Advertisement

Advances in understanding the genetic basis of diabetic kidney disease

  • Man Li
  • Marcus G. Pezzolesi
Review Article
Part of the following topical collections:
  1. Diabetic Nephropathy

Abstract

Diabetic kidney disease (DKD) is a devastating complication of Type 1 and Type 2 diabetes and leads to increased morbidity and mortality. Earlier work in families has provided strong evidence that heredity is a major determinant of DKD. Previous linkage analyses and candidate gene studies have identified potential DKD genes; however, such approaches have largely been unsuccessful. Genome-wide association studies (GWAS) have made significant contribution in identifying SNPs associated with common complex diseases. Thanks to advanced technology, new analytical approaches, and international research collaborations, many DKD GWASs have reported unique genes, highlighted novel biological pathways and suggested new disease mechanisms. This review summarizes the current state of GWAS technology; findings from GWASs of DKD and its related traits conducted over the past 15 years and discuss the future of this field.

Keywords

Diabetic kidney disease Diabetic nephropathy Genetics Genome-wide association study 

Introduction

Diabetic kidney disease (DKD), also known as diabetic nephropathy, affects nearly 40% of all patients with diabetes [1, 2, 3]. Despite improved management of DKD, including near universal implementation of renoprotective therapies, DKD remains the leading cause of end-stage renal disease (ESRD) worldwide and is associated with excess morbidity and premature mortality in patients with diabetes [1, 2, 3]. As the incidence of diabetes continues to rise globally, so too has the incidence of DKD.

DKD is a complex disease that is known to cluster in families and the heritability (i.e., the proportion of total phenotypic variation due to genetic effects) of DKD and DKD-related traits is estimated at 30–75% [4, 5, 6, 7, 8]. For more than 25 years, investigators have been working to identify the genes that underlie DKD susceptibility. Early gene mapping strategies relied on sparse sets of genetic markers to localize chromosomal regions shared among the related cases. Although these linkage studies identified several loci containing potential DKD genes, including AGTR1, which encodes the angiotensin II receptor 1, a component of the renin–angiotensin–aldosterone system (RAAS), and the CNDP1 gene, which encodes carnosinase and is thought to have a protective role in DKD [9], overall, such approaches have largely been unsuccessful.

Completion of both the Human Genome Project [10], which determined the DNA sequence of the entire human genome, and International HapMap Project’s development of a map of the block-like structure of human genome [11], launched a new era in complex disease research by providing researchers with better tools to interrogate genetic variation across the genome. Genome-wide association studies (GWASs), a gene mapping approach that use commercially available arrays to assay millions of single nucleotide polymorphisms (SNPs) across the entire genome, have led to the discovery of genes, pathways, and biological mechanisms that underlie Type 2 diabetes (T2D), Crohn’s disease, schizophrenia, and cardiovascular traits and diseases. More recent advances in next-generation sequencing (NGS) technology, including whole exome sequencing (WES) and whole genome sequencing (WGS), are further empowering researchers to fully characterize genetic variation and genotype–phenotype relationships at a genome-wide scale.

Although there are few published reports where NGS have been applied to studies of DKD, GWASs have generated a great deal of optimism among researchers working to identify susceptibility genes for DKD. In this review, we review the current state of GWAS technology, summarize findings from GWASs of DKD and its related traits conducted over the past 15 years, and present our perspective on the future of this field.

GWAS: an evolving technology to advance genetic discovery

GWAS first emerged as a powerful tool for investigating the genetic architecture of human disease more than a decade ago. Focusing primarily on common SNPs, i.e., those having a minor allele frequency (MAF) of at least 5%, this un-biased, non-candidate-driven approach has proven to be incredibly successful, having identified more than 14,000 genome-wide significant associations with various diseases or traits (http://www.ebi.ac.uk/gwas). These successes support the notion that common alleles have a significant contribution to the phenotypic variation underlying many common diseases and traits, a hypothesis known as the common disease/common variant (CDCV) hypothesis [12].

To conduct a GWAS, researchers determine and compare the genotype distributions of millions of SNPs from participants with a particular phenotype. These phenotypes can be either categorical, e.g., cases with DKD and controls without DKD, or quantitative, e.g., urinary albumin-to-creatinine ratio (uACR) or estimated glomerular filtration rate (eGFR). Variants that differ in frequency between subjects help pinpoint regions of the genome that may influence the risk of disease (Fig. 1a). Importantly, however, although such variants are associated with the disease or phenotype, whether the variant is causal or serving as a proxy to some nearby, or even distant, variant requires further fine-mapping and functional analyses.

Fig. 1

GWAS analysis in DKD. a GWASs examine the association between a variant, or SNP, and dichotomous (i.e., case/control) and quantitative traits. Statistical significance (i.e., p value) and effect size estimate (i.e., odds ratio or beta) are reported for each SNP. Variants associated with increased disease risk will be found at a higher frequency in cases compared to controls. GWAS results are often displayed using Manhattan plots, with − log10 (p values) for each SNP plotted against its position in the genome. The Manhattan plot shown here is from the GWAS from Pezzolesi et al. [13] and is reprinted with permission from Elsevier. b Since the advent of GWAS, GWAS genotyping arrays from the two leading manufacturers, Affymetrix (orange) and Illumina (blue), have evolved from the initial arrays that included ~ 100,000 SNPs to conventional platforms that include > 1,000,000 SNPs and include both rare and low-frequency variants as well as CNVs. c Schematic representation comparing the genotype resolution of GWAS, computational imputation, and NGS technologies. The blue table shows the observed genotypes (0, 1, or 2; indicating the number of minor alleles at each SNP location) obtained in a set of individuals genotyped using GWAS. Only SNPs included on the array can be ascertained; those that are not are unobserved genotypes (question marks). The green table shows that computational imputation using genotype data from a reference panel (e.g., the 1000 Genomes Phase 3 reference panel) is able to assign genotypes to some SNPs that were not included on the GWAS array. Because of uncertainty in computationally inferring these variants, they are assigned a genotype probability or dosage between 0 and 2. Some unobserved genotypes are not able to be imputed due to lack of information from the reference panel used for imputation. The orange table shows that NGS is able to assign genotypes to all variants with certainty. An added advantage of NGS technology over GWAS and imputation is that this method is able to discovery novel variants that are not present on genotyping platforms or in the reference panel used for imputation that may be specific to the population or disease under study

GWAS genotyping arrays leverage linkage disequilibrium, i.e., the degree to which an allele of one SNP is correlated with an allele of another SNP, allowing researchers to survey nearly all common variation across the entire genome by genotyping ~ 500,000 judiciously chosen markers [14]. As genotyping technology has improved, the capacity of genotyping arrays has grown from a few thousand SNPs on first generation arrays to approximately 5 million on current releases (Table 1; Fig. 1b). These high-throughput arrays have further evolved to include ancestry informative markers that can be used to identify population sub-structure, mitochondrial variants, and human leukocyte antigen (HLA) tagging variants. Additionally, GWAS arrays are also useful for detecting structural variation, e.g., copy number variation (CNV), a type of structural variation that has been reported to be associated with a broad range of human diseases, including metabolic disease, Type 1 diabetes (T1D) [15], and obesity [16].

Table 1

Commonly used GWAS genotyping arrays

Company

Array

# of SNPs

Lowest captured MAF

Notes

Affymetrix

SNP 5.0

~ 500,000

0.05

Includes additional non-polymorphic probes to capture CNVs

SNP 6.0

~ 906,000

0.05

Increased genome coverage

Axiom BioBank Array

~ 800,000

0.01

Includes an option to add custom content

Axiom Population-Focused Array

~ 2,000,000

0.01–0.05a

Includes markers that are specific to HapMap populations and has high coverage of rare and common variants

Axiom Exome 319 Array

~ 319,000

0.005

Includes novel rare and common exonic variants and insertion/deletions derived from NHLBI’s ESP

Illumina

HumanOmniExpress

~ 700,000

0.05

Includes optimized haplotype tagging SNPs

HumanOmni2.5

~ 2,500,000

0.025

Targets both common and rare variants

HumanOmni5

~ 4,300,000

0.01

Includes optimized content for whole-genome genotyping and CNV applications

Multi-Ethnic Global Array

~ 1,700,000

0.005–0.02a

Includes optimized haplotype tagging SNPs to maximize imputation accuracy for rare and common variants across the most commonly studied super-populations

HumanExome BeadChip

~ 550,000

< 0.001

Includes focused coverage of exonic variants derived from NHLBI’s ESP

aDue to slightly different genomic coverages in various populations, the minimum MAF captured on each population-specific array differs

Contemporary genotyping arrays can also include custom content that is tailored for specific research projects. These including, for example, the Immunochip, a custom array that interrogates variants associated with autoimmune and inflammatory diseases, and the Metabochip, which focuses on SNPs associated with metabolic diseases, as well as researcher-designed arrays used to fine-map particular genomic regions or loci of interest. Most recently, with the completion of the National Heart, Lung, and, Blood Institute’s Exome Sequencing Project, commercially available arrays also now include rare (MAF < 0.5%) coding variants.

A key step in GWAS analysis after genotypes are obtained is the imputation of genotypes of SNPs not included on the array used in the study [17]. Statistical imputation of unobserved variants can recover some of the information lost because of imperfect linkage disequilibrium between observed genotypes and unobserved causal variants [18, 19, 20]. Genotype imputation takes advantage of known haplotypes from external high-density ‘reference panels’ to predict genotypes that have not been directly typed in a sample of individuals (Fig. 1c). Currently, the most commonly used reference panel is the 1000 Genomes Phase 3 release which includes 2504 individuals from multiple ethnic groups and over 88 million SNPs, 3.6 million short insertions/deletions, and 60,000 structural variants [21]. Implementing the same reference panel for imputation allows cohorts from different studies to be analyzed jointly (i.e., meta-analyzed) across the same set of SNPs, thereby increasing sample size and improving statistical power. However, in comparison to WES (i.e., sequencing the coding regions of the genome) and WGS (i.e., sequencing of the entire genome), even with imputation, GWAS has limited resolution for testing associations at rare functional variants that may have larger effects on the disease or trait due to challenges in accurately imputing such variants and the inability of these methods to discover novel variants that could contribute to the disease or trait.

Overall, GWASs have had an enormous impact on our understanding of the genetic architecture of complex diseases and traits. This cost-effective, high-throughput approach has facilitated several large-scale analyses including a number of studies aimed at elucidating the genetic basis of DKD.

DKD and its sub-phenotypes

DKD is characterized by glomerular and tubular basement membrane thickening, mesangial expansion, glomerulosclerosis, podocyte effacement, and ultimately, nephron loss [22]. Clinically, DKD manifests as abnormal urinary albumin excretion, also known as albuminuria, and a progressive loss of renal sufficiency. Patients with DKD, however, can present with varying levels of albuminuria, at various stages of chronic kidney disease (CKD), and experience vastly different rates of progression of renal function decline.

Albuminuria is often the first clinical sign in the pathogenesis of DKD. Approximately 50% of all patients with diabetes experience at least moderate increases in urinary albumin excretion levels [23]; a condition referred to as microalbuminuria, where uACR ranges from 30 to 300 mg/g. A subset of these individuals (approximately 25%) develop overt proteinuria (uACR > 300 mg/day), even when managed with renoprotective medications, such as angiotensin receptor blockers (ARBs) or angiotensin converting enzyme (ACE) inhibitors. In addition to increased albuminuria, as DKD progresses, renal function deteriorates from normal levels (i.e., an eGFR > 90 ml/min/1.73 m2) to impaired renal function (eGFR < 60 ml/min/1.73 m2). Ultimately, for as many as 10–15% of diabetic patients, renal function declines to < 15 ml/min/1.73 m2, or ESRD, at which point renal replacement therapy in the form of dialysis or a kidney transplant is needed [23, 24, 25].

Genetics studies of DKD frequently use albuminuria- and eGFR-based sub-phenotypes to dichotomize subjects for association analyses. Commonly, normoalbuminuric subjects are used as ‘controls’ while subjects with proteinuria and/or ESRD, i.e., those with advanced DKD, are used as ‘cases’. Several other phenotypic definitions, including various quantitative traits (e.g., uACR and eGFR), have also been considered (Table 2). Unfortunately, however, although a subset of DKD patients progress to ESRD in the absence of clinically relevant proteinuria, no study to date has examined the genetic basis underlying this phenotype.

Table 2

Phenotypic comparisons used in GWASs of DKD

Category

Phenotype

Control definition

Case definition

Albuminuria-based phenotypes (uACR; mg/g)

NA vs. MA

< 30

30–300

NA vs. MA/PR

< 30

≥ 30

NA vs. PR

< 30

> 300

eGFR-based phenotypes (ml/min/1.73 m2)

Non-CKD vs. CKD

≥ 60

< 60

Albuminuria-(uACR; mg/g) and eGFR-based phenotypes (ml/min/1.73 m2)

NA vs. MA/PR/ESRD

< 30

≥ 30 or ESRD

NA vs. PR/ESRD

< 30

> 300 or ESRD

NA vs. ESRD

< 30

ESRD

Non-ESRD vs. ESRD

Non-ESRD

ESRD

 

CKD–DKD

eGFR ≥ 60 and uACR < 30

eGFR < 45 and uACR ≥ 30

Quantitative traits

uACR

 

eGFR

 

eGFR slope

NA normoalbuminuria; MA microalbuminuria; PR proteinuria; ESRD dialysis/transplant/eGFR < 15 ml/min/1.73 m2; CKD chronic kidney disease; uACR urinary albumin-to-creatinine ratio; eGFR estimated glomerular filtration rate

Identifying genes for DKD through GWAS

Since the launch of the GWAS era, significant progress has been made toward mapping genes for DKD and DKD-related traits. Over time, the field has witnessed growth in the size of DKD GWASs, fueled by ever increasing collaboration among leading investigators in this field, and more comprehensive interrogation of variation across the genome (Fig. 2). These studies support a role for multiple genes in DKD susceptibility; those identified thus far, however, have only modest effects on DKD risk (Fig. 3). Cumulatively, these variants account for only a small proportion of the heritability of DKD, suggesting that additional DKD loci remain to be identified. The major findings reported to date are summarized in Table 3.

Fig. 2

The history of GWASs for DKD. Since the first GWAS for DKD, the field has witnessed immense growth in the sample sizes of these studies. As sample sizes have increased, statistical power to discovery associations has improved resulting in the discovery of several novel loci over the past few years. The year of publication of these studies is represented on the x-axis; the total sample size is shown on the y-axis. PubMed identifiers for each study are shown at the base of the figure. The inner (darker) circles are scaled in proportion to discovery sample size, whereas the outer lighter circles are scaled in proportion to the total (discovery + replication) sample size. DKD loci with p values < 1.0 × 10− 5 are presented

Fig. 3

Effect sizes for DKD loci identified through GWAS. Odds ratios (OR) and beta estimates for dichotomous (orange) and quantitative DKD traits (blue), respectively, are presented for all DKD GWAS loci with p value < 1.0 × 10− 5. Of note, the effect sizes for the SSB, UMOD, and PRKAG2 loci are from associations with eGFR while those for HS6ST1 and RAB38 are from associations with uACR

Table 3

Summary of major loci identified through GWASs of DKD and DKD-related traits (p value < 1.0 × 10− 5)

Chr

Position

SNP

Nearest gene

p value

Effect size (OR/beta)

DKD phenotype

Population

Study design

N samples

N samples’ discovery cohort

N markers

References

2

100,460,654

rs7583877

AFF3

1.2 × 10− 8

1.29

ESRD vs. non-ESRD

Caucasian

Case–control

11,847

6652

2.4 million

[26]

2

129,027,961

rs13427836

HS6ST1

6.3 × 10− 7

0.19

uACR

Caucasian

Quantitative

7399

5509

2.2 million

[27]

2

170,646,916

rs1974990

SSB

1.4 × 10− 6

3.17

eGFR

Caucasian/Asian

Quantitative

14,828

13,158

37 million

[28]

2

213,168,768

rs7588550

ERBB4

2.1 × 10− 7

0.66

PR/ESRD vs. NA

Caucasian

Case–control

11,847

6231

2.4 million

[26]

4

87,529,078

rs61277444

PTPN13

1.9 × 10− 6

1.41

ESRD vs. non-ESRD

Caucasian

Case–control

12,540

5150

37 million

[29]

4

87,529,078

rs61277444

PTPN13

6.0 × 10− 6

1.42

ESRD vs. no DKD

Caucasian

Case–control

12,540

3406

37 million

[29]

6

89,948,232

rs9942471

GABRR1

4.5 × 10− 8

1.25

MA vs. NA

Caucasian

Case–control

4801

4227

37 million

[28]

6

154,947,408

rs955333

SCAF8/CNKSR3

1.3 × 10− 8

0.73

PR/ESRD vs. NA

Trans-ethnic

Case–control

13,736

6197

906,600

[30]

6

154,954,420

rs12523822

SCAF8/CNKSR3

5.7 × 10− 9

0.57

PR/ESRD vs. NA

American Indians

Case–control

2154

857

906,600

[30]

7

29,255,470

rs39059

CHN2/CPVL

5.0 × 10− 6

1.39

PR/ESRD vs. NA

Caucasian

Case–control

1705

359,193; 2.4 million

[13, 31]

7

36,917,995

rs741301

ELMO1

8.0 × 10− 6

2.67

PR/ESRD vs. NA

Japanese

Case–control

920

188

81,315

[32]

7

148,141,082

rs1989248

CNTNAP2

6.0 × 10− 7

1.26

MA/PR/ESRD or eGFR < 45 vs. NA

Caucasian

Case–control

12,540

3135

37 million

[29]

7

148,141,082

rs1989248

CNTNAP2

1.8 × 10− 6

1.29

ESRD vs. no DKD

Caucasian

Case–control

12,540

3406

37 million

[29]

7

151,415,041

rs10224002

PRKAG2

2.7 × 10− 8

2.01

eGFR

Caucasian/Asian

Quantitative

17,696

13,158

37 million

[28]

9

86,164,176

rs10868025

FRMD3

5.0 × 10− 7

1.45

PR/ESRD vs. NA

Caucasian

Case–control

1705

359,193; 2.4 million

[13, 31]

10

83,291,690

rs72809865

NRG3

7.4 × 10− 6

1.17

MA/PR/ESRD vs. NA

Caucasian

Case–control

12,540

5150

37 million

[29]

10

97,284,081

rs1326934

SORBS1

5.7 × 10− 7

0.84

PR/ESRD vs. NA

Caucasian

Case–control

7801

1462

11.1 million

[33]

11

3,060,725

rs451041

CARS

3.1 × 10− 6

1.36

PR/ESRD vs. NA

Caucasian

Case–control

1705

359,193; 2.4 million

[13, 31]

11

88,008,251

rs649529

RAB38

5.8 × 10− 7

-0.14

uACR

Caucasian

Quantitative

7787

5825

2.2 million

[27]

13

110,252,160/110,252,608

rs1411766/

rs17412858

MYO16/IRS2

1.8 × 10− 6

1.41

PR/ESRD vs. NA

Caucasian

Case–control

1705

359,193; 2.4 million

[13, 31]

15

94,141,833

rs12437854

RGMA/MCTP2

2.0 × 10− 9

1.80

ESRD vs. non-ESRD

Caucasian

Case–control

11,847

6652

2.4 million

[26]

16

20,400,839

rs11864909

UMOD

2.1 × 10− 12

2.22

eGFR

Caucasian

Quantitative

16,304

13,158

37 million

[28]

22

36,708,483

rs5750250

MYH9

7.7 × 10− 8

1.27

PR/ESRD vs. NA

African-American

Case–control

6108

3221

906,600

[30]

22

36,657,432

rs136161

APOL1

5.2 × 10− 7

1.36

PR/ESRD vs. NA

African-American

Case–control

6108

3221

906,600

[30]

Chr chromosome; NA normoalbuminuria; MA microalbuminuria; PR proteinuria; ESRD dialysis/transplant/eGFR < 15 ml/min/1.73 m2; uACR urinary albumin-to-creatinine ratio; eGFR estimated glomerular filtration rate

*Positions reported for all SNPs are relative to the human reference sequence genome (NCBI Build GRCh37/hg19)

In the very first GWAS for DKD, researchers from Japan performed a low-density, two-stage scan using fewer than 100,000 gene-based SNPs and a modest sample size that included only 188 T2D patients (94 cases with either proteinuria or ESRD and 94 normoalbuminuric controls) in the initial (i.e., discovery phase) analysis [34]. After testing the top associations from their discovery phase in 466 additional cases and 266 additional controls, SLC12A3, a thiazide-sensitive Na–Cl co-transporter, emerged as the first candidate DKD gene in the GWAS era. This top signal, however, was not statistically significant after adjusting for the multiple comparisons performed as part of this analysis. A second GWAS performed by these same investigators using a similar approach subsequently identified a common variant located in the ELMO1 gene that was found to be even more strongly associated with DKD [32]. Shortly after these two landmark studies, higher-throughput genotyping arrays, with > 500,000 or more SNPs, became mainstream and marked the launch of the GWAS revolution in DKD.

The first of several high-throughput GWAS of DKD was conducted in the Genetics of Kidneys in Diabetes (GoKinD) study cohorts as part of the Genetic Association Information Network (GAIN) [35]. In this study, more than 2.4 million SNPs (~ 360,000 genotyped SNPs and 2.1 million imputed SNPs) were examined in 820 case subjects (284 with proteinuria and 536 with ESRD) and 885 control subjects with T1D [13, 31]. The GoKinD study identified four genomic loci that were strongly associated with DKD. The strongest association occurred at the FRMD3 gene (p value = 5.0 × 10− 7). Three additional genomic regions near the CHN2/CPVL, CARS, and MYO16/IRS2 genes were also associated with DKD. Since reporting these findings, evidence of replication for associations at FRMD3, CARS, and MYO16/IRS2 with DKD-related traits in T2D have been observed among Caucasians [36], African Americans [37], and Asians [38], suggesting that these genes appear to be true susceptibility loci for kidney disease in both T1D and T2D.

Since the GoKinD study, the formation of large-scale international consortia has continued to move the genetic discovery of DKD genes forward with studies that have included increasingly larger sample sizes. The first such effort was the Genetics of Nephropathy: an International Effort (GENIE) consortium which performed the first meta-analysis GWAS of T1D DKD using 6691 individuals from three collections in its discovery phase; the Ireland Warren 3 Genetics of Kidneys in Diabetes UK Collection (UK-ROI), the Finnish Diabetic Nephropathy Study (FinnDiane), and the GoKinD study [26]. A total of 5873 individuals from 9 additional cohorts were then used in its replication stage. Genotype data from these collections were harmonized using ~ 2.4 million SNPs derived from imputation with the HapMap reference panel. In their combined meta-analysis, two signals observed at the AFF3 and RGMA/MCTP2 genes reached genome-wide significant associations (p value < 5 × 10− 8) with risk of ESRD. Interestingly, a novel variant on chromosome 2q (rs4972593) between the SP3 and CDC7 genes was identified as a gender-specific genetic variant associated with ESRD in female patients with T1D in the GENIE consortium [39]. GWASs for quantitative DN-related traits have also been reported by the GENIE consortium, including a strong association at the GLRA3 gene with albuminuria; this association, however, was not confirmed in a meta-analysis of seven independent cohorts [40].

The GENIE consortium recently expanded their meta-GWAS to include a more comprehensive set of genetic variants (~ 37 million SNPs), a larger number of subjects (12,540 individuals), and analyses across a range of albuminuria- and eGFR-based sub-phenotypes [29]. Despite its increased sample size, however, no variant reached genome-wide significance. A strong association was observed near the CNTNAP2 gene (p value = 6.0 × 10− 7) when patients with ESRD were compared with those without ESRD (i.e., normoalbuminuria, microalbuminuria, or proteinuria), suggesting that this gene may increase risk of ESRD. A strong association was also observed in the PTPN13 gene (p value = 1.9 × 10− 6) when cases with CKD and microalbuminuria, proteinuria, or ESRD were compared to controls with no sign of renal complications (i.e., normoalbuminuria and normal eGFR).

The Family Investigation of Nephropathy and Diabetes (FIND) conducted the only trans-ethnic GWAS of DKD, a study that included more than 13,000 unrelated T2D individuals of European–American, African–American, Mexican–American, or Indian American ancestry [30]. In this study, conventional genome-wide significance was observed in an ethnic group specific analysis of Mexican–Americans at rs12523822 on chromosome 6q (p value = 5.7 × 10− 9) between the SCAF8 and CNKSR3 genes. This same signal was also strongly associated with DKD in a trans-ethnic meta-analysis across all ethnic groups. In the African–American sub-group, an association at the APOL1/MYH9 locus approached genome-wide significance (p value = 5.2 × 10− 7). As APOL1 is well-known to be a major susceptibility locus for ESRD in individuals of African ancestry, this association is likely due to the inclusion of unrecognized non-DKD among cases included in this analysis [41, 42, 43].

Most recently, the Surrogate markers for Micro- and Macro-vascular hard endpoints for Innovative diabetes Tools (SUMMIT) consortium, a large-scale effort to identify genetic and non-genetic markers for DKD in T2D, performed the largest GWAS of DKD to date, involving more than 30,000 T2D subjects in its discovery and replication phases and 8 dichotomous and quantitative DKD phenotypes [28]. As part of these analyses, a novel signal near GABRR1 (p value = 4.5 × 10− 8) associated with microalbuminuria. Additionally, joint analysis of 16,304 T1D and T2D subjects with eGFR data confirmed the well-established association between this phenotype and variants near UMOD (p value = 4.4 × 10− 12), an observation that was previously reported by Pattaro et al. in > 11,000 diabetic individuals included in the CKDGen Consortium [44].

From association to function: further support for several DKD loci identified through GWAS

For many complex diseases, studies to elucidate the biology underlying statistical associations identified through GWAS have taken a back seat to increasingly larger and better powered studies aimed at uncovering yet unknown susceptibility loci [45]. Although we are witnessing this same trend in our own disease area, several studies have begun linking the statistical associations identified through GWAS of DKD with functionality (Table 4).

Table 4

Post-GWAS functional studies in DKD

Gene/locus

GWAS report

Functional report

Functional approach

Potential function mechanism

ELMO1

[32]

[32]

In situ hybridization of kidney tissue in diabetic and non-diabetic mice; over-expression of ELMO1 in cultured COS cells

Increased ELMO1 expression in glomerular and tubular epithelial cells in diabetic mice; increased expression of ECM protein genes in response to ELMO1 over-expression

ELMO1

[32]

[46]

In situ hybridization of kidney tissue in CKD rat model; over-expression of ELMO1 in cultured COS cells

Increased ELMO1 expression in glomerular epithelial cells of CKD rats; increased expression of ECM proteins and inhibited cell adhesion to ECM in cells over-expressing ELMO1

FRMD3

[13, 31]

[47]

Gene expression analysis; pathway and transcriptional pattern-based promoter modeling; EMSA

Decreased expression of FRMD3 in DKD; identified potential homeodomain transcription factor binding site at DKD-associated SNP; increased binding of glomerular nuclear protein(s) to DKD-associated SNP

AFF3

[26]

[26]

Over-expression of TGF-β1 in HK-2 cells

Increased expression of AFF3 upon TGF-β1 stimulation

SORBS1

[33]

[33]

Gene expression analysis

Increased expression of SORBS1 in tubules of DKD patients

HS6ST1

[27]

[27]

Gene expression analysis

Increased expression of HS6ST1 in tubules of DKD patients

RAB38

[27]

[27]

Gene expression analysis of kidney tissue; genetically modified rat models of diabetes

Increased RAB38 expression in tubules of DKD patients; higher urinary albumin concentrations and reduced amounts of megalin and cubilin at the proximal tubule cell surface in Rab38 knockout diabetic rats

ECM extracellular matrix; EMSA electromobility shift assay

Following their discovery of ELMO1 as a novel DKD gene, Shimazaki et al. demonstrated increased expression of ELMO1 in the presence of high glucose and showed that ELMO1 over-expression contributes to chronic glomerular injury by promoting excess TGF-β1, collagen type 1, and fibronectin expression and dysregulation of renal extracellular matrix (ECM) metabolism [32, 46]. ELMO1 is a soluble cytoplasmic protein that functionally cooperates with CRKII and DOCK 180 to mediate cytoskeletal rearrangements during phagocytosis of apoptotic cells and cell motility in mammalian cells [48]. Further supporting its likely role in the DKD, strong associations of variants in ELMO1 have been reported with DKD in multiple independent cohorts of diverse ethnic background, including Chinese [49], African-American [50], Pima Indian [51], and Caucasian cohorts [52]. Each of these studies is consistent with ELMO1’s role in DKD and suggests that extensive allelic heterogeneity exists across this locus.

Associations at the FRMD3 locus that were first reported in the GoKinD collection have also been replicated in several independent studies [37, 53, 54]. The potential role of FRMD3 in DKD is further supported by the observation that FRMD3 gene expression is significantly different in kidney tissue from subjects with advanced DKD relative to those with early DKD [47]. Adding to this, Martini et al. investigated the functional context of the DKD-associated SNP identified in this GWAS using an in silico approach that integrates pathway analyses with transcriptional pattern-based promoter modeling. This framework identified a link between the lead SNP located in the promoter region of FRMD3 and the BMP-signaling pathway. Additional functional assays identified a potential homeodomain factor transcription factor binding site at this SNP that affects protein binding in glomerular extracts from C57 black 6 mice.

Additionally, although neither gene had previously been implicated in DKD, functional analyses performed by Sandholm et al. [26] and Germain et al. [33] suggest that both AFF3 and SORBS1 likely have functional roles in DKD susceptibility.

Further driving the discovery of DKD genes

Over the past 15 years, tremendous progress has been made in identifying the genetic factors that underlie DKD. These efforts have been hastened by advances in genotyping technologies, larger GWAS sample sizes, and increased collaboration among leading investigators in this field. These continued efforts will likely lead to further DKD gene discovery.

Building on the success of these large-scale studies (with samples sizes > 10,000), the Juvenile Diabetes Research Foundation’s Diabetic Nephropathy Collaborative Research Initiative (JDRF-DNCRI) recently brought together investigators from across the globe and assembled the world’s largest collection of T1D patients for studies on the genetics of DKD. Nearly 30,000 samples included in this cohort have been genotyped using the latest genotyping technology. These data are currently being analyzed for association with several dichotomous and quantitative DKD phenotypes. This JDRF-DNCRI is also working closely with investigators from SUMMIT to combine resources and further empower DKD gene discovery. Additionally, the availability of ‘mega’ cohorts from the general population, such as the UK Biobank (N = ~ 500,000) [55] and Million Veterans Program (MVP) (N = ~ 450,000) [56], are likely to provide additional insights on the genetic architecture of kidney function in diabetic populations.

While there is hope that increased sample size from these and other efforts will yield novel discoveries of DKD genes, the question of how large of a sample size is needed to fully define the genetic factors responsible for DKD remains. Despite a combined sample size that exceeded 40,000 for some analyses, the recent GWAS from SUMMIT uncovered only a single novel DKD locus. Do GWASs for DKD require 50,000 samples to discover all genetic associations that account for its heritability? Or perhaps sample sizes in excess of 100,000, or more, are needed. Although increased statistical power is afforded by these larger scale studies, at what point does increased sample size begin bearing diminished returns?

In surveying this filed, it appears that the fundamental challenge now facing researchers studying the genetics of DKD is not how to increase the sample size of our GWAS, but rather, how to best account for the vast phenotypic heterogeneity of this disease. The current paradigm, based on albuminuria- and eGFR-defined sub-phenotypes, while convenient, fails to consider the true natural history of DKD. Despite their frequent use, it remains to be seen whether the various sub-phenotypes of DKD share a common genetic etiology. It is possible that the genetic factors that predispose diabetic patients to microalbuminuria (early DKD) could differ from those lead to proteinuria (advanced DKD) and the genetic factors that affect albuminuria may be distinct from those that increase a patient’s risk of renal function decline.

Longitudinal investigations of DKD have shed new light on the natural history of this disease and its heterogeneous nature [23, 57, 58, 59, 60, 61]. It is now well-established that while some patients experience a progressive increase in urinary albumin excretion over the course of their disease, many do not. In fact, patients frequently revert from albuminuria to normoalbuminuria [57]. Additionally, the rate of renal function decline varies widely among DKD patients, and perhaps, genetic factors influence whether this decline is slow, and for example, ESRD is only reached after decades of diabetes, or whether this decline is rapid and progression to ESRD occurs a few years after the onset of diabetes [23, 60, 61]. Because of the heterogeneity of the sub-phenotypes of DKD, considering such disparate patients simultaneously, even in GWASs that include tens of thousands of subjects, likely poses challenges to uncovering the genetic basis of DKD.

With relatively few exceptions, cohorts used to investigate genes that contribute to DKD have been cross-sectional in nature. However, because patients with DKD can present with varying levels of albuminuria, at various stages of CKD, and experience different rates of progression of DKD, such designs fail to account for the variable nature of this disease. To begin addressing this, two parallel subprojects from the JDRF-DNCRI are focusing on identifying genetic variants that are associated with longitudinal changes in renal function decline and repeated measures of renal function. However, more studies in this area are needed.

Additionally, the recent emergence of NGS technology is beginning to facilitate more comprehensive interrogation of variants across the genome. To date, only a single study has been reported using this technology [29]. Despite leveraging extreme phenotypes that included early onset advanced DKD cases and long-duration T1D controls, no exome-wide significant associations (p value < 5 × 10− 7) were observed. Nonetheless, this study is a major step toward interrogating low and rare frequency disease predisposing variation that contributes to the risk of DKD. Because such variants are more common among affected relatives compared to unrelated individuals, family-based studies may be more fruitful in this regard and it is likely that renewed interest in such studies will assist in defining the full spectrum of genetic variation that accounts for the heritability of DKD and DKD-related traits.

Lastly, genetic variation does not function in isolation. Indeed, several recent studies have identified novel predictors of DKD, including circulating levels of tumor necrosis factor (TNF) receptor 1 and TNF receptor 2 [62, 63, 64], and uncovered new mechanisms of kidney injury in DKD, including an inflammatory basis of DKD [65, 66]. These investigations are providing insight to the pathogenic mechanisms of DKD as well as beginning to inform improved treatments for patients at risk of DKD [67]. The rapid evolution of ‘multi-omic’ technologies is further enabling researchers to combine ‘omic’ data sets (i.e., genomics, transcriptomics, proteomics, metabolomics, etc.) to better understand the risk factors for DKD. Gene expression varies within an organism by cell type and genetic variations can be linked with gene expression regulation (i.e., expression quantitative trait loci, eQTL). Many publicly available resources provide eQTL annotations for known genetics variants. GTEx is a curated database of gene expression across > 50 tissue types from > 8000 samples, however, this database does not include data from kidney tissue [68]. NephroSeq (http://www.nephroseq.org), an integrative data mining platform of comprehensive renal disease gene expression data sets, is the most comprehensive resource of publicly available data for renal-related gene expression studies. Epigenetic mechanisms, including DNA methylation and acetylation, might also contribute to the risk and progression of DKD [69]. In addition, metabolomics [70, 71, 72] and proteomic [73, 74, 75] profiles are revealing intriguing biomarker signatures. As various DKD ‘omic’ data sets are developed, integrating these data may facilitate SNP prioritization and DKD gene discovery.

Conclusion

Significant progress has been made toward understanding the genetic causes of DKD. Since the first GWAS in 2003, continued advances in genomics and evolving technologies have revolutionized our ability to interrogate genetic variation across the genome. Although several candidate loci and associated variants have been identified to date, our understanding of the genetic basis of DKD is far from complete. Increasingly larger sample sizes are likely to uncover additional risk variants; however, these are expected to only modestly affect risk. Given the complexity of this disease, to continue advancing this area of research and further drive discovery of genes for DKD, the most fruitful approaches to identifying variants for DKD are likely to be those that focus on homogeneous sub-phenotypes of this disease and comprehensively interrogate genetic variation across the entire genome, including low-frequency and rare variants that are not well-captured by current GWAS platforms. As we transition from the GWAS-era to the NGS-era and beyond, it is clear that integrative ‘multi-omics’ platforms will help in fully illuminating the basis for DKD.

Notes

Compliance with ethical standards

Conflict of interest

All authors declare that they have no conflict of interest.

Statement of Human and Animal Rights

All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008 (5).

Statement of Informed Consent

Informed consent was obtained from all patients for being included in the study.

References

  1. 1.
    Krolewski ASWJ (1997) Clinical features and epidemiology of diabetic nephropathy. In: Pickup JCWG (ed) Textbook of diabetes, vol 2, 2nd edn. Blackwell Scientific Publications, Oxford, pp 53.51–53.13Google Scholar
  2. 2.
    Parving HHMM, Ritz E (2004) Diabetic nephropathy. In: BM B (ed) Brenner and Rector’s the kidney, 7th edn. Elsevier, Philadelphia, pp 1777–1818Google Scholar
  3. 3.
    Jones CA, Krolewski AS, Rogus J, Xue JL, Collins A, Warram JH (2005) Epidemic of end-stage renal disease in people with diabetes in the United States population: do we know the cause? Kidney Int 67(5):1684–1691CrossRefPubMedGoogle Scholar
  4. 4.
    Fogarty DG, Rich SS, Hanna L, Warram JH, Krolewski AS (2000) Urinary albumin excretion in families with type 2 diabetes is heritable and genetically correlated to blood pressure. Kidney Int 57(1):250–257CrossRefPubMedGoogle Scholar
  5. 5.
    Forsblom CM, Kanninen T, Lehtovirta M, Saloranta C, Groop LC (1999) Heritability of albumin excretion rate in families of patients with Type II diabetes. Diabetologia 42(11):1359–1366CrossRefPubMedGoogle Scholar
  6. 6.
    Hunter DJ, Lange M, Snieder H, MacGregor AJ, Swaminathan R, Thakker RV, Spector TD (2002) Genetic contribution to renal function and electrolyte balance: a twin study. Clin Sci (Lond) 103(3):259–265CrossRefGoogle Scholar
  7. 7.
    Krolewski AS, Poznik GD, Placha G, Canani L, Dunn J, Walker W, Smiles A et al (2006) A genome-wide linkage scan for genes controlling variation in urinary albumin excretion in type II diabetes. Kidney Int 69(1):129–136CrossRefPubMedGoogle Scholar
  8. 8.
    Langefeld CD, Beck SR, Bowden DW, Rich SS, Wagenknecht LE, Freedman BI (2004) Heritability of GFR and albuminuria in Caucasians with type 2 diabetes mellitus. Am J Kidney Dis 43(5):796–800CrossRefPubMedGoogle Scholar
  9. 9.
    Hipkiss AR, Preston JE, Himsworth DT, Worthington VC, Keown M, Michaelis J, Lawrence J et al (1998) Pluripotent protective effects of carnosine, a naturally occurring dipeptide. Ann N Y Acad Sci 854:37–53CrossRefPubMedGoogle Scholar
  10. 10.
    Lander ES, Linton LM, Birren B, Nusbaum C, Zody MC, Baldwin J, Devon K et al (2001) Initial sequencing and analysis of the human genome. Nature 409(6822):860–921.  https://doi.org/10.1038/35057062 CrossRefPubMedGoogle Scholar
  11. 11.
    International HapMap C (2005) A haplotype map of the human genome. Nature 437(7063):1299–1320.  https://doi.org/10.1038/nature04226 CrossRefGoogle Scholar
  12. 12.
    Reich DE, Lander ES (2001) On the allelic spectrum of human disease. Trends Genet 17(9):502–510CrossRefPubMedGoogle Scholar
  13. 13.
    Pezzolesi MG, Skupien J, Mychaleckyj JC, Warram JH, Krolewski AS (2010) Insights to the genetics of diabetic nephropathy through a genome-wide association study of the GoKinD collection. Semin Nephrol 30(2):126–140.  https://doi.org/10.1016/j.semnephrol.2010.01.004 CrossRefPubMedPubMedCentralGoogle Scholar
  14. 14.
    Wang WY, Barratt BJ, Clayton DG, Todd JA (2005) Genome-wide association studies: theoretical and practical concerns. Nat Rev Genet 6(2):109–118.  https://doi.org/10.1038/nrg1522 CrossRefPubMedGoogle Scholar
  15. 15.
    Grayson BL, Smith ME, Thomas JW, Wang L, Dexheimer P, Jeffrey J, Fain PR et al (2010) Genome-wide analysis of copy number variation in type 1 diabetes. PLoS One 5(11):e15393.  https://doi.org/10.1371/journal.pone.0015393 CrossRefPubMedPubMedCentralGoogle Scholar
  16. 16.
    Wheeler E, Huang N, Bochukova EG, Keogh JM, Lindsay S, Garg S, Henning E et al (2013) Genome-wide SNP and CNV analysis identifies common and low-frequency variants associated with severe early-onset obesity. Nat Genet 45(5):513–517.  https://doi.org/10.1038/ng.2607 CrossRefPubMedPubMedCentralGoogle Scholar
  17. 17.
    Marchini J, Howie B (2010) Genotype imputation for genome-wide association studies. Nat Rev Genet 11(7):499–511.  https://doi.org/10.1038/nrg2796 CrossRefPubMedGoogle Scholar
  18. 18.
    Browning SR, Browning BL (2007) Rapid and accurate haplotype phasing and missing-data inference for whole-genome association studies by use of localized haplotype clustering. Am J Human Genet 81(5):1084–1097.  https://doi.org/10.1086/521987 CrossRefGoogle Scholar
  19. 19.
    Marchini J, Howie B, Myers S, McVean G, Donnelly P (2007) A new multipoint method for genome-wide association studies by imputation of genotypes. Nat Genet 39(7):906–913.  https://doi.org/10.1038/ng2088 CrossRefPubMedGoogle Scholar
  20. 20.
    Li Y, Willer CJ, Ding J, Scheet P, Abecasis GR (2010) MaCH: using sequence and genotype data to estimate haplotypes and unobserved genotypes. Genet Epidemiol 34(8):816–834.  https://doi.org/10.1002/gepi.20533 CrossRefPubMedPubMedCentralGoogle Scholar
  21. 21.
    Auton A, Brooks LD, Durbin RM, Garrison EP, Kang HM, Korbel JO, Marchini JL et al (2015) A global reference for human genetic variation. Nature 526(7571):68–74.  https://doi.org/10.1038/nature15393 CrossRefPubMedGoogle Scholar
  22. 22.
    Najafian B, Mauer M (2012) Morphologic features of declining renal function in type 1 diabetes. Semin Nephrol 32(5):415–422.  https://doi.org/10.1016/j.semnephrol.2012.07.003 CrossRefPubMedPubMedCentralGoogle Scholar
  23. 23.
    Krolewski AS, Bonventre JV (2012) High risk of ESRD in type 1 diabetes: new strategies are needed to retard progressive renal function decline. Semin Nephrol 32(5):407–414.  https://doi.org/10.1016/j.semnephrol.2012.07.002 CrossRefPubMedPubMedCentralGoogle Scholar
  24. 24.
    Finne P, Reunanen A, Stenman S, Groop PH, Gronhagen-Riska C (2005) Incidence of end-stage renal disease in patients with type 1 diabetes. JAMA 294(14):1782–1787.  https://doi.org/10.1001/jama.294.14.1782 CrossRefPubMedGoogle Scholar
  25. 25.
    Harjutsalo V, Maric C, Forsblom C, Thorn L, Waden J, Groop PH (2011) Sex-related differences in the long-term risk of microvascular complications by age at onset of type 1 diabetes. Diabetologia 54(8):1992–1999.  https://doi.org/10.1007/s00125-011-2144-2 CrossRefPubMedGoogle Scholar
  26. 26.
    Sandholm N, Salem RM, McKnight AJ, Brennan EP, Forsblom C, Isakova T, McKay GJ et al (2012) New susceptibility loci associated with kidney disease in type 1 diabetes. PLoS Genet 8(9):e1002921.  https://doi.org/10.1371/journal.pgen.1002921 CrossRefPubMedPubMedCentralGoogle Scholar
  27. 27.
    Teumer A, Tin A, Sorice R, Gorski M, Yeo NC, Chu AY, Li M et al (2016) Genome-wide association studies identify genetic loci associated with albuminuria in diabetes. Diabetes 65(3):803–817.  https://doi.org/10.2337/db15-1313 CrossRefPubMedGoogle Scholar
  28. 28.
    van Zuydam NR, Ahlqvist E, Sandholm N, Deshmukh H, Rayner NW, Abdalla M, Ladenvall C et al (2018) A genome-wide association study of diabetic kidney disease in subjects with Type 2 diabetes. Diabetes.  https://doi.org/10.2337/db17-0914 PubMedGoogle Scholar
  29. 29.
    Sandholm N, Van Zuydam N, Ahlqvist E, Juliusdottir T, Deshmukh HA, Rayner NW, Di Camillo B et al (2017) The genetic landscape of renal complications in Type 1 diabetes. J Am Soc Nephrol 28(2):557–574.  https://doi.org/10.1681/ASN.2016020231 CrossRefPubMedGoogle Scholar
  30. 30.
    Iyengar SK, Sedor JR, Freedman BI, Kao WH, Kretzler M, Keller BJ, Abboud HE et al (2015) Genome-wide association and trans-ethnic meta-analysis for advanced diabetic kidney disease: family investigation of nephropathy and diabetes (FIND). PLoS Genet 11(8):e1005352.  https://doi.org/10.1371/journal.pgen.1005352 CrossRefPubMedPubMedCentralGoogle Scholar
  31. 31.
    Pezzolesi MG, Poznik GD, Mychaleckyj JC, Paterson AD, Barati MT, Klein JB, Ng DP et al (2009) Genome-wide association scan for diabetic nephropathy susceptibility genes in type 1 diabetes. Diabetes 58(6):1403–1410.  https://doi.org/10.2337/db08-1514 CrossRefPubMedPubMedCentralGoogle Scholar
  32. 32.
    Shimazaki A, Kawamura Y, Kanazawa A, Sekine A, Saito S, Tsunoda T, Koya D et al (2005) Genetic variations in the gene encoding ELMO1 are associated with susceptibility to diabetic nephropathy. Diabetes 54(4):1171–1178CrossRefPubMedGoogle Scholar
  33. 33.
    Germain M, Pezzolesi MG, Sandholm N, McKnight AJ, Susztak K, Lajer M, Forsblom C et al (2015) SORBS1 gene, a new candidate for diabetic nephropathy: results from a multi-stage genome-wide association study in patients with type 1 diabetes. Diabetologia 58(3):543–548.  https://doi.org/10.1007/s00125-014-3459-6 CrossRefPubMedGoogle Scholar
  34. 34.
    Tanaka N, Babazono T, Saito S, Sekine A, Tsunoda T, Haneda M, Tanaka Y et al (2003) Association of solute carrier family 12 (sodium/chloride) member 3 with diabetic nephropathy, identified by genome-wide analyses of single nucleotide polymorphisms. Diabetes 52(11):2848–2853CrossRefPubMedGoogle Scholar
  35. 35.
    Mueller PW, Rogus JJ, Cleary PA, Zhao Y, Smiles AM, Steffes MW, Bucksa J et al (2006) Genetics of Kidneys in Diabetes (GoKinD) study: a genetics collection available for identifying genetic susceptibility factors for diabetic nephropathy in type 1 diabetes. J Am Soc Nephrol 17(7):1782–1790.  https://doi.org/10.1681/ASN.2005080822 CrossRefPubMedPubMedCentralGoogle Scholar
  36. 36.
    Pezzolesi MG, Poznik GD, Skupien J, Smiles AM, Mychaleckyj JC, Rich SS, Warram JH et al (2011) An intergenic region on chromosome 13q33.3 is associated with the susceptibility to kidney disease in type 1 and 2 diabetes. Kidney Int 80(1):105–111.  https://doi.org/10.1038/ki.2011.64 CrossRefPubMedPubMedCentralGoogle Scholar
  37. 37.
    Palmer ND, Ng MC, Hicks PJ, Mudgal P, Langefeld CD, Freedman BI, Bowden DW (2014) Evaluation of candidate nephropathy susceptibility genes in a genome-wide association study of African American diabetic kidney disease. PloS One 9(2):e88273.  https://doi.org/10.1371/journal.pone.0088273 CrossRefPubMedPubMedCentralGoogle Scholar
  38. 38.
    Maeda S, Araki S, Babazono T, Toyoda M, Umezono T, Kawai K, Imanishi M et al (2010) Replication study for the association between four Loci identified by a genome-wide association study on European American subjects with type 1 diabetes and susceptibility to diabetic nephropathy in Japanese subjects with type 2 diabetes. Diabetes 59(8):2075–2079.  https://doi.org/10.2337/db10-0067 CrossRefPubMedPubMedCentralGoogle Scholar
  39. 39.
    Sandholm N, McKnight AJ, Salem RM, Brennan EP, Forsblom C, Harjutsalo V, Makinen VP et al (2013) Chromosome 2q31.1 associates with ESRD in women with type 1 diabetes. J Am Soc Nephrol 24(10):1537–1543.  https://doi.org/10.1681/ASN.2012111122 CrossRefPubMedPubMedCentralGoogle Scholar
  40. 40.
    Sandholm N, Forsblom C, Makinen VP, McKnight AJ, Osterholm AM, He B, Harjutsalo V et al (2014) Genome-wide association study of urinary albumin excretion rate in patients with type 1 diabetes. Diabetologia 57(6):1143–1153.  https://doi.org/10.1007/s00125-014-3202-3 CrossRefPubMedGoogle Scholar
  41. 41.
    Kao WH, Klag MJ, Meoni LA, Reich D, Berthier-Schaad Y, Li M, Coresh J et al (2008) MYH9 is associated with nondiabetic end-stage renal disease in African Americans. Nat Genet 40(10):1185–1192CrossRefPubMedGoogle Scholar
  42. 42.
    Kopp JB, Smith MW, Nelson GW, Johnson RC, Freedman BI, Bowden DW, Oleksyk T et al (2008) MYH9 is a major-effect risk gene for focal segmental glomerulosclerosis. Nat Genet 40(10):1175–1184CrossRefPubMedPubMedCentralGoogle Scholar
  43. 43.
    Genovese G, Friedman DJ, Ross MD, Lecordier L, Uzureau P, Freedman BI, Bowden DW et al (2010) Association of trypanolytic ApoL1 variants with kidney disease in African Americans. Science 329(5993):841–845.  https://doi.org/10.1126/science.1193032 CrossRefPubMedPubMedCentralGoogle Scholar
  44. 44.
    Pattaro C, Teumer A, Gorski M, Chu AY, Li M, Mijatovic V, Garnaas M et al (2016) Genetic associations at 53 loci highlight cell types and biological pathways relevant for kidney function. Nat Commun 7:10023.  https://doi.org/10.1038/ncomms10023 CrossRefPubMedPubMedCentralGoogle Scholar
  45. 45.
    Gallagher MD, Chen-Plotkin AS (2018) The post-GWAS era: from association to function. Am J Human Genet 102(5):717–730.  https://doi.org/10.1016/j.ajhg.2018.04.002 CrossRefGoogle Scholar
  46. 46.
    Shimazaki A, Tanaka Y, Shinosaki T, Ikeda M, Watada H, Hirose T, Kawamori R et al (2006) ELMO1 increases expression of extracellular matrix proteins and inhibits cell adhesion to ECMs. Kidney Int 70(10):1769–1776CrossRefPubMedGoogle Scholar
  47. 47.
    Martini S, Nair V, Patel SR, Eichinger F, Nelson RG, Weil EJ, Pezzolesi MG et al (2013) From single nucleotide polymorphism to transcriptional mechanism: a model for FRMD3 in diabetic nephropathy. Diabetes 62(7):2605–2612.  https://doi.org/10.2337/db12-1416 CrossRefPubMedPubMedCentralGoogle Scholar
  48. 48.
    Gumienny TL, Brugnera E, Tosello-Trampont AC, Kinchen JM, Haney LB, Nishiwaki K, Walk SF et al (2001) CED-12/ELMO, a novel member of the CrkII/Dock180/Rac pathway, is required for phagocytosis and cell migration. Cell 107(1):27–41CrossRefPubMedGoogle Scholar
  49. 49.
    Wu HY, Wang Y, Chen M, Zhang X, Wang D, Pan Y, Li L et al (2013) Association of ELMO1 gene polymorphisms with diabetic nephropathy in Chinese population. J Endocrinol Invest 36(5):298–302.  https://doi.org/10.3275/8525 PubMedGoogle Scholar
  50. 50.
    Leak TS, Perlegas PS, Smith SG, Keene KL, Hicks PJ, Langefeld CD, Mychaleckyj JC et al (2009) Variants in intron 13 of the ELMO1 gene are associated with diabetic nephropathy in African Americans. Ann Hum Genet 73(2):152–159CrossRefPubMedPubMedCentralGoogle Scholar
  51. 51.
    Hanson RL, Millis MP, Young NJ, Kobes S, Nelson RG, Knowler WC, DiStefano JK (2010) ELMO1 variants and susceptibility to diabetic nephropathy in American Indians. Mol Genet Metab 101(4):383–390CrossRefPubMedGoogle Scholar
  52. 52.
    Pezzolesi MG, Katavetin P, Kure M, Poznik GD, Skupien J, Mychaleckyj JC, Rich SS et al (2009) Confirmation of genetic associations at ELMO1 in the GoKinD collection supports its role as a susceptibility gene in diabetic nephropathy. Diabetes 58(11):2698–2702CrossRefPubMedPubMedCentralGoogle Scholar
  53. 53.
    Freedman BI, Langefeld CD, Lu L, Divers J, Comeau ME, Kopp JB, Winkler CA et al (2011) Differential effects of MYH9 and APOL1 risk variants on FRMD3 association with diabetic ESRD in African Americans. PLoS Genet 7(6):e1002150.  https://doi.org/10.1371/journal.pgen.1002150 CrossRefPubMedPubMedCentralGoogle Scholar
  54. 54.
    Pezzolesi MG, Jeong J, Smiles AM, Skupien J, Mychaleckyj JC, Rich SS, Warram JH et al (2013) Family-based association analysis confirms the role of the chromosome 9q21.32 locus in the susceptibility of diabetic nephropathy. PloS One 8(3):e60301.  https://doi.org/10.1371/journal.pone.0060301 CrossRefPubMedPubMedCentralGoogle Scholar
  55. 55.
    Bycroft C, Freeman C, Petkova D, Band G, Elliott LT, Sharp K, Motyer A et al (2017) Genome-wide genetic data on ~ 500,000 UK Biobank participants. BioRxiv.  https://doi.org/10.1101/166298 Google Scholar
  56. 56.
    Gaziano JM, Concato J, Brophy M, Fiore L, Pyarajan S, Breeling J, Whitbourne S et al (2016) Million Veteran Program: A mega-biobank to study genetic influences on health and disease. J Clin Epidemiol 70:214–223.  https://doi.org/10.1016/j.jclinepi.2015.09.016 CrossRefPubMedGoogle Scholar
  57. 57.
    Perkins BA, Ficociello LH, Silva KH, Finkelstein DM, Warram JH, Krolewski AS (2003) Regression of microalbuminuria in type 1 diabetes. N Engl J Med 348(23):2285–2293.  https://doi.org/10.1056/NEJMoa021835 CrossRefPubMedGoogle Scholar
  58. 58.
    Perkins BA, Ficociello LH, Ostrander BE, Silva KH, Weinberg J, Warram JH, Krolewski AS (2007) Microalbuminuria and the risk for early progressive renal function decline in type 1 diabetes. J Am Soc Nephrol 18(4):1353–1361.  https://doi.org/10.1681/asn.2006080872 CrossRefPubMedGoogle Scholar
  59. 59.
    Perkins BA, Ficociello LH, Roshan B, Warram JH, Krolewski AS (2010) In patients with type 1 diabetes and new-onset microalbuminuria the development of advanced chronic kidney disease may not require progression to proteinuria. Kidney Int 77(1):57–64.  https://doi.org/10.1038/ki.2009.399 CrossRefPubMedPubMedCentralGoogle Scholar
  60. 60.
    Skupien J, Warram JH, Smiles AM, Niewczas MA, Gohda T, Pezzolesi MG, Cantarovich D et al (2012) The early decline in renal function in patients with type 1 diabetes and proteinuria predicts the risk of end-stage renal disease. Kidney Int 82(5):589–597.  https://doi.org/10.1038/ki.2012.189 CrossRefPubMedPubMedCentralGoogle Scholar
  61. 61.
    Krolewski AS (2015) Progressive renal decline: the new paradigm of diabetic nephropathy in type 1 diabetes. Diabetes Care 38(6):954–962.  https://doi.org/10.2337/dc15-0184 CrossRefPubMedPubMedCentralGoogle Scholar
  62. 62.
    Niewczas MA, Gohda T, Skupien J, Smiles AM, Walker WH, Rosetti F, Cullere X et al (2012) Circulating TNF receptors 1 and 2 predict ESRD in type 2 diabetes. J Am Soc Nephrol 23(3):507–515.  https://doi.org/10.1681/asn.2011060627 CrossRefPubMedPubMedCentralGoogle Scholar
  63. 63.
    Gohda T, Niewczas MA, Ficociello LH, Walker WH, Skupien J, Rosetti F, Cullere X et al (2012) Circulating TNF receptors 1 and 2 predict stage 3 CKD in type 1 diabetes. J Am Soc Nephrol 23(3):516–524.  https://doi.org/10.1681/asn.2011060628 CrossRefPubMedPubMedCentralGoogle Scholar
  64. 64.
    Pavkov ME, Weil EJ, Fufaa GD, Nelson RG, Lemley KV, Knowler WC, Niewczas MA et al (2016) Tumor necrosis factor receptors 1 and 2 are associated with early glomerular lesions in type 2 diabetes. Kidney Int 89(1):226–234.  https://doi.org/10.1038/ki.2015.278 CrossRefPubMedPubMedCentralGoogle Scholar
  65. 65.
    Bassi R, Fornoni A, Doria A, Fiorina P (2016) CTLA4-Ig in B7-1-positive diabetic and non-diabetic kidney disease. Diabetologia 59(1):21–29.  https://doi.org/10.1007/s00125-015-3766-6 CrossRefPubMedGoogle Scholar
  66. 66.
    Baragetti I, El Essawy B, Fiorina P (2017) Targeting immunity in end-stage renal disease. Am J Nephrol 45(4):310–319.  https://doi.org/10.1159/000458768 CrossRefPubMedGoogle Scholar
  67. 67.
    Dellepiane S, Ben Nasr M, Assi E, Usuelli V, Letizia T, D’Addio F, Zuccotti GV et al (2018) Sodium glucose cotransporters inhibitors in type 1 diabetes. Pharmacol Res 133:1–8.  https://doi.org/10.1016/j.phrs.2018.04.018 CrossRefPubMedGoogle Scholar
  68. 68.
    The Genotype-Tissue Expression (GTEx) project (2013) Nat Genetics 45(6):580–585.  https://doi.org/10.1038/ng.2653 CrossRefGoogle Scholar
  69. 69.
    McKnight AJ, McKay GJ, Maxwell AP (2014) Genetic and epigenetic risk factors for diabetic kidney disease. Adv Chron Kidney Dis 21(3):287–296.  https://doi.org/10.1053/j.ackd.2014.03.010 CrossRefGoogle Scholar
  70. 70.
    Hirayama A, Nakashima E, Sugimoto M, Akiyama S, Sato W, Maruyama S, Matsuo S et al (2012) Metabolic profiling reveals new serum biomarkers for differentiating diabetic nephropathy. Anal Bioanal Chem 404(10):3101–3109.  https://doi.org/10.1007/s00216-012-6412-x CrossRefPubMedGoogle Scholar
  71. 71.
    Pena MJ, Lambers Heerspink HJ, Hellemons ME, Friedrich T, Dallmann G, Lajer M, Bakker SJ et al (2014) Urine and plasma metabolites predict the development of diabetic nephropathy in individuals with Type 2 diabetes mellitus. Diabet Med 31(9):1138–1147.  https://doi.org/10.1111/dme.12447 CrossRefPubMedGoogle Scholar
  72. 72.
    Stec DF, Wang S, Stothers C, Avance J, Denson D, Harris R, Voziyan P (2015) Alterations of urinary metabolite profile in model diabetic nephropathy. Biochem Biophys Res Commun 456(2):610–614.  https://doi.org/10.1016/j.bbrc.2014.12.003 CrossRefPubMedGoogle Scholar
  73. 73.
    Zurbig P, Jerums G, Hovind P, Macisaac RJ, Mischak H, Nielsen SE, Panagiotopoulos S et al (2012) Urinary proteomics for early diagnosis in diabetic nephropathy. Diabetes 61(12):3304–3313.  https://doi.org/10.2337/db12-0348 CrossRefPubMedPubMedCentralGoogle Scholar
  74. 74.
    Zubiri I, Posada-Ayala M, Sanz-Maroto A, Calvo E, Martin-Lorenzo M, Gonzalez-Calero L, de la Cuesta F et al (2014) Diabetic nephropathy induces changes in the proteome of human urinary exosomes as revealed by label-free comparative analysis. J Proteom 96:92–102.  https://doi.org/10.1016/j.jprot.2013.10.037 CrossRefGoogle Scholar
  75. 75.
    Caseiro A, Barros A, Ferreira R, Padrao A, Aroso M, Quintaneiro C, Pereira A et al (2014) Pursuing type 1 diabetes mellitus and related complications through urinary proteomics. Transl Res 163(3):188–199.  https://doi.org/10.1016/j.trsl.2013.09.005 CrossRefPubMedGoogle Scholar

Copyright information

© Springer-Verlag Italia S.r.l., part of Springer Nature 2018

Authors and Affiliations

  1. 1.Division of Nephrology and Hypertension, Department of Internal Medicine,University of Utah School of MedicineSalt Lake CityUSA
  2. 2.VA Boston Healthcare SystemVA Cooperative Studies ProgramBostonUSA
  3. 3.Diabetes and Metabolism CenterUniversity of Utah School of MedicineSalt Lake CityUSA
  4. 4.Department of Human GeneticsUniversity of Utah School of MedicineSalt Lake CityUSA

Personalised recommendations