Genetics of Diabetic Nephropathy

Are common causes of progressive kidney disease regulated by genes?

Many common diseases, including nephropathy, cluster in families, and genetic variants seem likely to regulate disease pathogenesis (1). Until recently, convincing evidence that common disease genes exist has been lacking. Much of the difficulty in identifying genes for common diseases, such as diabetic nephropathy, sporadic FSGS, and nephrosclerosis, arises from the genetic architecture responsible for common diseases, which differs from that of Mendelian disorders.

Most of us learned about Mendel and his peas in medical school. The success of gene mapping for Mendelian disorders, such as polycystic kidney disease, is common knowledge to nephrologists and other physicians. Mendelian traits have a simple correspondence between the gene and the disease phenotype, and mapping strategies have been highly successful in identifying genetic causes of disease with understandable patterns of inheritance.

In contrast, complex traits, like chronic kidney disease, do not have recognizable inheritance patterns, a finding that has made gene mapping more difficult. The frequency of the causal variant in the population and the size of its contribution to the disease phenotypes (the effect size) determine optimal gene mapping strategies (Figure 1). Investigators exploring genetic causes of common diseases still have not reached consensus on the underlying genetic architecture for these disorders.

Figure 1.

Relationship between the frequency of a genetic variant in the population, its effect size, and gene mapping strategy


Two models are dominant. The “common disease, common variant” hypothesis proposes that common diseases are caused by genetic variations having a frequency greater than 5 percent in the population. The functional effects of these variants impair but do not prevent protein function. An illustrative analogy would be that these common variants cause a light bulb to dim, but not go out. Case-control mapping strategies should be successful in identifying causal variants if this hypothesis is correct.

The competing hypothesis is that common diseases are caused by rare variants with moderate effect sizes that generate dysfunctional proteins with little or no wild type function. In this model, common diseases would be caused by a multitude of rare variants in multiple genes. Current mapping strategies are unlikely to identify these gene variants but new technologies—such as exon and whole genome sequencing—seem likely to identify causal rare variants. These two hypotheses are not mutually exclusive. Age-related macular degeneration has been associated with a common variant that causes an amino acid substitution in a complement regulatory protein. In contrast, rare genetic variants have been found to regulate blood pressure and dyslipidemia phenotypes.

Genes and diabetic nephropathy

Over the past decade, many studies have reported association of variants in biologically plausible candidate genes for diabetic nephropathy using case-control analytic strategies. However, few of these candidate gene associations have been replicated owing to study population heterogeneity, small sample sizes, inconsistent phenotype criteria between studies, and incomplete knowledge of disease pathophysiology.

In addition to candidate gene association studies, families with diabetic nephropathy have been collected for linkage analysis in both the United States and Europe (1). In contrast to focus on specific genes, linkage analysis is unbiased; no specific gene is hypothesized to be causal. Consistent linkage signals across studies, composed of African American, American Indian, and European populations, have been identified on chromosomes 3q, 7p, 10p, and 18q (1,2).

Further mapping of the 18q regions subsequently identified carnosinase 1, a gene that encodes an enzyme whose substrate inhibits ACE activity and advanced glycation end product formation, as a diabetic nephropathy susceptibility gene. Variants within the engulfment and motility 1 (ELMO1) gene, whose protein regulates cell migration and matrix and TGFβ expression, have been associated with diabetic nephropathy phenotypes and are located within the linkage signal on 7p. The linkage signal on 3q has been attributed to variants with adiponectin (ADIPOQ), which encodes an adipose-tissue-derived protein with anti-diabetic, anti-atherogenic and anti-inflammatory functions, and NCK1 (3), a gene that encodes an adapter protein NCK1 that links actin and nephrin in podocyte foot processes.

More recently, genewide association analysis (GWA), another unbiased strategy. has been applied to gene mapping for common disease. This approach uses a group of patients with disease and controls numbering in the thousands and takes advantage of the technical advances in genotyping platforms, which incorporate genetic variants from the International HapMap Project, an atlas of common genetic similarities and differences in human beings of differing ancestries. These studies have reproducibly identified common variants with highly statistically significant associations with common disease phenotypes, although the underlying biology responsible for the association can be obscure.

The first GWA data for diabetic nephropathy have now been reported by the Genetics of Kidneys in Diabetes (GoKinD) consortium, which included over 800 patients with Type 1 diabetic nephropathy as cases and 800 diabetic patients without nephropathy as controls (4). In this study, the most significant associations were within noncoding regions or close to the genes CHN2 (encodes chimerin 2 located on chromosome 7p, a region identified in linkage analyses, noted above, FRMD3 (encodes a FERM domain-containing protein), and CARS (encodes cysteinyl-tRNA synthetase). An intergenic region on chromosome 13q was also associated with diabetic nephropathy.

The GOKinD results highlight a strength of the GWA approach: the results point to novel pathways not previously considered as causal for diabetic nephropathy. Although the biology responsible for the association of these genes with diabetic nephropathy has yet to be discovered, these results were replicated in a prospective cohort from the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications Research Group. Results from ongoing association studies from the Family Investigation of Diabetes and Nephropathy (FIND) and Finnish Diabetic Nephropathy Study consortia should be available soon.

Genes and other causes of chronic kidney diseases

Using ancestry mapping, two consortia—the U.S. National Institutes of Health (NIH) FSGS Genetic Study and FIND—reported that common variants in MYH9, a gene that encodes a ubiquitous intracellular motor protein, myosin 2a, are associated with nondiabetic but not diabetic nephropathy in African American patients (5). The NIH patient sample included individuals with idiopathic or HIV-1-associated FSGS; the FIND group included African American nondiabetic ESRD patients. Much of the excess risk for kidney disease in African American patients can be explained by four genetic variants within MYH9 noncoding regions, suggesting these common variants have large effect size (Figure 1). The findings have been replicated in multiple, independent African American populations. The mechanism(s) by which MYH9 gene variants cause nondiabetic kidney diseases in African American patients is under intense study, as are approaches to apply this finding to patient care.

Finally, the Cohorts for Heart and Aging Research in Genetic Epidemiology (CHARGE) consortium reported that all-cause CKD, defined as estimated GFR less than 60 mL/min, was associated with common genetic variants in noncoding regions of the gene that encodes Tamm-Horsfall protein (THP) or uromodulin (UMOD) (6). Uromodulin/THP provides a protein scaffold for the urinary casts used to diagnose kidney diseases in the clinic. Patients with diabetes were included in this sample. Interestingly, UMOD mutations cause medullary cystic disease and familial hyperuricemic nephropathy, suggesting both common and rare variants with the same gene may regulate different phenotypic presentations of kidney diseases. However, in contrast to MYH9 variants, UMOD variants only explain a small percentage of the variability in the CKD phenotype (Figure 1). A subsequent study from CHARGE has shown urinary uromodulin levels can predict development of nephropathy.

Will studies of common kidney disease genetics impact patient management?

The last seven years have been a time of breathtaking discovery and technical advancement in human genetics: The human genome sequence was reported in 2003. The atlas of common genetic variation in individuals of different ancestries—the Hap Map—was completed in 2005. The 1000 genomes project, which will expand our understanding of the range of human genetic variation, is about to publically release its first data set.

The kidney community is actively applying the most advanced gene mapping strategies to understand the underlying genetic architecture of common kidney disease, especially diabetic nephropathy. Understandably, both nephrologists and kidney disease patients are hopeful these findings will lead to new therapies and tests that identify individuals at risk for kidney disease progression.

Although novel pathways that potentially regulate the pathogenesis of diabetic nephropathy and other chronic kidney diseases are being identified, much work needs to be done. The common variants associated with most common diseases including kidney disease only explain a small percentage of overall risk. We need to understand the mechanisms responsible for the missing heritability. Although genetic tests are marketed directly to consumers, studies of their clinical validity and utility, which are required of all other laboratory tests, must to be established. Of equal importance, we must understand how our patients will respond to genetic risk information for kidney diseases. Despite these issues, I am confident that studies of genetic causes of kidney and other common chronic diseases will positively impact personal and public health and help stem the worldwide epidemic of chronic diseases (7).


[1] John R. Sedor, MD, is with the department of medicine, MetroHealth System Campus; CWRU Center for the Study of Kidney Disease and Biology; department of physiology and biophysics, School of Medicine, Case Western Reserve University, Cleveland, OH.



Iyengar SK, Freedman BI, Sedor JR. Mining the genome for susceptibility to diabetic nephropathy: the role of large-scale studies and consortia. Semin Nephrol 2007; 27:208–222.


Freedman BI, Bostrom M, Daeihagh P, Bowden DW. Genetic factors in diabetic nephropathy. Clin J Am Soc Nephrol 2007; 2:1306–1316.


He B, Osterholm AM, Hoverfalt A, et al. Association of genetic variants at 3q22 with nephropathy in patients with type 1 diabetes mellitus. Am J Hum Genet 2009; 84:5–13.


Pezzolesi MG, Poznik GD, Mychaleckyj JC, et al. Genome-wide association scan for diabetic nephropathy susceptibility genes in type 1 diabetes mellitus. Diabetes 2009; 1508–1514.


Freedman BI, Sedor JR. Hypertension-associated kidney disease: perhaps no more. J Am Soc Nephrol 2008; 19:2047–2051.


Sedor JR. Uromodulin and translational medicine: will the SNPs bring zip to clinical practice? J Am Soc Nephrol 2010; 21:204–206.


Daar AS, Singer PA, Leah PD, et al. Grand challenges in chronic non-communicable diseases. Nature 2007; 450:494–496.