a number in available databases, including the NCBI dbSNP138 and ClinVar (34). Splice acceptor pathogenic variant c.193-2A>C found in COL4A4 at the genomic position chr2:227985866 (variant described according to reference genome GRCh37) was confirmed with standard dye-terminator sequencing. Sanger sequencing was performed on ABI310 (Applied Biosystems) with BigDye v1.1 chemistry (Thermo Fisher Scientific, Waltham, MA, USA). The results were visualized with Vector NTI Software (Thermo Fisher Scientific, Waltham, MA, USA).

Patients' characteristics
The summary of patients' clinical characteristics is shown in Table 1, while the summary of probands' kidney biopsy findings is shown in Table 2.
As we previously reported (35), the first proband (HR 1.1.) was a boy aged two years and two months referred to the Nephrology Department of Children's Hospital Zagreb due to hematuria and proteinuria. The patient experienced a delay in psycho-motoric development and megalencephaly. Extensive workup performed at 10 months revealed karyiogram 46 XY and negative tests for fragile X. Organic acids in urine, homocystein, B12 and folic acid, acyl-carnitine profile, and amino acids in urine and serum were within the reference range. Brain magnetic resonance imaging was unremarkable. Megalencephaly was described as familial benign megalencephaly (his father had head circumference above the 95th centile). At the age of 1 year and 2 months, macrohematuria and proteinuria were recorded for the first time. Protein/creatinine was 284 mg/mmol; afterwards he had persistent proteinuria and microhematuria. At the age of two, he had protein/creatinine 189 mg/mmol and microhematuria. Tonal audiogram was unremarkable; eye exam did not reveal anterior lenticonus. Renal biopsy on light microscopy (LM) showed the kidney cortex with 69 glomeruli, one of which was globally sclerosed. Immature and partly immature glomeruli made 30% of all glomeruli ( Figure 1A). There was one small focus of interstitial fibrosis and tubular atrophy, affecting 1% of the cortical parenchyma ( Figure 1A). Changes characteristic for AS with areas of lamellation and basket-weave appearance of the GBM were present on EM ( Figure 1B). After starting 6 mg/ m 2 ramipril, proteinuria decreased and protein/creatinine was 73 mg/mmol, while microhematuria persisted. Both  proteinuria, and preserved kidney function) (35). His older sister (age 5) was negative for the mutation (Figure 2A).
In the years following our initial report on this mutation, we identified four more probands with the same mutation.
The second proband (HR 2.1.) was a 58-year-old man who presented with hematuria and proteinuria of 2.34 g in 24-hour urine, preserved kidney function (estimated glomerular filtration rate of 99 mL/min/1.73m 2 by CKD-EPI equation), and hypertension. Perihilar focal segmental glomerulosclerosis (FSGS) in one glomerulus was observed on LM ( Figure 3A); 35% of glomeruli were globally sclerotic, while others were enlarged. Interstitial fibrosis and tubular atrophy were present in 25% of cortical parenchyma. Arterioles showed marked hyalinosis, while arteries showed mild fibrointimal thickening ( Figure 2B). EM revealed thin GBM (average thickness 186 nm) with discrete lamellation in the areas of thickening ( Figure 3C). Podocyte foot process effacement was present in 10% of GBM surface. The  proband has two sons, aged 36 (HR 2.2.) and 33 (HR 2.3.), who both inherited the same heterozygous mutation (Figure 2B). The older son (HR 2.2.) had hematuria with no proteinuria and preserved kidney function, while the younger (HR 2.3.) had no signs of the disease (no hematuria, no proteinuria, and preserved kidney function).
The third proband (HR 3.1.) was a 26-year-old man who presented with hematuria and proteinuria of 0.30 g in 24-hour urine, preserved kidney function (estimated glomerular filtration rate of 108 mL/min/1.73m 2 by CKD-EPI equation), and hypertension. Tonal audiogram and eye exam did not reveal any changes characteristic for AS. LM showed five globally sclerosed glomeruli out of 26, and three with ischemic changes. There was no segmental sclerosis. Interstitial fibrosis and tubular atrophy were present in 10% of the cortical parenchyma. Blood vessels had normal morphology. Marked variations in GBM thickness (70-972 nm), with average thickness of 216 nm, were detected on EM. In the areas of thickening, lamellation and granular appearance of GBM were present (Figure 4). There was no podocyte foot process effacement.
The fourth proband (HR 4.1.) was a 60-year-old woman who presented with hematuria and proteinuria of 2.30 g in 24-hour urine, preserved kidney function (estimated glomerular filtration rate of 110 mL/min/1.73 m 2 by CKD-EPI equation), and hypertension. Perihilar FSGS was present in one out of six glomeruli on LM (Figure 5 A). One glomerulus was globally sclerosed. Interstitial fibrosis and tubular atrophy were present in 15% of the cortical parenchyma ( Figure 5A). Arterioles showed mild hyalinosis, while arteries were not found in the kidney biopsy specimen. Thin GBM (144-288 nm, average thickness 205 nm) with focal lamellation was observed on EM ( Figure 5B). Focal podocyte foot process effacement was present in 25% of the GBM surface. The proband's sister (age 53) was negative for the mutation.
The fifth proband (HR 5.1.) was a 58-year-old woman who presented with hematuria, proteinuria of 1.14 g in 24-hour urine, moderately reduced kidney function (estimated glomerular filtration rate of 45 mL/min/1.73m 2 by CKD-EPI equation), and hypertension. LM revealed perihilar FSGS ( Figure 6A) in 23% of glomeruli and global glomerulosclerosis in 15% of glomeruli. Other glomeruli were enlarged. Interstitial fibrosis and tubular atrophy were present in 30% of the cortical parenchyma. Arterioles showed marked wall hyalinosis ( Figure 6B), while arteries had normal morphology. Thin GBM (96-363 nm, average thickness 175 nm) was observed on EM ( Figure 6C). There was no podocyte foot process effacement. The proband has a son (age 19) (HR5.2.) and daughter (age 38). The son, who is asymptomatic (no hematuria, no proteinuria, and preserved kidney function), inherited the same heterozygous mutation, while the daughter is negative for the mutation (Figure 2C).

Genetic analysis
Next-generation sequencing (NGS) revealed splice site c.193-2A>C mutation in COL4A4 at the genomic position chr2:227985866 in all patients. The youngest patient was homozygous, while other patients and family members were heterozygous for this mutation. This mutation has not been described previously in the Human Gene Mutation Database (HGMD), Leiden Open (source) Variation Database (LOVD), and Ensembl genome database (35). software showed PVS1, PM2, and PP3 levels of certainty for pathogenicity according to the American College of Medical Genetics and Genomics (ACMG) (36). The de-scribed mutation was a null variant (within ±2 of canonical splice site) affecting gene COL4A4 (PVS1), the allele was not found in GnomAD despite good coverage (PM2), and

DisCussiON
The main goal of our study was to present the pathohistological and clinical characteristics of five Croatian families with Alport spectrum disorders caused by splice acceptor pathogenic variant c.193-2A>C found in COL4A4 at the genomic position chr2:227985866. One patient (the youngest, previously reported) was homozygous, while other patients were heterozygous for the mutation and presented with reasonably mild clinical phenotype and variable pathohistological findings (35). Not many potential splicing mutations have been identified within the first 10 nucleotides of the intron-exon boundaries for the CO-L4A3 and COL4A4 genes (37). Since this mutation has not been previously described in databases (HGMD, LOVD, and Ensembl genome database); since bioinformatical analysis showed PVS1, PM2, and PP3 levels of certainty for pathogenicity according to the ACMG; since the variant is not present among 50 healthy Croatian individuals; and since the characteristic pathohistological findings were found among the biopsied probands, we concluded this was a pathogenic variant causing autosomal Alport spectrum disorder (36). In the literature and databases, there are reports on COL4A4 splice site mutation causing autosomal AS and TBMN (38,39). Rosado et al found IVS3 + 1G>C, replacement of guanine to cytosine in the position 1+ of intron 3, in the splicing region, suggesting ADAS with a mild phenotype in which kidney disease manifests at a later age without progression to ESRD (38 (49,50). One could argue that hypertension acted as a modifying factor of disease phenotype in all our biopsied probands (12).
In conclusion, while our only homozygous patient had evident clinical and histological signs of AS at a very young age, heterozygous patients presented with reasonably mild clinical phenotype and variable pathohistological findings at a later age. Our study shows variability of changes in pathohistological findings adding to the pool of knowledge about Alport spectrum disorders. Competing interests All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organization for the submitted work; no financial relationships with any organizations that might have an interest in the submitted work in the previous 3 years; no other relationships or activities that could appear to have influenced the submitted work.