All fractures in P1 and P2 before introduction of ther­apy were treated conservatively. The clinical, laboratory and X- ray data are summarized in Table 1. There was an extremely good response to the therapy. After 2 months of therapy the parents of both children noticed that they be­came more active, probably because of the decrease of pain in the bones. After 26 and 18 months of therapy, respectively, neither patient had incurred any additional fractures, despite normal physical activity. Serum alkaline phosphatase activity decreased, and OSI was within nor­mal range for the age.

Cultured skin fibroblasts of patients and control syn­thesized and secreted types I and III procollagens as shown in Fig. 1(A). However, in control fibroblasts, type III pro­collagen accounted for 5-7% of total collagen, whereas in the patients the levels of type III and type I procollagens were comparable. This may mean increased synthesis of type III or decreased synthesis of type I procollagen. In the type I OI, increased ratio of type III to type I procollagen usually results from half normal amount of type I pro­collagen caused by one "null" allele of COL1A1 gene encoding the proa1(I) chain, while the level of type III procollagen is normal [14]. However, collagen biosyn­thesis, measured by [³H]proline incorporation into proteins susceptible to the action of bacterial collagenase, in cells from both patients was comparable to the control (Fig. 2). When we loaded onto each slab the same amount of radio­activity incorporated into collagen, we noted a decreased level of type I collagen in OI fibroblasts with respect to control [Fig. 1(B)], while the level of type III collagen was markedly increased. If only normal collagen is produced by cells from our patients but in reduced amounts as in type I OI, the increased proportion of type III collagen may be a compensatory response to the low production of type I collagen. However, the control cells secreted about 50% of total collagen into the medium, whereas fibroblasts from the OI patients secreted less than the control by ap­proximately 19 and 28%, respectively (Fig. 3). In addi­tion, the amount of intracellular collagen increased in OI cells relative to the control fibroblasts. Secretion of non collagenous protein by the OI cells is also less efficient than by the control, which may suggest a more generalized transport defect in OI cells. However, in contrast to the control, the level of secreted collagen in OI medium is lower than that of non collagenous protein. This may indi­cate the presence of mutated collagen in cells and delayed secretion. The marker of mutated collagen is usually de­layed electrophoretic migration of procollagen and/or collagen chains caused by their over modification [4,5], although a normal migration does not exclude the presence of structurally abnormal collagen. In both patients migra­tion of secreted procollagen and collagen chains did not differ markedly from the control (Fig. 1). Faster migrating bands under a2(I) chains [Fig. 1(B)] were present in me­dium from both patients and control, but absent in the cell layers. To identify a molecular defect, analysis of DNA is needed.

 

 

 

Figure 1. Sodium dodecyl sulphate (SDS)/polyacrylamide gel electrophoresis (PAGE) of procollagen (A) and collagen (B), synthesized and secreted into medium by the control (C) and OI patients’ (P1, P2) fibroblasts. Procollagens were separated in reducing, and collagens in non reducing conditions. Migration of procollagen and collagen chains and of fibronectin (FN) are indicated.

 

 

Figure 2. Collagen biosynthesis in the control (C) and patients’ (P1, P2) fibroblasts. Results are shown as combined values for cell plus medium fractions. Each value represents the mean for four independent experiments (repeated at different times from different passages of cells). The results were submitted to statistical analysis using the Student’s t-test, accepting p<0.05 as significant.

 

 

 

 

Figure 3. Distribution of collagen and non collagenous proteins in medium and cell layer. Results are shown as percent of total collagen (i.e., cells plus medium). Dotted bars show collagen and clear bars non collagenous proteins. Each value represents the mean for four independent experiments (repeated at different times from different passages of cells). The results were submitted to statistical analysis using the Student’s t-test, accepting p<0.05 as significant.