| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
MINI-SYMPOSIUM |

*
Department of Human Genetics, University Hospital Nijmegen, Nijmegen, The
Netherlands
Department of Pediatrics, St. Joseph Hospital, Veldhoven, The
Netherlands.
Correspondence to Dr. Nine Knoers, Department of Human Genetics, University Hospital Nijmegen, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands. Phone: 31-24-3613946; Fax: 31-24-3565026; E-mail: n.knoers{at}antrg.azn.nl
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
The most serious component of NPS is the nephropathy, but this does not occur in all patients with the syndrome. In a retrospective study analyzing renal features of 123 observations in the literature, Meyrier et al. (3) discovered renal involvement in more than 60% of cases. In most cases, the nephropathy manifests only by chronic benign proteinuria. In approximately 15% of cases, however, the disease developed toward end-stage renal disease. It is unknown which factors are responsible for progression to renal insufficiency, and therefore renal prognosis is unpredictable. The light microscopy findings of renal biopsies in NPS are nonspecific. At the ultrastructural level, however, the glomeruli show a uniform and characteristic picture: irregular thickening of the glomerular basement membrane (GBM) with electron-lucent areas giving it a so-called "moth-eaten appearance." Within the GBM and the mesangium, fibrillar collagen-like material is found, and finally there is fusion of epithelial foot processes. Remarkably, these ultrastructural changes have been found in all biopsied cases, independent of clinical involvement, and therefore do not seem to correlate with impaired renal function.
In the late 1960s, the gene for NPS was assigned to the distal end of the long arm of chromosome 9 by the establishment of linkage of NPS to the ABO blood group locus and the adenylate kinase gene (4,5). It lasted a few decades, however, before the localization of the NPS gene could be significantly refined to a genetic interval of 1 cM in region 9q34.1 (6,7,8,9).
Recently, Dreyer et al. (10) showed that NPS is caused by mutations in the LMX1B gene. The involvement of this gene in NPS was subsequently confirmed by others (11,12). Mutations in LMX1B were also found in families with NPS and glaucoma (11). LMX1B belongs to a family of highly related LIM-homeodomain transcription factors that are involved in pattern formation during development. These proteins contain two tandem LIM domains, which are cysteine-rich, zinc-binding domains that facilitate the interaction with other transcription factors: a homeodomain that has DNA-binding activity and a transcriptional activation domain (13). The identification of the LMX1B gene as being responsible for NPS was anticipated by investigations of the role of LMX1B homologous genes in limb development. In the chicken, it had been shown that lmx1 specifies the dorsal cell fate of the limb. Thus, absence of lmx1 expression leads to the development of biventral limbs (14). Mice with a homozygous deletion of exon 3 to 7 in the mouse lmx1 homologue, LmX1b, exhibit nail dysplasia and skeletal defects similar to those observed in NPS, such as iliac hypoplasia, aplasia of the patellae, and joint abnormalities marked by osseous misarticulations. In addition, ultrastructural examination of the kidneys showed prominent irregular thickening of the GBM with occasional regions of membrane discontinuity, reminiscent of the renal findings in NPS (15).
In this report, we present the identification of seven different mutations, including one novel variant, in the LMX1B gene in eight Dutch families with NPS and discuss the presumptive effect of these mutations on the function of the LMX1B protein, as well as the absence of convincing genotypephenotype correlations.
| Materials and Methods |
|---|
|
|
|---|
|
|
Mutation Detection Analysis
Genomic DNA was extracted from whole blood according to the procedure
described by Miller et al.
(17). To find mutations in the
LMX1B gene, we performed single-strand conformation polymorphism
(SSCP) analysis. The different LMX1B exons were amplified by PCR
using standard PCR buffer in a total volume of 25 ul, containing 50 ng of
genomic DNA, 15 pmol of each appropriate primer
(Table 2), 1 U of Taq
polymerase (Life Technologies, Breda, The Netherlands) and 0.5 mM of each
dNTP. Amplifications were performed in a Thermal Cycler (Perkin Elmer,
Norwalk, CT). Samples were denatured at 92°C for 5 min and then subjected
to 35 cycles of amplification: 1 min at 95°C, 1 min at annealing
temperature (Table 2), 1 min at
72°C. SSCP analysis was performed using the Genephor Gelsystem (Amersham
Pharmacia Biotech, Roosendaal, the Netherlands). PCR products were loaded on
precast, ready-to-use GeneGel Excel 12.5/24 polyacrylamide gels (Pharmacia).
After electrophoresis, gels were silver stained in an automated gel stainer
using the Plus-one DNA Silverstaining Kit (Pharmacia), according to the
protocol of the manufacturer. Samples were run in duple at 5°C and
15°C, respectively, to increase sensitivity. When a band shift was
detected, the corresponding fragment was bidirectionally sequenced to reveal
the nature of the mutation. In case no mutations were found through SSCP
analysis, all exons were sequenced. Automated DNA sequencing was performed on
an ABI PRISM 377 (PE Biosystems, Nieuwerkerk ald Yssel, The Netherlands) using
dye terminator chemistry.
|
| Results |
|---|
|
|
|---|
A) will probably lead
to a shorter LMX1B mRNA as a result of exon skipping. Because
LMX1B-expressing tissue, such as kidney, was not available, we were
not able to confirm this presumed effect of this mutation. In families 2, 3,
4, 7, and 8, the mutations cosegregate with the NPS phenotype. The seven
mutations were not found in a panel of 100 control chromosomes. In addition to
these putative loss-of-function mutations, five common polymorphisms, two
silent mutations (Ser219Ser; Glu124Glu), and three intron polymorphisms (IVS
2, +7 G
C; IVS 4, -49 C
T; IVS 8, -49 C
A) were found in our
series.
| Discussion |
|---|
|
|
|---|
Three of the mutations identified in our study are frequently occurring mutations. R198X has been found in 4%, R200Q in 5%, and A213P in 3% of all families tested. Although they did not show their data, McIntosh et al. (12) concluded, on the basis of haplotype analysis with polymorphic microsatellite markers from the 9q34 region, that these mutations recur in unrelated individuals. Haplotype analysis in our Dutch cohort has been performed only in the larger families, at the time when the gene responsible for NPS was not yet identified, to narrow the chromosomal region encompassing the disease gene. The patients carrying the R198X, R200Q, and A213P mutations did not belong to these previously analyzed families, either because they were sporadic patients or because the family was too small or only a few family members were available for study. Therefore, although unlikely, we cannot exclude the possibility that our patients are related to any of the other families carrying a similar mutation. The high rate of recurrence of the R198X and R200Q mutations might be explained by the fact that these mutations are the result of a transition at a CpG dinucleotide, which is a mutational hot spot because of the tendency for the cytosine to be methylated and subsequently deaminated to thymine. The recurrence of the A213P mutation, however, remains unexplained.
On the basis of the nature of the mutations and/or the mutations'
localization in functionally important domains of the LMX1B protein,
together with the consegregation of the mutations with the NPS phenotype
within the families and the absence of these mutations in 100 control
chromosomes, it is very likely that these seven mutations are indeed harmful
mutations and not innocuous polymorphisms. Thus, the two nonsense mutations
and the splice-site mutation are predicted to result in truncated
LMX1B proteins. The four remaining mutations are missense mutations,
and three of these amino-acid substitutions are localized in the homeodomain
of the protein, where they are likely to have an effect on DNA binding. R200Q
and A213P were shown to reduce and abolish DNA binding, respectively, in
in vitro DNA binding studies
(12). The homeodomain-DNA
interactions of homeobox genes have been intensively studied in many species
(for review, see reference
(20). In the course of the
evolution, the amino-acid sequence of the homeodomain has been conserved to a
high degree, reflecting the importance of this sequence for establishing DNA
binding. Because it is known which amino acids in the homeodomain are
important for DNA-binding specificity, direct DNA-backbone contacts, and
secondary structure, the putative effects of the identified mutations in the
homeodomain of LMX1B on DNA binding may be predicted. The R200Q
substitution affects a highly conserved amino acid located in the flexible
N-terminal arm of the homeodomain, which is important in establishing contact
to the minor groove of the DNA and as such significantly contributes to the
high DNA-binding affinity of the homeodomain. The alanine substitution in the
first helix of the homeodomain (A213P) is likely to disturb the formation of a
three-dimensional structure that is determined by the folding of the
helices I, II, and III into a tight globular structure
(21). Because this
three-dimensional structure of the homeodomain ensures the positioning of the
third (recognition) helix in the major groove of the DNA, where it establishes
contact with both strands of the DNA, it can be predicted that disturbance of
the three-dimensional structure by the A213P mutation will indirectly disrupt
direct DNA contact with the third helix and therefore DNA binding. The
substitution of an alanine in the third helix of the homeodomain (A249P) is
likely to hinder the base-specific contact with DNA in the major groove and as
such the specificity of DNA binding.
From the function and expression studies of lmx-1 in chicken and
Lmx1b in mice, respectively, it can be deduced that the skeletal
defects seen in NPS reflect a disruption of normal dorsoventral patterning of
the limb during development
(14,15).
Although it is known from these animal studies that Lmx1b is
expressed in fetal and adult kidney tissue, the exact role of this gene in the
development of the kidney is much less clear. On the basis of the renal
manifestations in NPS, it is likely that Lmx1b has an important role
in patterning and cell differentiations in the kidney also. Recent studies by
Morello et al. (22)
showed that Lmx1b localizes to the glomeruli of E15.5 and newborn
wild-type mice. In addition, these investigators demonstrated absence of
3 (IV) and
4 (IV) collagen expression in Lmx1b
/ mutant kidneys. It is widely known that during development,
there is a switch from an
1/
2 to an
3/
4 network of
collagen IV expression in the GBM
(23). Therefore, on the basis
of the findings in Lmx1b mutant kidneys, it was postulated that
Lmx1b might play an integral role in this coordinated transcriptional
switch (22). For humans with a
heterozygous mutation in LMX1B, however, the situation may be much
less clear. Thus, previous immunologic studies of the GBM using MCAP1
antibody, which recognizes the NC1 domain of
3 (IV), gave conflicting
results: normal labeling in three patients and no GBM labeling in two others
(24,25).
Consistent with the findings of other investigators
(11,12,18),
we have not found convincing evidence of a correlation between the presence
and severity of certain features of NPS and the type or location of the
LMX1B mutation. The absence of a genotypephenotype correlation
is not surprising given that there is not only interfamilial but also
intrafamilial variability in the expression and severity of symptoms. Family 8
is a very good example of intrafamilial phenotypic variability. It is
intriguing that a large percentage of the patients within this family have
nephropathy. Nevertheless, renal abnormalities are not present in all affected
individuals of this family; in addition, the severity of the nephropathy
varies among the patients, ranging from very mild proteinuria in nine patients
to end-stage renal disease in four of them. Patients from a family reported by
McIntosh et al. (12),
who carry the same splice-site mutation (672 +1G
A) as patients in family
8, do not have nephropathy. Therefore, the high incidence and the significant
difference in severity of nephropathy within family 8 cannot be attributed to
the LMX1B genotype. Recently, Farley et al.
(26), on the basis of genetic
studies in a very small family with NPS and quantified variable expression of
orthopaedic symptoms, found support for the old hypothesis of Renwick
(27), who proposed that the
severity of NPS symptoms is modulated by the allele contributed by the
unaffected parent. We believe that more genetic studies in clinically well
characterized large families with NPS are necessary to evaluate the possible
influence of modifying alleles on phenotypic expression and variability.
GBM abnormalities similar to those seen in NPS have been observed in several families without bone or nail abnormalities (28). It will be of interest to determine, by LMX1B mutation analysis, whether these cases represent partial expression of NPS or an independent genetic form of GBM disease. In case this NPS-like nephropathy turns out not to be due to LMX1B mutations, the possibility that genes encoding proteins that interact with LMX1B are involved should be investigated. We are performing yeast-two-hybrid analysis with the ultimate aim of identifying these genes.
| Acknowledgments |
|---|
| References |
|---|
|
|
|---|
3,
4, and
5
chains in rodent basal laminae: Sequence, distribution, association with
laminins, and developmental switches. J Cell Biol127
: 879-891,1994This article has been cited by other articles:
![]() |
L. Heidet, E. M. H. F. Bongers, M. Sich, S.-Y. Zhang, C. Loirat, A. Meyrier, M. Broyer, G. Landthaler, B. Faller, Y. Sado, et al. In Vivo Expression of Putative LMX1B Targets in Nail-Patella Syndrome Kidneys Am. J. Pathol., July 1, 2003; 163(1): 145 - 155. [Abstract] [Full Text] [PDF] |
||||
![]() |
E Sweeney, A Fryer, R Mountford, A Green, and I McIntosh Nail patella syndrome: a review of the phenotype aided by developmental biology J. Med. Genet., March 1, 2003; 40(3): 153 - 162. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |