UMOD And The Architecture Of Kidney Disease

Abstract 

The identification of genetic factors associated with the risk, onset, and progression of kidney disease has the potential to provide mechanistic insights and therapeutic perspectives. In less than two decades, technological advances yielded a trove of information on the genetic architecture of chronic kidney disease. The spectrum of genetic influence ranges from (ultra) rare variants with large effect sizes, involved in Mendelian diseases, to common variants, often non-coding and with small effect sizes, which contribute to polygenic diseases. Here, we review the paradigm of UMOD, the gene coding for uromodulin, to illustrate how a kidney-specific protein of major physiological importance is involved in a spectrum of kidney disorders. This new field of investigation illustrates the importance of genetic variation in the pathogenesis and prognosis of disease, with therapeutic implications.

CLICK HERE TO KNOW CISTANCHE FOR KIDNEY DISEASE TREATMENT

With an estimated global prevalence above 10%, a high individual and societal burden, and limited therapeutic options, chronic kidney disease (CKD) is a major public health issue. Even in its early stages, CKD is associated with multiple complications and adverse outcomes, and it is a major risk factor for accelerated cardiovascular disease and aging [15]. For decades, the lack of mechanistic understanding of the multiple functions ensured by the kidney slowed the development of pharmacologic interventions targeting CKD and its associated complications [10, 32].

The familial clustering of kidney disorders, the high heritability of kidney function parameters, and the variable susceptibility of inbred animal strains to kidney damage drive strong efforts to decipher the genetic architecture of kidney diseases [5, 15, 37, 60]. The first genetic breakthroughs in nephrology were the mapping of autosomal dominant polycystic kidney disease (ADPKD) in 1985 [51] and the identification of a mutation in COL4A5 causing Alport syndrome in 1990 [1]. Identification of genes involved in classic glomerular and tubular diseases quickly followed, including steroid-resistant nephrotic syndrome, nephropathic cystinosis, Dent disease, Bartter and Gitelman syndromes, nephrogenic diabetes insipidus, and ADPKD [10]. In less than two decades, the advent of new technologies including next-generation sequencing (NGS) and genome-wide association studies (GWAS) led to the identification of hundreds of genes causing inherited kidney disorders or associated with kidney function metrics and risk of CKD. In turn, these insights improved our understanding of processes operating in the normal and diseased nephron segments, thus providing novel therapeutic targets [10, 60, 67]. These discoveries were made possible thanks to collaborative studies gathering families with rare disorders and large, multi-ethnic cohorts for the discovery of rare and common genetic variants, respectively. In this brief review, we will use the paradigm of genetic variation in UMOD, the gene coding for uromodulin—the most abundant protein in the kidney and in normal urine—to illustrate how genetic approaches have the potential to yield insights into the architecture of CKD. These studies involved several Swiss cohorts, ranging from normal population to rare diseases, and technological developments supported by the Swiss National Center of Competence Kidney. CH.

The spectrum of allele frequency and effect size in disease 

Once the genetic influence on a specific aspect of kidney function has been established, specific investigations can decipher the genetic architecture of disorders related to that trait. Three criteria are important when considering disease-associated allelic variants: (i) the frequency of the variant in the population; (ii) the effect size of the variant on the phenotype; and (iii) the number of genetic variants acting on the phenotype [52]. These criteria are classically represented on an X–Y plot, with the prevalence of the genetic variant on the x-axis and the magnitude of its effect size on the y-axis (Fig. 1). This relationship visualizes the spectrum of allelic variation, going from (ultra-) rare variants (minor allele frequency< 1°/00) with large effect size, involved in Mendelian disorders, to common variants (frequency>5%), with small-effect size involved in common, polygenic disorders [34]. These genetic variants can be identified using new technologies including NGS and GWAS, respectively. Intermediate-effect variants, not captured by GWAS but amenable to sequencing approaches, are predicted to be part of the continuum (Fig. 1). Both population-based studies and rare disease cohorts are essential for such gene discovery.

SKIPOGH and CoLaus cohorts: heritability and GWAS studies 

Two population-based Swiss cohorts provided invaluable support to studies investigating the heritability of kidney function parameters and the role of UMOD variations in CKD (Fig. 2).

The CoLaus study (“Cohorte Lausannoise”) is a longitudinal, population-based cohort initiated in 2003 in the city of Lausanne. The CoLaus participants were randomly selected from the complete list of Lausanne inhabitants in 2003 [16]. Inclusion criteria were (i) written informed consent; (ii) age 35–75 years; (iii) willingness to take part in the examination and donate a blood sample; and (iv) Caucasian origin. By the end of the recruitment in 2006, the baseline CoLaus dataset included 6538 participants (participation rate: 34%). Participants were first seen from 2003 to 2006, then followed up from 2009 to 2012, 2014 to 2017, and 2018 to 2021. At all four visits, participants attended a morning medical examination at the Lausanne University Hospital after an overnight fast, which comprised a blood sampling, assessment of anthropometric features, and completion of a detailed health questionnaire inquiring about lifestyle factors, socioeconomic and marital status, personal and family history of disease, as well as medication intake [39]. The CoLaus project includes a series of “sub-studies” aimed at exploring specific health-related outcomes and environmental exposures (https://www.colaus-psycolaus.ch/).

Fig. 1 Genetic architecture of disease: spectrum of allele frequency and effect size. Variants in important genes may be involved in a continuum between rare and complex diseases, according to the risk allele frequency in the population (x-axis) and the strength of the effect size (odds ratio, y-axis). (Ultra)-rare alleles can be identified by next-generation sequencing (NGS), whereas common variants can be identified using genome-wide association studies (GWAS). Intermediate-size effect variants are predicted to complete the dichotomy, manifesting as a non-fully penetrant Mendelian disease or an oligo/ polygenic model modifying disease expressivity. Figure adapted from Manolio et al. [34]

Fig. 2 Infographics summarizing the design of the CoLaus and SKIPOGH population-based studies

The Swiss Kidney Project on Genes in Hypertension (SKIPOGH) is a family- and population-based cohort investigating the genetic and environmental determinants of renal function and blood pressure. SKIPOGH is a multicenter, longitudinal study with participants recruited in the city of Lausanne, and the cantons of Geneva and Bern [46, 49]. At inception, eligible subjects were randomly selected from the population-based CoLaus study in Lausanne and from the population-based Bus Santé study in Geneva [26]. In Bern, subjects were randomly selected using the cantonal phone directory. Inclusion criteria were (i) written informed consent; (ii) minimum 18 years of age; (iii) European descent; and (iv) at least one, and ideally three, first degree family members willing to participate to the study. The participation rate was 20% in Lausanne, 22% in Geneva, and 21% in Bern. The first SKIPOGH wave took place between 2009 and 2013 (baseline: SKIPOGH 1), whereas the second wave began in 2012 and ended in 2016 (follow-up: SKIPOGH 2). SKIP-OGH 1 included 1128 participants coming from 273 nuclear families. At both waves, study participants attended a morning medical visit after an overnight fast, which included blood sampling, kidney imaging, cardiac evaluation, and biobanking. Participants completed a detailed questionnaire about their medical history, medication intake, lifestyle factors, and socioeconomic characteristics (http://www.skipo gh.ch/index.php/Welcome_to_the_SKIPOGH_study!).

Estimating the relative contribution of genetic and environmental influences on kidney function parameters and serum and urinary electrolyte levels is a prerequisite for studies aiming at identifying genes involved in kidney function. Thanks to the application of a standardized protocol across several centers, the heritability (i.e., the proportion of variance of a given trait explained by genetic effects) of estimated glomerular filtration rate (eGFR) as well as of serum and urine concentrations, renal clearances, and fractional excretions of major electrolytes could be analyzed in the family-based SKIPOGH cohort [37]. High heritability estimates were observed for GFR estimated using the CKDEPI Eq. (46±6%), serum creatinine (49±6%), CKD-EPI cystatin C (58±5%), and serum urea (49±6%). The heritability of serum concentrations was highest for calcium, 37%, and lowest for sodium, 13%. All probability values were signifcant. These results provided a basis for further investigations of the genetic basis of kidney function and electrolyte homeostasis in the general population.

In addition to the heritability studies, data from participants of the SKIPOGH and/or CoLaus cohorts were included in GWAS which identified new loci for kidney function and CKD [70], uromodulin production and excretion [29, 40], albuminuria [58, 59], serum urate levels [61], magnesium and calcium homeostasis [6, 7], and osmoregulation [3].

UMOD variants in GWAS for kidney function and CKD

Genome-wide association studies are a useful, unbiased tool to uncover genomic regions (“loci”) associated with kidney function parameters and risk of CKD. Typically, GWAS require large sample sizes, yielding genome-wide signifcant (P value<5× 10−8) variants that are common and associated with a relatively small increase in disease risk [69]. The largest GWAS in the field of nephrology addressed traits like the eGFR based on serum creatinine (eGFRcrea) [13, 69, 70]. As discussed above, there is a strong genetic predisposition to CKD, as the heritability of the eGFR is close to 50% [37]. The largest and most recent GWAS for eGFR identifed>250 genetic loci and explained nearly 20% of that heritability [70]. Among those, variants in the UMOD locus display the largest effect size on the eGFR and CKD [70]. This effect thought to be linked to the genetically driven expression of uromodulin (see below), is consistent across different ethnic groups and is also observed for longitudinal traits including rapid decline of eGFR [23].

The UMOD locus includes single-nucleotide polymorphisms (SNPs; e.g., rs12917707 and rs4293393) that are in complete linkage disequilibrium (LD) in a large block encompassing the gene promoter. For these SNPs, the “risk” allele is the common allele, with a frequency ranging between 0.765 and 0.955 [65]. Critically, the UMOD promoter variants are associated with the expression of uromodulin in the kidney and its levels in the urine and blood. In fact, homozygous carriers of the UMOD risk allele have twofold higher levels of uromodulin in the urine, compared with homozygous carriers of the (minor) protective allele [29, 40, 64]. Most recent GWAS for eGFR identified a strong, independent signal within PDILT, the gene flanking UMOD on chromosome 16 [70]. Of interest, the PDILT intronic variant rs77924615 is associated with uromodulin urinary levels, but not with the expression of PDILT, suggesting that it regulates uromodulin expression [70]. This hypothesis is supported by the independent association of rs77924615 with urinary uromodulin levels in two meta-GWAS [29, 40]. The possible links between uromodulin expression, which has been experimentally validated for the UMOD-PDILT variants, and other transcripts correlated with variants in the same locus have not been explored [31].

The UMOD locus has also been associated with hypertension and incident cardiovascular disease [44], kidney stones [25], and gout/uric acid [25, 61], strengthening its importance for a set of common diseases. In these GWAS, the UMOD alleles associated with higher uromodulin expression/levels are associated with an increased risk of CKD, hypertension, and hyperuricemia, whereas they are protective against kidney stones [11, 53].

The biological relevance of the UMOD locus for CKD 

The relevance of the UMOD GWAS locus in relation to kidney function is immediate, as UMOD is a kidney-specific gene coding for uromodulin, the most abundant protein expressed in the kidney and excreted in normal urine [11, 13, 53]. Uromodulin is mainly produced by the thick ascending limb (TAL) of the loop of Henle and, to a lesser extent (about 10%) in the initial segment of the distal convoluted tubule (DCT) [63]. Uromodulin is a glycosylphosphatidylinositol (GPI)-anchored protein belonging to the family of zona pellucida (ZP) domain proteins (Fig. 3). The protein is heavily glycosylated (30% of the Mw) and it contains 48 conserved cysteine residues involved in 24 intramolecular disulfide bonds necessary for correct folding. Uromodulin contains four epidermal growth factor (EGF)-like domains, a cysteine-rich domain (D8C), and a bipartite ZP domain allowing protein polymerization. In TAL cells, the apical membrane-bound monomeric uromodulin is cleaved by the serine protease hepsin at a conserved cleavage site located near the C-terminus of the protein [4, 42]. The cleavage allows the polymerization of extracellular uromodulin into filaments forming a matrix-like structure in the urine, the main constituent of urinary casts [56]. Recently, the three-dimensional (3D) structure of native urinary uromodulin polymers was obtained by cryo-electron tomography, showing a zigzag-shaped backbone formed by polymerized ZP domains and protruding arms composed of the EGF and the D8C domains [68]. N-glycosylation mapping, biophysical assays, and imaging revealed that uromodulin binds the Escherichia coli type 1 pilus adhesin and that uromodulin filaments associate with uropathogens and mediate bacterial aggregation, potentially preventing adhesion and promoting clearance of the pathogens [68]. Recent studies also shed light on the regulation of uromodulin excretion in the urine, including a role played by the calcium-sensing receptor [62] and the type 1 keratin KRT40 [29], and the activity of transport processes operating in the TAL [50, 54, 62].

If the vast majority of uromodulin is apically targeted to form organized polymers in the urine, it should be noted that uromodulin is also detected in monomeric form in the serum, at levels that are at least 100-fold lower than in the urine [11, 53]. The mechanisms regulating the basolateral trafficking of uromodulin and its release into the interstitium and further into the circulation remain essentially unknown. A recent meta-GWAS of circulating uromodulin using different detection methods revealed that a missense variant (p.Cys466Arg) in the uromodulin-glycosylating enzyme B4GALNT2 was a loss-of-function allele leading to higher serum uromodulin levels, while other loci influence the glycosylation of circulating uromodulin [33]. The current view is that interstitial uromodulin may cross-talk with proximal tubule cells, contributing to regulating innate immunity, whereas circulating uromodulin may act as an anti-oxidant [53].

Studies in uromodulin knock-out (Umod−/−) mice have provided key insights into the roles of uromodulin and its relevance for various processes operating in the kidney. A detailed description of these roles is provided in recent reviews [11, 53]. In brief, uromodulin polymers in the urine protect against kidney stone formation and urinary tract infections (UTIs); uromodulin regulates sodium transport systems in the TAL and DCT, thus regulating blood pressure and urinary concentrating ability; it stabilizes TRPV5, TRPV6, and TRPM6 channels, thus increasing their activity in the DCT. The small amount of uromodulin released from the basolateral side of the TAL cells into the interstitium may regulate processes in neighboring proximal tubule cells and contribute to the innate response, e.g., by regulating the abundance and activity of resident mononuclear phagocytes, or enter the circulation and act as an antioxidant molecule [53]. All these processes are relevant when considering the onset of CKD, tubulointerstitial injury, and the complications arising from kidney failure.

Fig. 3 Site of production and structure of uromodulin. Uromodulin is mainly produced by the cells that line the thick ascending limb (TAL), a segment involved in the reabsorption of NaCl and divalent cations, while being not permeable to water. Uromodulin is a glycosylphosphatidylinositol (GPI)-anchored protein which traffics to the apical membrane of the cells, where it is cleaved by the serine protease hepsin and released in the urine where it forms large polymers. These polymers form the matrix of the urinary casts. The predicted structure of uromodulin contains a leader peptide (L); four EGF-like domains (I to IV); a cysteine-rich D8C domain; a bipartite C-terminal Zona Pellucida domain (ZP_N and ZP_C) connected by a linker; and a GPI-anchoring site at position 614. The seven N-glycosylation sites are indicated by triangles. Figure adapted from Devuyst et al. [11, 12]

The association of uromodulin expression levels with kidney damage was substantiated in a transgenic mouse model overexpressing wild-type uromodulin, mimicking the situation observed in carriers of the risk UMOD haplotype [65]. The transgenic Umod mice displayed a dose-dependent increase in systolic blood pressure, with salt-sensitive hypertension. Although kidney function remained normal, aging kidneys from the Umod transgenic mice showed focal lesions (e.g., tubular casts, cysts) and increased expression of kidney damage markers (e.g., lipocalin-2, Kim-1) and chemokines, that were not detected in the kidneys of control littermates. Similar focal lesions were evidenced in kidney biopsies from individuals aged>65 years homozygous for the risk UMOD haplotype, compared to individuals homozygous for protective variants [65]. Together, these results suggest that, in subjects carrying the UMOD risk alleles, genetically driven higher production of uromodulin becomes deleterious over time, promoting the onset of CKD. The deleterious effect of higher uromodulin production in the kidney could be due to increased metabolic demand, chronic activation of salt reabsorption systems, or defects in the lumen of the distal tubule or in the circulation, explaining why these loci are associated with a strong age-effect [11, 13]. Of note, the UMOD risk variants associated with higher uromodulin levels are also associated with activated NKCC2, and thus a higher response to furosemide [65]. A clinical trial testing the potential influence of these UMOD variants on the response to loop diuretics in individuals with hypertension is underway (ClinicalTrials.gov Identifier: NCT03354897).

Supportive Service:

Email:wallence.suen@wecistanche.com 

Whatsapp/Tel:+86 15292862950

Shop:

https://www.xjcistanche.com/cistanche-shop