BASIC RESEARCH www.jasn.org

Refined Mapping of the Renal Failure Rf-3 Quantitative Trait Locus

Caitlin C. O’Meara,*† Jozef Lazar,*‡ Matthew Hoffman,*† Carol Moreno,*† and Howard J. Jacob*†§

*Human and Molecular Genetics Center, †Department of Physiology, ‡Department of Dermatology, and §Department of Pediatrics, Medical College of Wisconsin, Milwaukee, Wisconsin

ABSTRACT Rf-3, a quantitative trait locus (QTL) on rat chromosome 3, affects the development of CKD in Fawn- Hooded Hypertensive (FHH) rats. This QTL spans 110 Mb and approximately 1400 genes; therefore, narrowing the position of this locus is necessary to elucidate potential candidate genes. Here, we used congenic models and comparative genomics to refine the Rf-3 candidate region. We generated congenic lines carrying smaller intervals (subcongenics) of the Rf-3 region and used these lines to reduce the Rf-3 candidate region by 94% (to 7.1 Mb). We used comparative genomics to identify QTL for both nephropathy and albuminuria in the syntenic region of this interval for both human and mouse. We also used the overlapping homologous regions to reduce the number of likely positional candidate genes to 13 known or predicted genes. By combining congenic models and cross-species studies, we narrowed the list of candidate genes to a level that we could sequence the whole interval to further identify the causative gene in future studies.

J Am Soc Nephrol 22: 518–525, 2011. doi: 10.1681/ASN.2010060661

Chronic kidney disease (CKD) is a growing health sive (FHH) rat is a well-established model for hy- risk in the United States and worldwide, with the pertension-associated kidney disease.16–20 This incidence continuing to rise at an alarming rate.1 particular strain spontaneously develops systolic Epidemiologic studies have shown that familial and and glomerular hypertension, and consequently, ethnic components contribute to an individual’s renal complications, as indicated by proteinuria, al- risk of developing renal complications as a result of buminuria, and glomerular sclerosis.16,18–23 Be- hypertension and/or diabetes.1–5 Human associa- cause of its robust phenotype, we crossed this strain tion and linkage studies have identified specific re- with the normotensive, renal failure–resistant Au- gions of the genome that significantly contribute to gust Copenhagen Irish (ACI) rat and performed F2 renal disease susceptibility6–12; however, the degree linkage analyses to identify regions of the rat ge- of genetic heritability accounted for by these genes nome that cause kidney disease susceptibility in the is only a small percentage of the total heritability.13 FHH rat.20,21 These linkage analyses showed the Consequently, there is a need to pursue other strat- presence of five renal failure quantitative trait loci egies for identifying genes and their associated (QTLs) called Renal failure 1 through 5 (Rf-1 pathways that are driving CKD. The rat model of- fers numerous advantages, such as an abundance of Received June 23, 2010. Accepted October 16, 2010. physiologic data on many well-characterized dis- Published online ahead of print. Publication date available at ease models and available consomic and congenic www.jasn.org. strains that can be used to study the genetic basis of Correspondence: Dr. Howard J. Jacob, HRC5200, 8701 Water- kidney disease.14,15 town Plank Road, Milwaukee, WI 53226. Phone: 414-456-4887; As in humans, genes play a role in the develop- Fax: 414-456-6516; E-mail: [email protected] ment of CKD in rats. The Fawn-Hooded Hyperten- Copyright © 2011 by the American Society of Nephrology

518 ISSN : 1046-6673/2203-518 J Am Soc Nephrol 22: 518–525, 2011 www.jasn.org BASIC RESEARCH

through Rf-5). Subsequent phenotypic analysis of single and RESULTS double congenic animals has shown a synergistic relationship between the various Rf QTLs. Specifically, an interaction was Assessment of Albumin Excretion and BP in Rf-1 ؉ identified between Rf-1 and Rf-3, and Rf-1 and Rf-4, whereas, 3؉4 Congenic Strains the Rf-3 and Rf-4 loci had little to no apparent effect on renal To physically narrow the candidate region of the Rf-3 QTL, we gen- function alone.24,25 erated and phenotyped a panel of subcongenic lines targeting the Rf-3 To elucidate the specific gene variant(s) causing the ob- QTL for UAV at 9 weeks of age after UNX. We found that congenic served phenotype, it is necessary to narrow the candidate lines containing the FHH genotype between genetic markers region to a manageable number of genes, because the Rf-3 D3Got102 and D3Got121 had higher UAV compared with other sub- region is 110 Mbp in size and contains Ͼ1400 genes. Our congenic lines (data not shown). To further study the contribution group has previously used this strategy to narrow the Rf-2 of this region to renal impairment, we selected two subcongenic QTL region and identify Rab38 as a candidate gene.22 Be- lines—Rf-1 ϩ 3ϩ4_a (ACI.FHH [D1Mit18-D1Rat90]/[D3Rat6- cause gene–gene interactions have been identified between D3Got149]/[D14Mit11-D14Rat33/D14Rat65-D14Rat90]) and the various Rf QTLs and because of the polygenic nature of Rf-1 ϩ 3ϩ4_b (ACI.FHH [D1Mit18-D1Rat90]/[D3Got102- renal disease, we received triple congenic animals from Dr. D3Got149]/[D14Mit11-D14Rat33/D14Rat65-D14Rat90])—for Abraham Provoost (Erasmus MC, Rotterdam, The Nether- phenotypic analysis. These overlapping congenic lines are geneti- lands) that have an ACI disease-resistant background and cally identical to Rf-1 ϩ 4 except for the Rf-3 region, where FHH-sensitive locus introgressed onto Rf-1 (D1Mit18- Rf-1 ϩ 3ϩ4_a is FHH from D3Rat6 to D3Got149 and Rf-1 ϩ D1Rat90), Rf-3 (D3Rat84-D3Rat59), and Rf-4 (D14Mit11- 3ϩ4_b is FHH from D3Got102 to D3Got149 (Figure 1). At 9 D14Rat33/D14Rat65-D14Rat90), called Rf-1 ϩ 3ϩ4, be- weeks of age, Rf-1 ϩ 3ϩ4_b had significantly higher levels of cause the presence of FHH alleles on the Rf-1 locus, and UAV than Rf-1 ϩ 3ϩ4_a (40.06 Ϯ 7.60 versus 15.40 Ϯ 2.40 possibly the Rf-4 locus, is necessary for Rf-3 to have a mea- mg/d, respectively, P ϭ 0.024) and also compared with surable effect on renal disease.24,25 In preliminary studies, Rf-1 ϩ 4 (11.18 Ϯ 3.18 mg/d, P ϭ 0.002; Figure 2), indicat- we observed that, after unilateral nephrectomy (UNX), albumin excretion (UAV) of Rf-1 ϩ 3ϩ4 is almost three times higher than that of the Rf-1 ϩ 4 double congenics (ACI.FHH [D1Mit18-D1Rat90]/[D14Mit11-D14Rat33/ D14Rat65-D14Rat90]), showing the utility of the triple congenic model for mapping disease- causing variant(s) in Rf-3 on a resistant genome background. In this study, we crossed Rf-1 ϩ 3ϩ4 animals to Rf-1 ϩ 4 animals to generate subcongenic lines targeting the Rf-3 QTL. Phenotypic analysis of these sub- congenic lines showed a 7.1-Mb region of rat chromosome 3 that significantly con- tributes to the early development of renal disease. Interestingly, three separate link- age analysis studies in human and mouse have mapped kidney disease loci con- cordant to this 7.1-Mb candidate re- gion,12,26,27 suggesting that the same ge- netic elements may play a role in renal disease susceptibility across species. By comparing the breakpoints of our candi- date region to the boundaries of mouse and human renal function QTLs, we were able to further narrow down the list of candi- Figure 1. Schematic representation of the flanking markers on rat chromosome 3 for Rf-1 ϩ 4, Rf-1 ϩ 3ϩ4_a, and Rf-1 ϩ 3ϩ4_b congenic lines and the homologous regions date genes from 1400 to just 13 known and in human and mouse. The flanking markers for the FHH rat chromosome 3 (RNO3) are predicted genes. In addition to cross-species D3Got102-D3Got149 for Rf-1 ϩ 3ϩ4_b and for Rf-1 ϩ 3ϩ4_a are D3Rat6-D3Got149. analysis, we also sequenced the entire 7.1-Mb The region differentiating lines a and b (Rf-3_b), from D3Got102 to D3Got121,is region to identify variants between ACI and homologous to two regions on human chromosome 20, from the p end of the FHH potentially responsible for the Rf-3 re- chromosome to approximately 1.3 Mb and from 29 to 37 Mb. Rf-3_b is homologous to nal phenotype. mouse chromosome 2 from approximately 151.4 to 158.1 Mb.

J Am Soc Nephrol 22: 518–525, 2011 Fine Mapping the Rf-3 QTL 519 BASIC RESEARCH www.jasn.org

Figure 3. Mean arterial pressure (MAP) is elevated in Rf- 1ϩ3ϩ4_b animals. Number of animals was 6, 7, and 6 for Rf-1 ϩ ϩ Figure 2. Rf-1 3 4_b animals excrete higher levels of albumin ϩ 4, Rf-1 ϩ 3ϩ4_a, and Rf-1 ϩ 3ϩ4_b, respectively. **P Ͻ 0.01. ϩ ϩ ϩ than Rf-1 4 and Rf-1 3 4_a animals at 9 weeks of age. Number Statistical comparison was done using a one-way ANOVA fol- ϩ ϩ ϩ of animals was 9, 13, and 16 for Rf-1 4, Rf-1 3 4_a, and lowed by a Holm-Sidak post hoc test. Rf-1 ϩ 3ϩ4_b, respectively. *P Ͻ 0.05; **P Ͻ 0.01. Datasets were not normally distributed, so each data point was log ϩ Ϯ Ͻ ϩ ϩ transformed before statistical analysis. Statistical comparison with Rf-1 4 (0.34 0.10, P 0.001) and Rf-1 3 4_a Ϯ ϭ was done using a one-way ANOVA followed by a Holm-Sidak (0.52 0.05, P 0.004) glomeruli (Figure 4C). post hoc test. Comparative Genomics ing that gene(s) in the 7.1-Mb region differentiating these The Rf-3_b region is comprised of 7.1 Mb containing 181 known lines (D3Got102-D3Got121 [called Rf-3_b]) are causing in- and predicted genes (Table S1) and is syntenic to two regions on creased UAV at an early age. human chromosome 20 (Figure 1). One megabase (Mb) region To assess whether differences in protein excretion between toward the p terminal of Rf-3_b (141.9 to 142.9 Mb) is homolo- congenic strains is an independent event or whether it is sec- gous to the p end of human chromosome 20 and extends approx- ondary to differences in BP that can accelerate renal damage in imately 1.3 Mb in the reverse orientation (p to q) compared with a susceptible background, we also measured mean arterial the rat. The human homolog to Rf-3_b from 143 to 149 Mb is also pressure (MAP) in the congenic strains. Rf-1 ϩ 3ϩ4_b showed located on human chromosome 20 spanning from approximately a slight but significantly higher MAP (124.71 Ϯ 1.62 mmHg) 29 to 37 Mb in the same orientation (p to q) as the rat. Rf-3_b is compared with Rf-1 ϩ 3ϩ4_a (110.82 Ϯ 3.45 mmHg, P ϭ syntenic to a 6.7-Mb contiguous region on mouse chromosome 2 0.002) and Rf-1 ϩ 4 strains (111.08 Ϯ 2.04 mmHg, P ϭ 0.003; (151.4 to 158.1 Mb) in the same orientation (p to q) as the rat. Figure 3). Two mouse26,27 and one human12 renal function QTLs spanning the Rf-3_b syntenic region have been previously Histologic Analysis mapped. Sheehan et al.26 performed an F2 linkage analysis by Histologic analysis showed significant differences in kidney crossing the C57BL/6J with the DBA/2J mouse strains and morphology between strains. Protein casts observed in stained identified an albuminuria QTL (Albq5) located on mouse kidney sections were larger and more abundant in the outer chromosome 2. The boundaries of Albq5 were compared with medulla and cortex of Rf-1 ϩ 3ϩ4_b compared with Rf-1 ϩ another mouse renal function QTL also located on chromo- 3ϩ4_a and Rf-1 ϩ 4 animals (Figure 4A). Quantification of the some 2.27 The concordant region of a human diabetic ne- observed difference was obtained using a color threshold ana- phropathy QTL identified in the Pima Indian population12 was lytical method. Rf-1 ϩ 3ϩ4_b had higher percent area of pro- compared with the mouse region, further narrowing the can- tein cast (3.29 Ϯ 0.43) in the outer stripe of the medulla and didate genes of the Albq5 region (137 to 152.4 Mb on mouse cortex compared with Rf-1 ϩ 4 (0.78 Ϯ 0.17, P Ͻ 0.001) and chromosome 2) to just 133 genes. A small piece of the Rf-3_b Rf-1 ϩ 3ϩ4_a (1.18 Ϯ 0.26, P ϭ 0.002) kidneys (Figure 4B). region, D3Got102 (141.9 Mb) to approximately 143.2 Mb, Furthermore, Rf-1 ϩ 3ϩ4_b kidney sections showed an in- overlaps the Albq5 candidate region defined by the mouse and creased presence of focal segmental glomerular sclerosis, as human. Forty known and predicted genes are located within this well as focal tubular interstitial fibrosis and tubular atrophy interval, and of these, 25 genes map to the p end of human chro- compared with Rf-1 ϩ 4 and Rf-1 ϩ 3ϩ4_a kidneys (Figure mosome 20 that does not overlap with the human diabetic ne- 4A). Average glomerular sclerosis score was significantly phropathy QTL, and two genes map to other regions of the hu- higher in Rf-1 ϩ 3ϩ4_b (1.10 Ϯ 0.12) glomeruli compared man genome. The remaining 13 genes, shown in Table 1, map to

520 Journal of the American Society of Nephrology J Am Soc Nephrol 22: 518–525, 2011 www.jasn.org BASIC RESEARCH

Figure 4. Rf-1ϩ3ϩ4_b kidneys have increased protein casting and glomerular sclerosis compared with Rf-1ϩ4 and Rf-1ϩ3ϩ4_a kidneys. (A) Histologic sections (10ϫ)of(1)Rf-1 ϩ 4,(2) Rf-1 ϩ 3ϩ4_a, and (3) Rf-1 ϩ 3ϩ4_b medulla and histologic sections (20ϫ) of (4) Rf-1 ϩ 4,(5) Rf-1 ϩ 3ϩ4_a, and (6) Rf-1 ϩ 3ϩ4_b cortex. All kidney sections were stained using Gomori’s one-step trichrome stain. Rf-1 ϩ 3ϩ4_b kidneys showed larger and more abundant protein casts (stained red) compared with Rf-1 ϩ 4 and Rf-1 ϩ 3ϩ4_a kidneys (panels 1 to 3). Panel 6 shows glomerular sclerosis (ϩ) and interstitial fibrosis (3)inRf-1 ϩ 3ϩ4_b cortex. (B) Quantification of percent area of protein casting in the outer stripe of the medulla and cortex for Rf-1 ϩ 4 (n ϭ 4), Rf-1 ϩ 3ϩ4_a (n ϭ 3), and Rf-1 ϩ 3ϩ4_b (n ϭ 4). (C) Average sclerosis score for Rf-1 ϩ 4 (n ϭ 4), Rf-1 ϩ 3ϩ4_a (n ϭ 3), and Rf-1 ϩ 3ϩ4_b (n ϭ 4) glomeruli. **P Ͻ 0.01; ***P Ͻ 0.001. Statistical comparisons were done using a one-way ANOVA followed by a Holm-Sidak post hoc test. chromosome 20 from 29 to 31 Mb, which is concordant with all and 96.87% of the 7.1-Mb target region for FHH and ACI, three QTLs for renal function in rat, mouse, and human. respectively. We identified a total of 9556 sequence variants between the two strains. Thirty-three variants resulted in non- Candidate Region Sequence Analysis synonymous amino acid changes in a total of 22 genes within Using the sequence capture array followed by GS-FLX 454 se- the Rf-3_b region (Tables S1 and S2). Within the candidate quencing, we were able to obtain a sequence covering 96.98 region deduced by comparative mapping, we found exon vari-

J Am Soc Nephrol 22: 518–525, 2011 Fine Mapping the Rf-3 QTL 521 BASIC RESEARCH www.jasn.org

Table 1. List of candidate genes in the Rf-3_b region that overlap the this a useful model for fine mapping of the syntenic mouse and human renal function QTLs Rf-3 gene(s). As the Rf-3 congenic region Symbol Name Start Stop alone does not confer susceptibility to mea- LOC690064 Similar to 40S ribosomal protein S10 142,853,462 142,853,996 sureable renal disease, it was necessary to Defb29 Defensin ␤ 29 142,860,674 142,865,955 incorporate the Rf-1 and/or Rf-4 regions in Defb21 Defensin ␤ 21 142,910,151 142,911,465 our congenic model to obtain a robust phe- Defb24 Defensin ␤ 24 142,913,877 142,920,010 notype for dissecting the Rf-3 locus. This Defb27 Defensin ␤ 27 142,932,437 142,937,373 triple congenic model shows the significant Defb36 Defensin ␤ 36 142,946,483 142,959,997 role of gene–gene interactions in the devel- Defb25 Defensin ␤ 25 142,972,019 142,972,859 opment and severity of complex disease. Rem1 RAS-like GTP binding 1 142,976,928 142,985,368 We found that Rf-1 ϩ 3ϩ4_b exhibited H13 Histocompatability 13 143,020,612 143,056,203 slightly elevated (10 mmHg) MAP com- Id1 Inhibitor of DNA binding 1 143,086,162 143,087,289 pared with Rf-1 ϩ 4 and Rf-1 ϩ 3ϩ4_a LOC499921 Similar to high mobility group protein 1 143,095,429 143,096,094 Cox4i2 Cytochrome c oxidase subunit IV isoform 2 143,103,348 143,114,236 congenic animals. Compared with Rf-1 and Bcl2l1 Bcl2-like1 143,129,087 143,180,199 Rf-3 single congenic animals, Van Dijk et 24 Start and stop are base positions on rat chromosome 3. al. also reported a slight but significant increase in systolic BP in the Rf-1 ϩ 3 ani- ants in 2 of the 13 candidate genes and found 21 intron or mals at 18 weeks after UNX. Because Rf-1 has been shown to

highly conserved intergenic variants potentially affecting tran- affect renal autoregulation resulting in an increased PGC,itis scription factor binding (Table S3). conceivable that a slight increase in BP coupled with impaired autoregulation could cause the observed renal impairment in the Rf-1 ϩ 3ϩ4_b animals, and we cannot formally exclude DISCUSSION this hypothesis. However, Van Dijk et al.24 administered N␻- Nitro-L-arginine methyl ester to raise BP to an average of 180 Van Dijk et al.24 reported that the Rf-3 region does not cause mmHg in combination with UNX on the Rf-1 ϩ 4 genetic severe renal damage in the rat without the presence of Rf-1, yet background, resulting in only a twofold increase in UAV. That the combination of Rf-1 and Rf-3 causes a remarkable augmen- our animals show a threefold increase in UAV at a young age tation in proteinuria. The original interacting Rf-3 congenic re- with slight (approximately 10 mmHg) increase in BP suggests gion spanned about 110 Mb and contained Ͼ1400 genes. In this that the Rf-3 gene(s) is directly causing the observed kidney study, we generated subcongenic lines targeting the Rf-3 95% damage. Also, the original F2 analysis significantly linked Rf-3 confidence interval of the QTL to physically narrow the with UAV and FGS but not with BP,21 and it has been previ- candidate region. By comparing renal damage susceptibility ously reported that, unlike the FHH rat, Rf-1, Rf-3, and Rf-1 ϩ of the Rf-1 ϩ 3ϩ4_b to that of the Rf-1 ϩ 3ϩ4_a, we suc- 3 two kidney animals do not show elevated BP,24 further sup- cessfully reduced the interval by 94% to a 7.1-Mb region of porting the hypothesis that Rf-3 does not directly increase BP. rat chromosome 3 that makes a significant contribution to Further studies are needed to determine whether Rf-3_b di- renal function in the FHH rat. rectly affects kidney function resulting in a secondary increase The Rf-1 region carries genes that cause impaired autoreg- in BP or whether the initial insult of elevated BP causes the ulation in the FHH rat,18,28 likely leading to an increase in observed renal damage.

glomerular capillary pressure (PGC). This insult is necessary to With the region reduced to just 7.1 Mb, we were able to use amplify the affects of Rf-3, which does not influence renal au- pyrosequencing technology to identify potentially damaging toregulation.24 These data suggest that the Rf-3_b gene (or sequence variants within the candidate region. Genetic vari- genes) directly confers susceptibility to dysfunction in the ants underlying QTLs can be caused by alterations in regula- nephron, affecting glomerular permeability or tubular reab- tory elements, such as transcription factor binding sites or sorption of proteins, and in the absence of a stressor such as polymorphisms that result in protein sequence changes. We

increased PGC (Rf-1), the dysfunction does not manifest. The analyzed the protein coding region of all annotated genes in presence of Rf-1 is also required to amplify the effects of Rf-4,25 this 7.1-Mb region and identified 22 genes with nonsynono- whereas the interaction between Rf-3 and Rf-4 has never been mous amino acid changes between ACI and FHH. These genes directly examined. We have combined the Rf-1, Rf-3, and Rf-4 can be considered promising candidates for future studies, and QTLs to generate a triple congenic model to study the effects of these sequence data provide an important resource for identi- genes in the Rf-3 region on renal function. The UAV fold fying genetic variants responsible for the Rf-1 ϩ 3ϩ4_b phe- change of Rf-1 ϩ 3ϩ4 versus Rf-1 ϩ 4 is similar to the observed notype. change in Rf-1 ϩ 3 animals compared with Rf-1 animals,24 Genetic studies of complex disease in rodent models have suggesting there is no profound epistasis between the Rf-3 and resulted in identification of causative gene variants that have Rf-4 regions. The triple congenic model exhibited significantly been shown to affect the corresponding disease in humans, higher levels of UAV than that of the Rf-1 ϩ 4 animals, making suggesting that genetic elements share function across spe-

522 Journal of the American Society of Nephrology J Am Soc Nephrol 22: 518–525, 2011 www.jasn.org BASIC RESEARCH

cies.29–31 We used comparative genomics to both validate the CONCISE METHODS utility of cloning this gene by position and further reduce the interval. This strategy has been used by several groups to re- Generation of Triple Congenic Overlapping duce complex disease QTL intervals in the mouse and human Subcongenic Strains for atherosclerosis,32–35 hypertension,32,36 and renal disease.26 Rats were housed in the Biomedical Resource Center of the Medical Sheehan et al.26 compared mouse and human homologous re- College of Wisconsin, an American Association for the Accreditation nal function QTL to narrow the candidate region, Albq5. The of Laboratory Animal Care–approved facility. All protocols used in original Rf-3 boundaries could not be used to further narrow these studies were approved by the local Animal Care and Use Com- their Albq5 candidate region because the Rf-3 breakpoints were mittee. We performed an initial cross between congenic strains ACI.FHH outside of the Albq5 interval. In this study, we used congenic (D1Mit18-D1Rat90)/(D14Mit11-D14Rat33/D14Rat65-D14Rat90), referred approaches to narrow the Rf-3 region and further minimize the to as Rf-1 ϩ 4,25 and ACI.FHH (D1Mit18-D1Rat90)/(D3Rat84-D3Rat59)/ candidate region that Sheehan et al. had identified. By compar- (D14Mit11-D14Rat33/D14Rat65-D14Rat90), referred to as Rf-1 ϩ 3ϩ4, ing the endpoint of Albq5 with our current Rf-3_b interval to generate an F1 generation. We intercrossed F1 littermates, and the breakpoints, we were able to preliminarily narrow the list of F2 generation was genotyped with 19 microsatellite markers across candidate genes to 13 known and predicted genes. We cannot 110 Mbp of the Rf-3 region by fluorescence genotyping, as described formally exclude the possibility that any of the 181 genes in the previously.43 Recombinant animals were selected for breeding to gen- Rf-3_b region may contribute to renal insufficiency in the FHH erate a panel of subcongenic lines targeting the Rf-3 95% confidence rat, but at this point, we can use this synteny to reduce the interval of the QTL. predicted interval further, without the need to generate addi- tional congenics. In Vivo Renal Damage Assessment One of these 13 candidate genes, Bcl2l1, belongs to the Bcl-2 Male rats between 5 and 6 weeks of age were anesthetized with keta- family, which is known to consist of anti-apoptotic genes ex- mine (30 mg/kg) plus xylazine (2.5 mg/kg) and acepromazine (0.6mg/ pressed in renal tubular cells.37–40 Transgenic mice deficient in kg), the right kidney was exposed by a flank incision, the renal artery Bcl-2 have renal hypoplasia and distal tubular damage, whereas and vein were ligated, and the kidney was removed. After surgery, the other regions of the nephron such as glomeruli and proximal animal was placed on a purified AIN-76A rodent diet containing 0.4% tubules seem unaffected in these mice.41 Bcl2l1 is an interesting NaCl (Dyets, Bethlehem, PA) and allowed to recover for 4 weeks. candidate gene in our model because the Rf-1 ϩ 3ϩ4_b line After recovery, the animals were placed in metabolic cages (Nalgene, shows severe tubular dilation and protein casting at an early Rochester, NY) and allowed to adapt to the cages for 2 days, followed age. Another gene in this region, Rem1, is functionally related by urine collection for two consecutive 24-hour periods. Albumin to the previously identified Rf-2 gene, Rab38, because both are concentration was determined using Albumin Blue 580 assay (Molec- GTPase members of the Ras superfamily,42 making this an in- ular Probes, Eugene, OR). teresting candidate as well. However, we did not find any po- tentially damaging sequence variants in these two genes, re- BP Measurement ducing the likelihood that they are functionally responsible for After the metabolic cage experiment, MAP was measured by radiote- renal insufficiency in the FHH rat. In the 13 gene interval, we lemetry (Data Sciences, St. Paul, MN) in 10- to 11-week-old rats. did identify highly conserved intergenic variants potentially Telemetry transmitters (TA11PA-C40) were implanted subcutane- affecting transcription factor binding, and we also found ously (under isoflurane anesthesia), and the catheter was inserted into highly conserved exon variants in the predicted genes the abdominal aorta via the femoral artery. Animals were allowed 4 LOC690064 and LOC100363271. The latter does not map to days for recovery after surgery, and BP was recorded in conscious, human or mouse homologous renal function QTL, but it is freely moving animals for 3 consecutive days at 500 Hz. Ten-second located in the intron of Bcl2l1. In future studies, we plan to intervals were continuously recorded every 2 minutes, and these data investigate whether these, or any of the 246 variants within this were averaged over a 3-hour period each day to estimate MAP. 13 gene interval, are causal. If not, we will return to the larger gene set. Histologic Examination In summary, we physically reduced the Rf-3 QTL to a Three-month-old uniphrectomized males were killed by isoflurane 7.1-Mb region of rat chromosome 3 that significantly contrib- overdose, and the left kidney was removed and immediately placed in utes to renal impairment in the FHH rat. We compared our 10% buffered formalin (Sigma-Aldrich) for fixation. Fixed kidneys candidate region to the concordant mouse and human renal were sectioned and stained using Gomori’s one-step trichrome stain function QTLs to narrow the list of 181 candidate genes in this for histologic analysis. Percent area of protein casting was quantified region down to 13 known and predicted genes. Additional in the outer stripe of outer medulla and cortex region. To quantify studies are needed to elucidate the specific variant(s) causing percent area of protein casting, stained kidney sections were scanned the observed phenotype and to determine how these candidate at 4000 dpi using a Nikon Coolscan V ED (Nikon Instruments) and genes affect kidney function in the FHH rat. In the interim, analyzed using MetaMorph version 7.1.3 (Molecular Devices). The these genes can be tested in subsets of genome wide association outer stripe of the outer medulla and cortex region was encircled on studies and other species or strains. the scanned images, and a color threshold was set to selectively detect

J Am Soc Nephrol 22: 518–525, 2011 Fine Mapping the Rf-3 QTL 523 BASIC RESEARCH www.jasn.org the deep red color that matched the hue of protein casts. The percent 4. Marin R, Gorostidi M, Fernandez-Vega F, Alvarez-Navascues R: Sys- protein casting was calculated by (Protein Cast Area/Total Encircled temic and glomerular hypertension and progression of chronic renal Area) ϫ 100. Glomeruli were scored based on percent sclerosis on a disease: The dilemma of nephrosclerosis. Kidney Int 68: S52–S56, 2005 scale from 0 (approximately 0%) to 4 (approximately 100%), and 5. Freedman BI, Iskandar SS, Appel RG: The link between hypertension scores for 30 glomeruli were averaged for each animal. and nephrosclerosis. Am J Kidney Dis 25: 207–221, 1995 6. Pattaro C, Aulchenko YS, Isaacs A, Vitart V, Hayward C, Franklin CS, Capture, Sequencing, and Analysis Polasek O, Kolcic I, Biloglav Z, Campbell S, Hastie N, Lauc G, Meit- ACI and FHH genomic DNA for the 7.1-Mb candidate region was inger T, Oostra BA, Gyllensten U, Wilson JF, Pichler I, Hicks AA, Campbell H, Wright AF, Rudan I, van Duijn CM, Riegler P, Marroni F, captured using a custom tiling 385K array (coordinates: rat chr3: Pramstaller PP; EUROSPAN Consortium: Genome-wide linkage anal- 141,890,119 to 149,038,368) designed and manufactured by Roche- ysis of serum creatinine in three isolated European populations. Kid- Nimblegen. Genomic DNA was sheared using nebulization, and ney Int 76: 297–306, 2009 adaptors were ligated to the resulting fragments. Linker mediated 7. Fox CS, Yang Q, Cupples LA, Guo C-Y, Larson MG, Leip EP, Wilson PCR was performed to amplify the library, and the amplified library PWF, Levy D: Genomewide linkage analysis to serum creatinine GFR, and creatinine clearance in a community-based population: The Fra- was hybridized to the custom array according to the manufacture mingham Heart Study. J Am Soc Nephrol 15: 2457–2461, 2004 protocol (Roche Applied Science/Nimblegen). Unbound DNA was 8. Kottgen A, Glazer NL, Dehghan A, Hwang SJ, Katz R, Li M, Yang Q, washed away, and hybridized DNA was eluded off of the chip. One Gudnason V, Launer LJ, Harris TB, Smith AV, Arking DE, Astor microgram of target-enriched DNA was sequenced on the Roche GS- BC, Boerwinkle E, Ehret GB, Ruczinski I, Scharpf RB, Chen YD, de Boer FLX sequencer running 454 sequencing technology (Roche Applied IH, Haritunians T, Lumley T, Sarnak M, Siscovick D, Benjamin EJ, Levy D, Upadhyay A, Aulchenko YS, Hofman A, Rivadeneira F, Uitterlinden Science). Sequence reads were assembled, and variants were detected AG, van Duijn CM, Chasman DI, Pare´ G, Ridker PM, Kao WH, Witte- using the gsMapper software (Roche Applied Science). Highly con- man JC, Coresh J, Shlipak MG, Fox CS: Multiple loci associated with served variants were identified using the VISTA genome browser indices of renal function and chronic kidney disease. Nat Genet 41: (http://pipeline.lbl.gov/cgi-bin/gateway2), and transcription factor 712–717, 2009 binding sites were found using the TFSearch website algorithm (http:// 9. Schelling JR, Abboud HE, Nicholas SB, Pahl MV, Sedor JR, Adler SG, Arar NH, Bowden DW, Elston RC, Freedman BI, Goddard KA, Guo X, molsun1.cbrc.aist.go.jp/research/db/TFSEARCH.html). Hanson RL, Ipp E, Iyengar SK, Jun G, Kao WH, Kasinath BS, Kimmel PL, Klag MJ, Knowler WC, Nelson RG, Parekh RS, Quade SR, Rich SS, Statistical Analysis Saad MF, Scavini M, Smith MW, Taylor K, Winkler CA, Zager PG, Shah Data are presented as mean Ϯ SEM. We analyzed data by t test or a VO; Family Investigation of Nephropathy and Diabetes Research one-way ANOVA followed by the Holm-Sidak multiple comparison Group: Genome-wide scan for estimated glomerular filtration rate in test using Sigma Plot 11.0 software. Because albumin data failed the multi-ethnic diabetic populations. Diabetes 57: 235–243, 2008 10. Mottl AK, Vupputuri S, Cole SA, Almasy L, Go¨ring HH, Diego VP, equal variance test, we transformed the data by taking the logarithm of Laston S, Franceschini N, Shara NM, Lee ET, Best LG, Fabsitz RR, each value and performed a one-way ANOVA followed by the Holm- MacCluer JW, Umans JG, North KE: Linkage analysis of glomerular Sidak multiple comparison test. filtration rate in American Indians. Kidney Int 74: 1185–1191, 2008 11. Rogus JJ, Poznik GD, Pezzolesi MG, Smiles AM, Dunn J, Walker W, Wanic K, Moczulski D, Canani L, Araki S, Makita Y, Warram JH, Krolewski AS: High-density single nucleotide polymorphism genome- ACKNOWLEDGMENTS wide linkage scan for susceptibility genes for diabetic nephropathy in type 1 diabetes. Diabetes 57: 2519–2526, 2008 This study was performed with financial support from the National 12. Imperatore G, Hanson RL, Pettitt DJ, Kobes S, Bennett PH, Knowler Heart, Lung, and Blood Institute (NHLBI-5R01HL069321) to H.J.J. WC: Sib-pair linkage analysis for susceptibility genes for microvascular complications among Pima Indians with type 2 diabetes. Pima Diabe- The authors thank Mike Tschannen and Jaime Wendt Andrae for tes Genes Group. Diabetes 47: 821–830, 1998 excellent technical assistance. 13. Altshuler D, Daly MJ, Lander ES: Genetic mapping in human disease. Science 322: 881–888, 2008 14. Mullins L, Mullins J: Insights from the rat genome sequence. Genome Biol 5: 221, 2004 DISCLOSURES 15. Mattson DL, Dwinell MR, Greene AS, Kwitek AE, Roman RJ, Cowley None. AW, Jacob HJ: Chromosomal mapping of the genetic basis of hyper- tension and renal disease in FHH rats. Am J Physiol Renal Physiol 293: F1905–F1914, 2007 REFERENCES 16. Kriz W, Hosser H, Hahnel B, Simons JL, Provoost AP: Development of vascular pole-associated glomerulosclerosis in the Fawn-hooded rat. J Am Soc Nephrol 9: 381–396, 1998 1. Coresh J, Selvin E, Stevens LA, Manzi J, Kusek JW, Eggers P, Van 17. Simons JL, Provoost AP, De Keijzer MH, Anderson S, Rennke HG, Lente F, Levey AS: Prevalence of chronic kidney disease in the United Brenner BM: Pathogenesis of glomerular injury in the fawn-hooded States. JAMA 298: 2038–2047, 2007 rat: Effect of unilateral nephrectomy. J Am Soc Nephrol 4: 1362–1370, 2. Lindner TH, Monks D, Wanner C, Berger M: Genetic aspects of dia- 1993 betic nephropathy. Kidney Int 63: S186–S191, 2003 18. Van Dokkum RPE, Alonso-Galicia M, Provoost AP, Jacob HJ, Roman 3. Freedman BI, Spray BJ, Tuttle AB, Buckalew VM: The familial risk of RJ: Impaired autoregulation of renal blood flow in the fawn-hooded end-stage renal disease in African Americans. Am J Kidney Dis 21: rat. Am J Physiol Regul Integr Comp Physiol 276: R189–R196, 1999 387–393, 1993 19. Van Dokkum RPE, Sun C-W, Provoost AP, Jacob HJ, Roman RJ:

524 Journal of the American Society of Nephrology J Am Soc Nephrol 22: 518–525, 2011 www.jasn.org BASIC RESEARCH

Altered renal hemodynamics and impaired myogenic responses in the 31. Rapp JP, Wang SM, Dene H: A genetic polymorphism in the renin fawn-hooded rat. Am J Physiol Regul Integr Comp Physiol 276: R855– gene of Dahl rats cosegregates with blood pressure. Science 243: R863, 1999 542–544, 1989 20. Brown DM, Provoost AP, Daly MJ, Lander ES, Jacob HJ: Renal disease 32. Chang Y-PC, Liu X, Kim JD, Ikeda MA, Layton MR, Weder AB, Cooper susceptibility and hypertension are under independent genetic con- RS, Kardia SL, Rao DC, Hunt SC, Luke A, Boerwinkle E, Chakravarti A: trol in the fawn-hooded rat. Nat Genet 12: 44–51, 1996 Multiple genes for essential-hypertension susceptibility on chromo- 21. Shiozawa M, Provoost AP, Van Dokkum RPE, Majewski RR, Jacob HJ: some 1q. Am J Hum Genet 80: 253–264, 2007 Evidence of gene–gene interactions in the genetic susceptibility to 33. Wang X, Ishimori N, Korstanje R, Rollins J, Paigen B: Identifying novel renal impairment after unilateral nephrectomy. J Am Soc Nephrol 11: genes for atherosclerosis through mouse-human comparative genet- 2068–2078, 2000 ics. Am J Hum Genet 77: 1–15, 2005 22. Rangel-Filho A, Sharma M, Datta YH, Moreno C, Roman RJ, Iwamoto 34. Burgess-Herbert SL, Cox A, Tsaih S-W, Paigen B: Practical applications Y, Provoost AP, Lazar J, Jacob HJ: Rf-2 gene modulates proteinuria of the bioinformatics toolbox for narrowing quantitative trait loci. and albuminuria independently of changes in glomerular permeability Genetics 180: 2227–2235, 2008 in the Fawn-Hooded Hypertensive rat. J Am Soc Nephrol 16: 852– 35. Yu X, Teng H, Marques A, Ashgari F, Ibrahim SM: High resolution 856, 2005 mapping of Cia3: A common arthritis quantitative trait loci in different 23. Brown DM, Van Dokkum RPE, Korte MR, McLauglin MG, Shiozawa M, species. J Immunol 182: 3016–3023, 2009 Jacob HJ, Provoost AP: Genetic control of susceptiblity for renal damage 36. DiPetrillo K, Tsaih S-W, Sheehan S, Johns C, Kelmenson P, Gavras H, in Hypertensive Fawn-Hooded rats. Renal Failure 20: 407–411, 1998 Churchill GA, Paigen B: Genetic analysis of blood pressure in C3H/ 24. Van Dijk SJ, Specht PAC, Lazar J, Jacob HJ, Provoost AP: Synergistic HeJ and SWR/J mice. Physiol Genomics 17: 215–220, 2004 QTL interactions between Rf-1 and Rf-3 increase renal damage sus- 37. Gobe G, Zhang XJ, Willgoss DA, Schoch E, Hogg NA, Endre ZH: ceptibility in double congenic rats. Kidney Int 69: 1369–1376, 2006 Relationship between expression of Bcl-2 genes and growth factors in 25. Van Dijk SJ, Specht PAC, Lutz MM, Lazar J, Jacob HJ, Provoost AP: ischemic acute renal failure in the rat. J Am Soc Nephrol 11: 454–467, Interaction between Rf-1 and Rf-4 quantitative trait loci increases 2000 susceptibility to renal damage in double congenic rats. Kidney Int 68: 38. Basile DP, Liapis H, Hammerman MR: Expression of bcl-2 and bax in 2462–2472, 2005 regenerating rat renal tubules following ischemic injury. Am J Physiol 26. Sheehan S, Tsaih S-W, King BL, Stanton C, Churchill GA, Paigen B, Renal Physiol 272: F640–F647, 1997 DiPetrillo K: Genetic analysis of albuminuria in a cross between 39. Valde´s F, Pa´saro E, Díaz I, Centeno A, Lo´pez E, García-Doval S, C57BL/6J and DBA/2J mice. Am J Physiol Renal Physiol 293: F1649– Gonza´lez-Roces S, Alba A, Laffon B: Segmental heterogeneity in Bcl-2, F1656, 2007 Bcl-xL and Bax expression in rat tubular epithelium after ischemia- 27. Shike T, Gohda T, Tanimoto M, Kobayashi M, Makita Y, Funabiki K, reperfusion. Nephrology 13: 294–301, 2008 Horikoshi S, Hirose S, Shirai T, Tomino Y: Chromosomal mapping of a 40. Takase O, Minto AW, Puri TS, Cunningham PN, Jacob A, Hayashi M, quantitative trait locus for the development of albuminuria in diabetic Quigg RJ: Inhibition of NF-[kappa]B-dependent Bcl-xL expression by KK/Ta mice. Nephrol Dial Transplant 20: 879–885, 2005 clusterin promotes albumin-induced tubular cell apoptosis. Kidney Int 28. Lopez B, Ryan RP, Moreno C, Sarkis A, Lazar J, Provoost AP, Jacob HJ, 73: 567–577, 2007 Roman RJ: Identification of a QTL on chromosome 1 for impaired 41. Sorenson CM, Padanilam BJ, Hammerman MR: Abnormal postpartum autoregulation of RBF in fawn-hooded hypertensive rats. Am J Physiol renal development and cystogenesis in the bcl-2(-/-) mouse. Am J Renal Physiol 290: F1213–F1221, 2006 Physiol Renal Physiol 271: F184–F193, 1996 29. O’Brien SJ, Menotti-Raymond M, Murphy WJ, Nash WG, Wienberg J, 42. Pochynyuk O, Stockand JD, Staruschenko A: Ion channel regulation by Stanyon R, Copeland NG, Jenkins NA, Womack JE, Marshall Graves Ras, Rho, and Rab small GTPases. Exp Biol Med 232: 1258–1265, JA: The promise of comparative genomics in mammals. Science 286: 2007 458–481, 1999 43. Moreno C, Kennedy K, Andrae JW, Jacob HJ: Genome-wide scanning 30. Bottger A, van Lith HA, Kren V, Krenova´D, Bíla´V, Vorlícek J, Zídek V, with SSLPs in the rat. Methods Mol Med 108: 131–138, 2005 Musilova´ A, Zdobinska´ M, Wang JM, van Zutphen BF, Kurtz TW, Pravenec M: Quantitative trait loci influencing cholesterol and phos- pholipid phenotypes map to chromosomes that contain genes regu- lating blood pressure in the spontaneously hypertensive rat. J Clin Supplemental information for this article is available online at http://www. Invest 98: 856–862, 1996 jasn.org/.

J Am Soc Nephrol 22: 518–525, 2011 Fine Mapping the Rf-3 QTL 525

Non-Synonymous Gene Symbol Gene Name Start Stop Polymorphism Sdcbp2 Syndecan binding protein (syntenin) 2 141,891,232 141,919,905 I82V Snph Syntaphilin 141,921,547 141,961,697 A411T, 415_420* Rad21l1 RAD21-like 1 (S. pombe) 141,984,729 142,009,665 LOC100363280 mCG140681-like 142,022,959 142,030,458 RGD1564048 Similar to Protein C20orf46 142,043961 142,047,717 Similar to Proteasome inhibitor PI31 LOC689852 142,058,695 142,083,908 subunit LOC100363380 Hypothetical LOC100363380 142,139,020 142,143,826 Rspo4 R-spondin family, member 4 142,185,028 142,216,015 Angpt4 Angiopoietin 4 142,249,115 142,282,307 Family with sequence similarity 110, Fam110a 142,326,013 142,328,572 member A LOC689884 Hypothetical protein LOC689884 142,335,923 142,336,396 LOC100363434 Ribosomal protein S29-like 142,339,596 142,341,169 RGD1304644 Similar to RIKEN cDNA 2310046K01 142,360,661 142,365,285 Scratch homolog 2, zinc finger protein Scrt2 142,434,272 142,445,984 (Drosophila) Srxn1 Sulfiredoxin 1 homolog (S. cerevisiae) 142,457,352 142,462,725 Tcf15 Transcription factor 15 142,491,664 142,497,446 Csnk2a1 Casein kinase 2, alpha 1 polypeptide 142,564,754 142,609,311 Tbc1d20 TBC1 domain family, member 20 142,623,713 142,640,197 RanBP-type and C3HC4-type zinc finger Rbck1 142,644,156 142,660,799 containing 1 Trib3 Tribbles homolog 3 (Drosophila) 142,664,641 142,670,135 Nrsn2 Neurensin 2 142,692,988 142,702,131 Sox12 SRY (sex determining region Y)-box 12 142,714,836 142,715,855 Zcchc3 Zinc finger, CCHC domain containing 3 142,727,648 142,730,465 LOC502684 Hypothetical protein LOC502684 142,745,094 142,764,314 Defb23 Defensin beta 23 142,780,126 142,784,338 Defb20 Defensin beta 20 142,804,804 142,807,427 Defb22 Defensin beta 22 142,813,671 142,817,950 Defb26 Defensin beta 26 142,834,066 142,837,469 Defb28 Defensin beta 28 142,838,348 142,853,944 LOC690064 Similar to 40S ribosomal protein S10 142,853,462 142,853,996 Defb29 Defensin beta 29 142,860,674 142,865,955 Defb19 Defensin beta 19 142,903,282 142,905,194 Defb21 Defensin beta 21 142,910,151 142,911,465 Defb24 Defensin beta 24 142,913,877 142,920,010 Defb27 Defensin beta 27 142,932,437 142,937,373 Defb36 Defensin beta 36 142,964,483 142,959,997 Defb25 Defensin beta 25 142,972,019 142,972,859 Rem1 RAS-like GTP binding 1 142,976,928 142,985,368 H13 Histocompatability 13 143,020,612 143,056,203 Mcts2 Malignant T cell amplified sequence 2 143,038,079 143,039,285 Id1 Inhibitor of DNA binding 1 143,086,162 143,087,289 LOC499921 Similar to high mobility group protein 1 143,095,429 143,096,094 Cox4i2 Cytochrome c oxidase subunit IV isoform 2 143,103,348 143,114,236 Bcl2l1 Bcl2-like1 143,129,087 143,180,199 LOC100363271 Ribosomal protein S2-like 143,149,520 143,150,401 Tpx2 TPX2, microtubule-associated, homolog 143,194,067 143,236,185 (Xenopus laevis) Mylk2 Myosin light chain kinase 2 143,252,183 143,263,849 Foxs1 Forkhead box S1 143,272,507 143,273,776 Dusp15 Dual specificity phosphatase 15 143,285,894 143,294,571 Tubulin tyrosine ligase-like family, Ttll9 143,304,937 143,351,359 member 9 Pdrg1 p53 and DNA damage regulated 1 143,353,054 143,359,223 XK, Kell blood group complex subunit- Xkr7 143,376,730 143,400,522 related family, member 7 RGD1305202 Similar to Protein C20orf160 143,409,027 143,423,744 Hck Hemopoietic cell kinase 143,447,697 143,490,281 Transmembrane 9 superfamily protein Tm9sf4 143,502,896 143,557,285 member 4 Tspyl3 TSPY-like 3 143,574,208 143,576,284 Plagl2 Pleiomorphic adenoma gene-like 2 143,578,877 143,592,055 Pofut1 Protein O-fucosyltransferase 1 143,592,269 143,614,466 LOC690246 Similar to 60S ribosomal protein L27a 143,631,638 143,632,082 LOC366233 Similar to 60S ribosomal protein L21 143,636,966 143,637,446 Kif3b Kinesin family member 3B 143,642,175 143,681,902 LOC690274 Hypothetical protein LOC690274 143,695,573 143,700,082 RGD1561878 Similar to mKIAA0978 protein 143,703,295 143,767,523 Similar to RIKEN cDNA 8430427H17 RGD1563510 143,774,195 143,904,844 gene Commd7 COMM domain containing 7 143,997,788 144,012,407 DNA (cytosine-5-)-methyltransferase 3 Dnmt3b 144,030,737 144,069,265 beta LOC690309 Similar to DNA methyltransferase 3B 144,080,177 144,112,035 Similar to Alpha-enolase (2-phospho-D- LOC689804 glycerate hydro-lyase) (Non-neural 144,109,924 144,111,554 enolase) (Enolase 1) Microtubule-associated protein, RP/EB Mapre1 144,120,532 144,148,856 family member 1 RGD1560257 Similar to hypothetical protein A630008I04 144,161,795 144,225,579 Efcab8 EF-hand calcium binding domain 8 144,163,986 144,169,818 Spag4l Sperm associated antigen 4-like 144,236,963 144,257,526 S10L Bactericidal/permeability-increasing Bpil1 144,260,247 144,279,234 protein-like 1 Bactericidal/permeability-increasing Bpil3 144,293,297 144,297,498 protein-like 3 Rya3 Antimicrobial peptide RYA3 144,303,258 144,317,875 RY2G5 Potential ligand-binding protein 144,325,761 144,350,230 LOC100359775 Splunc6-like 144,367,912 144,380,189 Psp Parotid secretory protein 144,395,541 144,403,499 Smgb Neonatal submandibular gland protein B 144,423,417 144,435,625 Similar to Short palate, lung and nasal LOC690402 epithelium carcinoma-associated protein 3 144,511,823 144,521,745 homolog precursor Palate, lung, and nasal epithelium Plunc 144,529,855 144,535,598 associated Similar to NIMA (never in mitosis gene a)- LOC690426 144,536,781 144,546,511 related kinase 2 Similar to Palate lung and nasal carcinoma- RGD1559748 144,548,148 144,554,452 like protein precursor Similar to von Ebner minor salivary gland RGD1563047 144,601,201 144,635,308 M352T protein LOC690479 Hypothetical protein LOC690479 144,638,920 144,652,523 E196D, N205S, P308L, I319M LOC690486 Similar to 60S ribosomal protein L7 144,659,980 144,660,754 D196E, S205N, LOC690497 Similar to MORF-related gene X 144,712,002 144,712,695 L308P, M319I LOC690507 Similar to Vomeromodulin 144,723,203 144,733,928 Similar to HMG-1 (High mobility group LOC690521 144,762,561 144,763,202 protein B1) CDK5 regulatory subunit associated Cdk5rap1 144,803,213 144,840,056 protein 1 Snta1 Syntrophin, acidic 1 144,843,678 144,874,238 Core-binding factor, runt domain, alpha Cbfa2t2 144,904,533 145,007,851 subunit 2; translocated to, 2 LOC690027 Similar to spermine synthase 144,911,210 144,912,792 LOC683008 Similar to spermine synthase 144,911,661 144,912,847 LOC100360105 Zinc finger, FYVE domain containing 21 144,981,179 144,999,578 N-terminal EF-hand calcium binding Necab3 145,017,420 145,031,905 162_163** protein 3 Similar to chromosome 20 open reading RGD1561517 145,022,077 145,023,268 frame 144 E2f1 E2F transcription factor 1 145,032,489 145,043,729 Pxmp4 Peroxisomal membrane protein 4 145,061,123 145,078,370 LOC296300 Similar to zinc finger protein 341 145,086,839 145,105,846 Eukaryotic translation elongation factor 1 LOC100360150 145,106,442 145,114,354 alpha 2 Zfp341 Zinc finger protein 341 145,108,413 145,124,023 LOC100359642 rCG37275-like 145,138,905 145,176,471 Chmp4b Chromatin modifying protein 4B 145,178,263 145,179,851 LOC690589 Similar to 60S ribosomal protein L7a 145,187,694 145,204,782 LOC690597 Similar to 60S ribosomal protein L38 145,254,907 145,255,116 Raly RNA binding protein, autoantigenic 145,275,832 145,337,808 LOC100360250 Hypothetical LOC100360250 145,327,005 145,327,215 Eukaryotic translation initiation factor 2, Eif2s2 145,346,392 145,367,117 subunit 2 beta Asip Agouti signaling protein 145,445,175 145,536,831 36_44* LOC100360304 Hypothetical LOC100360304 145,467,378 145,467,743 Ahcy Adenosylhomocysteinase 145,544,834 145,560,058 LOC366235 Similar to 60S ribosomal protein L12 145,587,354 145,587,927 LOC690622 Hypothetical protein LOC690622 145,621,455 145,622,913 Itchy E3 ubiquitin protein ligase homolog Itch 145,644,315 145,707,190 (mouse) Dynlrb1 Dynein light chain roadblock-type 1 145,718,890 145,740,187 Microtubule-associated protein 1 light Map1lc3a 145,758,939 145,760,532 chain 3 alpha Phosphatidylinositol glycan anchor Pigu 145,760,693 145,841,916 biosynthesis, class U Tumor protein p53 inducible nuclear Tp53inp2 145,878,005 145,886,088 protein 2 Ncoa6 Nuclear receptor coactivator 6 145,886,874 145,957,928 258_263* Ggt7 Gamma-glutamyltransferase 7 145,987,531 146,010,922 Acyl-CoA synthetase short-chain family Acss2 146,013,995 146,057,119 member 2 Gss Glutathione synthetase 146,057,517 146,087,820 K16E Myosin, heavy chain 7B, cardiac muscle, Myh7b 146,115,687 146,139,441 beta Transient receptor potential cation channel, Trpc4ap 146,139,493 146,206,625 Y166C subfamily C, member 4 associated protein Similar to Glyceraldehyde-3-phosphate LOC690264 146,189,074 146,190,106 dehydrogenase ER degradation enhancer, mannosidase Edem2 146,215,243 146,241,062 alpha-like 2 Procr Protein C receptor, endothelial 146,268,567 146,273,718 LOC100360401 Hypothetical LOC100360401 146,284,682 146,285,087 Cep250 Centrosomal protein 250kDa 146,378,670 146,407,642 M387V LOC100359643 mCG58586-like 146,411,621 146,412,464 Ergic3 ERGIC and golgi 3 146,416,599 146,426,325 Fer1l4 Fer-1-like 4 (C. elegans) 146,427,611 146,463,147 H467R, A623S Spag4 Sperm associated antigen 4 146,472,609 146,476,953 Rbm12 RNA binding motif protein 12 146,504,101 146,521,424 771_777** Similar to Glyceraldehyde-3-phosphate LOC362250 146,527,591 146,528,570 dehydrogenase NFS1 nitrogen fixation 1 homolog (S. Nfs1 146,528,820 146,551,156 cerevisiae) Romo1 Reactive oxygen species modulator 1 146,551,220 146,552,820 RNA-binding region containing protein 2- LOC100360432 146,554,528 146,554,855 like Rbm39 RNA binding motif protein 39 146,555,202 146,587,650 Phf20 PHD finger protein 20 146,601,763 146,707,185 Scand1 SCAN domain-containing 1 146,710,153 146,711,017 LOC100360625 rCG37452-like 146,710,451 146,711,796 RGD1311678 Similar to 4921517L17Rik protein 146,726,507 146,793,519 Similar to Ferritin light chain 1 (Ferritin L LOC679618 146,806,878 146,821,365 subunit 1) LOC679637 Hypothetical protein LOC679637 146,851,544 146,854,442 Epb4.1l1 Erythrocyte protein band 4.1-like 1 146,876,451 146,944,603 D1091G, 852_854* Similar to RIKEN cDNA 0610011L14 RGD1311066 146,962,602 146,983,950 K184E gene Discs, large homolog-associated protein 4 Dlgap4 147,081,222 147,176,024 (Drosophila) Myl9 Myosin, light chain 9, regulatory 147,190,252 147,193,865 RGD1564927 Similar to TGFB-induced factor 2 147,239,123 147,254,254 RGD1560115 Similar to TGFB-induced factor 2 147,254,612 147,257,397 RGD1307752 Similar to RIKEN cDNA 1110008F13 147,266,788 147,276,798 Sla2 Src-like-adaptor 2 147,277,668 147,290,729 Similar to N-myc downstream regulated LOC679863 147,298,239 147,320,575 gene 3 Similar to N-myc downstream regulated LOC679885 147,402,020 147,422,644 gene 3 Ndrg3 N-myc downstream regulated gene 3 147,454,138 147,571,926 LOC679915 Similar to 40S ribosomal protein S2 147,573,852 147,575,831 DSN1, MIND kinetochore complex Dsn1 147,575,110 147,586,968 component, homolog R1151Q, T1252P, LOC311578 Hypothetical protein LOC311578 147,596,447 147,667,671 T1261A RGD1561343 Similar to C20orf118 147,675,172 147,687,339 Samhd1 SAM domain and HD domain, 1 147,689,421 147,727,429 96_97** Rbl1 Retinoblastoma-like 1 (p107) 147,734,622 147,796,630 G1018R LOC100360777 Hypothetical protein LOC100360777 147,804,746 147,976,705 LOC100360178 mCG6121-like 147,835,752 147,976,273 Rpn2 Ribophorin II 147,976,544 148,023,743 Ghrh Growth hormone releasing hormone 148,027,235 148,046,352 Manbal Mannosidase, beta A, lysosomal-like 148,066,721 148,096,306 Src v-src sarcoma viral oncogene homolog 148,157,256 148,170,524 Bladder cancer associated protein homolog Blcap 148,265,615 148,275,959 (human) Nnat Neuronatin 148,269,404 148,271,781 Ctnnbl1 Catenin, beta like 1 148,440,757 148,602,378 LOC680040 Similar to brain protein 44-like 148,462,608 148,463,497 Similar to chromosome 20 open reading RGD1305725 148,624,668 148,651,409 frame 102 RGD1561297 Similar to Rpl7a protein 148,651,424 148,665,266 RGD1562582 Similar to KIAA0406-like protein 148,692,148 148,720,469 Regulation of nuclear pre-mRNA domain Rprd1b 148,736,594 148,783,743 containing 1B Tgm2 Transglutaminase 2, C polypeptide 148,832,866 148,862,385 Q29E Similar to hypothetical protein RGD1563354 148,911,549 148,936,434 E588D, G743R D630003M21 Bactericidal/permeability-increasing Bpi 148,970,965 148,998,795 protein Lbp Lipopolysaccharide binding protein 149,016,605 149,043,530

Snph AA 413 414 415 416 417 418 419 420 ACI A A A A A A T T GCC GCC GCC GCC GCC GCC ACC ACC FHH A A A - - - - T GCC GCC GCC - - - - ACC

Necab3 AA 160 161 162 163 164 ACI A Q A R G Q R GCC CAG GCC CGT GGC CAG CGG FHH A Q A - - Q R GCC CAG GCC - - CAG CGG

ASIP AA 35 36 37 38 39 40 41 42 43 44 45 ACI S L K S N S S I N S L AGT CTA AAG AGC AAC TCT TCC ATC AAC TCA CTG FHH S L ------T H AGT CTA ------ACT CAC

Ncoa6 AA 257 258 259 260 261 262 263 264 ACI F P Q L Q Q Q Q TTT CCC CAG CTG CAG CAA CAG CAG FHH F P - - - - Q Q TTT CCC - - - - CAG CAG

Rbm12 AA 770 771 772 773 774 775 776 777 778 ACI P A - - - - - M P CCC GCA - - - - - ATG CCT FHH P A M P G P A M P CCC GCA ATG CCT GGT CCC GCA ATG CCT

Epb4.1l1 AA 852 853 854 855 856 ACI E E A W T GAA GAA GCC TGG ACT FHH E - A W T GAA - GCC TGG ACT

Samhd1 AA 95 96 97 98 ACI D P - - E S GAT CCC - - GAA AGC FHH D P D S E S GAT CCC GAC TCC GAA AGC

Gene Locus Start Stop ACI/FHH Exon/Intron/Intergenic NCBI ID LOC690064 142,853,483 142,853,483 T/C Exon ss262803283 LOC690064 142,853,547 142,853,547 T/C Exon ss262803284 LOC690064 142,853,579 142,853,579 G/A Exon ss262803285 LOC690064 142,860,696 142,860,696 A/G Exon ss262803286 Defb29 142,862,192 142,862,192 A/G Intron- TF Binding ss262803287 Defb29 142,862,798 142,862,798 G/A Intron- TF Binding ss262803288 Defb29 142,863,107 142,863,107 G/A Exon ss262803289 Defb29 142,863,319 142,863,319 A/G Intron- TF Binding ss262803290 Defb29 142,863,661 142,863,661 T/C Intron- TF Binding ss262803291 Defb29 142,864,148 142,864,148 C/T Intron- TF Binding ss262803292 Defb29 142,865,637 142,865,637 T/C Intron- TF Binding ss262803293 Defb36 142,946,958 142,946,958 -/G Intron- TF Binding ss262803294 Defb25 142,972,458 142,972,458 T/- Intron- TF Binding ss262803295 Rem1 142,982,920 142,982,922 CTC/T Intron- TF Binding ss262803340 (Intergenic) 142,991,271 142,991,273 GTG/TTGT Intergenic- TF Binding ss262803315 (Intergenic) 142,991,310 142,991,310 T/- Intergenic- TF Binding ss262803316 H13 143,034,387 143,034,387 T/G Intron- TF Binding ss262803296 H13 143,034,391 143,034,391 G/T Intron- TF Binding ss262803297 H13 143,044,057 143,044,057 G/A Intron- TF Binding ss262803298 Cox4i2 143,109,051 143,109,051 T/- Intron- TF Binding ss262803299 Bcl2l1 143,135,727 143,135,727 CAA/- Intron- TF Binding ss262803300 Bcl2l1 143,135,741 143,135,741 -/ACAAA Intron- TF Binding ss262803301 Bcl2l1 143,143,669 143,143,669 A/T Intron- TF Binding ss262803302 Bcl2l1( LOC100363271) 143,149,747 143,149,747 A/G Intron (Exon) ss262803303 Bcl2l1( LOC100363271) 143,149,834 143,149,834 G/A Intron (Exon) ss262803304 Bcl2l1( LOC100363271) 143,149,897 143,149,897 T/C Intron (Exon) ss262803305 Bcl2l1( LOC100363271) 143,149,948 143,149,948 A/C Intron (Exon) ss262803306 Bcl2l1( LOC100363271) 143,150,006 143,150,006 A/G Intron (Exon) ss262803307 Bcl2l1( LOC100363271) 143,150,011 143,150,011 T/A Intron (Exon) ss262803308 Bcl2l1( LOC100363271) 143,150,015 143,150,015 T/C Intron (Exon) ss262803309 Bcl2l1( LOC100363271) 143,150,098 143,150,098 G/T Intron (Exon) ss262803310 Bcl2l1( LOC100363271) 143,150,243 143,150,243 A/G Intron (Exon) ss262803311 Bcl2l1( LOC100363271) 143,151,881 143,151,881 G/A Intron (Exon) ss262803312 Bcl2l1 143,158,066 143,158,066 G/A Intron- TF Binding ss262803313 (Intergenic) 143,158,066 143,158,066 G/A Intergenic- TF Binding ss262803317 Bcl2l1 143,159,771 143,159,771 -/GG Intron- TF Binding ss262803314