UMOD And The Architecture Of Kidney Disease Ⅱ

GWAS for urinary levels of uromodulin 

The availability of reliable assays and protocols to measure uromodulin levels allowed for analyzing the genetic factors associated with its excretion [71]. A first meta-GWAS approach performed on 10,884 individuals of European descent from six cohorts identified common variants within the promoter of UMOD as the single genome-wide significant locus associated with the levels of uromodulin in urine [40]. The top UMOD promoter variant, rs12917707, was associated in a dose-dependent fashion with the urinary levels of uromodulin, and in strong LD with the CKD variants [40].

A meta-analysis designed to increase the power to detect novel loci was recently conducted in 29,315 individuals of European ancestry from 13 cohorts including CoLaus [29]. Two genome-wide signifcant signals were identified for the urinary levels of uromodulin: a novel locus over the KRT40 gene coding for keratin-40 (KRT40), a type 1 keratin expressed in the kidney, and the UMOD-PDILT locus, with two independent sets of single nucleotide polymorphisms spread over UMOD and the adjacent PDILT. In follow-up experiments, KRT40 was shown to colocalize with uromodulin in TAL cells, while knock-down of KRT40 expression in primary mTAL cells affected uromodulin processing and excretion, providing a biological counterpart for the GWAS association [29]. Keratins are intermediate filaments that form the cytoskeleton in epithelial cells. As cytoskeletal proteins, keratins are involved in maintaining the physical integrity, mechanical stability, and intracellular organization within cells—for example, trafficking of proteins to the plasma membrane [8, 28]. That altered expression of KRT40 affects uromodulin (and also ROMK) processing in TAL cells suggests a role of specific cytokeratins on the sorting of proteins in kidney tubular cells [29]

CLICK HERE TO GET CISTANCHE TO IMPROVE KIDNEY FUNCTION

Uromodulin as a biomarker of kidney functional mass 

Since it is exclusively produced in tubular cells, uromodulin may have a specific biomarker value for kidney function. The factors associated with uromodulin excretion were analyzed in the SKIPOGH and CoLaus cohorts [50]. In both studies, positive associations were found between uromodulin and urinary sodium, chloride, and potassium excretion and osmolality. In SKIPOGH, 24-h uromodulin excretion was positively associated with kidney length and volume and with creatinine excretion and urine volume. It was negatively associated with age and diabetes. Both spot uromodulin concentration and 24-h uromodulin excretion were linearly and positively associated (multivariate analyses) with eGFR<90 ml/min per 1.73 m2. Age, creatinine excretion, diabetes, and urinary volume are independent clinical correlates of urinary uromodulin excretion. The associations of uromodulin excretion with markers of tubular functions and kidney dimensions suggest that it may reflect the distal tubular transport activity (e.g., reabsorption of NaCl and/or divalent cations) in the general population [50].

Further analyses in the SKIPOGH cohort evidenced that, in multiple linear regression analysis, 24-h urine uromodulin excretion was associated with the same predictors of nephron mass as identified by Denic et al. [9]. Age, female sex, and uric acid were negative predictors whereas height and birth weight were positive predictors of 24-h urine uromodulin excretion [45]. These data substantiate the measurement of urinary uromodulin levels as a useful surrogate marker for nephron mass. As such, higher uromodulin levels may indicate a higher functional reserve of the kidney with a lower risk of acute kidney injury (AKI): the more uromodulin you have, the higher functional reserve you have in case of post-operative AKI for instance [11, 53]. This situation should not be confused with the results of the GWAS studies indicating that the risk variants at the UMOD locus, which drive higher production of uromodulin reflected by higher levels in urine and blood, are consistently associated with an increased risk of CKD.

Mendelian randomization to assess the causality of uromodulin and CKD and hypertension 

Following the GWAS and follow-up investigations pointing at UMOD variants driving higher uromodulin expression (and thus higher excretion in urine) as associated with an increased risk of CKD, the issue of causality was raised. In other words: does higher production of uromodulin drives a higher risk of kidney damage?

Since uromodulin is exclusively produced by kidney tubular cells, urinary uromodulin levels positively correlate with eGFR (<90 ml/min/1.73 m2 ) and with kidney length and volume, whereas 24-h urinary uromodulin excretion is considered as a proxy of nephron mass [45, 50]. In line, higher urinary uromodulin levels, reflecting a higher functional reserve, are inversely correlated with the risk of kidney function decline in at-risk cohorts [20, 57]. The fact that lower urinary levels of uromodulin reflect decreased kidney functional mass is typical of reverse causation, i.e., when the disease affects the investigated risk factor. These elements constitute a bias when evaluating the causality of the urinary uromodulin levels, and thus of the UMOD variants, on the risk of CKD [13]. The facts that uromodulin also regulates blood pressure and that blood pressure and kidney function are interconnected further complicate the analysis [44, 47]

The use of Mendelian randomization (MR) provides a way to assess whether the production of uromodulin, reflected by its urinary levels, is a true risk factor for CKD and whether this potential association is related to blood pressure (Fig. 4). The MR method uses common genetic variants associated with the exposure (e.g., through GWAS), to test whether a given risk factor causes or aggravates a disease [55]. Recently, MR was used to clarify the causality between urinary uromodulin levels, kidney function, and blood pressure in individuals of European descent [48]. The link between urinary uromodulin levels and eGFR was frst investigated in CoLaus (n=3851 available data). In observational data, higher urinary uromodulin is associated with higher eGFR. Conversely, when using the UMOD rs12917707 as an instrumental variable in one-sample Mendelian randomization, higher uromodulin levels were strongly associated with eGFR decline

Fig. 4 Mendelian randomization analyses support the association of higher levels of urinary uromodulin with lower kidney function and higher blood pressure. a Two-sample MR to assess the bidirectional causal effects between urinary uromodulin (uUMOD) and eGFR and between CKD and blood pressure (BP). The analyses were performed in the meta–GWAS for uUMOD involving 10,884 individuals; the CKD Genetics (CKDGen) consortium, a meta-analysis of 121 GWASs including 567,460 individuals of European ancestry; and a combined analysis of the UK Biobank (UKB) and the International Consortium of Blood Pressure (ICBP) GWAS, which amounted to 757,601 individuals. b Multivariable MR to assess direct and indirect effects of uUMOD on BP through eGFR and of uUMOD on eGFR through BP based on MR causal effects between the exposure, mediator, and outcome in the 2-sample MR analyses. The analysis suggests that the association of uUMOD with higher BP is partially through decreased kidney function, whereas BP does not appear to mediate the association of uUMOD with low kidney function. Modifed from Ponte et al. [48] and Turner and Staplin [66]

Interpretation of MR studies should integrate that genetic variants may have other effects beyond the studied biomarker—a situation called genetic pleiotropy. Since uromodulin modulates sodium transport and influences salt sensitivity [47], our MR analysis substantiated the genetic association between the main UMOD variant and blood pressure, in addition to CKD and eGFR [48]. We thus applied two-sample Mendelian randomization on four GWAS consortia to explore causal links between urinary uromodulin levels and eGFR, CKD risk (567,460 individuals), and blood pressure (757,461 individuals). Higher uromodulin levels are significantly associated with lower eGFR, higher odds for eGFR decline or CKD, and higher systolic or diastolic blood pressure. Although the causal effects of urinary uromodulin levels on the risk of CKD are independent of blood pressure, the effect on blood pressure is mediated by eGFR [48]. These data support that genetically driven levels of uromodulin have a direct, causal, and adverse effect on kidney function outcome in the general population, not mediated by blood pressure (Fig. 4). They provide a strong rationale for investigating the mechanism by which elevated urinary uromodulin may contribute to CKD and whether a suitable modulator of uromodulin production/secretion may be identified [48, 66].

The UMOD locus and adaptation against uropathogens 

As discussed above, the UMOD variants associated with the risk of hypertension and CKD in the general population increase the expression and urinary excretion of uromodulin [11, 13, 53]. Yet, population genetics investigations indicated that the T allele of the top UMOD GWAS variant, rs4293393, associated with CKD risk, is the ancestral allele and is kept at high frequency in most modern populations [22, 65]. In fact, the distribution of the UMOD ancestral allele does not follow the ancestral susceptibility model observed for variants associated with salt-sensitive hypertension, i.e., a higher prevalence of salt-retaining alleles in African be compared to non-African populations, reflecting purifying selection outside of Africa [52]. Instead, the risk variant showed a signifcant correlation with pathogen diversity (bacteria, helminths) and the prevalence of antibiotic-resistant UTIs. An inverse correlation between urinary levels of uromodulin and markers of UTIs was detected in CoLaus [22]. A prospective cohort study of elderly community-dwelling individuals found that those with urinary uromodulin concentrations in the highest quartile had a lower risk of UTI events than those in the lowest quartile, independent of classical UTI risk factors [19].

Uromodulin is known to show antimicrobial properties in urine, and Umod-knockout mice are more prone to UTIs induced by type-1 fimbriated E. coli (FimH adhesin) than are their wild-type littermates [2, 35]. UTIs are among the most common bacterial infections, with high incidence and risk of recurrence in young women and potential complications that may reduce fitness. Pathogens and infectious diseases have imposed a strong selective pressure throughout human history [18]. Thus, the UMOD ancestral allele, driving higher urinary excretion of uromodulin, may have been kept at a high frequency because of its protective efect against UTIs [22]. This hypothesis is supported by structural and biochemical studies showing that the high mannose N-glycans at residue Asn275 in the cysteine-rich D8C domain of uromodulin are critical for FimH binding and thus prevention of the colonization of bladder epithelia by UPEC. The capacity of uromodulin to bind and aggregate bacteria is prevented by the removal of the high mannose N-glycan or by the competitive effect of an excess of d-mannose [68]. Of note, uromodulin was also able to aggregate other bacteria involved in human UTIs, including Klebsiella pneumoniae, Pseudomonas aeruginosa, and Streptococcus mitis [68]. These data demonstrated the role for uromodulin in defense against UTI through the capacity to aggregate uropathogens and to prevent their adhesion to the urothelium.

Rare mutations in UMOD cause autosomal dominant tubulointerstitial kidney disease 

In addition to common genetic variants associated with the risk of CKD and hypertension in GWAS, dominantly inherited mutations in UMOD are causing a rare form of kidney disorder leading to kidney failure.

Autosomal dominant tubulointerstitial kidney disease (ADTKD, MIM #16,200) is an increasingly recognized cause of end-stage kidney disease, characterized by tubular damage and interstitial fibrosis of the kidney in the absence of glomerular lesions. Affected individuals present with urinary concentrating defects, progressive chronic kidney disease (CKD), normal-to-mild proteinuria, and normal-sized kidneys, often with a positive family history [12, 14]. A relatively specific feeling is hyperuricemia due to low fractional excretion of uric acid, causing gout, usually before the onset of CKD [41]. The disease invariably progresses to end-stage kidney disease (ESKD) in adulthood, with a penetrance of 100%. Rare (MAF<10–4), dominant mutations in UMOD represent the most frequent cause of ADTKD (ADTKDUMOD). More than 95% of the UMOD mutations associated with ADTKD are missense, often targeting cysteine residues and leading to the formation of uromodulin aggregates within the endoplasmic reticulum (gain-of-toxic function) with a sharp decrease of its excretion in urine. Over 100 mutations in UMOD have been associated with ADTKDUMOD, with an overall prevalence of ~2% in patients with kidney failure, representing one of the most common mono-genic kidney diseases [21, 24].

The pathogenic mechanism of ADTKD-UMOD is due to the toxic accumulation of mutant uromodulin in TAL cells, with ER expansion and a parallel decrease in the urinary levels of the protein [12]. Studies of mouse models carrying uromodulin mutations confirm that intracellular accumulation of mutant uromodulin leads to ER stress, induction of the unfolded protein response, and subsequent tubular damage and interstitial fibrosis—substantiating the gain of toxic function mechanism in ADTKD-UMOD [53]. This mechanism suggests that decreasing the production of the mutant protein, for instance by using antisense oligonucleotides, could be a strategy to slow the disease course. The kidney-specific expression of UMOD, and the mild phenotype in Umod knockout mice support the potential value of such an approach [12].

UMOD is thus implicated at both extremes of the genetic disease spectrum: ultrarare variants with large effect size (ADTKD-UMOD) and common GWAS variants associated with reduced eGFR and risk of CKD in the general population. Thus, disorders involving uromodulin production and/ or excretion are of widespread relevance.

Conclusions and perspectives 

Thirty years after the identification of the frst gene involved in inherited kidney disease, the use of increasingly efficient and affordable genetic tools has allowed increasing diagnosis efficiency for rare kidney disorders, to clarify genetic heterogeneity and disease ontology, and to discover modifier genes involved in intrafamilial variability [10, 60, 67, 69]. The use of whole genome, SNP genotyping, and phenotype data has also helped to elucidate the role of complex genetic variation in the missing heritability observed for CKD and kidney-related traits. For instance, a recent study evidenced the influence of variable nucleotide tandem repeats (VNTRs) in MUC1 with multiple kidney phenotypes [38].

Discovering new genes will drive multi-level studies substantiating cellular mechanisms and possible drug targets. These analyses should further decipher the continuum of genetic kidney disease risk, with genes involved from rare Mendelian disorders to common variations in the general population. Examples include genes involved in NaCl handling at the kidney tubule level (e.g., SLC12A3, KCNJ1, SLC12A1, UMOD), also relevant for blood pressure regulation in the population; genes involved in receptor-mediated endocytosis in the proximal tubule (e.g., LRP2, CUBN, DAB2), shown by GWAS to affect renal function and risk of CKD; genes involved in rare disorders of Ca2+ and Mg2+ handling (e.g., CASR, TRPM6, CLDN14, CNNM2), also associated with mineral homeostasis and metabolic traits in the general population [67]. The mechanisms sustaining the effect of rare and common variants in these genes or in additional genes identified by GWAS will provide insights into various aspects of kidney function.

Analysis of large genomic datasets suggests that genetic variants with intermediate effect sizes must bridge the gap between rare, high-effect variants causing Mendelian disorders and frequent, low-effect variants involved in complex diseases (Fig. 1). These intermediate-effect variants can lead to either non-fully penetrant Mendelian disease or to an oligo/polygenic model modifying disease expressivity [30]. Such intermediate-effect variants may also be part of the genetic continuum underlying CKD, based on the lack of rare, pathogenic variants in the majority of CKD patients [24] and the missing heritability in GWAS [70]. Recently, we identified and characterized intermediate-effect variants in UMOD contributing to CKD by crossing general population datasets with curated variants reported in ADTKD; analyzing biological and phenotypical effect sizes using in silico modeling, cell systems, databases, and biobanks; and validating the impact on kidney failure in the 100,000 Genomes Project and UK Biobank [43].

Obtaining genetic information in patients with rare kidney diseases is already substantiating precision medicine and will probably increase rapidly [10, 17, 36]. With larger GWAS and advanced statistical methods, polygenic risk scores (PRS) will develop, with the perspective of the early identification of subjects at risk of developing complex kidney diseases, before eGFR decline has manifested—offering the possibility of early intervention [27, 70, 72].

References 

1. Barker DF, Hostikka SL, Zhou J et al (1990) Identification of mutations in the COL4A5 collagen gene in Alport syndrome. Science 248:1224–1227 

2. Bates JM, Raf HM, Prasadan K et al (2004) Tamm-Horsfall protein knockout mice are more prone to urinary tract infection. Kidney Int 65:791–797 

3. Böger CA, Gorski M, McMahon GM et al (2017) NFAT5 and SLC4A10 loci associate with plasma osmolality. J Am Soc Nephrol 28:2311–2321 

4. Brunati M, Perucca S, Han L et al (2015) The serine protease pepsin mediates urinary secretion and polymerization of Zona 

5. Pellucida domain protein uromodulin. Elife 4:e08887 Buf R, Korstanje R. (2022) The impact of genetic background on mouse models of kidney disease. Kidney Int (in press) 

6. Corre T, Arjona FJ, Hayward C et al (2018) Genome-wide meta-analysis unravels interactions between magnesium homeostasis and metabolic phenotypes. J Am Soc Nephrol 29:335–348

7. Corre T, Olinger E, Harris SE et al (2017) Common variants in CLDN14 are associated with differential excretion of magnesium over calcium in the urine. Pfugers Arch 469:91–103 

8. Davezac N, Tondelier D, Lipecka J et al (2004) Global proteomic approach unmasks involvement of keratins 8 and 18 in the delivery of cystic fibrosis transmembrane conductance regulator (CFTR)/deltaF508-CFTR to the plasma membrane. Proteomics 4:3833–3844 

9. Denic A, Mathew J, Lerman LO et al (2017) Single-Nephron glomerular filtration rate in healthy adults. N Engl J Med 376:2349–2357 

10. Devuyst O, Knoers NV, Remuzzi G, Schaefer F (2014) Rare inherited kidney diseases: challenges, opportunities, and perspectives. Lancet 383:1844–1859

Supportive Service:

Email:wallence.suen@wecistanche.com 

Whatsapp/Tel:+86 15292862950

Shop:

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