Mineral particle size in children with osteogenesis imperfecta type I is not increased independently of specific collagen mutations.

Osteogenesis imperfecta (OI) type I represents the mildest form of OI and is usually caused by two classes of autosomal dominant mutations in collagen type I: haploinsufficiency leading to a reduced quantity of structurally normal collagen (quantitative mutation), or sequence abnormalities generating structurally aberrant collagen chains (qualitative mutation). An abnormally high bone matrix mineralization has been observed in all OI cases investigated so far, independently of mutation type. This raises the question whether the increased amount of mineral is due to mineral particles growing to larger sizes or to a higher number of more densely packed particles. For this reason, we revisit the problem by investigating the mineral particle size in cancellous bone from two subsets of the previously analyzed biopsies (patient's age: 2-4.2 and 7.6-11years) comparing OI quantitative mutations (n=5), OI qualitative mutations (n=5) and controls (n=6). We used a combined small-angle X-ray scattering (SAXS) and wide-angle X-ray diffraction (WAXD) setup with a beam diameter of 10μm of synchrotron radiation, which allows the determination of mineral particle characteristics in 10μm thick sections at the same positions where the matrix mineralization density was previously determined. The thickness parameter of mineral particles (T) was obtained from SAXS data and the mineral volume fraction was calculated from the mean calcium content of the bone matrix determined by quantitative back-scattered electron imaging (qBEI). The combination of these two quantities allowed calculating the true particle width (W) of the plate-like mineral crystals. T was larger in the older than in the younger age-group independently of genotype (p<0.004) and was larger in the controls than in each OI group. The qBEI results showed that the mineral volume fraction increased from 32.45wt.% in controls to 36.44wt.% in both OI groups (corresponding to a 12% increase in relative terms). Combining these data, we find that also W was larger in the older than in the younger age-group (p<0.002), but stayed equal or smaller in both OI genotypes (controls: 2.3nm±0.04, OI qualitative: 2.2±0.05; OI quantitative 2.3±0.04, mean±SEM). A linear regression analysis even suggests a slower increase of W in qualitative OI as compared to quantitative OI and controls, where the particle sizes stayed similar at all ages. We thus conclude that the high mineral density in human OI is not due to increased particle size but rather to increased particle packing density. The lack of an observed difference between the two classes of mutations suggests the occurrence of a bone cell defect downstream of the collagen mutation.