Date Published: April 06, 2017
Publisher: John Wiley and Sons Inc.
Author(s): Kristen Wigby, Stephen R. F. Twigg, Ryan Broderick, Katherine P. Davenport, Andrew O. M. Wilkie, Stephen W. Bickler, Marilyn C. Jones.
Curry–Jones syndrome (CJS) is a pattern of malformation that includes craniosynostosis, pre‐axial polysyndactyly, agenesis of the corpus callosum, cutaneous and gastrointestinal abnormalities. A recurrent, mosaic mutation of SMO (c.1234 C>T; p.Leu412Phe) causes CJS. This report describes the gastrointestinal and surgical findings in a baby with CJS who presented with abdominal obstruction and reviews the spectrum of gastrointestinal malformations in this rare disorder. A 41‐week, 4,165 g, female presented with craniosynostosis, pre‐axial polysyndactyly, and cutaneous findings consistent with a clinical diagnosis of CJS. The infant developed abdominal distension beginning on the second day of life. Surgical exploration revealed an intestinal malrotation for which she underwent a Ladd procedure. Multiple small nodules were found on the surface of the small and large bowel in addition to an apparent intestinal duplication that seemed to originate posterior to the pancreas. Histopathology of serosal nodules revealed bundles of smooth muscle with associated ganglion cells. Molecular analysis demonstrated the SMO c.1234 C>T mutation in varying amounts in affected skin (up to 35%) and intestinal hamartoma (26%). Gastrointestinal features including structural malformations, motility disorders, and upper GI bleeding are major causes of morbidity in CJS. Smooth muscle hamartomas are a recognized feature of children with CJS typically presenting with abdominal obstruction requiring surgical intervention. A somatic mutation in SMO likely accounts for the structural malformations and predisposition to form bowel hamartomas and myofibromas. The mutation burden in the involved tissues likely accounts for the variable manifestations.
Curry–Jones syndrome (CJS; OMIM #601707) is a pattern of malformation that includes craniosynostosis, pre‐axial polysyndactyly, agenesis of the corpus callosum, cutaneous and gastrointestinal abnormalities (Temple et al., 1995). The first two cases were presented as unknown syndromes at the David W. Smith workshop on malformations in 1987 (Curry, 1987). Cohen (1988) recognized the cases as a pattern of malformation and termed the CJS after the physicians who had described the first two cases. Eleven unrelated cases have been reported in the literature presenting with craniofacial abnormalities, patchy cutaneous findings, and polysyndactyly (Grange et al., 2008; Mingarelli, Mokini, Castriota‐Scanderbeg, & Dallapiccola, 19999; Temple et al., 1995; Thomas et al., 2006; Twigg et al., 2016).
The proband is a female infant born at 41 weeks EGA to an 18‐year‐old G1P0 Hispanic‐Samoan mother and non‐consanguineous 18‐year‐old Hispanic father. The mother had routine prenatal care and there were no significant exposures. Pregnancy was complicated by pyelectasis and absent cavum septum pellucidum detected on ultrasound during the second trimester fetal anatomy survey. Upon transfer of care at 35 weeks (after the family relocated), she was started on weekly non‐stress tests due to the presence of suspected fetal anomalies. A repeat ultrasound obtained at 39 3/7 weeks GA showed accelerated fetal growth (gestational age equivalents: biparietal diameter: 40 0/7 weeks, head circumference 40 6/7 weeks, abdominal circumference 39 6/7 weeks, femur length 37 1/7 weeks), and apparent fusion of the anterior lateral ventricles, concerning for alobar holoprosencephaly. The infant was delivered by cesarean section after a failed induction of labor with Apgars of 9 and 9 at 1 and 5 min, respectively. The infant was large for gestational age with a birth weight of 4,165 g (96th percentile, z‐score 1.88), length of 53.3 cm (98th percentile, z‐score 2.23), and occipitofrontal head circumference of 38.5 cm (∼100th percentile, z‐score 3.90) (WHO 0–2 years female growth chart).
Informed consent was obtained with an IRB‐approved consent form prior to participation in research. For analysis of SMO, we collected normal and affected skin samples as well as formalin‐fixed paraffin‐embedded (FFPE) sections of the intestinal smooth muscle hamartoma. Skin samples were placed in tissue culture dishes to isolate fibroblasts and keratinocytes. After 2 days of culture, fibroblasts were removed by trypsin digestion (and subsequently passaged twice before analysis), while the keratinocytes (which surround the skin sample) were harvested using a cell scraper. DNA was extracted from skin (including isolated fibroblasts and keratinocytes) by Proteinase K digestion followed by phenol/chloroform extraction and ethanol precipitation, while DNA from FFPE material (five 10 μM sections) was isolated using the QIAamp DNA FFPE Tissue kit (Qiagen). Cellular identities of keratinocytes and fibroblasts were confirmed with RT‐PCR analysis of expression of the two distinguishing FGFR2 isoforms (IIIb is only expressed in keratinocytes while IIIc is only expressed in fibroblasts, data not shown). The same mutant allele of SMO (c.1234C>T, p.Leu412Phe) found in other cases of CJS (Twigg et al., 2016) was identified. Figure 4 presents the distribution of mutant and wild‐type allele for affected and unaffected skin samples. The mutant allele was found in both affected dermis (represented by fibroblasts; T allele frequency 16.2%) and at a higher level in affected epidermis (represented by keratinocytes; T allele frequency 34.7%). The mutant T allele was present also in unaffected dermis and epidermis but at a lower level compared to affected skin. In the intestinal hamartoma the mutant allele was present in 25.6% of the sample.
Important insights into the clinical and molecular pathogenesis of CJS can be drawn from the present case. From a molecular standpoint, a recurrent mosaic mutation in SMO results in malformations of the gastrointestinal tract and skin. Analyses of affected skin and intestinal smooth muscle hamartoma show an increased frequency of the mutant SMO allele compared with unaffected tissues. In all affected tissues studied, the mutant allele is present at a frequency less than 50%, which reflects tissue mosaicism. Twigg et al. (2016) recently identified a mosaic gain of function mutation in SMO (c.1234C>T, p.Leu412Phe) in seven other cases with CJS in multiple affected tissues including skin, FFPE samples of medulloblastoma, intestinal myofibromas, or smooth muscle hamartomas. Within the skin, the mutant allele frequency varied with the phenotypic findings of the skin (hyperpigmented or affected compared with unaffected) but also with the layers of the skin. The factors that drive a higher mutant allele frequency within the epidermis are unknown. In addition, detection of the mutant allele within apparently unaffected skin at a low level suggests that the mutation arose early in skin development. Indeed, the presence of mosaicism as well as detection of the mutant allele in multiple different tissue types suggests that the mutation occurred as a post‐zygotic error early in embryonic development.