We describe detailed phenotype and genetic analysis of three cases of RD. Two newborns were found to have homozygous mutations in ZMPSTE24
, while one stillborn male did not have any disease causing mutation in either ZMPSTE24
. Most of the features of our cases, such as IUGR, joint contractures, skin breakdown, micrognathia, and acro-osteolysis, are typical of RD. Histologically, skin findings, including parallel collagen bundles and an absence of elastic fibers, are also consistent with previously reported cases of RD confirmed to be due to ZMPSTE24
). However, butterfly thoracic 5 vertebra has not been previously noted. Also, although clavicles are often reported as hypoplastic or dysmorphic (2
), the bulbous appearance of the distal portion of the clavicles in RD500.3 is a novel observation.
Homozygous or compound heterozygous ZMPSTE24
mutations were first reported in 10 newborns with RD by Navarro et al.
in 2005 (4
). Since then, near all other cases of RD have also harbored null mutations in ZMPSTE24
). Interestingly, more than 50% of the patients with RD and ZMPSTE24
mutations harbored a homozygous c.1085dupT mutation (1
) while an additional 17% were compound heterozygotes involving a c.1085dupT mutation on one allele (4
). One of our patients had homozygous 1085dupT mutation while another had a novel homozygous c.1020G>A (p.Trp340X) mutation.
mutations also cause autosomal recessive mandibuloacral dysplasia (MAD; OMIM 248370 and 608612) (13
). In contrast to RD, MAD patients with ZMPSTE24
mutations harbor a null mutation on one allele and a missense mutation on the other allele (13
). We have previously documented, using a yeast halo assay, that null mutations have no ZMPSTE24 activity whereas the missense mutants have partial loss or nearly normal activity (13
). Thus, the variable manifestations of the two disorders could be explained by varying amounts of prelamin A accumulation.
Only eight cases of MAD due to ZMPSTE24
mutations have been reported thus far (13
). Both MAD and RD patients with ZMPSTE24
mutations manifest prematurity, micrognathia, small pinched nose, sparse or absent hair, enlarged fontanelles, dysplastic clavicles and acro-osteolysis (). However, MAD due to ZMPSTE24 deficiency is not lethal within the newborn period with death reported at ages 3, 28, and 37 years (13
). MAD patients do not present with IUGR, fixed facial expression, mouth in the “o
” position, or skin erosions and denudations (). Development of contractures and joint stiffness is also delayed in MAD patients.
Comparison of clinical features of previously published and our patients with restrictive dermopathy and mandibuloacral dysplasia due to ZMPSTE24 mutations.
Navarro et al.
) have reported heterozygous abnormal splice inducing mutations (c.IVS11+1G>A and c.1824C>T) in the LMNA
gene in two patients age 5 months and 6 months with presumed diagnosis of RD. However, both these mutations cause Hutchinson-Gilford progeria syndrome (HGPS) (20
) and therefore, these may have been misdiagnosed. More recently, a newborn with RD was reported to carry a heterozygous c.1821G>A, (p.Val607Val) LMNA
); however, whether this mutation induces an alternative splicing was not demonstrated.
The phenotype of the stillborn fetus was consistent with a diagnosis of RD except for the presence of tapering rather than rounded, bulbous digits. The lack of mutations in LMNA and ZMPSTE24 in this fetus suggests additional loci for RD. The presence of two heterozygous SNPs in ZMPSTE24 reduces the likelihood of a large deletion of one allele. Regardless, a heterozygous deletion or c.−1355G>C promoter region variant in ZMPSTE24, without a concomitant null mutation, are not sufficient to cause RD by themselves. Furthermore, we did not find any disease-causing nucleotide alterations in the proximal (approximately 2 kb) promoter regions of LMNA and ZMPSTE24. However, lack of RNA precludes us to determine if this patient harbored any homozygous cryptic intronic mutation in ZMPSTE24. We were also unable to exclude the possibility for small deletions in LMNA.
There is a significant risk of recurrence of RD in subsequent pregnancies. As illustrated in our cases, early diagnosis of RD is difficult given that most affected fetuses have an unremarkable prenatal course. Ultrasound findings of polyhydramnios, decreased fetal movements, and growth retardation are suggestive of RD, but are nonspecific and late findings. A small fixed open mouth on ultrasound is also a late finding (23
) and as such is not useful in prenatal genetic counseling. Skin biopsy at 20 weeks gestation has been attempted (24
), but failed to predict the presence of RD likely because fetal skin development is not complete by then. Thus, the best option for early prenatal diagnosis is genetic testing of DNA obtained via chorionic villus sampling or amniocentesis. This would be mostly done in families who have previously had newborns with RD. In one such case (2
), prenatal diagnosis revealed the presence of a homozygous ZMPSTE24
mutation; the pregnancy was subsequently terminated.
In conclusion, we report a novel and a previously reported homozygous null mutation in ZMPSTE24 in two newborns with RD. We also report a stillborn fetus with features of RD but no mutations in LMNA or ZMPSTE24, suggesting an additional locus may exist for RD.