Experiment
In the present study, genome-wide linkage analysis was performed using backcross progeny from susceptible ICGN and resistant C57BL/6J (B6) mice carrying Tns2nph to identify the chronic kidney disease (CKD) resistance genes and elucidate the mechanism by which Tns2 deficiency leads to CKD.
Blood samples were collected from the postcava under isoflurane anesthesia. Hemoglobin concentration (HGB), hematocrit (HCT) and blood urea nitrogen (BUN) were measured. Formalin-fixed and paraffin-embedded kidney blocks were sectioned. Two types of histopathological scores were adopted to quantitate the severities of renal damage. A glomerular index, in which the severities were based on glomerular injury, was calculated from the scores for twenty glomeruli randomly selected from each sample. The tubular index was the sum of two histopathological parameters, urinary cast and tubule dilation.
Two hundred and forty-six (male =124, female = 122) Tns2 (nph/nph) N2 backcross ((ICGN x C57BL/6J)F1 x ICGN) mice were used for a genome-wide scan. Tns2 genotypes were determined by PCR using genomic DNA from the tail tips. Tns2nph homozygous mice were identified using the primers AGACACCACCAGCACCTTCT and CGATCCAGCTCCTGTCTTTC, which distinguished the Tns2nph allele by the 8-base deletion in exon 18 of Tns2. For QTL analysis, a total of 92 informative microsatellite markers and coat color were used for the genotyping analysis. Map positions of the microsatellite markers were based on information from the Mouse Genome Informatics of Jackson Laboratory (MGI; http://www.informatics.jax.org).
QTL analysis was performed using the Map Manager QTXb20 software program that uses a maximum likelihood algorithm with interval mapping and simultaneous search. For each Chr, the LRS values were calculated by 10000 random permutations of the trait values relative to the genotypes of the marker loci. Genome-wide significance thresholds were calculated in terms of LRS by 10000 random permutation tests based on the established guidelines. The logarithm of odds (LOD) score was calculated by dividing the LRS value by 4.61. Significant linkage was determined according to the traditional threshold (LOD score>3.3).
HGB, HCT, BUN, glomerular index and tubular index at 16 weeks of age were used as indicators for CKD. Table 1:
QTL Tubisq1 (tubulointerstitial injury severity, QTL 1) was identified from analysis of the tubular index, which revealed the intensity of tubulointerstitial injury. The locus mapped to Chromosome 2 peaking at 31 cM with marker D2Mit369 with a LOD score of 5.36, p<0.001 and accounted for 10% of trait variance. Heterozygous (ICGN/B6) mice exhibited CKD resistance at Tubisq1. Two suggestive loci mapped to Chr 10 (LOD=2.27 at 64 cM) and Chr X (LOD=2.65 at 67 cM), respectively.
QTL Hbbcq3 (hemoglobin concentration QTL 3) was identified from analysis of hemoglobin concentration (HGB), LOD=4.47. The locus mapped to Chromosome 2 peaking at 36 cM with marker D2Mit380, Student's T test, p<0.001. Heterozygous (ICGN/B6) mice exhibited CKD resistance at Hbbcq3.
QTL Hctq6 (hematocrit QTL 6) was identified in analysis of percent hematocrit and also mapped to Chromosome 2 peaking at 36 cM with marker D2Mit380, Student's T test, p<0.001 with a LOD score of 3.58. Heterozygous (ICGN/B6) mice exhibited CKD resistance at Hctq6.
QTL Gssq1 (glomerulosclerosis severity QTL 1) was identified in analysis of the glomerular index mapping to Chromosome 13 peaking at 14 cM with marker D13Mit60, LOD=3.49, p<0.001. ICGN/B6 heterozygous mice showed statistically milder glomerular injury than ICGN/ICGN homozygous mice. A second, suggestive locus mapped to Chr 2 (LOD=2.41 at 51 cM).
Epistatic interactions were searched between all of the tested markers by MapManager QTXb20. Three significant interactions in the tubular index were identified associated with markers on Chrs 2 and 10, which showed significant or suggestive linkages to tubular index, suggesting multiple QTL for tubular index on Chr 2.
To identify candidate genes, the QTLs on Chrs 2 and 13 were screened for podocyte-related genes using XPodNet, and protein-protein interactions in the podocytes. In total, 46 candidate genes were listed (Table 2 and 3).
Additional QTL associated with blood traits, independent of CKD, were also identified, Table S2:
QTL Mchq17 (mean corpuscular hemoglobin QTL 17) mapped to Chromosome 2 peaking at 51 cM nearest marker D2Mit66 with a LOD score of 3.71, p<0.001. The B6 allele was the high allele at this locus.
QTL Mchq18 (mean corpuscular hemoglobin QTL 18) mapped to Chromosome 9 peaking at 60 cM nearest marker D9Mit212 with a LOD score of 15.81, p<0.001. The ICGN allele was the high allele at this locus.
QTL Mchq19 (mean corpuscular hemoglobin QTL 19) mapped to Chromosome 11 peaking at 16 cM nearest marker D11Mit230 with a LOD score of 4.77, p<0.001. The ICGN allele was the high allele at this locus.
QTL Mchcq2 (mean corpuscular hemoglobin concentration QTL 2) mapped to Chromosome 9 peaking at 60 cM nearest marker D9Mit212 with a LOD score of 6.51, p<0.001. The ICGN allele was the high allele at this locus.
QTL Mcvq14 (mean corpuscular volume QTL 14) mapped to Chromosome 6 peaking at 63 cM nearest marker D6Mit194 with a LOD score of 4.59, p<0.001. The ICGN allele was the high allele at this locus.
QTL Mcvq15 (mean corpuscular volume QTL 15) mapped to Chromosome 11 peaking at 16 cM nearest marker D11Mit230 with a LOD score of 4.01, p<0.001. The ICGN allele was the high allele at this locus.
QTL Hbbcq4 (hemoglobin concentration QTL 4) mapped to Chromosome 7 peaking at 60 cM with marker D7Mit220 with a LOD score of 3.49, p<0.001. The ICGN allele was the high allele at this locus.
QTL Hbbcq5 (hemoglobin concentration QTL 5) mapped to Chromosome 9 peaking at 60 cM with marker D9Mit212 with a LOD score of 4.32, p<0.001. The B6 allele was the high allele at this locus.
QTL Hctq7 (hematocrit QTL 7) mapped to Chromosome 9 peaking at 60 cM nearest marker D9Mit212 with a LOD score of 8.98, p<0.001. The B6 allele was the high allele at this locus.
QTL Hctq8 (hematocrit QTL 8) mapped to Chromosome 18 peaking at 20 cM nearest marker D18Mit94 with a LOD score of 4.08, p<0.001. The ICGN allele was the high allele at this locus.
QTL Rbc9 (red blood cell count 9) mapped to Chromosome 9 peaking at 60 cM with marker D9Mit212 with a LOD score of 13.9, p<0.001. The B6 allele was the high allele at this locus.
Analysis Tools