Twenty-five fetuses with limb body wall complex (LBW complex) were evaluated. The diagnosis was based on two out of three of the following: exencephaly/encephalocele with facial clefts; thoraco- and/or abdominoschisis; and limb defect. Ninety-five percent (24/25) of the fetuses had associated internal structural defects. In 72% (18/25) the internal defects have been recognized as being secondary to vascular disruption. Concordance was not found between the side and location of the body wall defect versus the limb, internal, and cranial defects. In 85% there was evidence for persistence of the extraembryonic coelom by examination of the placenta. In this same group (85%) there was persistence of the ectodermal-amnion margin, with the amnion being continuous with the skin of the body wall defect. In 40% (10/25) there were tags and amniotic adhesions at other sites. There was no difference in the types or incidence of internal defects between those with and those without amniotic bands. The abnormalities in this collection and experimental animal models support vascular disruption during 4-6 weeks' gestation as an etiology for LBW complex. There is disruption and loss of existing tissues, persistence of embryonic structures, and secondary malformations. Persistence of the extraembryonic coelom may lead to the typical amniotic tags, ring constrictions, and adhesions seen in some specimens.

Osteogenesis imperfecta is a clinically and genetically heterogeneous brittle bone disorder that results from defects in the synthesis, structure, or posttranslational modification of type I procollagen. Dominant forms of OI result from mutations in COL1A1 or COL1A2, which encode the chains of the type I procollagen heterotrimer. The mildest form of OI typically results from diminished synthesis of structurally normal type I procollagen, whereas moderately severe to lethal forms of OI usually result from structural defects in one of the type I procollagen chains. Recessively inherited OI, usually phenotypically severe, has recently been shown to result from defects in the prolyl-3-hydroxylase complex that lead to the absence of a single 3-hydroxyproline at residue 986 of the alpha1(I) triple helical domain. We studied a cohort of five consanguineous Turkish families, originating from the Black Sea region of Turkey, with moderately severe recessively inherited OI and identified a novel locus for OI on chromosome 17. In these families, and in a Mexican-American family, homozygosity for mutations in FKBP10, which encodes FKBP65, a chaperone that participates in type I procollagen folding, was identified. Further, we determined that FKBP10 mutations affect type I procollagen secretion. These findings identify a previously unrecognized mechanism in the pathogenesis of OI.

On page 555 under the section titled Mutations in FKBP10 cause Recessive OI, there are two errors in the nomenclature for the identified mutations. The FKBP10 (NM_021939.3) mutation isolated in the Turkish cases (proband R06-113A) is c.321_353 del and is predicted to result in the deletion of eleven amino acids in the protein, p.Met107_Leu117 del. In the second paragraph of the subheading, the mutation in the Mexican-American family (proband R93-188) should be

Limb defects from 25 fetuses with limb-body wall (LBW) complex were evaluated to determine the mechanism of limb damage. The limb defects could be divided into 3 pathogenetic groups: (1) secondary to disruption of embryonic vessels and surrounding tissue (84%), (2) secondary to amniotic bands or adhesions (16%), and (3) deformation versus hemorrhage (44% with club feet), with some fetuses having more than one pathogenetic mechanism causing limb defects. The hypothesis that the majority of limb defects resulted from disruption of embryonic vessels was supported by the following findings: 96% of the LBW complex fetuses had limb defects; the lower limbs were at greater risk of damage than the upper limbs (28% rt arm, 52% lt arm, 60% rt leg, 72% lt leg); there was a distal to proximal progression of limb damage in 92% of the fetuses; statistical analysis of comparing the location of the most severe limb defect and the body wall defect did not find concordance between the side (p = 1.0) and the region (p = 0.18) of the body wall defect; and limb defects found in the human specimens were similar to those produced in experimental animals following disruption of embryonic vessels at a corresponding gestation. In the specimens with amniotic band related limb defects (16%), the most likely pathogenesis is mechanical rupture through the amnion in the presence of a persistent extraembryonic coelom or from adhesion of the amnion to necrotic embryonic tissue after the initial disruptive event. Club feet were present in 44% and may be due either to disruption of embryonic vessels or to deformation. Further studies are needed to resolve this question.

Vertebrates have four clusters of Hox genes (HoxA, HoxB, HoxC, and HoxD). A variety of expression and mutation studies indicate that posterior members of the HoxA and HoxD clusters play an important role in vertebrate limb development. In humans, mutations in HOXD13 have been associated with type II syndactyly or synpolydactyly, and, in HOXA13, with hand-foot-genital syndrome. We have investigated two unrelated children with a previously unreported pattern of severe developmental defects on the anterior-posterior (a-p) limb axis and in the genitalia, consisting of a single bone in the zeugopod, either monodactyly or oligodactyly in the autopod of all four limbs, and penoscrotal hypoplasia. Both children are heterozygous for a deletion that eliminates at least eight (HOXD3-HOXD13) of the nine genes in the HOXD cluster. We propose that the patients' phenotypes are due in part to haploinsufficiency for HOXD-cluster genes. This hypothesis is supported by the expression patterns of these genes in early vertebrate embryos. However, the involvement of additional genes in the region could explain the discordance, in severity, between these human phenotypes and the milder, non-polarized phenotypes present in mice hemizygous for HoxD cluster genes. These cases represent the first reported examples of deficiencies for an entire Hox cluster in vertebrates and suggest that the diploid dose of human HOXD genes is crucial for normal growth and patterning of the limbs along the anterior-posterior axis.