Chronic kidney disease (CKD) is a serious disease associated with high morbidity and mortality and increasing health expenditure per patient. However, CKD etiology is often not firmly established and, in many patients, remains unknown.1 The etiology of CKD can be due to either environmental, genetic factors and/or a combination of both.2,3 Renal diseases due to genetic factors may follow an autosomal dominant or recessive pattern of inheritance.

For years a group of autosomal dominant renal diseases has been described that mainly affects tubular and cystic level, characterized by progressive deterioration of renal function accompanied by tubulointerstitial fibrosis, being UMOD one of the genes responsible for these diseases.

This work is focused on the phenotypic manifestations caused by alterations in UMOD gene. This gene encodes the uromodulin protein (also known as Tamm-Horsfall glycoprotein), and is specifically expressed at renal level.4

Under physiological conditions, uromodulin is the most abundant protein in urine, and it is secreted by the epithelial cells of the thick ascending limb of loop of Henle in a bidirectional manner into the urine and circulation.4 Uromodulin can be found polymerized forming filaments, or in its non-polymeric form.4 This non-polymeric form has different renal and systemic functions, this is why its site of action is important 4,5 The main functions of this protein are regulation of ionic transport in the kidney, urinary and systemic homeostasis, as highlighted by Trachtman et al.6 in their study. In addition, uromodulin has a protective effect against calcium nephrolithiasis and urinary tract infections, with a crucial role in local and systemic immunomodulation.7–9

Its protein structure differentiates 4 EGF-like domains (EGF-like I, EGF-like II, EGF-like III and EGF-like IV), a domain with 10 conserved cystines (D10C), and a bipartite zona pellucida (ZP) in ZP-N and ZP-C.10–14 These domains allow junctions that make the protein to fold three-dimensionally and form its functional structure and polymerization.7,11,12,14,15

Mutations in UMOD gene can cause defects in the structure, maturation, secretion and/or function of uromodulin, depending on the type of variant and affected domain . Most of pathogenic alterations in this gene affect the correct structural conformation, resulting in an abnormal protein. This anomalous protein is retained and accumulated in the endoplasmic reticulum, causing cell death in the renal tubule.7,16,17 This phenomenon can trigger endoplasmic reticulum stress, activation of inflammatory responses, renal interstitial fibrosis and secondary mitochondrial dysfunction,7,18 together with compensatory reduced expression of the sodium-potassium-chloride cotransporter (Na+–K+–2Cl–) (NKCC2),7,13,19 contributes to increased proximal urate reabsorption and progressive deterioration of renal function due to the oxidative stress associated with the accumulation of this abnormal protein, leading to hyperuricemia, episodes of gout, fibrosis and renal failure7,18 (Fig. 1).

Fig. 1.

Schematic diagram of the pathophysiology of UMOD alterations. On the left, the restricted expression of UMOD to the ascending branch of Henle and the first sections of the distal convoluted tubule is depicted. On the right, the altered processes due to expression of mutant UMOD. Source: adapted from Mabillard et al. (2023).7 NKCC2: sodium–potassium–chlorine cotransporter (Na+–K+–2Cl–); ER: endoplasmic reticulum; uUMOD: UMOD in urine; UPR: unfolded protein response.

(0.31MB).

In the literature, a great phenotypic variability has been associated with alterations in UMOD, even among individuals who are carriers of the same alteration. However, there are other genetic and environmental modifying factors that may influence the presentation and progression of the disease.

Patients with alterations in UMOD, may present a decrease in the fractional excretion of urate, which causes hyperuricemia, and sporadic glomerular involvement with manifestations such as microhematuria. Proteinuria is usually mild or absent. The presence of renal cysts has also been observed in patients with alterations in this gene, although to a much lesser extent than other manifestations such as hiperuricemia.20–24

It has been observed that individuals with mutations in UMOD reach end-stage renal disease between the ages of 25 and 70 years or more, and according to the study by Ayasreh et al.25,26, those with a history of gout experienced its onset between the ages of 3 and 51 years.

Clinical studies have shown that kidney biopsies from affected patients reveal interstitial fibrosis and tubular atrophy, confirming the pattern of tubulointerstitial damage associated with these mutations.27

Genes associated with Autosomal Dominant Tubulointerstitial Disease (ADTKD) development are HNF1B, MUC1, REN, SEC61A1, and UMOD. All of them exhibit phenotypic variability (diabetes, liver cancer, renal cysts, and psychiatric disorders, among others), with ADTKD being the common manifestation that links them, but with incomplete penetrance. Variants identified in the UMOD gene are mostly associated with ADTKD.5,7,28–31 Patients with ADTKD-UMOD show a progressive course of kidney failure with early onset of hyperuricemia, and although the vast majority of cases are asymptomatic, the development of gout has also been described.28,30–32

Recent studies in closed populations have identified several recurrent mutations in UMOD. These recurrent mutations are of particular interest due to their high prevalence in specific populations and their association with severe clinical phenotypes.7 The study of recurrent mutations offers a unique window to understand the pathogenic mechanisms underlying the function of the altered gene, characterize the phenotypic manifestations, establishing a correlation between the genetic mutation and disease progression, and develop targeted therapeutic strategies.

An association has been described between the age of onset of end-stage renal disease, the underlying UMOD mutation and the domain in which the mutation is found.32 In this sense, analyzing mutations in UMOD and their relationship to inherited kidney disease provides a solid basis for early identification, genetic diagnosis and clinical management of these patients.

In the last 25 years, our team has focused on studying and establishing the genetic map of hereditary diseases in Galicia divided into 3 large groups: cystic disease, glomerular disease and tubulointerstitial disease. This comprehensive study includes 2538 families and 4094 individuals studied genetically, with the aim of establishing a genotype-phenotype correlation. Within the group of tubulointerstitial diseases, 371 families were genetically characterized and 81 of them with ADTKD, of which 31 are families with mutations in UMOD. This study identifies the presence in the Galician population of 2 recurrent genetic variants responsible for most cases of ADTKD without familial relationship (UMOD, transcript NM_001008389.3: p.C255Y and p.Q316P, transcribed NM_001008389.3). In this article a genotype-phenotype correlation is established in 37 patients from 15 unrelated families in the Galician community, in order to establish phenotypic patterns associated with the presence of these variants.

This is a retrospective, multicenter, observational cohort study that analyzed NefroCHUS cohort of 4094 patients with suspected hereditary kidney diseases. This cohort includes patients from different hospitals throughout Spain, with an enrichment in the Galician population.

Inclusion criteria for patients' selection: 1) genetic testing requested by a nephrologist; 2) initial clinical suspicion of ADTKD; 3) genetic sequencing of genes associated with ADTKD (HNF1B, MUC1, REN, SEC61A1, and UMOD) at least; and 4) patients carrying one of the two most recurrent mutations of UMOD in Galician community: p.C255Y or p.Q316P.

A total of 37 patients from 15 families were selected. Ten of these families carried p.C255Y mutation, and the remaining five carried p.Q316P.

For the clinical characterization of the patients in this study, clinical data such as hypertension, estimated glomerular filtration rate (eGFR), CKD stage, hyperuricemia (considered when serum uric acid levels were greater than 7 mg/dl), fractional excretion of uric acid, gout episode, presence of renal cysts, albuminuria, hematuria, type of renal replacement therapy and kidney transplant were collected.

Patient DNA was extracted from peripheral blood leukocytes using the commercial Chemagic Blood 1k-3k kits using the Chemagen extraction robot (PerkinElmer, Waltham, Massachusetts, USA) and the QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Germany), following the manufacturers' recommended protocols. Quality and concentration of the extracted DNA was assessed using the Qubit 3 fluorometer, Broad Range kit (Thermo Fisher Scientific, Waltham, Massachusetts, USA).

All patients included in this selection underwent genetic testing, either using Next-Generation Sequencing (NGS) or targeted sequencing using Sanger sequencing. Whenever technology allowed, the index cases in each family were genetically tested using NGS technology (in 86.67% of index cases, 13/15 probands).

The NGS technique used in the patients in this study is the sequencing of a panel of known genes associated with the most prevalent tubular and cystic renal pathologies33. Obtaining the sequence of the exons and flanking regions of selected genes. This identifies point variants, small insertions/deletions (indels), and the presence of structural variants (copy number variants [CNVs]). Raw sequencing reads were processed according to GATK best practice guidelines and aligned to the human reference genome (GRCh37) using bwa version 0.7.17-r1188. Low-quality reads were removed from the primary dataset using fastp version 0.20. Variant calling was performed using GATK version 4.1.9, Pindel version 0.2.5b9, and ExomeDepth version 1.1.15, in accordance with best practice recommendations. Variant annotation was performed using an in-house process, combining functional gene annotations from SnpEff and ANNOVAR to retrieve population frequencies (1000 Genomes Project, gnomAD, and an in-house database, among others), functional prediction scores (SIFT, CADD, etc.), and clinical information from databases such as ClinVar and OMIM.

Sanger sequencing is a targeted sequencing technique that sequences short DNA regions of interest with high accuracy. This technique is used in most relatives of index cases to study the candidate alteration segregation. By identifying the candidate variant that could cause pathology in index cases, sequencing can be targeted to that region and more quickly identify whether relatives are carriers of the variant.

The variants were classified independently by two geneticists specialized in hereditary kidney diseases following the guidelines and recommendations of the American College of Medical Genetics (ACMG).34

A literature search was conducted for the 2 recurrent variants in this study using different search tools such as PubMed, ClinVar, Human Gene Mutation Database (HGMD) and MetaDome to determine whether they have been previously reported, their molecular involvement and their phenotypic associations.

The statistical significance of the results was assessed for nominal variables using the Chi-square test or Fisher's exact test for variables with a frequency greater or less than 5, respectively, and for quantitative variables, the Mann-Whitney-Wilcoxon test using RStudio. Due to the small sample size, p-values ​​were adjusted using the Benjamini-Hochberg method to control the false discovery rate (FDR). This adjusted p-value is the q-value; statistical significance is considered a q-value greater than 0.05.

The log-rank statistical test was applied to the Kaplan–Meier survival curves to study the differences between variant types and establish the p-value.

At molecular level, the p.C255Y alteration is classified as probably pathogenic. It affects the D10C10 domain; this domain is highly conserved, so genetic alterations affecting this domain are highly likely to impact its function.13–15 When studying tolerance to the change of reference amino acid at position 255 using MetaDome program,35 we found that it is intolerant to change, with a tolerance score of 0.22 (scores lower than 0.7 are considered intolerant to amino acid changes). Therefore, this recurrent alteration, which changes the cystine residue to that a tyrosine, has a negative impact at protein level.35

This is the most recurrent UMOD variant in our Galician cohort of patients with suspected hereditary kidney disease, with a total of 28 carriers. Of these, 64.29% are men (18/28).

Twelve patients (44.44%) presented with hypertension (HT), either at the time of diagnosis or during the course of the disease. Furthermore, 77.77% of patients had some degree of CKD (considered to be stage 2 or higher) at the time of the study, with a mean age of 57 years. Eight patients (33.33%) progressed to End-Stage Renal Disease (ESRD), starting dialysis at a mean age of 61 years.

Of note is the case of the female patient F4.S3, a homozygous carrier of the p.C255Y alteration. F4.S3 began renal replacement therapy (RRT) at 46 years old, a decade or more earlier than the rest of the patients with RRT in the cohort (Table 2). This suggests that alterations in homozygosity are associated with more accelerated and severe progression of kidney disease.

Seventy-six percent of cases met the criteria of hyperuricemia (19 of 25 patients). Hyperuricemia diagnosis and initiation of treatment with allopurinol occurred at a mean age of 41.38 years (range: 16–67 years). Urinary sediment was generally unremarkable, with only 4 cases of albuminuria (17.39%) and 2 cases of microhematuria (8%). Furthermore, cyst development was observed in 48% of patients.

Among the 32% of patients with gout, the mean age at onset of the first gout episode was 38.43 years (range: 22–63 years). Fig. 3 shows the survival curve of the ages at the first gout episode. Despite the onset in early adulthood, several elderly patients have not developed gout, which could be interpreted as indicating that this event is not highly related to age.

Fig. 3.

Kaplan–Meier survival curve showing the age at which the first episode of gout occurred. No patients with the p.Q316P alteration had episodes of gout.

(0.06MB).

Of these carrier patients, 13 belong to family F4, which has a history of hyperuricemia and ESRD, conditions likely associated with this mutation. Two individuals in this family have undergone a kidney transplant. At intrafamilial level, we observed characteristics with a high familial prevalence, such as an estimated glomerular filtration rate lower than 90 ml/min/1.73 m2, indicative of kidney disease.

Additionally, it has been found variability in the development of hyperuricemia, with this characteristic being observed in patients of different ages.

It was also observed some age-related variability in other characteristics, with approximately half of the older relatives presenting the trait whereas the other half of younger relatives do not. These characteristics include: the onset of advanced CKD (individuals under 54 years of age in this family have not progressed), RRT (the cut-off age for this trait is 60 years), and presence of renal cysts (with a cutting age of 54 years).

It is also worth noting that six members of this family are also carriers of a heterozygous pathogenic alteration in PKHD1gene, c.383delC, p.Thr128fs (F4.S1, F4.S2, F4.S6, F4.S7, F4.S9, and F4.S13). Alterations in this gene are primarily associated with the development of renal cysts, with a primarily autosomal recessive inheritance, although a cystic phenotype has also been described in patients with heterozygous alterations.36 In the family F4, there are a total of 5 individuals with renal cysts, 3 of whom are carriers of the PKHD1 alteration. Therefore, we cannot confirm or rule out that this alteration in PKHD1 modulates the appearance of renal cysts in patients who are carriers of alterations in UMOD and PKHD1.

UMOD p.Q316P alteration is also molecularly classified as likely pathogenic. It affects the EGF-like domain IV.10–12 This domain is highly conserved and it is involved in the bisulfate bonds that contribute to the three-dimensional structure of uromodulin.10 This position is also intolerant to change, with a tolerance score of 0.3637.

This variant was found in 9 patients, of whom 66.67% were men (6/9). In this cohort, it was observed a higher percentage of patients with hypertension, present in 5 patients (62.5%). At the time of diagnosis, 50% of patients had some degree of CKD (considered from stage 2 onward), at a mean age of 54 years. To date, 3 patients (75%) have achieved ESRD, with onset on RRT at a mean age of 63.67 years.

Fifty percent of the cohort met criteria for hyperuricemia. Diagnosis of hyperuricemia and initiation of allopurinol treatment occurred at a mean age of 49.33 years (range: 32–60 years). To date, no episodes of gout have been reported in these patients, although there is a family history of gout.

The urinary sediment of this second cohort was also unremarkable, with microalbuminuria present in 3 out of the 8 patients (one of them with albumin creatinine ratio> 300 mg/g) and only one case of microhematuria. Similar to the first subcohort, cystic development was observed in 50% of patients.

Family F12 has the highest number of p.Q316P carriers, includes 5 patients, and 4 of these present hypertension. The effect of age on renal impairment is also evident in this family, as the two youngest members of the family only present a pathological phenotype of hypertension, without any additional characteristics, while the older patients have a glomerular filtration rate of 35.2 ml/min/1.73 m2; renal cysts; hyperuricemia; and have reached advanced CKD.