Nephrol Dial Transplant (2008) 23: 3184–3191 doi: 10.1093/ndt/gfn215 Advance Access publication 28 April 2008
Original Article
A genome-wide search for linkage to chronic kidney disease in a community-based sample: the SAFHS Nedal H. Arar1,3 , Venkata S. Voruganti2 , Subrata D. Nath1 , Farook Thameem1 , Richard Bauer1,3 , Shelley A. Cole2 , John Blangero2 , Jean W. MacCluer2 , Anthony G. Comuzzie2 and Hanna E. Abboud1,3 1
Department of Medicine, University of Texas Health Science Center, 2 Department of Genetics, Southwest Foundation for Biomedical Research, San Antonio and 3 South Texas Veterans Health Care System (STVHCS), TX, USA
Abstract Background. Chronic kidney disease (CKD) phenotypes such as albuminuria measured by urinary albumin creatinine ratio (ACR), elevated serum creatinine (SrCr) and/or decreased creatinine clearance (CrCl) and glomerular filtration rate (eGFR) are major risk factors for renal and cardiovascular diseases. Epidemiological studies have reported that CKD phenotypes cluster in families suggesting a genetic predisposition. However, studies reporting chromosomal regions influencing CKD are very limited. Therefore, the purpose of this study is to identify susceptible chromosomal regions for CKD phenotypes in Mexican American families enrolled in the San Antonio Family Heart Study (SAFHS). Methods. We used the variance components decomposition approach (implemented in the software package SOLAR) to perform linkage analysis on 848 participants from 26 families. A total of 417 microsatellite markers were genotyped at an average interval of 10 cM spanning 22 autosomal chromosomes. Results. All phenotypes were measured by standard procedures. Mean ± SD values of ACR, SrCr, CrCl and eGFR were 0.06 ± 0.38, 0.85 ± 0.72 mg/dl, 129.85 ± 50.37 ml/min and 99.18 ± 25.69 ml/min/1.73 m2 body surface area, respectively. All four CKD phenotypes exhibited significant heritabilities (P < 0.0001). A genome-wide scan showed linkage on chromosome 2p25 for SrCr, CrCl and eGFR. Significant linkage was also detected on chromosome 9q21 for eGFR [logarithm of the odds (LOD) score = 3.87, P = 0.00005] and SrCr (LOD score = 2.6, P = 0.00026). ACR revealed suggestive evidence for linkage to a region on chromosome 20q12 (LOD score = 2.93, P = 0.00020).
Correspondence and offprint requests to: Nedal Arar, Department of Medicine/Nephrology, University of Texas Health Science Center, South Texas Veterans Health Care System, 7400 Merton Minter Blvd, San Antonio, TX 78229-4404, USA. Tel: +1-210-567-0075; Fax: +1-210567-4712; E-mail:
[email protected]
Conclusion. Findings indicate that chromosomal regions 2p25, 9q21 and 20q12 may have functional relevance to CKD phenotypes in Mexican Americans. Keywords: chronic kidney disease; genome-wide search; SAFHS
Introduction Chronic kidney disease (CKD) is becoming a major public health concern [1,2]. It is associated with many systematic complications including renal and cardiovascular diseases (CVDs) [3]. CKD is a manifestation of a gradual but permanent loss of kidney function over time, eventually progressing to end-stage renal disease (ESRD) and requiring renal replacement therapy [4]. CKD can be assessed by several clinical indicators such as albuminuria measured by the urinary albumin creatinine ratio (ACR), elevated serum creatinine (SrCr) and/or decreased glomerular filtration rate (GFR) [3]. A number of genetic epidemiological studies have shown that the CKD indicators/phenotypes cluster in families enriched with diabetes and/or hypertension, suggesting that there is a genetic predisposition of these phenotypes [5,6]. A genetic analysis of urinary ACR in Pima Indians with type 2 diabetes suggested an existence of a major gene effect with Mendelian inheritance [7,8]. Genome-wide scan to identify susceptibility loci for urinary ACR revealed suggestive evidence for linkage on various chromosomes [9–12]. Some studies have identified 7q susceptibility loci in scans employing DN phenotypes defined by albuminuria [12,13]. Krolewski et al. analyzed albuminuria as a quantitative trait and discovered a linkage peak at 7q36.2, which yielded a logarithm of the odds (LOD) score in excess of 3.0 [12]. Genetic analysis conducted with creatinine clearance (CrCl) and estimated GFR (eGFR) have demonstrated that they are highly heritable [14–18]. Hunt et al. reported heritabilities of 0.33, 0.36 and 0.53 for three successive CrCl
C The Author [2008]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please e-mail:
[email protected]
CKD in Mexican Americans
examinations in a healthy Utah pedigree [19]. DeWan et al. calculated heritability of CrCl as 0.17 in African Americans and 0.18 in non-Hispanic White hypertensive cohorts [20]. In regard to eGFR, using the Framingham Heart Study, Fox et al. determined that the eGFR heritability was 0.36 [21]. In a study of type 2 diabetic non-Hispanic White subjects, Langefeld et al. calculated the heritability of eGFR, using the MDRD (Modification of Diet in Renal Disease Study) equation, to be 0.75 after adjusting for age, gender, blood pressure, medications and haemoglobin A1c (HbA1c) levels [22]. Furthermore, in families with hypertension, Bochud et al. estimated the heritability of 0.52 for CrCl, estimated using the Cockcroft–Gault formula [23]. More recently, the Joslin group demonstrated significant eGFR heritability (h2 = 0.45) in families enriched for T2DM [24]. Many of these heritability studies simultaneously performed linkage analysis to identify chromosomal regions affecting regulation of quantitative CrCl and eGFR phenotypes. In the HyperGEN study involving 1100 hypertensive non-Hispanic White and African American subjects, a locus on 3q27 was identified [20]. Follow-up studies by this same group, with added subjects and denser genotyping, revealed linkage to 3p [25] and 7q [26] in the African American cohort. In two studies by Hunt et al., which examined pedigrees from non-Hispanic White families in Utah, loci on chromosomes 2q21.3 [14] and 10q23.3 [20] were identified for CKD phenotypes. The Framingham Heart Study identified linkage on chromosomes 3q26.3 with CrCl and on 4q23.3 with eGFR [21]. Altogether, these data indicate that CKD phenotypes are heritable in families enriched with diabetes, hypertension or both in African Americans, non-Hispanic White and Native Americans. However, the genetic influence on variation in CKD phenotypes in Mexican American families drawn from a community-based sample remains uninvestigated. The San Antonio Family Heart Study (SAFHS) aims to identify genes influencing the risk of CVD among low-income Mexican American families. This paper presents the first genome scan for localizing susceptible chromosomal regions for CKD phenotypes (ACR, SrCr, CrCl, eGFR) in families enrolled in SAFHS.
Methods Population characteristics The SAFHS is the first population-based project aimed to identify genes influencing the risk of CVD among Mexican American families. Probands for the SAFHS were selected randomly from a census tract in San Antonio of low-income Mexican Americans regardless of any preexisting medical conditions. Probands who were 40–60 years of age, had a spouse who was willing to participate and had at least six offspring who were ≥16 years of age were recruited. Family members including all first-, second- and third-degree relatives of probands ≥16 years of age and their spouse were invited to participate. Information regarding medical history, sociodemographics and anthropometrics was obtained in the first phase of SAFHS data collection [27,28].
3185
Institutional review board from the University of Health Science Center at San Antonio approved the study and informed consent was taken from all participants. Clinical variables CKD measures and calculations In this study, a total of 848 individuals were phenotyped for serum levels of creatinine, and urinary albumin and creatinine. Single-void morning urine sample was collected from each participant. Albumin in urine was quantitatively determined by turbidimetric method (Beckman Synchron LX 20, Beckman Coulter, Inc., Fullerton, CA, USA). Urine creatinine was measured using the modified kinetic Jaff´e method [29]. Urinary microalbumin was indexed to urinary creatinine as the urinary urine ACR in order to account for differences in urine concentration. Urinary ACR is a validated, reliable single-sample measure of urinary albumin excretion that is highly correlated with albumin excretion rates assessed by 24-h urine collection [30,31]. Blood samples were collected from all participants after an overnight fast and plasma was prepared and stored at −80◦ C until analyzed. SrCr was estimated by the modified kinetic Jaffe reaction (Beckman Synchron LX System). Creatinine clearance (CrCl) was estimated using the Cockcroft–Gault equation: CrCl (ml/min) = [(140 − age) × weight]/(72 × SrCr) (× 0.85 if female). GFR was estimated by the MDRD equation: eGFR (ml/min/1.73 m2 body surface area) = 186 × (SrCr) × age × (−0.203) × (0.742 if female) × (1.210 if black). Genotyping Genomic DNA was prepared from the lymphocytes of all enrolled subjects [32]. All family members were genotyped for 417 microsatellite markers that were spaced at an average interval of 10 cM. In addition, polymorphic markers in candidate genes were also included in the map as described previously by Comuzzie et al. [33]. The screening set and genotyping protocols are available at the website of the Center for Medical Genetics, Marshfield Medical Research Foundation (http://research.marshfieldclinic.org/genetics). After completion of the genotyping for the 10 cM marker map, the data were searched for spurious double recombination and those that were reflected by a high posterior probability of genotyping error were corrected. This correction significantly reduced the map expansion and led to the calculation of more accurate multipoint identity by descent (IBD) probabilities. Quantitative genetic analyses A multipoint linkage analysis was employed to detect a quantitative trait locus (QTL) or loci (QTLs) that might influence the variation in phenotypes related to CKD such as ACR, SrCr, CrCl and GFR. This technique is implemented in the software package SOLAR [34]. It is an extension of the variance components approach in which variance due to a specific QTL is added to the basic model. It is based
3186
N. H. Arar et al.
on estimating the effect of a specific QTL on the variation in phenotype and can be modelled as a function of the IBD relationship at the marker locus between family members. The phenotypic correlations between family members can be described as the cumulative effect of a specific QTL associated with a marker, and residual genetic and environmental effects [35]. A model under the null hypothesis in which the additive genetic variance for a specific QTL equals zero was tested against a model under an alternate hypothesis in which the additive variance was estimated. This is known as a likelihood ratio test (LRT), and the resultant LRT statistic in this particular case was distributed asymptotically as a 12 : 12 mixture of a χ2 variable with one degree of freedom and a point mass at zero [36]. Traditionally, a LOD score, which is computed directly from the LRT, is reported in linkage analyses. A LOD score of 3 implies that the probability that a QTL on a certain chromosome is affecting the variation in the trait of interest is 1000:1 and is considered to be significant, and a LOD score >2 (probability of linkage is 100:1) is considered as evidence of suggestive linkage [34].
Table 1. Relative pairs used in this study Relationship
Relative pairs
Phenotype
ACR
SrCr
CrCl
eGFR
Parent–offspring Siblings Grandparent–grandchild Avuncular Half-siblings Great grandparent–grandchild Grand avuncular Half avuncular First cousins First cousins, once removed Second cousins Other pairs Total
503 573 111 1122 110 – 262 148 1363 1257 586 387 6422
565 642 127 1237 123 2 279 159 1513 1394 628 442 7111
519 585 116 1108 109 2 255 127 1305 1163 519 467 6275
559 633 127 1211 122 2 277 159 1489 1364 619 442 7004
ACR: albumin creatinine ratio; SrCr: serum creatinine (mg/dl); CrCl: creatinine clearance (ml/min); eGFR: estimated glomerular filtration rate (eGFR, ml/min/1.73 m2 body surface area). Table 3. Clinical characteristics of the participants (%) Variables
Frequency (%) (N = 848)
Males Smokers Consume alcohol Diabetic Hypertensive Hypertension medication Diabetic medication Variables Age (years) BMI (kg/m2 ) Waist circumference (cm) Waist-to-hip ratio Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg)
383 (37) 166 (20) 364 (43) 183 (22) 443 (52) 212 (25) 180 (22) Mean ± SE 47.87 ± 14.8 31.76 ± 7.2 105.4 ± 70.8 0.96 ± 0.08 124.41 ± 19.03 69.80 ± 10.3
Results This study was conducted with 848 participants (men = 313, women = 535) for whom we have both the genotypic and CKD phenotypic data. Relative pairs used for this study are listed in Table 1. Relative pairs included in this study varied according to the trait being analyzed. A total of 6422, 7111, 6275 and 7004 pairs were used in the analysis of urinary ACR, SrCr, CrCl and eGFR, respectively. These included pairs such as parent–offspring, siblings, first cousins, second cousins, etc. The mean age of participants was 47.8 (SD ± 14.8) years and their BMI was 31.7 (SD ± 7.1) kg/m2 . Mean values for urinary ACR, SrCr, CrCl and eGFR were 0.06 ± 0.38, 0.85 ± 0.72 mg/dl, 129.85 ± 50.37 ml/min and 99.18 ± 25.69 ml/min/1.73 m2 body surface area, respectively. Of the CKD phenotypes, only SrCr was significantly different between sexes with men having higher levels of SrCr than women (P < 0.001). Our findings showed a relationship between ACR, CrCl and eGFR in diabetic and hypertensive subjects (Table 2). The P-value for Table 2 was obtained from t-statistic which was computed from Student’s t-test. This t-test was used to compare the
phenotypic values between men and women. Of the study participants 37% were males, 52% hypertensives, 22% diabetics and ∼14% had albuminuria (Table 3). We conducted quantitative genetic analysis of CKD phenotypes (ACR, SrCr, CrCl and eGFR) to determine genetic influence on their variation. Analysis showed all the above-mentioned phenotypes to be significantly heritable, as shown in Table 2. Heritabilities obtained in this study
Table 2. Descriptive statistics of some selected clinical variables by chronic kidney disease phenotypes and heritabilities Traits
ACR SrCr CrCl EGFR
h2 (SE)
0.24 (0.08) 0.19 (0.07) 0.25 (0.07) 0.21 (0.07)
P-value
1.5e−04 3.6e−04 6.0e−06 1.0e−04
Sex
Diabetes status
Hypertension status
Men
Women
Absent
Present
Absent
Present
0.087 (0.56) 1.01 (0.94) 131.61 (48.7) 98.86 (0.24)
0.04 (0.23) 0.76# (0.54) 128.8 (51.3) 99.37 (0.27)
0.02 (0.16) 0.81 (0.67) 131.38 (49) 101.23 (23)
0.19# (0.75) 0.97 (0.89) 124.2$ (56) 91.7$ (31)
0.024 (0.24) 0.77 (0.28) 136.53 (45) 103.74 (22)
0.09$ (0.48) 0.92$ (0.96) 123.60$ (52) 95.59$ (27)
All values are presented as mean (standard deviation). # P < 0.05. $ P < 0.001. ACR: albumin creatinine ratio; SrCr: serum creatinine (mg/dl); CrCl: creatinine clearance (ml/min); eGFR: estimated glomerular filtration rate (eGFR, ml/min/1.73 m2 body surface area).
CKD in Mexican Americans
3187
Table 4. Linkage results for renal phenotypes (LOD >
1)∗
Phenotypes
Chromosome
Nearest marker
Max location (cM)
LOD score (>1)
P-value
ACR
1 13 14 17 20 2 5 6 7 9 15 2 4 10 12 13 2 4 8 9 15
D1S200 D13S317 D14S742 D17S974 D20S107 D2S1780 D5S1017 D6S1056 D7S1830 D9S922 D15S642 D2S1780 D4S2639 & D4S3244 D10S2470 D12S330 D13S779 D2S1780 D4S403 D8S1145 D9S1122 D15S642
79 74 13 28 65 4 69 103 82 79 131 12 44 112 123 102 4 30 33 75 134
1.3785 1.6640 2.63 1.5589 2.9383 1.7822 1.3466 2.1967 1.4820 2.6153 2.2194 2.0480 1.1840 1.4319 1.1357 1.2142 1.5373 1.2377 1.2686 3.8731 1.6181
0.00587 0.00282 0.00025 0.00369 0.00209 0.00209 0.00638 0.00073 0.00449 0.00026 0.00069 0.00107 0.00902 0.00502 0.01110 0.00902 0.00390 0.00848 0.00782 0.00001 0.00317
SrCr
CrCl
eGFR
SrCr: serum creatinine (mg/dl); CrCl: creatinine clearance (ml/min); SEM: standard error of mean; SE: standard error; ACR: albumin creatinine ratio; eGFR: estimated glomerular filtration rate (ml/min/1.73 m2 body surface area); BMI: body mass index; CRP: C-reactive protein; GM-CSF: granulocyte-macrophage colony-stimulating factor. ∗ Highest LOD score for each trait has been depicted in bold.
Fig. 1. Multipoint LOD scores for phenotypes (CKD) related to chronic kidney disease on chromosome 20. Plot of the genome scan for CKD phenotypes as a continuous trait. The genetic distance along the chromosome is plotted on the X-axis. The strength of the linkage signal is plotted on the Y-axis. SrCr: serum creatinine; eGFR: estimated glomerular filtration rate; CrCl: creatinine clearance; ACR: albumin creatinine ratio. The Xaxis represents the chromosomal position and the Y-axis represents LOD scores.
for CKD phenotypes were moderate and ranged between 0.19 and 0.25 (P < 0.001). The best model for the genomewide scan for these phenotypes was obtained with age, sex, their higher terms and interactions, BMI, blood pressure medication and diabetic duration as covariates. Using this model, we subsequently performed multipoint linkage analysis to localize the chromosomal region(s) influencing these quantitative traits of CKD. Linkage mapping for
Fig. 2. Multipoint LOD scores for phenotypes related to chronic kidney disease on chromosome 9. Plot of the genome scan for CKD phenotypes as a continuous trait. The genetic distance along the chromosome is plotted on the X-axis. The strength of the linkage signal is plotted on the Y-axis. SrCr: serum creatinine; eGFR: estimated glomerular filtration rate; CrCl: creatinine clearance; ACR: albumin creatinine ratio. The X-axis represents the chromosomal position and the Y-axis represents LOD scores.
urine ACR showed five distinct genetic locations that did not overlap with the remaining CKD phenotypes. These regions exhibited a LOD score >1 on chromosomes 1, 13, 14, 17 and 20 (Table 4). The highest linkage signal was on chromosome 20 near the marker D20S107 (Figure 1). Chromosomal regions associated with SrCr were 2, 5, 6, 7, 9 and 15 (Table 4). The P-value for Table 4 (linkages) was calculated from a chi-square statistic usually used to test for variances. The chi-square statistic was computed from the
3188
Fig. 3. Multipoint LOD scores for phenotypes related to chronic kidney disease on chromosome 2. Plot of the genome scan for CKD phenotypes as a continuous trait. The genetic distance along the chromosome is plotted on the X-axis. The strength of the linkage signal is plotted on the Y-axis. SrCr: serum creatinine; eGFR: estimated glomerular filtration rate; CrCl: creatinine clearance; ACR: albumin creatinine ratio. The X-axis represents the chromosomal position and the Y-axis represents LOD scores.
difference between the null model and the test model. Of these, signals on chromosomes 6, 9 and 15 showed evidence of suggestive linkage (LOD > 2), whereas the highest signal was on chromosome 9 (LOD = 2.61, near marker the D9S922) (Figure 2). Similarly, chromosomal regions on 2, 4, 10, 12, and 13 were associated with CrCl (Table 4). Of these, the highest signal was found to occur at the marker D2S1780 on chromosome 2 (Figure 3). A strong signal was obtained for eGFR on chromosome 9 (Figure 1b; LOD = 3.87, near the marker D9S1122) whereas other signals with LOD > 1 were localized to chromosomes 2, 4, 8 and 15 (Table 4). QTL on chromosome 2 (near the marker D2S1780) was common to SrCr, CrCl and eGFR. Regions on chromosomes 2, 9 and 15 were common for both SrCr and eGFR; in fact their highest signals were just 3 cM apart (Table 4). Several positional candidate genes exist within the 1LOD support interval for signals for the examined CKD phenotypes. The 1-LOD support interval is a standardized approach and is equivalent to a 95% confidence interval under the appropriate settings. The 1-LOD change corresponds to a change in the chi-square statistic of 4.6 with 1 df and a significance value of 0.04. To increase our chances in covering a larger interval, we extend the search to include the 1-LOD support interval before and after the threshold of the linkage peak [34,36]. Table 5 depicts the 1-LOD support interval for QTLs for these phenotypes and genes that are present within this interval. The 1-LOD support intervals for urinary ACR and CrCl were 27 cM and 22 cM, respectively. SrCr and eGFR had overlapping 1-LOD support intervals of 13 cM each (Table 5).
Discussion In the current report, all the four examined CKD (ACR, SrCl, CrCl and eGFR) phenotypes were significantly heri-
N. H. Arar et al.
table and showed evidence of linkage on various chromosomal locations. Heritability is a measure of trait-sharing among relatives due to inherited factors and ranges between 0 and 1 [37]. Studies estimating heritabilities for urine ACR and eGFR found them to be in the range of 0.3–0.44, which is similar to what we found in our study. Even though our heritabilities are relatively moderate, we were able to reject the null hypothesis of no genetic effect on the variation in CKD phenotypes (P < 0.05). Additionally, we found several major chromosomal regions; the most prominent at 2p25, 20q12 and 9q21 were significant contributors to CKD phenotypes among low-income Mexican American families enrolled in the SAFHS. Albuminuria as estimated by urinary ACR was present in ∼14% of the participants. It is a major risk factor for renal as well as CVDs. We had previously conducted the genomewide scan with a smaller study sample and had a significant linkage of 3.5 on chromosome 20 (20q12) [38]. In this study sample, the addition of more individuals resulted in a decrease of the LOD score to 2.9 (P < 0.001), although the linkage remained in the same region. Replication of the linkage result shows that this region is of considerable importance with respect to albuminuria in this population. Previous studies reporting genome-wide scans for urinary ACR have identified loci on chromosomes 5q, 7, 8, 16, 17, 18, 19 and 22q [12,13,16,26,39]. However, none of these studies have identified urinary ACR in the same region as ours. The difference in results might be due to differences in ascertainment methods, sample size and study designs. There are some interesting candidate genes in the 1-LOD support interval for this signal. Hepatic nuclear factor 4 alpha (HNF4alpha) is one among them, whose mutations have been associated with maturity onset diabetes of the young, type 1 (MODY 1) [40]. In Caucasian patients with T2DM, a single-nucleotide polymorphism in the promoter region of HNF4alpha was found to co-segregate with diabetes as well as renal target organ damage [41]. Other relevant candidate genes in this region are protein tyrosine phosphatase-1B (PTP-1B) [42], protein kinase inhibitor gamma (PKIG) [43] and phospholipase C, gamma1 (PLCG1) [44]. In our study, genome-wide scans for SrCr and eGFR identified QTLs for both of them in the same chromosomal region (9q21). Genetic studies exploring genomic loci regulating variation in GFR and SrCr have shown significant linkage on chromosomes 2, 3, 4, 7, 10, 14 and 19 [14,16,17,26]. In a study conducted in Pima Indian sibling pairs, 9q22 was associated with diabetic nephropathy [13,45]. The genes of potential interest in the region of 9q21–22 are Cathespin L (CTSL), implicated in the renal tubular response to proteinuria and myocardial ischaemia [46], proprotein convertase 5 (PC5), which is colocalized with alpha V-integrin in atherosclerotic plaques and thus may have a role in inflammation or be associated with inflammation-related phenotypes [47] and transient receptor potential (TRP) cation channel, subfamily M, member 3 (TRPM3), a member of TRP family, which has a role in volume-regulated activity and renal calcium homeostasis [48]. Mutations in one of the TRP family members, TRPC6, have been implicated in focal segmental glomerulosclerosis [49].
CKD in Mexican Americans
3189
Table 5. One-LOD support intervals for CKD-related phenotypes and respective candidate genes located within the interval Phenotype
Chromosome
One-LOD support interval (cM)
Positional candidate genes
ACR
20q11–12
55–82
SrCr
9q21
70–83
CrCl eGFR
2p25 9q21
0–22 69–82
PLCG1 (Phospholipase C, gamma1) PPP1R16B [Protein phosphatase 1, regulatory (inhibitor) subunit 16B] HNF4 (Hepatic nuclear factor 4) CTSL (Cathepsin L) PC5 (Proprotein convertase) TRPM3 (Transient receptor potential cation channel, subfamily M, member 3) KLF 11 (Kruppel-like factor 11) CTSL (Cathepsin L) PC5 (Proprotein convertase 5) TRPM3 (Transient receptor potential cation channel, subfamily M, member 3)
ACR: albumin creatinine ratio; SrCr: serum creatinine (mg/dl); CrCl: creatinine clearance (ml/min); eGFR: estimated glomerular filtration rate (eGFR, ml/min/1.73 m2 body surface area).
In this study, we found CrCl to be under significant genetic influence and obtained evidence for a suggestive linkage on chromosome 2 (2p25). This region harbours Kruppel-like factor 11 (KLF11) whose mutations tend to cause MODY type 7. The chromosomal region on chromosome 2 was consistent with all three phenotypes (same marker D2S1780) and similarly QTLs on chromosomes 9 and 15 were common for SrCr and eGFR. Altogether, the above results suggest a common genetic link between CrCl, SrCr and eGFR. This finding is expected since all the three phenotypes include SrCr as an important variable in their assessment. Interestingly, urinary ACR revealed distinct chromosomal regions as compared to SrCr, CrCl and eGFR. We speculate that urinary ACR and the remaining CKD phenotype may be regulated by different genes. In this study, we found that the genes in the 1-LOD support interval surrounding the identified QTLs or the QTLs by themselves were related to CVD risk. For example, QTLs identified in this study on 2p25 and 20q12 were also associated with essential hypertension [50,51]. A product of the gene prostacyclin synthase (PTGIS) in the region of 20q12–13, prostaglandin I2 is a vasodilator and inhibitor of platelet aggregation [52]. Nakayama et al. [53] found an association between exon 9 deletion in this gene and essential hypertension. Similarly, a gene in the region of 9q22–31, ATP-binding cassette, subfamily A, member 1 (ABCA1) has an important function as a cholesterol efflux pump in the cellular lipid removal pathway. Mutations in the ABCA1 gene have been associated with very low levels of HDL cholesterol and excessive tissue deposition of cholesterol esters [54]. Thus, regions around QTLs for CKD phenotypes 2p25, 20q12 and 9q21 harbour genes that confer risk for CVD. Of the very few whole-genome scans conducted for CKD, there is some overlap between chromosomal regions identified in this study and reported by others. Puppala et al. [55] analyzed data from a Mexican American population that was at high risk for T2DM and its complications, to examine the genetic determinants of eGFR and determine the genotype by diabetes interaction influences on variation in eGFR. They found that the phenotypes GFR using the Cockcroft–Gault equation and GFR using the MDRD equation are significantly heritable (h2 = 40% and 36%, respectively) in the Mexican-American population. Additionally,
a strong evidence for linkage of eGFR was mapped to a genetic location on chromosome 9q in between the markers D9S922 (80 cM, 9q21.31) and D9S1120 (89 cM, 9q21.33). Interestingly, we identified similar linkage of eGFR to chromosome 9q (D9S1120). We speculate that unique eGFR alleles for the Mexican Americans cluster within these loci. Fine mapping of the 9q21 region would allow identification of common genetic determinants influencing eGFR in Mexican Americans. In contrast, our findings did not overlap with the first scan by Hunt et al. who identified linkage of CrCl to chromosome 10q in families from Utah, which were ascertained for CVD risk with a relatively normal GFR [19]. Also, our findings did not overlap with results from the second scan done by the same group, which reported an evidence for linkage of GFR to chromosome 2q [14]. Additionally, our findings and Hunt et al.’s results did not overlap with those of DeWan et al. who analyzed CrCl in a mixed African American and European American population ascertained for hypertension and at an increased risk for CVDs. DeWan et al. identified linkage on chromosome 3q27, but not chromosome 2 or 10 [20]. Fox et al. employed a multipoint variance component linkage analysis in 330 families from the community-based Framingham Heart Study offspring cohort, for SrCl, CrCl and eGFR, measured from 1998 to 2001 [21]. CrCl was estimated using the Cockcroft–Gault equation and eGFR was estimated using the simplified MDRD equation. The peak LOD scores for SrCr, eGFR and CrCl were 2.28 at 176 cM on chromosome 4, 2.19 at 78 cM on chromosome 4 and 1.91 at 103 cM on chromosome 3, respectively. Recently, Placha et al. reported a genome scan of eGFR estimated by serum cystatin C, and MDRD and Cockroft–Gault equations, in 63 extended Caucasian pedigrees (406 individuals with T2DM, 428 without diabetes) with normal eGFR values [24]. This group detected evidence for linkage among diabetic relatives on chromosome 2q and suggestive evidence on 10q, in proximity to the region previously identified by Hunt et al. [19], and 18p. Despite the similarities in phenotype definitions in these studies, linkage peaks identified in the current study are contradictory to the previous findings owing to differences in ethnic composition of the populations studied and/or study designs. A potential limitation of this study is that some quantitative measures such as urinary ACR and eGFR were
3190
determined on the basis of a solo random collection. Although validation with sequential measures would be an ideal approach, yet quality control studies demonstrated negligible variability on repeated creatinine measurements of the same serum sample (personal communication; Parekh et al., unpublished data). Further future studies will be performed and statistically analyzed in the genome-wide linkage approach to confirm our findings. In conclusion, several linkage peaks for the phenotypes of CKD were identified in a Mexican American cohort (SAFHS). After adjustment for diabetes duration and hypertension medication, the peaks on chromosomes 2, 9 and 20 were significant for linkage strongly suggesting that unique CKD alleles cluster within these loci. Fine mapping of these regions would eventually identify the genetic variants responsible for the linkage. Additional investigations examining the association of CKD and genetic loci in large populations are justified to verify our conclusions. Acknowledgements. We wish to thank all participants of the San Antonio Family Heart Study for their cooperation and generous participation. This study was supported by PO1 HL4522 from NHLBI and MH59490 from NIH. We also acknowledge the fund from NIH/NIDDK (P50 DK061597). This work was supported by VHA-Merit Review from BLR&D CSR&D (PI: H.A.). The Fredric C. Bartter General Clinical Research Center, UTHSCSA (M01-RR01346) provided clinical support for this project. Conflict of interest statement. None declared.
References 1. Schieppati A, Remuzzi G. Chronic renal diseases as a public health problem: epidemiology, social, and economic implications. Kidney Int Suppl 2005; 98: S7–S10 2. Jones CA, McQuillan GM, Kusek JW et al. Serum creatinine levels in the US population: Third National Health and Nutrition Examination Survey [published correction appears in Am J Kidney Dis 2000; 35: 178]. Am J Kidney Dis 1998; 32: 992–999 3. McCollough PA, Jurkovitz CT, Pergola PE et al. Independent components of chronic kidney disease as a cardiovascular risk state: results from the Kidney Early Evaluation Program (KEEP). Arch Intern Med 2007; 167: 1122–1129 4. Basile JN. Recognizing the link between CKD and CVD in the primary care setting: accurate and early diagnosis for timely and appropriate intervention. Med J 2007; 100: 499–505 5. Forsblom CM, Kanninen T, Lehtovirta M et al. Heritability of albumin excretion rate in families of patients with type II diabetes. Diabetologia 1999; 42: 1359–1366 6. Pontiroli AE, Monti LD, Pizzini A et al. Familial clustering of arterial blood pressure, HDL cholesterol, and pro-insulin but not of insulin resistance and microalbuminuria in siblings of patients with type 2 diabetes. Diabetes Care 2000; 23: 1359–1364 7. Imperatore G, Knowler WC, Pettitt DJ et al. Segregation analysis of diabetic nephropathy in Pima Indians. Diabetes 2000; 49: 1049–1056 8. Fogarty DG, Rich SS, Hanna L et al. Urinary albumin excretion in families with type 2 diabetes is heritable and genetically correlated to blood pressure. Kidney Int 2000; 57: 250–257 9. Imperatore G, Knowler WC, Nelson RG et al. Genetics of diabetic nephropathy in the Pima Indians. Curr Diab Rep 2001; 1: 275– 281 10. Freedman BI, Beck SR, Rich SS et al. (HyperGEN Investigators). A genome-wide scan for urinary albumin excretion in hypertensive families. Hypertension 2003; 42: 291–296 11. Fox CS, Yang Q, Guo CY et al. Genome-wide linkage analysis to urinary microalbuminuria in a community-based sample: the Framingham Heart Study. Kidney Int 2005; 67: 70–74
N. H. Arar et al. 12. Krowleski AS, Poznik GD, Placha G et al. A genome-wide linkage scan for genes controlling variation in urinary albumin excretion in type 2 diabetes. Kidney Int 2006; 69: 129–136 13. Imperatore G, Hanson RL, Pettitt DJ et al. (Pima Diabetes Genes Group). Sib-pair linkage analysis for susceptibility genes for microvascular complications among Pima Indians with type 2 diabetes. Diabetes 1998; 47: 821–830 14. Hunt SC, Coon H, Hasstedt SJ et al. Linkage of serum creatinine and glomerular filtration rate to chromosome 2 in Utah pedigrees. Am J Hypertens 2004; 17: 511–515 15. Freedman BI, Bowden DW, Rich SS et al. A genome scan for allcause end-stage renal disease in African-Americans. Nephrol Dial Transplant 2005; 20: 712–718 16. Turner ST, Kardia SLR, Mosley TH et al. Influence of genomic loci on measures of chronic kidney disease in hypertensive sibships. J Am Soc Nephrol 2006; 17: 2048–2055 17. Chen G, Adeyemo AA, Zhou J et al. A genome-wide search for linkage to renal function phenotypes in West Africans with type 2 diabetes. Am J Kidney Dis 2007; 49: 394–400 18. Iyengar SK, Abboud HE, Goddard KA et al. (Family Investigation of Nephropathy and Diabetes Research Group). Genome-wide scans for diabetic nephropathy and albuminuria in multiethnic populations: the Family Investigation of Nephropathy and Diabetes (FIND). Diabetes 2007; 56: 1577–1585 19. Hunt SC, Hasstedt SJ, Coon H et al. Linkage of creatinine clearance to chromosome 10 in Utah pedigrees replicates a locus for end-stage renal disease in humans and renal failure in the fawn-hooded rat. Kidney Int 2002; 62: 1143–1148 20. DeWan AT, Arnett DK, Atwood LD et al. A genome scan for renal function among hypertensives: the HyperGEN study. Am J Hum Genet 2001; 68: 136–144 21. Fox CS, Yang Q, Cupples LA et al. Genomewide linkage analysis to serum creatinine, GFR, and creatinine clearance in a communitybased population: the Framingham Heart Study. J Am Soc Nephrol 2004; 15: 2457–2461 22. Langefeld CD, Beck SR, Bowden DW et al. Heritability of GFR and albuminuria in Caucasians with type 2 diabetes mellitus. Am J Kidney Dis 2004; 43: 796–800 23. Bochud M, Elston RC, Maillard M et al. Heritability of renal function in hypertensive families of African descent in the Seychelles (Indian Ocean). Kidney Int 2005; 67: 61–69 24. Placha G, Poznik GD, Dunn J et al. A genome-wide linkage scan for genes controlling variation in renal function estimated by serum cystatin C levels in extended families with type 2 diabetes. Diabetes 2006; 55: 3358–3365 25. DeWan AT, Arnett DK, Miller MB et al. Refined mapping of suggestive linkage to renal function in African Americans: the HyperGEN study. Am J Hum Genet 2002; 71: 204–205 26. Leon JM, Freedman BI, Miller MB et al. Genome scan of glomerular filtration rate and albuminuria: the HyperGen study. Nephrol Dial Transplant 2007; 22: 763–771 27. MacCluer JW, Stern MP, Almasy L et al. Genetics of atherosclerosis risk factors in Mexican Americans. Nutr Rev 1999; 57: S59–S65 28. Mitchell BD, Kammerer CM, Blangero J et al. (the San Antonio Family Heart Study). Genetic and environmental contributions to cardiovascular risk factors in Mexican Americans. Circulation 1996; 94: 2159–2170 29. Kasiske BL, Keane WF. Laboratory assessment of renal disease: clearance, urinalysis, and renal biopsy. In: Brenner BM, Rector FC Jr (eds). The Kidney. Philadelphia, PA: W.B. Saunders, 1996, 1137– 1173 30. Bakker AJ. Detection of microalbuminuria. Receiver operating characteristic curve analysis favors albumin-to-creatinine ratio over albumin concentration. Diabetes Care 1999; 22: 307–313 31. Nathan DM, Rosenbaum C, Protasowicki VD. Single-void urine samples can be used to estimate quantitative microalbuminuria. Diabetes Care 1987; 10: 414–418 32. Miller SA, Dykes DD, Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 1988; 16: 1215
CKD in Mexican Americans 33. Comuzzie AG, Hixson JE, Almasy L et al. A major quantitative trait locus determining serum leptin levels and fat mass is located on human chromosome 2. Nat Genet 1997; 15: 273–276 34. Almasy L, Blangero J. Multipoint quantitative-trait linkage analysis in general pedigrees. Am Hum Genet 1998; 62: 1198–1211 35. Blangero J, Almasy L. Mulitpoint oligogenic linkage analysis of quantitative traits. Genet Epidemiol 1997; 14: 959–964 36. Self S, Liang K. Asymptotic properties of maximum likelihood-ratio tests under nonstandard conditions. J Am Stat Assoc 1987; 82: 605– 610 37. Rogers J, Mahaney MC, Almasy L et al. Quantitative trait linkage mapping in anthropology. Am J Phys Anthropol 1999; 129: 127– 151 38. Arar NH, Nath SD, Thameem F et al. Genome-wide scans for microalbuminuria in Mexican Americans: the San Antonio family heart study. Genet Med 2007; 9: 80–87 39. Placha G, Canani LH, Warram JH et al. Evidence for different susceptibility genes for proteinuria and ESRD in type 2 diabetes. Adv Chronic Kidney Dis 2005; 12: 155–169 40. Ryffel GU. Mutations in the human genes encoding the transcription factors of the hepatocyte nuclear factor (HNF)1 and HNF4 families: functional and pathological consequences. J Mol Endocrinol 2001; 27: 11–29 41. Price JA, Fossey SC, Sale MM et al. Analysis of the HNF4 alpha gene in Caucasian type II diabetic nephropathic patients. Diabetologia 2000; 43: 364–372 42. Forsell PA, Boie Y, Montalibet J et al. Genomic characterization of the human and mouse protein tyrosine phosphatase-1B genes. Gene 2000; 260: 145–153 43. Collins SP, Uhler MD. Characterization of PKIgamma, a novel isoform of the protein kinase inhibitor of cAMP-dependent protein kinase. J Biol Chem 1997; 272: 18169–18178 44. Bristol A, Hall SM, Kriz RW et al. Phospholipase C-148: chromosomal location and deletion mapping of functional domains. Cold Spring Harb Symp Quant Biol 1988; 53: 915–920
3191 45. Sale MM, Freedman BI. Genetic determinants of albuminuria and renal disease in diabetes mellitus. Nephrol Dial Transplant 2006; 21: 13–16 46. Joseph LJ, Chang LC, Stamenkovich D et al. Complete nucleotide and deduced amino acid sequences of human and murine preprocathepsin L: an abundant transcript induced by transformation of fibroblasts. J Clin Invest 1988; 1621–1629 47. Cao H, Mok A, Miskie B et al. Single-nucleotide polymorphisms of the proprotein convertase subtilisin/kexin type 5 (PCSK5) gene. J Hum Genet 2001; 46: 730–732 48. Grimm C, Kraft R, Sauerbruch S et al. Molecular and functional characterization of the melastatin-related cation channel TRPM3. J Biol Chem 2003; 278: 21493–21501 49. Reiser J, Polu KR, Moller CC et al. TRPC6 is a glomerular slit diaphragm-associated channel required for normal renal function. Nat Genet 2005; 37: 739–744 50. Angius A, Petretto E, Maestrale GB et al. A new essential hypertension susceptibility locus on chromosome 2p24–p25, detected by genomewide search. Am J Hum Genet 2002; 71: 893–905 51. Wallace C, Xue MZ, Newhouse SJ et al. Linkage analysis using cophenotypes in the BRIGHT study reveals novel potential susceptibility loci for hypertension. Am J Hum Genet 2006; 79: 323–331 52. Yokoyama C, Yabuki T, Inoue H et al. Human gene encoding prostacyclin synthase (PTGIS): genomic organization, chromosomal localization, and promoter activity. Genomics 1996; 36: 296–304 53. Nakayama T, Soma M, Watanabe Y et al. Splicing mutation of the prostacyclin synthase gene in a family associated with hypertension. Biochem Biophys Res Commun 2002; 297: 1135–1139 54. Lawn RM, Wade DP, Garvin MR et al. The Tangier disease gene product ABC1 controls the cellular apolipoprotein-mediated lipid removal pathway. J Clin Invest 1999; 104: R25–R31 55. Puppala S, Arya R, Thameem F et al. Genotype by diabetes interaction effects on the detection of linkage of glomerular filtration rate (GFR) to a region on chromosome 2q in Mexican Americans. Diabetes 2007; 56: 2818–2828 Received for publication: 26.11.07 Accepted in revised form: 26.3.08
