Genetic disorders of
bone and extracellular matrix
Joan
C. Marini, MD, PhD, Head, Section on
Connective Tissue Disorders Armando
Flor, MD, Clinical Associate Aarthi
Ashok, PhD, Postdoctoral Fellow Thomas
Uveges, PhD, Postdoctoral Fellow Antonella
Forlino, PhD, Contractor Anne
Letocha, MSN, CRNP, Senior Research
Assistant Wayne
A. Cabral, AB, Chemist Aileen
M. Barnes, MS, Research Associate Sarah Milgrom, BS, Postbaccalaureate Fellow |
|
The
OI/EDS region of the alpha1(I) collagen chain Cabral, Letocha, Marini; in collaboration
with Leikin A distinct subset of osteogenesis imperfecta
patients are those with OI/EDS. In addition to the skeletal fragility of OI,
they have characteristics of EDS, including severe laxity of large and small
joints and early-onset scoliosis. In seven children with OI/EDS, we
delineated mutations in the first 90 residues of the helical region of the
alpha1(I) chain. We determined that these collagen mutations cause abnormal
N-propeptide processing, incorporation of pN-collagen into matrix, and
decreased diameter of dermal fibrils. These data provide a mechanism for the
children’s EDS symptoms while the helical changes per se are
responsible for bone fragility. Thus, the mechanism of the subject
children’s EDS is shared with patients with EDS VIIA and B attributable
to the absence of the N-proteinase cleavage site from the alpha1(I) or
alpha2(I) chain, respectively. We
identified seven mutations by direct sequencing of RT-PCR amplification
products of alpha1(I) mRNA. In contrast to normal procollagen and collagen,
which have an identical melting peak, the thermal stabilities of each of the
mutant collagens differed from those of the corresponding procollagens. Our
observations also stand in contrast to collagen mutations beyond the first 90
amino acids; their differential scanning calorimetry tracings usually show
both normal and lower stability peaks, but the melting curves of procollagen
and collagen are identical for each mutation. In vitro cleavage with N-proteinase processed only 25 percent of
proband pro-alpha1(I) chains for exon 7 mutations and 65 to 85 percent of
pro-alpha1(I) chains for exon 8-11 mutations. The pericelluar processing of
the seven mutants was also delayed. The pN-collagen is incorporated into
matrix deposited by cultured fibroblasts, with pN-alpha1(I) collagen
prominently present in the newly incorporated and immaturely cross-linked
fractions. Electron microscopy of dermal fibrils of six patients revealed
that fibril diameters of all six were significantly smaller than those of
matched controls, as seen in EDS VII. The above assays defined a folding region of
alpha1(I) in which mutations cause a distinct OI/ED phenotype by altering the
triple helical structure and secondary structure of the N-proteinase cleavage
site. The retention of N-propeptide in a substantial proportion of collagen
chains limits fibril diameter. The abnormal fibrils may cause laxity of
joints and paraspinal ligaments either directly by reduced resistance to
shearing forces or indirectly by altering interactions between collagen and
other matrix components in the overlap zone of the D-periods. Type
I collagen C-propeptide mutations Marini, Barnes, Ashok Mutations in the C-propeptide of type I
collagen have been found in a small number of patients with OI. The phenotype
of these patients ranges from lethal to moderately severe. The mutations are
of special interest because they are located in a region that is cleaved from
procollagen before collagen fibril assembly. Therefore, the mutations per se
are not expected to be present in collagen fibrils in tissues. The
implication is that the pathophysiological mechanism of these mutations must
differ from that associated with mutations in the helical region of the alpha
chains; the latter mutations are incorporated into matrix and exert a
dominant negative effect. In the collagen of children with types III and
IV OI, we identified four novel C-propeptide mutations at conserved residues.
All mutations delayed incorporation of alpha1 chains into heterotrimers, with
delay ranging from two- to six-fold the chain incorporation time of normal
control collagen. A pericellular processing assay suggests a delay in C-propeptide
removal from secreted collagens containing these mutations. Mutant collagens
are incorporated into fibroblast matrix in culture and form mature
cross-links. Using immunofluorescence assays, we
investigated the intracellular interaction of the mutant procollagen
molecules with ER chaperones in OI fibroblasts by comparing the behavior of
normal control molecules with that of molecules with a C-terminal helical
mutation. Our results demonstrate a clear correlation between the
presence/type of mutation and the subcellular localization pattern of
procollagen. Alendronate
treatment of Brtl mouse Marini, Uveges; in collaboration with
Goldstein, Gronowicz Bisphosphonate drugs are widely administered
to children with OI, but the drugs’ effects on OI bone containing
abnormal type I collagen have not been directly examined. The Brtl mouse
model for type IV OI has a glycine substitution (G349C) knocked into one
COL1A1 allele. We treated Brtl and wild-type offspring of Brtl x CD-1 matings
from two to 14 weeks of age with either alendronate (0.219 mg/kg/wk, gift of
Merck) or saline placebo. Brtl mouse weight and femor length were
significantly smaller than in wild-type mice and were unchanged by
alendronate. Whole-bone density of femurs and lumbar
vertebrae, measured by using a Lunar Piximus, were significantly increased in
both treated Brtl and wild-type mice; treated Brtl samples attained the
average untreated wild-type bone mineral density (BMD). We tested the mechanical properties of femurs
in four-point bending. Alendronate treatment increased femoral stiffness and
decreased pre-yield displacement in wild-type mice, indicating that the
treatment is not benign for normal bone. Stiffness, pre-yield displacement,
and yield load were unchanged in treated Brtl femora. In both genotypes, we
observed an increase in the ultimate load at which femurs fracture. Unfortunately,
alendronate has a negative impact on several aspects of bone quality. First,
treatment reduces the predicted material strength and modulus of Brtl and
wild-type bone. Second, the brittleness (post-yield displacement) of treated
Brtl femurs did not improve; post-yield displacement decreased further in
treated Brtl as compared with untreated wild-type mice. After the yield point
is reached, Brtl femurs fracture with similar additional load and deformation
as if untreated. Third, the metaphyses of treated Brtl femurs exhibit
increased remnants of mineralized cartilage. The matrix discontinuities
caused by the presence of mineralized cartilage in the bone may increase the
risk of fracture initiation and account for some of the observed increase in BMD.
Fourth, we observed a detrimental effect on bone cells. After 12 weeks of
alendronate treatment, bone formation rates (BFR/BS), mineral apposition rate
(MAR), and mineralized surface (MS/BS) decline to less than 25 percent of
pretreatment values in Brtl and wild-type mice. At that time, the percent of
osteoblast surface is significantly lower in both genotypes while the percent
of osteoclast surface remains stable. In addition, the morphology of the Brtl
osteoblasts changes from the plump cuboidal osteoblasts seen in untreated
femurs to an intermediate morphology, evidence of a toxic effect on the
cells. Our interpretation is that alendronate treatment of Brtl improves bone
geometry and increases loading before fracture but decreases predicted bone
material quality and alters osteoblast surface and morphology. The data
suggest that limited treatment duration may be optimal for obtaining improved
bone geometry and minimizing the detrimental effect of extended treatment on
bone quality. Pamidronate
treatment of children with types III and IV OI Letocha, Marini; in collaboration with
Gerber, Paul Uncontrolled trials of bisphosphonates in OI
children report increased vertebral bone density and height, improved
strength and functional level, and decreased fractures and bone pain. We
undertook a randomized controlled trial of pamidronate in children with Types
III and IV OI. The first study year was controlled; children in the treatment
group received pamidronate (10 mg/m2/day for three days every three
months); children in both treatment and control groups underwent quarterly
rehabilitation and physical therapy assessments, including measurements of
function, strength, and pain. Children in the treatment group received
pamidronate for an additional six to 21 months. All patients underwent L1-L4
DEXA, spine qCT, spine radiographs, and musculoskeletal and functional
testing. In the controlled phase,
treated patients experienced a significant increase in vertebral BMD z-score
as compared with the controls. They also had significant increases in L1-L4
mid-vertebral height and total vertebral area as compared with the controls.
The treatment group did not experience decreased long bone fractures. In the
extended treatment phase, DEXA z-scores and vertebral heights and areas did
not increase significantly beyond the 12-month value. In the context of maximized
physical rehabilitation, we did not see an additional functional effect from
bisphosphonate treatment. In contrast to reports from uncontrolled trials, we
found no significant changes in ambulation level, lower extremity strength,
or pain in OI children treated with pamidronate. We assessed motor skills
related to ambulation with the 10-point Brief Assessment of Motor Function
(BAMF). At initiation, the BAMF of the treatment group was 6.1±1.8 versus
6.7±1.9 at 12 months. At initiation, the BAMF of the control group was 6.6±2
versus 7.02±1.31 at 12 months. Manual muscle testing was assessed as the sum
(total points 110) of abdominal, straight leg raise, hip abduction,
extension, and flexion, and quadriceps strength. Lower extremity muscle
strength did not change. There was no significant decrease in pain on a
four-point scale. Some patients reported increased endurance or decreased
back pain, but most reported no perceptible changes. The previously reported
changes in these measures appear to have been placebo effects in the
uncontrolled trials. Interestingly, the response
within the treatment group showed considerable variety. Some children had a
robust response in all measurements; others had increased bone density, but
not increased area or height. Changes in DEXA z-scores ranged from less than
1 SD to more than 3 SD. The variability of response has not been previously
reported and is presumably related to differences in bone matrix caused by
the underlying collagen mutations. Chernoff E, Letocha AD,
Marini JC. Osteogenesis imperfecta. In: Allanson J, Cassidy S, eds. Clinical Management of Common Genetic
Syndromes. Kozloff KM, Carden A, Berwitz
C, Forlino A, Uveges TE, Morris MD, Marini JC, Goldstein SA. Brittle IV mouse
model for osteogenesis imperfecta IV demonstrates post pubertal adaptations
to improve whole bone strength. J Bone
Min Res 2004;19:614-622. Kuznetsova NV, Forlino A,
Cabral WA, Marini JC, Leikin S. Structure, stability and interactions of type
I collagen with Gly 349 Cys substitution in alpha1(I) chain in a murine
osteogenesis imperfecta model. Matrix
Biol 2004;23:101-112. Walker LC, Overstreet MA,
Willing MC, Marini JC, Cabral WA, Pals G, Bristow J, Atsawasuwan P, Yamauchi
M, Yeowell HN. Heterogeneous basis of the type VIB form of Ehlers-Danlos
syndrome (EDS VIB) that is unrelated to decreased collagen lysyl
hydroxylation. Am J Med Genet
2004;131A:155-162. COLLABORATORS Antonella Forlino,
PhD, Lynn Gerber, MD, Rehabilitation Medicine, NIH Clinical
Center, Steven Goldstein,
PhD, Gloria Gronowicz, PhD, University of Sergey Leikin,
PhD, Unit on Molecular Forces and
Structure, NICHD, Scott Paul, MD, Rehabilitation Medicine, NIH Clinical
Center, For further
information, contact mailto:marinij@mail.nih.gov |