Case Report Genomic and Epigenomic Analyses of Monozygotic Twins Discordant for Congenital Renal Agenesis Meiling Jin, MS,1,2,* Shida Zhu, MB,3,* Panpan Hu, MB,1,* Dongbing Liu, MB,3 Qinggang Li, MD, PhD,1 Zuoxiang Li, MS,1 Xueguang Zhang, MD, PhD,1 Yuansheng Xie, MD, PhD,1 and Xiangmei Chen, MD, PhD1 Monozygotic twins have been widely studied to distinguish genetic and environmental factors in the pathogenesis of human diseases. For renal agenesis, the one-sided absence of renal tissue, the relative contributions of genetic and environmental factors to its pathogenesis are still unclear. In this study of a pair of monozygotic twins discordant for congenital renal agenesis, the genomic profile was analyzed from a set of blood samples using high-throughput exome-capture sequencing to detect single-nucleotide polymorphisms (SNPs), copy number variations (CNVs), and insertions and deletions (indels). Also, an epigenomic analysis used reduced-representation bisulfite sequencing to detect differentially methylated regions (DMRs). No discordant SNPs, CNVs, or indels were confirmed, but 514 DMRs were detected. KEGG analysis indicated the DMRs localized to 10 signaling pathways and 25 genes, including the mitogen-activated protein kinase pathway and 6 genes (FGF18, FGF12, PDGFRA, MAPK11, AMH, CTBP1) involved in organ development. Although methylation results from our adult patient and her sister may not represent the pattern that was present during kidney development, we could at least confirm a lack of obvious differences at the genome level, which suggests that nongenetic factors may be involved in the pathogenesis of renal agenesis. Am J Kidney Dis. -(-):---. ª 2014 by the National Kidney Foundation, Inc. INDEX WORDS: Renal agenesis; monozygotic twins; genome; epigenome; congenital abnormalities of the kidney and urinary tract (CAKUT); methylation; nephrogenesis.
C
ongenital abnormalities of the kidney and urinary tract (CAKUT), which are defined as structural and functional malformations resulting in defective morphogenesis of the kidney and/or urinary tract,1 occur in approximately 1 per 500 newborns and constitute approximately 20%-30% of anomalies found in neonates.2,3 They are the most frequent cause of chronic kidney disease in children.4,5 Renal agenesis is a common CAKUT6,7 and is defined as the one-sided absence of renal tissue, which results from the failure of embryonic development of the metanephros or the absence of the nephrogenic cord.8 Its cause is believed to be complex and may be affected by genetic and environment factors.9 In recent years, monozygotic twins have become an important experimental model to distinguish genetic and environmental factors in the pathogenesis of multifactorial diseases.10,11 Monozygotic twins are presumed to be identical in genotype and different in phenotype due to environmental factors. Recent studies have found genetic and epigenetic differences between monozygotic twins.12,13 Genetic differences may result from somatic mosaicism,12 and environmental factors may contribute to the epigenetic changes, especially CpG methylation.14 In this study, single-nucleotide polymorphisms (SNPs), copy number variations (CNVs), insertions and deletions (indels), and differentially methylated regions (DMRs) were analyzed in a pair of monozygotic twins discordant for congenital renal agenesis. This analysis aimed to dissect the relative contributions of Am J Kidney Dis. 2014;-(-):---
genetics and environment to the pathogenesis of renal agenesis.
CASE REPORT A 30-year-old Chinese woman given a diagnosis of congenital renal agenesis of the left kidney at birth had a healthy monozygotic twin sister who did not have renal agenesis. The patient had no particular past or family history and no notable clinical symptoms. This study was conducted in accordance with the principles of the Declaration of Helsinki and has been reviewed and approved by the Ethics Committee of the Chinese PLA General Hospital (ethics approval number 2012-001). Written informed consent was obtained from both women. To confirm that the twins were monozygotic, we analyzed short tandem repeat (STR) loci and amelogenin with a DNA sequencer (detailed methods for this and all other procedures described in this section are provided in Item S1, available as online
From the 1State Key Laboratory of Kidney Disease, Institute of Nephrology, Chinese PLA General Hospital, Beijing; 2Medical College, Nankai University, Tianjin; and 3BGI-Shenzhen, Shenzhen, China. * M.J., S.Z., and P.H. contributed equally to this article. Received October 21, 2013. Accepted in revised form January 6, 2014. Address correspondence to Yuansheng Xie, MD, PhD, State Key Laboratory of Kidney Disease, Institute of Nephrology, Chinese PLA General Hospital, 28th Fuxing Road, Beijing 100853, China (e-mail: [email protected]) or Xiangmei Chen, MD, PhD, State Key Laboratory of Kidney Disease, Institute of Nephrology, Chinese PLA General Hospital, 28th Fuxing Road, Beijing 100853, China (e-mail: [email protected]). 2014 by the National Kidney Foundation, Inc. 0272-6386/$36.00 http://dx.doi.org/10.1053/j.ajkd.2014.01.423 1
Jin et al supplementary material). As shown in Table 1, STR loci were consistent between the 2 samples, which suggests that this pair of twins was monozygotic. Next we looked for SNPs that might differ between the patient and her twin. This analysis was done using the NimbleGen exome capture and sequencing system, followed by data analysis with GATK (Genome Analysis Toolkit15). As seen in Tables S1 and S2, we detected 97,039 SNPs in the patient and 97,814 SNPs in her twin; of these, 93,663 SNPs were identical between the 2 samples. First, as a check of the reliability of our SNP data, we compared results from our patient and her twin with 5 other twin pairs (Table S3). The genotype consistencies we observed in the twins in our study were within the range of other monozygotic twins and differed from data relating to a dizygotic twin pair. Thus, we concluded that the SNP results in this study are of acceptable accuracy. Next, we identified putative differential SNPs by determining the appropriate cutoff value denoting a substantial difference between the 2 samples in the fraction of reads supporting the variant allele (DiffSuppReads in Item S1). In all, we detected 30 putatively significant differential SNPs (Table S4); however, all were found to be false positives after manual inspection (Table S5). We then analyzed the data using MuTect,16 which usually is used to detect somatic point mutations in next-generation sequencing data of cancer genomes. These point mutations between paired samples are referred to as single-nucleotide variations (SNVs). In the patient and her sister, we detected 95 and 64 SNVs, respectively. However, only a small fraction (,5%) of the reads of the loci
Table 1. Short Tandem Repeat Typing Results for the Patient and Her Twin Sister Patient Marker
Allele 1
AMEL
Twin Sister
Allele 2
Allele 1
X
Allele 2
X
D3S1358
15
15
15
15
D1S1656
16
16.3
16
16.3
D6S1043
12
15
12
15
D13S317
10
11
10
11
Penta E D16S539
12 10
20 11
12 10
20 11
D18S51
12
14
12
14
D2S1338
18
24
18
24
CSF1P0
11
12
11
12
Penta D
9
11
9
11
THO1
7
9
7
9
vWA
14
17
14
17
D21S11 D7S820
29 9
32.2 10
29 9
32.2 10
D5S818
11
13
11
13
8
11
8
11
D8S1179
13
15
13
15
D12S391
18
21
18
21
D19S433
13
13
13
13
FGA
23
26
23
26
TPOX
Note: Analysis performed using a human identification kit that amplifies 20 short tandem repeat loci and the amelogenin locus (for sex typing). Values shown are the number of short tandem repeats detected at each allele of each locus. 2
containing putative SNVs indicated a difference between the 2 samples (each locus is sequenced multiple times; only a low proportion of these reads gave a different sequence in one twin vs the other). This low mutation frequency means that there is a high possibility of the putative SNV being a false positive; these putative SNVs were discarded in subsequent data filtering. Thus, the MuTect analysis provided further evidence of a lack of single-nucleotide sequence differences between the 2 samples in this study. Another way that genomes can differ is through CNVs. Using ExomeCNV,17 we detected 3,535 focal CNVs; however, the CNVs did not significantly differ after CONTRA (copy number analysis for targeted resequencing) filtering18 (P 5 0.3 in the minimum target zone; Table S6), suggesting the absence of bona fide CNVs between the twins. In addition, we used GATK to detect somatic indels. For this analysis, we defined the patient’s sample as “normal” and her twin’s as “tumor” (the process is intended for detecting indels that arise in tumor tissue). Sixty indels were detected by GATK, but all were deemed to be false positives after verification (Table S7). To look for epigenomic differences between the patient and her sister, we focused on DMRs. We performed this analysis by reduced-representation bisulfite sequencing19,20 (Fig 1; Tables S8 and S9). We then performed a KEGG (Kyoto Encyclopedia of Genes and Genomes) analysis and found that 10 pathways and 25 genes were involved, as seen in Table S10. Among the 10 pathways, the mitogen-activated protein kinase (MAPK) pathway (including MAP kinase, JNK/p38 MAP kinase pathway, and ERK5 pathway) is well known to take part in the regulation of proliferation, differentiation, and apoptosis of cells. However, an association between the 9 other pathways and kidney development is not supported by existing evidence. Among the 25 genes, 6 (FGF18, FGF12, PDGFRA, MAPK11, AMH, and CTBP1) are involved in organ development, although there is no evidence that they are associated with kidney development.
DISCUSSION This study had 2 innovative aspects. First, cases of monozygotic twins discordant for congenital renal agenesis are very rare. To our knowledge, this study constitutes the first analysis of SNPs, CNVs, indels, and DMRs to directly observe the differences resulting from genetic and environmental factors in monozygotic twins discordant for congenital renal agenesis. Second, this study provides evidence suggesting a lack of differences at the sequence level (SNPs, CNVs, and indels) and the presence of epigenetic differences (at least in terms of differential methylation) in a pair of monozygotic twins who were discordant for congenital renal agenesis. Thus, we suggest that environmental factors, not genetic factors, may lead to the pathogenesis of renal agenesis by epigenetic changes. In week 5 of gestation, the ureteric bud, originating from the Wolffian duct, penetrates the metanephric blastema. After 2 further weeks of gestation, the ureteric bud induces nephrogenesis. By week 20, the ureteric bud has branched and created the collecting duct system in its entirety. However, at that stage, nephrogenesis is just 30% complete, and there is further maturation until week 36.21 Because methylation changes over time, the DMRs detected in this study of 30-year-old twins may not represent the Am J Kidney Dis. 2014;-(-):---
Genetic and Environmental Factors in the Pathogenesis of Renal Agenesis
Figure 1. Landscape of differentially methylated regions (DMRs). We identified 514 DMRs in the studied pair of twins, ranging from 3-52 per chromosome (denoted by bars, which are associated with the left vertical axis). DMR frequency, defined as the length of the DMR divided by the length of the enzyme digestion fragment, ranged from 0.03% to .0.13% in different chromosomes (denoted by the line, which is associated with the right vertical axis).
DMRs present during kidney development. Thus, a larger number of cases would be needed to more conclusively link these DMRs to effects on kidney development. Also, because epigenetic changes accumulate during aging, epigenetic analysis of younger twins, especially newborns, would be preferable to study the pathogenesis of renal agenesis. In addition, although our study revealed genomic similarity and epigenetic differences between the twins, other effects also could contribute to phenotype diversity. To confirm or eliminate these effects, more pairs of samples should be recruited in future work. Previous studies have found that gene mutations could induce CAKUT, including renal agenesis.1,22-24 It has been reported that some candidate genes may associate with renal agenesis.1 Previous studies have suspected that the pathogenesis of renal agenesis is multifactorial (including genetic and environmental factors), but this notion was only a hypothesis. In this study, we failed to identify an obvious mutation on genome level. Although the methylation results would need further verification, our findings suggest that nongenetic factors could cause renal agenesis, which may provide new insight into the pathogenesis of renal agenesis. In this study, blood samples from a pair of monozygotic twins discordant for renal agenesis were analyzed using high-throughput exome-capture sequencing (including analysis of SNPs, CNVs, and indels) and an epigenomic analysis using reducedrepresentation bisulfite sequencing to detect DMRs. We saw no evidence of discordances in SNPs, CNVs, and indels. However, 514 DMRs were detected in reduced-representation bisulfite sequencing. Thus, we suggest that nongenetic (environmental) factors may be important in the pathogenesis of renal agenesis. Am J Kidney Dis. 2014;-(-):---
ACKNOWLEDGEMENTS We thank the volunteers for participating in this study. Support: This work was supported by a grant (2011CB944004) from the National Basic Research Program of China. Financial Disclosure: The authors declare that they have no other relevant financial interests.
SUPPLEMENTARY MATERIAL Table S1: Overview of data production. Table S2: Summary of SNPs in exome-capture sample. Table S3: Comparison of SNP consistency between 6 pairs of twins. Table S4: Differential SNPs detected in primary screening. Table S5: Manual checking of the 30 putatively differential SNPs. Table S6: Processing of CONTRA results. Table S7: Manual checking of the 60 putatively differential indels. Table S8: Summary of DMRs. Table S9: Gene locations of DMRs. Table S10: KEGG analysis of DMRs. Item S1: Detailed methods. Note: The supplementary material accompanying this article (http://dx.doi.org/10.1053/j.ajkd.2014.01.423) is available at www.ajkd.org
REFERENCES 1. Saisawat P, Tasic V, Vega-Warner V, et al. Identification of two novel CAKUT-causing genes by massively parallel exon resequencing of candidate genes in patients with unilateral renal agenesis. Kidney Int. 2012;81(2):196-200. 2. Pohl M, Bhatnagar V, Mendoza SA, Nigam SK. Toward an etiological classification of developmental disorders of the kidney and upper urinary tract. Kidney Int. 2002;61(1):10-19. 3. Woolf AS. A molecular and genetic view of human renal and urinary tract malformations. Kidney Int. 2000;58(2): 500-512. 4. Sanna-Cherchi S, Ravani P, Corbani V, et al. Renal outcome in patients with congenital anomalies of the kidney and urinary tract. Kidney Int. 2009;76(5):528-533. 3
Jin et al 5. Mansoor O, Chandar J, Rodriguez MM, et al. Long-term risk of chronic kidney disease in unilateral multicystic dysplastic kidney. Pediatr Nephrol. 2011;26(4):597-603. 6. Jelen Z. The value of ultrasonography as a screening procedure of the neonatal urinary tract: a survey of 1021 infants. Int Urol Nephrol. 1993;25(1):3-10. 7. Stoll C, Alembik Y, Roth MP, Dott B, Sauvage P. Risk factors in internal urinary system malformations. Pediatr Nephrol. 1990;4(4):319-323. 8. Kerecuk L, Schreuder MF, Woolf AS. Renal tract malformations: perspectives for nephrologists. Nat Clin Pract Nephrol. 2008;4(6):312-325. 9. Parikh CR, McCall D, Engelman C, Schrier RW. Congenital renal agenesis: case-control analysis of birth characteristics. Am J Kidney Dis. 2002;39(4):689-694. 10. Baranzini SE, Mudge J, van Velkinburgh JC, et al. Genome, epigenome and RNA sequences of monozygotic twins discordant for multiple sclerosis. Nature. 2010;464(7293): 1351-1356. 11. Furukawa H, Oka S, Matsui T, et al. Genome, epigenome and transcriptome analyses of a pair of monozygotic twins discordant for systemic lupus erythematosus. Hum Immunol. 2013;74(2): 170-175. 12. Bruder CE, Piotrowski A, Gijsbers AA, et al. Phenotypically concordant and discordant monozygotic twins display different DNA copy-number-variation profiles. Am J Hum Genet. 2008;82(3):763-771. 13. Javierre BM, Fernandez AF, Richter J, et al. Changes in the pattern of DNA methylation associate with twin discordance in systemic lupus erythematosus. Genome Res. 2010;20(2):170-179. 14. Heijmans BT, Tobi EW, Stein AD, et al. Persistent epigenetic differences associated with prenatal exposure to famine in humans. Proc Natl Acad Sci U S A. 2008;105(44):17046-17049.
4
15. McKenna A, Hanna M, Banks E, et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing nextgeneration DNA sequencing data. Genome Res. 2010;20(9): 1297-1303. 16. Cibulskis K, Lawrence MS, Carter SL, et al. Sensitive detection of somatic point mutations in impure and heterogeneous cancer samples. Nat Biotechnol. 2013;31(3):213-219. 17. Sathirapongsasuti JF, Lee H, Horst BA, et al. Exome sequencing-based copy-number variation and loss of heterozygosity detection: ExomeCNV. Bioinformatics. 2011;27(19):2648-2654. 18. Li J, Lupat R, Amarasinghe KC, et al. CONTRA: copy number analysis for targeted resequencing. Bioinformatics. 2012;28(10):1307-1313. 19. Meissner A, Gnirke A, Bell GW, Ramsahoye B, Lander ES, Jaenisch R. Reduced representation bisulfite sequencing for comparative high-resolution DNA methylation analysis. Nucleic Acids Res. 2005;33(18):5868-5877. 20. Gu H, Smith ZD, Bock C, Boyle P, Gnirke A, Meissner A. Preparation of reduced representation bisulfite sequencing libraries for genome-scale DNA methylation profiling. Nat Protoc. 2011;6(4):468-481. 21. Yalavarthy R, Parikh CR. Congenital renal agenesis: a review. Saudi J Kidney Dis Transpl. 2003;14(3):336-341. 22. Abdelhak S, Kalatzis V, Heilig R, et al. Clustering of mutations responsible for branchio-oto-renal (BOR) syndrome in the eyes absent homologous region (eyaHR) of EYA1. Hum Mol Genet. 1997;6(13):2247-2255. 23. Lu W, van Eerde AM, Fan X, et al. Disruption of ROBO2 is associated with urinary tract anomalies and confers risk of vesicoureteral reflux. Am J Hum Genet. 2007;80(4):616-632. 24. Weber S, Taylor JC, Winyard P, et al. SIX2 and BMP4 mutations associate with anomalous kidney development. J Am Soc Nephrol. 2008;19(5):891-903.
Am J Kidney Dis. 2014;-(-):---
C
ongenital abnormalities of the kidney and urinary tract (CAKUT), which are defined as structural and functional malformations resulting in defective morphogenesis of the kidney and/or urinary tract,1 occur in approximately 1 per 500 newborns and constitute approximately 20%-30% of anomalies found in neonates.2,3 They are the most frequent cause of chronic kidney disease in children.4,5 Renal agenesis is a common CAKUT6,7 and is defined as the one-sided absence of renal tissue, which results from the failure of embryonic development of the metanephros or the absence of the nephrogenic cord.8 Its cause is believed to be complex and may be affected by genetic and environment factors.9 In recent years, monozygotic twins have become an important experimental model to distinguish genetic and environmental factors in the pathogenesis of multifactorial diseases.10,11 Monozygotic twins are presumed to be identical in genotype and different in phenotype due to environmental factors. Recent studies have found genetic and epigenetic differences between monozygotic twins.12,13 Genetic differences may result from somatic mosaicism,12 and environmental factors may contribute to the epigenetic changes, especially CpG methylation.14 In this study, single-nucleotide polymorphisms (SNPs), copy number variations (CNVs), insertions and deletions (indels), and differentially methylated regions (DMRs) were analyzed in a pair of monozygotic twins discordant for congenital renal agenesis. This analysis aimed to dissect the relative contributions of Am J Kidney Dis. 2014;-(-):---
genetics and environment to the pathogenesis of renal agenesis.
CASE REPORT A 30-year-old Chinese woman given a diagnosis of congenital renal agenesis of the left kidney at birth had a healthy monozygotic twin sister who did not have renal agenesis. The patient had no particular past or family history and no notable clinical symptoms. This study was conducted in accordance with the principles of the Declaration of Helsinki and has been reviewed and approved by the Ethics Committee of the Chinese PLA General Hospital (ethics approval number 2012-001). Written informed consent was obtained from both women. To confirm that the twins were monozygotic, we analyzed short tandem repeat (STR) loci and amelogenin with a DNA sequencer (detailed methods for this and all other procedures described in this section are provided in Item S1, available as online
From the 1State Key Laboratory of Kidney Disease, Institute of Nephrology, Chinese PLA General Hospital, Beijing; 2Medical College, Nankai University, Tianjin; and 3BGI-Shenzhen, Shenzhen, China. * M.J., S.Z., and P.H. contributed equally to this article. Received October 21, 2013. Accepted in revised form January 6, 2014. Address correspondence to Yuansheng Xie, MD, PhD, State Key Laboratory of Kidney Disease, Institute of Nephrology, Chinese PLA General Hospital, 28th Fuxing Road, Beijing 100853, China (e-mail: [email protected]) or Xiangmei Chen, MD, PhD, State Key Laboratory of Kidney Disease, Institute of Nephrology, Chinese PLA General Hospital, 28th Fuxing Road, Beijing 100853, China (e-mail: [email protected]). 2014 by the National Kidney Foundation, Inc. 0272-6386/$36.00 http://dx.doi.org/10.1053/j.ajkd.2014.01.423 1
Jin et al supplementary material). As shown in Table 1, STR loci were consistent between the 2 samples, which suggests that this pair of twins was monozygotic. Next we looked for SNPs that might differ between the patient and her twin. This analysis was done using the NimbleGen exome capture and sequencing system, followed by data analysis with GATK (Genome Analysis Toolkit15). As seen in Tables S1 and S2, we detected 97,039 SNPs in the patient and 97,814 SNPs in her twin; of these, 93,663 SNPs were identical between the 2 samples. First, as a check of the reliability of our SNP data, we compared results from our patient and her twin with 5 other twin pairs (Table S3). The genotype consistencies we observed in the twins in our study were within the range of other monozygotic twins and differed from data relating to a dizygotic twin pair. Thus, we concluded that the SNP results in this study are of acceptable accuracy. Next, we identified putative differential SNPs by determining the appropriate cutoff value denoting a substantial difference between the 2 samples in the fraction of reads supporting the variant allele (DiffSuppReads in Item S1). In all, we detected 30 putatively significant differential SNPs (Table S4); however, all were found to be false positives after manual inspection (Table S5). We then analyzed the data using MuTect,16 which usually is used to detect somatic point mutations in next-generation sequencing data of cancer genomes. These point mutations between paired samples are referred to as single-nucleotide variations (SNVs). In the patient and her sister, we detected 95 and 64 SNVs, respectively. However, only a small fraction (,5%) of the reads of the loci
Table 1. Short Tandem Repeat Typing Results for the Patient and Her Twin Sister Patient Marker
Allele 1
AMEL
Twin Sister
Allele 2
Allele 1
X
Allele 2
X
D3S1358
15
15
15
15
D1S1656
16
16.3
16
16.3
D6S1043
12
15
12
15
D13S317
10
11
10
11
Penta E D16S539
12 10
20 11
12 10
20 11
D18S51
12
14
12
14
D2S1338
18
24
18
24
CSF1P0
11
12
11
12
Penta D
9
11
9
11
THO1
7
9
7
9
vWA
14
17
14
17
D21S11 D7S820
29 9
32.2 10
29 9
32.2 10
D5S818
11
13
11
13
8
11
8
11
D8S1179
13
15
13
15
D12S391
18
21
18
21
D19S433
13
13
13
13
FGA
23
26
23
26
TPOX
Note: Analysis performed using a human identification kit that amplifies 20 short tandem repeat loci and the amelogenin locus (for sex typing). Values shown are the number of short tandem repeats detected at each allele of each locus. 2
containing putative SNVs indicated a difference between the 2 samples (each locus is sequenced multiple times; only a low proportion of these reads gave a different sequence in one twin vs the other). This low mutation frequency means that there is a high possibility of the putative SNV being a false positive; these putative SNVs were discarded in subsequent data filtering. Thus, the MuTect analysis provided further evidence of a lack of single-nucleotide sequence differences between the 2 samples in this study. Another way that genomes can differ is through CNVs. Using ExomeCNV,17 we detected 3,535 focal CNVs; however, the CNVs did not significantly differ after CONTRA (copy number analysis for targeted resequencing) filtering18 (P 5 0.3 in the minimum target zone; Table S6), suggesting the absence of bona fide CNVs between the twins. In addition, we used GATK to detect somatic indels. For this analysis, we defined the patient’s sample as “normal” and her twin’s as “tumor” (the process is intended for detecting indels that arise in tumor tissue). Sixty indels were detected by GATK, but all were deemed to be false positives after verification (Table S7). To look for epigenomic differences between the patient and her sister, we focused on DMRs. We performed this analysis by reduced-representation bisulfite sequencing19,20 (Fig 1; Tables S8 and S9). We then performed a KEGG (Kyoto Encyclopedia of Genes and Genomes) analysis and found that 10 pathways and 25 genes were involved, as seen in Table S10. Among the 10 pathways, the mitogen-activated protein kinase (MAPK) pathway (including MAP kinase, JNK/p38 MAP kinase pathway, and ERK5 pathway) is well known to take part in the regulation of proliferation, differentiation, and apoptosis of cells. However, an association between the 9 other pathways and kidney development is not supported by existing evidence. Among the 25 genes, 6 (FGF18, FGF12, PDGFRA, MAPK11, AMH, and CTBP1) are involved in organ development, although there is no evidence that they are associated with kidney development.
DISCUSSION This study had 2 innovative aspects. First, cases of monozygotic twins discordant for congenital renal agenesis are very rare. To our knowledge, this study constitutes the first analysis of SNPs, CNVs, indels, and DMRs to directly observe the differences resulting from genetic and environmental factors in monozygotic twins discordant for congenital renal agenesis. Second, this study provides evidence suggesting a lack of differences at the sequence level (SNPs, CNVs, and indels) and the presence of epigenetic differences (at least in terms of differential methylation) in a pair of monozygotic twins who were discordant for congenital renal agenesis. Thus, we suggest that environmental factors, not genetic factors, may lead to the pathogenesis of renal agenesis by epigenetic changes. In week 5 of gestation, the ureteric bud, originating from the Wolffian duct, penetrates the metanephric blastema. After 2 further weeks of gestation, the ureteric bud induces nephrogenesis. By week 20, the ureteric bud has branched and created the collecting duct system in its entirety. However, at that stage, nephrogenesis is just 30% complete, and there is further maturation until week 36.21 Because methylation changes over time, the DMRs detected in this study of 30-year-old twins may not represent the Am J Kidney Dis. 2014;-(-):---
Genetic and Environmental Factors in the Pathogenesis of Renal Agenesis
Figure 1. Landscape of differentially methylated regions (DMRs). We identified 514 DMRs in the studied pair of twins, ranging from 3-52 per chromosome (denoted by bars, which are associated with the left vertical axis). DMR frequency, defined as the length of the DMR divided by the length of the enzyme digestion fragment, ranged from 0.03% to .0.13% in different chromosomes (denoted by the line, which is associated with the right vertical axis).
DMRs present during kidney development. Thus, a larger number of cases would be needed to more conclusively link these DMRs to effects on kidney development. Also, because epigenetic changes accumulate during aging, epigenetic analysis of younger twins, especially newborns, would be preferable to study the pathogenesis of renal agenesis. In addition, although our study revealed genomic similarity and epigenetic differences between the twins, other effects also could contribute to phenotype diversity. To confirm or eliminate these effects, more pairs of samples should be recruited in future work. Previous studies have found that gene mutations could induce CAKUT, including renal agenesis.1,22-24 It has been reported that some candidate genes may associate with renal agenesis.1 Previous studies have suspected that the pathogenesis of renal agenesis is multifactorial (including genetic and environmental factors), but this notion was only a hypothesis. In this study, we failed to identify an obvious mutation on genome level. Although the methylation results would need further verification, our findings suggest that nongenetic factors could cause renal agenesis, which may provide new insight into the pathogenesis of renal agenesis. In this study, blood samples from a pair of monozygotic twins discordant for renal agenesis were analyzed using high-throughput exome-capture sequencing (including analysis of SNPs, CNVs, and indels) and an epigenomic analysis using reducedrepresentation bisulfite sequencing to detect DMRs. We saw no evidence of discordances in SNPs, CNVs, and indels. However, 514 DMRs were detected in reduced-representation bisulfite sequencing. Thus, we suggest that nongenetic (environmental) factors may be important in the pathogenesis of renal agenesis. Am J Kidney Dis. 2014;-(-):---
ACKNOWLEDGEMENTS We thank the volunteers for participating in this study. Support: This work was supported by a grant (2011CB944004) from the National Basic Research Program of China. Financial Disclosure: The authors declare that they have no other relevant financial interests.
SUPPLEMENTARY MATERIAL Table S1: Overview of data production. Table S2: Summary of SNPs in exome-capture sample. Table S3: Comparison of SNP consistency between 6 pairs of twins. Table S4: Differential SNPs detected in primary screening. Table S5: Manual checking of the 30 putatively differential SNPs. Table S6: Processing of CONTRA results. Table S7: Manual checking of the 60 putatively differential indels. Table S8: Summary of DMRs. Table S9: Gene locations of DMRs. Table S10: KEGG analysis of DMRs. Item S1: Detailed methods. Note: The supplementary material accompanying this article (http://dx.doi.org/10.1053/j.ajkd.2014.01.423) is available at www.ajkd.org
REFERENCES 1. Saisawat P, Tasic V, Vega-Warner V, et al. Identification of two novel CAKUT-causing genes by massively parallel exon resequencing of candidate genes in patients with unilateral renal agenesis. Kidney Int. 2012;81(2):196-200. 2. Pohl M, Bhatnagar V, Mendoza SA, Nigam SK. Toward an etiological classification of developmental disorders of the kidney and upper urinary tract. Kidney Int. 2002;61(1):10-19. 3. Woolf AS. A molecular and genetic view of human renal and urinary tract malformations. Kidney Int. 2000;58(2): 500-512. 4. Sanna-Cherchi S, Ravani P, Corbani V, et al. Renal outcome in patients with congenital anomalies of the kidney and urinary tract. Kidney Int. 2009;76(5):528-533. 3
Jin et al 5. Mansoor O, Chandar J, Rodriguez MM, et al. Long-term risk of chronic kidney disease in unilateral multicystic dysplastic kidney. Pediatr Nephrol. 2011;26(4):597-603. 6. Jelen Z. The value of ultrasonography as a screening procedure of the neonatal urinary tract: a survey of 1021 infants. Int Urol Nephrol. 1993;25(1):3-10. 7. Stoll C, Alembik Y, Roth MP, Dott B, Sauvage P. Risk factors in internal urinary system malformations. Pediatr Nephrol. 1990;4(4):319-323. 8. Kerecuk L, Schreuder MF, Woolf AS. Renal tract malformations: perspectives for nephrologists. Nat Clin Pract Nephrol. 2008;4(6):312-325. 9. Parikh CR, McCall D, Engelman C, Schrier RW. Congenital renal agenesis: case-control analysis of birth characteristics. Am J Kidney Dis. 2002;39(4):689-694. 10. Baranzini SE, Mudge J, van Velkinburgh JC, et al. Genome, epigenome and RNA sequences of monozygotic twins discordant for multiple sclerosis. Nature. 2010;464(7293): 1351-1356. 11. Furukawa H, Oka S, Matsui T, et al. Genome, epigenome and transcriptome analyses of a pair of monozygotic twins discordant for systemic lupus erythematosus. Hum Immunol. 2013;74(2): 170-175. 12. Bruder CE, Piotrowski A, Gijsbers AA, et al. Phenotypically concordant and discordant monozygotic twins display different DNA copy-number-variation profiles. Am J Hum Genet. 2008;82(3):763-771. 13. Javierre BM, Fernandez AF, Richter J, et al. Changes in the pattern of DNA methylation associate with twin discordance in systemic lupus erythematosus. Genome Res. 2010;20(2):170-179. 14. Heijmans BT, Tobi EW, Stein AD, et al. Persistent epigenetic differences associated with prenatal exposure to famine in humans. Proc Natl Acad Sci U S A. 2008;105(44):17046-17049.
4
15. McKenna A, Hanna M, Banks E, et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing nextgeneration DNA sequencing data. Genome Res. 2010;20(9): 1297-1303. 16. Cibulskis K, Lawrence MS, Carter SL, et al. Sensitive detection of somatic point mutations in impure and heterogeneous cancer samples. Nat Biotechnol. 2013;31(3):213-219. 17. Sathirapongsasuti JF, Lee H, Horst BA, et al. Exome sequencing-based copy-number variation and loss of heterozygosity detection: ExomeCNV. Bioinformatics. 2011;27(19):2648-2654. 18. Li J, Lupat R, Amarasinghe KC, et al. CONTRA: copy number analysis for targeted resequencing. Bioinformatics. 2012;28(10):1307-1313. 19. Meissner A, Gnirke A, Bell GW, Ramsahoye B, Lander ES, Jaenisch R. Reduced representation bisulfite sequencing for comparative high-resolution DNA methylation analysis. Nucleic Acids Res. 2005;33(18):5868-5877. 20. Gu H, Smith ZD, Bock C, Boyle P, Gnirke A, Meissner A. Preparation of reduced representation bisulfite sequencing libraries for genome-scale DNA methylation profiling. Nat Protoc. 2011;6(4):468-481. 21. Yalavarthy R, Parikh CR. Congenital renal agenesis: a review. Saudi J Kidney Dis Transpl. 2003;14(3):336-341. 22. Abdelhak S, Kalatzis V, Heilig R, et al. Clustering of mutations responsible for branchio-oto-renal (BOR) syndrome in the eyes absent homologous region (eyaHR) of EYA1. Hum Mol Genet. 1997;6(13):2247-2255. 23. Lu W, van Eerde AM, Fan X, et al. Disruption of ROBO2 is associated with urinary tract anomalies and confers risk of vesicoureteral reflux. Am J Hum Genet. 2007;80(4):616-632. 24. Weber S, Taylor JC, Winyard P, et al. SIX2 and BMP4 mutations associate with anomalous kidney development. J Am Soc Nephrol. 2008;19(5):891-903.
Am J Kidney Dis. 2014;-(-):---
