Tensin2 (Tns2), a member of the tensin family that includes Tns1, Tns3 and Tns4, is a focal adhesion-localized multidomain protein that possesses protein tyrosine phosphatase (PTP) and Src homology 2 (SH2)-phosphotyrosine binding (PTB) domains [1]. Tns2 mRNA is detected in the brain, heart, kidney, skeletal muscle, liver, lung, colon, small intestine and other tissues [2,3,4,5,6]. Stable expression of Tns2 in HEK293 cells alters cell morphology and reduces actin filaments. Overexpression of Tns2 in HEK293 cells reduces cell proliferation in a functional PTP domain-dependent manner and inhibits cell migration [7]. In contrast, stable expression of Tns2 protein with its C terminus fused to green fluorescent protein in HEK293 cells promotes cell migration [2]. In a podocyte cell line, it was shown that Tns2 knockdown enhances actin stress fiber formation and cell migration [8]. Taken together, these studies suggest that Tns2 function is involved in the reorganization of the actin cytoskeleton.
However, in vivo, Tns2 dysfunction only leads to apparent abnormalities in the kidney glomerulus [4, 8]. Newborn ICGN mice, carrying a Tns2-null mutation (hereafter nph), showed normal glomerular development, whereas podocyte foot process effacement and abnormal thickening of the glomerular basement membrane (GBM) were observed by 14 days after birth [9]. GBM thickening, which begins during glomerular maturation in ICGN mice, is considered to be a result of the abnormal extracellular matrix accumulation caused by podocytes but not by endothelial cells [9]. As observed in chronic kidney disease, ICGN mice then develop glomerular injury, including mesangial expansion, and massive albuminuria followed by tubulointerstitial injury with age [10, 11].
The podocyte-related manifestations of Tns2 dysfunction also depend on murine strain. For example, in response to nph mutation, albuminuria or podocyte foot process effacement was observed in FVB/NJ (FVB) and DBA/2J, but not in C57BL/6J (B6) and 129+Ter/SvJcl mice [12,13,14,15,16]. This disparity among strains indicates the presence of modifier genes. A comparison of the differences in the genetic backgrounds between the Tns2-deficiency resistant and susceptible strains can help in identifying these modifier genes, which may also help in elucidating the mechanism underlying Tns2 function in podocytes. Indeed, we previously performed quantitative trait loci (QTL) analysis and congenic analysis using resistant B6 and susceptible ICGN mice, and identified a modifier locus associated with podocyte injury caused by the nph mutation, Tpir, on chromosome 2 [17, 18]. ICGN.B6 congenic strains carrying the B6 allele of Tpir showed alleviated podocyte foot process effacement, GBM thickening and albuminuria, although the degree of improvement was inferior to that in the B6 genetic background itself, which shows almost no podocyte alteration [18]. This result indicates the existence of other modifier genes. The following points may explain why our previous QTL analysis failed to find the other significant loci: (1) our previous analysis focused on the severity of chronic kidney disease, and evaluated the traits related to the late-phase of chronic kidney disease in 16-week-old mice, such as tubulointerstitial injury, renal anemia and renal failure, but not albuminuria; (2) the quantitative traits, except for the histological scores, varied greatly in the ICGN mice themselves; and (3) the glomerular injury score focused on mesangial matrix accumulation, which is primarily caused by abnormal extracellular matrix metabolism in mesangial cells [19]. A further problem is that complete genetic information on the ICGN mouse genome remains unavailable.
In this study, we focused on podocyte injury and undertook a genome-wide linkage analysis of backcrosses between two Tns2-deficient mouse strains, B6.ICGN-Tns2nph (B6-Tns2nph) and FVB.ICGN-Tns2nph (FVB-Tns2nph) to identify modifier loci that may have been missed in previous studies.