Quantitative trait loci for proteinuria in the focal
glomerulosclerosis mouse model
Laboratory of Animal Genetics, Division of Applied Genetics and Physiology, Graduate School of Bioagricultural Sciences,
Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Aichi 464-8601, Japan
Genetic Resources Center, Korea Research Institute of Bioscience and Biotechnology (KRIBB), Taejon, 305-333, Korea
Department of Medical Genetics/Experimental Animal Center, College of Medicine, Hallym University, Chunchon, 200-702, Korea
Received: 26 October 2004 / Accepted: 6 January 2005
The FGS/Kist strain of mice, a new animal model for
focal glomerulosclerosis (FGS) in humans, was pre-
viously established by recurrent selection for high
proteinuria, which is a principal marker of FGS, from
descendants of CBA/Nga and RFM/Nga strains. We
performed a genome-wide scan for quantitative trait
loci (QTLs) affecting proteinuria in a population of
356 backcross progeny derived from a cross between
FGS/Kist and the standard normal strain, C57BL/6J.
Five proteinuria QTLs (Ptnu1–5) were detected at the
genome-wide 5% or less level. Ptnu1 and Ptnu2, lo-
cated on Chromosomes (Chrs) 8 and 17, respectively,
had main effects on proteinuria and also interacted
epistatically with each other. However, Ptnu3 on
Chr 9 and Ptnu4 and Ptnu5 both on Chr 15 had
epistatic interaction effects only. Except for the
epistatic interaction effect of Ptnu4 and Ptnu5, all
alleles derived from FGS/Kist were responsible for
the high proteinuria. These results indicated that the
genetic control of proteinuria is complex and the
identified QTLs may provide new insights into the
pathogenesis of FGS in mice as well as in humans.
Focal glomerulosclerosis (FGS) is one of the most
common and nonspecific patterns of glomerular in-
jury in humans. It is characterized clinically by
proteinuria and progressive renal impairment and
histologically by focal and segmental areas of glo-
merular sclerosis and tuft collapse (Harris 1999;
Sharma et al. 1999). In developing countries, FGS
accounts for about 20% of children and 5% of adults
with end-stage renal disease (Winn et al. 1999).
Possible genetic contributions to the development of
FGS have been reported in multiple families and
sibling pairs (Winn et al. 1999; Mathis et al. 1998).
To date, podocin (Boute et al. 2000) and nephrin
(Kestila et al. 2000) are reported as causal genes for
congenital nephritic syndrome which leads to FGS.
a-actinin 4 (Kaplan et al. 2000) and CD2-associated
protein (Kim et al. 2003) are also reported as being
associated with development of FGS. Apart from
these monogenic forms of FGS, very few causal
genes have been identified in the polygenic forms
that are influenced by strong environmental factors.
The progressive stages of FGS phenotypic heteroge-
neity also deter identification of the causal genes.
Genetic dissection using animal models repre-
sents an alternative approach to understanding the
genetic basis of complex renal disease (Brown et al.
1996; Rapp 2000; Schulz el al. 2002). Brown et al.
(1996) mapped quantitative trait loci (QTLs) for renal
impairment in the fawn-hooded rat, accounting for
about half of the genetic variation. Shiozawa et al.
(2000) reported three independently segregating
QTLs for proteinuria in the same rat. Of these loci,
Rf2 (renal failure-2) is located in the region syntenic
to human Chromosome (Chr) 19q13 (Mathis et al.
1998), where a-actinin 4, described as a causal gene
for the monogenic form of FGS, is located (Kaplan et
The FGS/Kist mouse, established by recurrent
selection for high proteinuria from F5 progeny of a
cross between CBA/Nga and RFM/Nga strains, is
characterized by rapid decrease in body weight, high
proteinuria at 4–5 months of age, and death within 1
month after onset of the wasting syndrome (Hyun et
Correspondence to: Akira Ishikawa; E-mail: email@example.com
DOI: 10.1007/s00335-004-3023-7 Volume 16, 242–250 (2005) Springer Science+Business Media, Inc. 2005