Leri-Weill dyschondrosteosis (LWD) is a pseudoautosomal dominant skeletal dysplasia, characterised by mesomelic short stature and bilateral Madelung deformity (BMD). The LWD is associated with haploinsufficiency of the Short Stature Homeobox-containing gene (SHOX), located on the X- and Y-chromosomes. This thesis aimed to evaluate if SHOX had a significant role in human skeletal growth and development. The unique tissue resources utilised included staged foetal growth plate and growth plate and bone marrow fibroblasts (BMF) from LwD/SHOX haploinsufficient subjects. A unique genotype/phenotype analysis of a large cohort with BMD and family members was evaluated for genetic and clinical heterogeneity of BMD and LWD. Surgical and medical interventions in subjects with LWD were also assessed.
Two subjects with UND/SHOX haploinsufficiency underwent pysiolysis (Vickers procedure) due to wrist pain and limited wrist movement. Histology revealed disruption of physeal anatomy within the zone of dysplasia and a generalised abnormality in the remaining physis. In addition, hypertrophic osteoid with micro-enchondromata in the metaphysis suggested abnormal endochondral ossification. The Vickers ligament was also described. Physiolysis reduced wrist pain and improved wrist movements in the two subjects.
An investigation of SHOX immunohistochemical staining within skeletal tissue, growth plate (foetal and childhood) and bone marrow fibroblasts was undertaken. In foetal growth plate SHOX immunohistochemical staining was demonstrated from 12 weeks gestation. Initially, SHOX immunostaining was demonstrated in the reserve, proliferative and hypertrophic zones. From 28 weeks gestation, immunostaining decreased in the proliferative and hypertrophic zones. In childhood, immunostaining persisted in the reserve zone, decreased in the proliferative chondrocytes and became evident again in the hypertrophic zone. This was reflected in SHOX haploinsufficient growth plate. These results, in combination with the LWD histopathology results, suggested that SHOX may maintain the proliferative phenotype.
In bone marrow fibroblasts (BMF) SHOX immunostaining was demonstrated predominantly within the nucleus, supporting SHOX acting as a transcription regulator. In SHOX haploinsufficient BMF the number of cells with nuclear SHOX immunostaining was decreased, as was the density of SHOX immunostaining within each BMF.
By comparing the phenotype of 26 SHOX haploinsufficient subjects with 45 relatives and population standards, SHOX was demonstrated to be a major growth gene. Compared to siblings, the SHOX haploinsufficient cohort was 2.14 SDS (3.8 cm) shorter at birth, 2.1 SDS shorter through childhood and at final height females were 2.4 SDS (14.4 cm) shorter and males 0.8 SDS (5.3 cm) shorter. The family height analysis suggests that the effect of SHOX haploinsufficiency on growth may have been previously underestimated at birth and overestimated in males at final height. SHOX haploinsufficiency lead to short arms in 92%, bilateral Madelung deformity in 73% and short stature in 54%. Females were more severely affected than males.
The effect of twenty-four months of recombinant human growth hormone (rhGH) therapy on the stature and BMD of two females with LWD was evaluated. Both subjects showed an increase in height standard deviation score (SDS) and height velocity SDS. Subject 1 demonstrated a relative increase in arm-span and upper segment measurements with rhGH while subject 2 demonstrated a relative increase in lower limb length. Bone age advanced appropriately, no adverse events were observed, and there was no clinical deterioration in BMD.
Phenotypic comparisons of SHOX wild-type and SHOX haploinsufficient BMD cohorts confirmed the molecular heterogeneity of LWD and BMD. The similar appearance of BMD between the cohorts suggested a common aetiological basis for the deformity in both groups, independent of SHOX. The role of the Vickers ligament in the development of BMD was also highlighted. It remains unclear why there is a preponderance of females with BMD, although a higher oestrogen concentration in females is unlikely to be the sole reason.
Through evaluating the phenotypic features of the SHOX haploinsufficient and wild-type cohorts, the best predictor of an individual having a SHOX mutation was found to be the combination of short stature and BMD. As a screening tool for the presence of a SHOX mutation, short arm-span was the most sensitive phenotypic feature. When available, SHOX mutation analysis should be offered to BMD probands, all first-degree relatives and short individuals with short arms or stature disproportion.
This thesis adds significantly to the literature on the biological actions and clinical effects of the SHOX gene and protein. It also raises further questions and introduces potential areas of future research.