© 1999 Oxford University Press

Human Molecular Genetics, 1999, Vol. 8, No. 10 Review 1839–1846

Osteopetrosis and osteoporosis: two sides of the same coin Francesca Lazner, Maxine Gowen1, Durda Pavasovic and Ismail Kola+ Centre for Functional Genomics and Human Disease, Institute of Reproduction and Development, Monash University, Monash Medical Centre, 246 Clayton Road, Clayton, Victoria 3168, Australia and 1Department of Bone and Cartilage Biology, SmithKline Beecham Pharmaceuticals, King of Prussia, PA 19406, USA Received April 20, 1999; Accepted May 13, 1999

Together, osteoporosis and osteopetrosis comprise a substantial proportion of the bone diseases that severely affect humans. In order to understand and effectively treat these disorders, an understanding of the mechanisms controlling bone remodelling is essential. While numerous animal models of bone disease have been generated, the lack of correlation between these animal models and human disease has limited their utility in terms of defining therapeutic strategies. The generation and analysis of cathepsin K knockout mice has resulted in a model for pycnodysostosis, a rare human osteopetrotic disease, and is now providing considerable insights into both osteoclast function and potential therapeutic strategies for the treatment of bone disease. This review highlights the importance of genes such as cathepsin K in understanding bone remodelling and illustrates a new trend towards understanding bone disease as a complete entity rather than as a series of unrelated disorders.

INTRODUCTION Bone development and homeostasis is a complex process in which a balance between bone formation and resorption is delicately maintained. The major effector cells of bone formation and resorption are the osteoblasts and osteoclasts. These cells and their precursors are regulated by a vast array of autocrine and paracrine factors. Perturbation of the balance between bone formation and resorption can result in the formation of either too much bone (osteopetrosis) or too little bone (osteoporosis). The term osteopetrosis in humans is used to define a number of distinct diseases that can be classified on the basis of severity and age of onset into three major groups: infantile malignant osteopetroses, intermediate mild osteopetroses and adult onset osteopetroses. Similarly, the term osteoporosis defines a group of (apparently) aetiologically distinct diseases. The numerous forms of both osteopetrosis and osteoporosis in humans are shown in Table 1. While osteopetrotic diseases are relatively rare in humans, the prevalence of osteoporosis, some estimates being as high as 40% in Caucasian women over the age of 80 (1), makes bone disease one of the major health issues of present day society. UNDERSTANDING THE BASIS OF HUMAN DISEASE The key to understanding the basis of each of these bone diseases is to determine whether the primary defect resides in the osteoblast or osteoclast lineage or indeed whether the skeletal phenotype arises as a consequence of other immunological and/or endocrinological defects. Osteoclasts, as macrophage+To

like cells, are derived from the haematopoietic compartment by a differentiation pathway that diverges at a late stage from that of the monocyte/macrophage lineage (2–5). Thus, mutations that alter the differentiation potential of haematopoietic stem cells or precursor cells often result in skeletal abnormalities. Bone marrow transplantation is the most direct method used to determine where the primary defect resides in osteopetrotic diseases. This technique has provided considerable information regarding the lineage responsible for numerous osteopetrotic diseases (4,6–9) and has also been of therapeutic value in the treatment of some human osteoclast-related disorders. For example, bone marrow transplantation has been used in a number of individuals with infantile malignant osteopetrosis (10–12), but the results have been variable; some individuals have responded to this treatment with a reversal (complete or partial) of the skeletal phenotype, whilst others have shown no response. These and other studies (13–15) suggest, firstly, that the aetiology of disease within this group of phenotypically similar individuals may not be the same. In accord with these findings, morphological investigation of osteoclasts derived from individuals with infantile malignant osteopetrosis show marked heterogeneity with individuals having increased, decreased or normal numbers of osteoclasts, increased volume or number of nuclei and/or lack of ruffled borders and clear zones (13). Secondly, whilst defective cross-talk between cell types is likely, the observation of defective osteoblasts as well as osteoclasts in two cases of human malignant osteopetrosis (15) for which bone marrow transplantation was successful raises the possibility of a multigenic aetiology (15) for this disorder. The difficulties associated with bone marrow transplan-

whom correspondence should be addressed. Tel: +61 3 9594 7202; Fax: +61 3 9594 7212; Email: [email protected]

1840 Human Molecular Genetics, 1999, Vol. 8, No. 10 Review

Table 1. Various forms of osteopetrosis and osteoporosis that affect humans Type

Subtype

Cause

Osteopetrosis Infantile malignant Intermediate mild

Autosomal dominant

Responsive to bone marrow transplantation

Haematopoietic origin

Non-responsive to bone marrow transplantation

Non-haematopoietic origin

Carbonic anhydrase II deficiency

Mutation in carbonic anhydrase gene

Pycnodysostosis

Mutation in cathepsin K gene

Other

Unknown

Type I, uniform increase in bone density

Unknown

Type II, non-uniform increase in bone denisty

Unknown

Post-menopausal osteoporosis

Loss of sex steroids either directly or indirectly

Senile osteoporosis

Unknown

Endocrine disorders

Hyperparathyroidism

Osteoporosis Primary Secondary

Cushings syndrome Diabetes Pregnancy and lactation Hyperthyroidism Hypogonadism Genetic disorders

Osteogenesis imperfecta Menkes syndrome Homocysteinuria Adult hypophosphatasia

Neoplastic disorders

Multiple myeloma Systemic mastocytosis Waldenström macroglobulinaemia

Nutritional causes

Intestinal malabsorption Protein malnutrition Scurvy

Drug related

Heparin Anticonvulsants

Other

Immobilization Primary biliary cirrhosis Rheumatoid arthritis Chronic pulmonary disease

tation and the variability of success rates, however, limit its use as a therapeutic strategy for the majority of osteopetrotic diseases. No information exists on the value of bone marrow transplantation as a therapeutic strategy in osteoporosis. Our understanding of human osteopetroses has been greatly aided by the significant number of animal models. In addition to providing clues to the aetiology of this disease in humans, a number of animal models have also provided considerable insight both into the mechanisms of bone formation and resorption and into the ontogeny of the osteoclast. Finding the exact source responsible for the genesis of osteoporosis has proved more difficult. Genetic, environmental and lifestyle factors have all been associated with the pathogenesis of osteoporosis (1). Despite the abundance of literature on the subject, even the definition of what constitutes primary

osteoporosis remains controversial (1). According to the World Health Organisation, osteoporosis is defined as bone mineral density (BMD) (in the spine) >2.5 standard deviations below the ‘young normal’ mean value (16). The inadequacy of this definition is highlighted by evidence that for a given BMD the risk of fracture increases with age (17). The second factor that has hampered the study of osteoporosis is the scarcity of appropriate animal models. One of the most striking features of the ‘bone disease field’ is the lack of continuity between studies into osteopetrosis and osteoporosis. These two types of disorder affect the same system, the skeletal system, with virtually opposing effects and yet past and present literature (with few exceptions) has failed to explore any overlap. This review will detail a number of animal models of bone disease which have provided considerable

Human Molecular Genetics, 1999, Vol. 8, No. 10 Review 1841

Table 2. Summary of the major phenotypic features of a number of spontaneous mutant strains of mice which display an osteopetrotic phenotype

Lifespan

op/op osteopetrotic gl/gl grey lethal

mi/mi microphthalmic

oc/oc osteosclerotic

Reduced

Variable