 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
The Journal of Bone and Joint Surgery (American). 2005;87:2550-2564. doi:10.2106/JBJS.E.00019 © 2005 The Journal of Bone and Joint Surgery, Inc.
Molecular Biology in Orthopaedics: The Advent of Molecular Orthopaedics
1 Center for Molecular Orthopaedics, Harvard Medical School, 221 Long-wood
Avenue, BLI-152, Boston, MA 02115. E-mail address:
cevans{at}rics.bwh.harvard.edu
The workshop on which this review article is based was supported, in part,
by an R13 award from the National Institute of Arthritis and Musculoskeletal
and Skin Diseases; National Institutes of Health conference grant; the
American Academy of Orthopaedic Surgeons; the Orthopaedic Research and
Educational Foundation; Merck, Inc.; Genetics Institute, Inc.; Immunex Corp.;
Millenium Pharmaceuticals; and Pfizer Pharmaceuticals. In addition, C.H. Evans
received payments or other benefits or a commitment or agreement to provide
such benefits from a commercial entity. (He is a member of the Scientific
Advisory Board of TissueGene, Inc., and Orthogen, AG. He also received
research support from TissueGene, Inc; Orthogen, AG; and Osiris, Inc.) No
commercial entity paid or directed, or agreed to pay or direct, any benefits
to any research fund, foundation, educational institution, or other charitable
or nonprofit organization with which the authors are affiliated or
associated.
Molecular biology is the study, at the molecular level, of how genetic information is stored, inherited, and expressed and how it influences the structure and function of cells. Although molecular biology approaches have been used for decades in orthopaedic research, they are only now beginning to influence clinical practice. A variety of sophisticated techniques permit rapid and affordable DNA sequencing, gene expression profiling, gene cloning, gene manipulation, gene transfer, recombinant protein production, and other technologies of enormous biomedical importance. Success in genomics has spawned additional ambitious endeavors, including proteomics, pharmacogenetics, and bioinformatics. These techniques are providing new diagnostic, staging, prognostic, and therapeutic opportunities in all areas of medicine, including orthopaedics. With the use of molecular criteria, treatment of the orthopaedic patient may become more individualized, and greater emphasis will be placed on preventative strategies based on the patient's genetic makeup. Both surgical and nonsurgical decisions will increasingly accommodate molecular criteria.
Depending on when you start counting, the field of molecular biology is approximately fifty years old1. For much of this time, it has been an esoteric pursuit by laboratory scientists interested in determining the molecular basis of life. In recent years, however, this has dramatically changed and molecular biology has entered mainstream medicine2. Molecular screening for disease-related genetic polymorphisms, molecular diagnostics, and recombinant proteins as drugs are already available; gene therapy and other molecular therapies lie ahead. Some future applications, such as germ-line gene therapy3 and genetic profiling4, have raised enormous ethical issues. Although molecular biology has been used for some time as a tool in orthopaedic research, the practice of orthopaedic surgery has been little affected. This is about to change. The first evidence of this is the utilization of recombinant bone morphogenetic proteins (BMPs), recently approved for use to enhance long-bone healing5 and spinal fusion6. Much more lies ahead, and molecular biology will dramatically alter the way in which orthopaedists interact with their patients. Anticipating the arrival of molecular orthopaedics, the American Academy of Orthopaedic Surgeons and the National Institute of Arthritis and Musculoskeletal and Skin Diseases recently held a workshop called Molecular Biology in Orthopaedics, at which experts in the field defined the state of the art, identified problems and research questions, and mapped out a research agenda. This Current Concepts Review is based on the conclusions of that workshop. The complete transactions of the workshop are available in book form7.
As Lehninger8 commented, "Living things are composed of lifeless molecules." In this sense, practically any aspect of subcellular biology can be considered molecular. The term molecular biology, however, is usually reserved for a component of biology that might more accurately be called molecular genetics—that is, the study, at the molecular level, of how genetic information is stored, inherited, and expressed and how it influences the structure and function of cells. However, as reflected in this review, in a medical setting molecular biology needs to be discussed in the broader context of the biochemistry and biology of the organism. Deoxyribonucleic acid (DNA) serves as the repository of genetic information in all but a few specialized instances, such as certain viruses that use ribonucleic acid (RNA) instead. A molecule of DNA comprises two nucleotide chains that combine in an anti-parallel fashion to form the well-known double helix, held together by hydrogen bonding between purine (adenine [A] and guanine [G]) and pyrimidine (thymine [T] and cytosine [C]) bases that project from the sugar-phosphate backbone. The two strands separate ("melt") under certain conditions (e.g., warming) and reanneal with precision when the constraint is removed (e.g., with cooling). This ability of two nucleotide chains to hybridize reversibly in a highly specific manner is a very important property that is the basis of many fundamental techniques in molecular biology. In eukaryotic cells, nearly all of the DNA is contained within the nucleus, where it exists in the form of chromosomes. There are forty-six chromosomes in every nucleated, human cell: twenty-two pairs of somatic chromosomes and two sex chromosomes. Each chromosome is a single, enormously long molecule of DNA. Chromosomes are numbered according to their size, with chromosome 1, the largest, containing approximately 250 million base pairs of DNA and chromosome 22, the smallest, containing approximately 40 million base pairs. A gene is a sequence of DNA that gives rise to (codes for) the synthesis of a specific molecule of RNA or protein. Data from the recently completed Human Genome Project suggest that there are about 20,000 to 25,000 genes contained within the DNA of humans (the human genome)—far less than was originally estimated9. Often overlooked is the fact that >98% of human DNA does not encode an RNA or protein product. In fact, at least 50% of this DNA consists of short, repeating sequences. This is often dismissed as "junk DNA," but recent research indicates that this DNA may have important roles in organizing the genome and regulating gene expression10. Each time that a cell undergoes mitosis, its DNA is replicated by DNA polymerases working in conjunction with a number of additional proteins in a structure known as a synthesome11. The first step in gene expression is the synthesis of RNA molecules. Like DNA, RNA molecules comprise a sugarphosphate backbone with bases attached, but RNA contains uracil (U) instead of thymine. Moreover, RNA molecules are largely single-stranded, although important short regions of intermolecular and intramolecular double strands, and other higher-order structures, can form. RNA molecules are synthesized by a process of transcription during which the DNA double helix locally unwinds, and an enzyme, RNA polymerase, makes copies of one strand of the DNA. Three main classes of RNA are synthesized in this way. Approximately 80% of the newly synthesized RNA is ribosomal (r) RNA, a major constituent of the ribosomes. A second form of RNA, transfer (t) RNA, is used to transfer amino acids to the ribosome for protein synthesis. The final form, messenger (m) RNA, gives rise to protein. Although mRNA accounts for only a small percentage of cellular RNA, it has received the most attention because it leads to the synthesis of specific proteins. Each gene that gives rise to mRNA is responsible for the synthesis of one, or a limited number of, related proteins. The genetic information responsible for the specificity of protein synthesis resides in the precise sequence of purine and pyrimidine bases contained within the gene. One way in which several different proteins can be made from the same gene reflects the fact that the genes of eukaryotes (organisms whose cells have a membrane-bound nucleus), unlike those of prokaryotes (organisms, such as bacteria, whose cells lack a membrane-bound nucleus), are discontinuous, with the coding sequences of the genes (exons) interrupted by non-coding sequences (introns). In some cases, the non-coding sequences can be very large. Dystrophin, the longest human gene that we know of, has more than 2.4 million base pairs, but only 0.5% code for protein; the rest comprises seventy-nine introns. After the gene has been transcribed into a complementary RNA copy, this transcript undergoes splicing to remove introns. Different exons from the same gene can be spliced together, yielding different mRNA molecules and, eventually, different proteins. Splicing and other chemical changes to the RNA are known collectively as RNA processing. Processed mRNA molecules leave the nucleus and engage the ribosomes, where protein synthesis occurs by a process known as translation. During translation, the nucleotide sequence within each mRNA molecule is read sequentially. Each run of three nucleotides, known as a codon, specifies the insertion of a particular amino acid into the growing peptide chain. A single initiation codon, AUG (which codes for methionine), begins the process of translation. Other codons, stop codons (UGA, UAA, and UAG), terminate elongation of the protein chain, which is then released from the ribosome. The amino acid specificity of each codon is embodied in the genetic code. Nascent protein molecules may undergo a variety of posttranslational modifications, including specific cleavage by proteinases; the covalent addition of various moieties such as phosphate, lipid, and sugar molecules; as well as additional chemical modification of certain amino acids. Thus, by a combination of mRNA splicing and posttranslational modification, a single gene may give rise to several different proteins. The identification of the entire complement of proteins synthesized by the body (the proteome) is the aim of a new discipline called proteomics12, so named because it is analogous to the field of genomics, which aims to identify and understand the body's entire repertoire of genes (the genome). Other neologisms include the transcriptome (the repertoire of RNA transcripts) and the metabolome (all of the metabolic activities of cells). There will doubtless be others in this expanding world of "-omics."
With the exception of specific cells, such as lymphocytes, that undergo genetic rearrangements, all nucleated cells in the body have essentially the same genome. They achieve and maintain their individuality by selectively expressing different subsets of genes from this genetic repertoire. In some cases, this can be very specific. The gene encoding the  -chains of
type-II collagen (Col2A1), for example, is expressed almost exclusively in
chondrocytes. Other genes, such as those encoding the cytoskeletal protein
actin, are expressed in almost all cells. Puzas et
al.13 coined the
term orthopaedic genome to signify the component of the genome of
relevance to orthopaedics. Gene expression can also vary temporally. For
example, certain genes are switched on or off when the cell divides or is
exposed to inflammation or injury. The expression of other genes is not
greatly affected by such influences, and those genes are often referred to as
housekeeping genes. These genes play important roles in maintaining
homeostatic cell physiology and are experimentally useful references against
which to measure changes in the expression of more responsive genes. From the preceding section, it should be apparent that gene expression may be regulated at many levels, particularly transcription and translation. By far, the most attention has been paid to transcriptional regulation, although posttranscriptional mechanisms, such as mRNA stability, translational efficiency, and proteolysis, are also important.
The expression of many genes is modulated in response to external stimuli, such as growth factors, hormones, and mechanical forces. In most cases, the process is initiated when the growth factor, for example, binds to a specific cell-surface receptor. This triggers the process of signal transduction (Fig. 1), during which the occupied receptor initiates a chain of intracellular, biochemical changes that result in altered gene expression. Often, the first detectable response is phosphorylation of the receptor itself, or of other proteins with which it associates, at specific serine, threonine, or tyrosine amino-acid residues. The enzymes responsible for doing this are known collectively as protein kinases. Their importance is reflected in the human genome, nearly 2% of which encodes different protein kinases. Much research activity is dedicated to the identification and characterization of protein kinases, which has led to the development of successful new drugs for the treatment of various cancers. For example, Gleevec (matinib mesylate) and Iressa (gefitinib), which inhibit specific tyrosine kinases14,15, are used in the treatment of chronic myeloid leukemia and non-small-cell lung cancer, respectively.
In many cases, the signal transduction cascade leads to the generation of
small, diffusible second-messenger molecules such as cyclic AMP and
phosphoinositols, or the influx or intracellular release of calcium ions.
Usually, the end result is a phosphorylated protein that enters the nucleus to
initiate changes in gene expression. Examples that engage the attention of
orthopaedic researchers include the Smad family, important in transforming
growth factor-ß (TGF-ß) and BMP
signaling16; the
STAT17 (signal
transduction and activator of transcription) family, important in the
signaling of many inflammatory cytokines and immunomodulators; and nuclear
factor Steroid hormones classically signal in a different manner19. Because they are lipid soluble, they can diffuse through the cell membrane and bind to receptors present in the cytoplasm, instead of on the cell surface like the classic growth factors. This complex then enters the nucleus to influence gene expression. Within the nucleus, the final products of the signal transduction cascade interact with various transcription factors to regulate transcription (Fig. 1). To do this, they engage nucleotide sequences (response elements) contained within a region of DNA immediately in front of ("upstream" to) the gene, known as a promoter. Transcription factors possess a DNA-binding domain and an activator domain. The former binds to specific response elements in the promoter region, while the activator domain recruits other proteins, including RNA polymerase, forming an initiation complex that starts transcription. Inhibitory transcription factors also exist, and whether or not a gene is expressed reflects the balance between the two types of factors. The interaction of specific transcription factors with their cognate response elements within promoters forms the basis for tissue-specific gene expression. The level of expression may be further modified by regions of DNA that act as enhancers or repressors.
Analyzing Genes and Gene Expression It is now possible to determine the nucleotide base sequence of very large stretches of DNA quickly and inexpensively. Much of the current sequencing technology was developed as part of the Human Genome Project, which required the entire human genome of approximately three gigabases to be sequenced in an ambitious time frame and within budget. These methods are now commonplace, and most modern university medical centers offer DNA sequencing at a core facility for a few cents per base. Indeed, the problem now is not so much the generation of data as its analysis. Such bottlenecks have created the new field called bioinformatics20, in which advanced computing approaches are used to handle vast amounts of biological data. To give an example of how much computing power is needed, the University of Rochester Medical Center uses close to 25% of the university's total computing capacity just to run the microarray core facility (A. Brooks, personal communication). In addition to sequencing entire genomes, or fragments thereof, it is also possible to isolate and replicate them, a process known as cloning. Although entire genes can be cloned, they can be very large, so it is easier to generate a complementary DNA (cDNA) that lacks introns. This is achieved by using an enzyme called reverse transcriptase (RT) that copies mRNA into cDNA. In some cases, the entire mRNA population of a cell is reverse transcribed to form a cDNA library. In theory, this contains copies of all of the genes being expressed by the cell of origin. The cDNA generated by reverse transcription can then be spliced into plasmids, which are circular molecules of DNA that exist within bacterial cells. Plasmids often accumulate to several hundred copies per bacterium, and they are replicated as their bacterial hosts divide. Because bacteria grow very rapidly to high densities in nutrient medium, this provides a convenient way to grow large amounts of plasmid DNA containing cDNA(s) of interest. The ability to manipulate DNA in this way relies on the properties of restriction enzymes. These cleave DNA at specific locations in a highly predictable manner, on the basis of their ability to recognize specific base sequences. These generate a limited number of fragments, which can be spliced together with use of DNA ligases. The specific cleavage and ligation of DNA in this fashion, and its amplification in plasmids, form the basis of much of recombinant DNA technology. Arguably the most famous technique in molecular biology, the polymerase chain reaction (PCR) enables the rapid, selective amplification of precise regions of DNA. The key to the polymerase chain reaction is the combination of two synthetic oligonucleotide primers. (Oligonucleotides are short sequences of nucleic acid.) The primers flank the region of DNA to be amplified, one on each strand, and direct the activities of DNA polymerase selectively to the region between them. This reaction continues in multiple rounds or cycles, each of which doubles the amount of DNA synthesized. In this fashion, multiple copies of the DNA region of interest are rapidly produced from tiny amounts of starting material. For example, a single copy of DNA subjected to twenty-five cycles of polymerase chain reaction generates 225 (approximately 34 million) copies. The polymerase chain reaction amplification of DNA is useful for genotyping and cloning, but its greatest application is probably found in conjunction with the enzyme reverse transcriptase in a technique known as reverse transcriptase-polymerase chain reaction (RT-PCR). This is used to detect and measure the expression of specific mRNA molecules. Reverse transcriptase copies RNA into cDNA, which is then amplified in the manner indicated above. A derivative of reverse transcriptase-polymerase chain reaction, known as real-time polymerase chain reaction, has been developed as a more rapid, quantifiable way of measuring mRNA abundance. It is based on an increase in fluorescence that occurs with each round of amplification.
Although it is a powerful technique, reverse transcriptasepolymerase chain reaction is not well suited to measuring the abundances of large numbers of different mRNA molecules. The technique of choice for doing this is based on the use of glass substrates, or chips, a few centimeters square, on which are spotted an array of different cDNAs or oligonucleotides21. Each will hybridize with, and thus capture, a specific mRNA. Sophisticated detection devices and associated software identify the captured mRNA and measure its abundance. Because of their small size, these chips are known as microarrays, and they can be customized for selective investigation of the expression of genes related to a particular disease or organ. One company recently developed a "chondrochip" to examine gene expression in chondrocytes. However, chips covering the entire repertoire of transcripts are becoming available for certain species, including humans. Microarrays are revolutionizing the study of gene expression, especially as it relates to disease. Miniaturizing DNA analysis in this way has led to the development of similar approaches for expediting research in other areas of biology, biochemistry, and bioengineering. Activity in this area is so great that a new journal, Lab on a Chip, has been launched to accommodate the resulting surge in publications.
For certain purposes, it is necessary to identify the cells that produce
specific transcripts. A technique known as in situ hybridization
makes this possible. It involves the synthesis of a probe, usually an
oligonucleotide, which hybridizes specifically with the mRNA of interest. The
probe is labeled in some manner that permits it to be visualized within cells
expressing its cognate message. As demonstrated in
Figure 2, in situ hybridization
permits the examination of individual cells within a complex tissue. In the
example shown in Figure 2, the
sequential expression of genes that identify articular chondrocytes (type-II
collagen [Col2]), hypertrophic chondrocytes (type-X collagen [Col10]), and
osteoblasts (core binding factor
Proteomics In recent years, there has been a rapid expansion in the field of proteomics, in which the aim is to identify and measure all proteins produced by an organism12. This is far more challenging than the pursuit of genomics, as there are approximately 25,000 different human genes but an estimated one million different human proteins. Moreover, the techniques for separating and analyzing proteins are not as sophisticated as the corresponding techniques for genomic analysis. Nevertheless, the field is a very active area of research and progress will be rapid. Protein "chips," analogous to the DNA chips presently used for examining RNA expression, are under development. Alternatively, proteins can be separated by mass spectrometry, and the pattern of emerging signals can be read like a bar code. These methods should be of great future relevance to orthopaedics, especially in the areas of diagnosis, staging, prognosis, and optimizing treatment strategies22.
Manipulation of Gene Expression The efficiency of gene transfer into mammalian cells is greatly increased when recombinant viruses are used as vectors. Gene transfer with use of a virus is known as transduction, and it takes advantage of the natural ability of viruses to transfer their own genes to their host cells. As well as providing useful experimental tools, gene transfer vectors are of potential use in gene therapy, as described below.
Regulatory RNA Molecules Lack of specificity is one problem with the antisense approach. Better precision is achieved by using a type of RNA molecule, known as a ribozyme26, that degrades other RNA molecules in a sequence-specific manner. Much excitement has recently been generated by identification of a process called RNA interference (RNAi)27. This is a naturally occurring regulatory process, first detected in nematode worms, in which short sequences of RNA (small interfering RNAs [siRNAs]), generated from larger precursors, bind in a specific fashion to target mRNAs and bring about their degradation. This process is remarkably efficient and is being used increasingly as an experimental tool as well as being a potential therapeutic strategy26. Inhibiting gene expression with siRNA molecules is known as gene knockdown.
Genetically Modified Animals
Genes may also be deliberately inactivated (knocked out) by the targeted introduction of a disrupting DNA insert. In this case, embryonic stem cells lacking a functional copy of the disrupted gene lead to the birth of knock-out animals sharing this genetic alteration. A recent modification of this approach introduces genes encoding siRNA molecules to knock down gene expression in the animal. Use of such methods makes it possible to express normal or mutant genes or to prevent expression of specific genes in selected tissues. This can occur at specific times in development or in a way that can be turned on and off as needed for the experiment. Such capabilities permit detailed study of the roles of genes and their products in living animals and allow the generation of novel animal models of disease. In one of the earliest orthopaedic examples, disruption of the proto-oncogene c-src in mice did not produce the expected cancers or brain abnormalities, but instead led to osteopetrosis30.
There is scarcely any area of orthopaedic research that remains, or will remain, untouched by advances in molecular biology. The following discussion provides some important examples but is not exhaustive.
Growth and Development The embryonic axial and appendicular skeleton is initially cartilaginous. The cartilaginous anlage then undergoes endochondral ossification in the appropriate regions to produce the fetal skeleton. The future articular cartilage resists ossification and is retained within regions where the diarthrodial joints will form. As the first step in chondrogenesis, undifferentiated mesenchymal cells aggregate (condense) into clusters. The cell-to-cell contact that this involves occurs through specific proteinaceous binding molecules, particularly cadherins33 and other cell adhesion molecules (CAMs), expressed on the cell surface as well as components of the extracellular matrix. The process is regulated by a variety of growth factors, involving members of the TGF-ß superfamily (including the BMPs), fibroblast growth factors (FGFs), hedgehog proteins, and signaling molecules belonging to a family known as Wnt (pronounced wint)34-37. The interaction of these growth factors with their receptors triggers signal transduction events that culminate in the expression of specific genes. Data from genetically modified mice indicate the importance of many of these factors and their relevance to human disease. Mice lacking the BMP family member growth and differentiation factor 5 (GDF-5), for example, have severely shortened limbs and loss of phalangeal elements38. Similar mutations in humans produce Hunter-Thompson or Grebe types of chondrodysplasias39.
Condensation is associated with the expression of a number of genes whose
products help to regulate the process. Homeobox (Hox) genes are activated in
response to retinoic acid and are thought to initiate limb bud outgrowth and
limb field
positioning40.
Another group of genes, the T-box (Tbx) genes, are involved in specification
of the limb bud
field41.
Chondrogenesis within these aggregates is accompanied by the expression of
chondrocyte-specific genes. The Sox family of transcription factors,
especially Sox 9, Sox 5, and Sox 6, is important in this
regard42,43.
Mutations in the gene encoding Sox 9 result in the human disease camptomelic
dysplasia44. These
and other transcription factors bind to response elements in the promoter
regions of a number of genes encoding chondrocyte-specific proteins, including
the
The onset of endochondral ossification is signaled by the expression of
bone-related genes, notably the expression of a transcription factor Cbfa1
(core binding factor The manner in which these and other genes are expressed in exquisite temporal and spatial precision can be appreciated with use of in situ hybridization techniques. As shown in Figure 2, for example, the Col2 gene is expressed shortly after the mesenchymal cell condensation that occurs on embryonic day 12 in the mouse forelimb. By embryonic day 13, expression of Col2 becomes widespread throughout the digits, humerus, radius, and ulna, but it is absent by embryonic day 14.5. Instead, Col10 is found in hypertrophic cartilage51. Expression of Cbfa1, osteocalcin, Indian hedgehog, and the transcription factor gli 1 also shows representative patterns of expression51. Endochondral ossification, along the lines described above, forms the axial and appendicular skeleton. The craniofacial bones, in contrast, form by intramembranous ossification, the direct differentiation of mesenchymal progenitors into osteoblasts52. It is interesting to note that bone, one of the few adult tissues able to undergo efficient scarless repair after injury, does so by recapitulating the embryonic events that occur in utero. Using a mouse fracture model, Miclau et al.52 showed that the early stages of repair of a nonstabilized fracture involve mesenchymal cell condensation, chondrogenesis, with the expression of Col2 and related genes, followed by Cbfa1, osteocalcin, and endochondral ossification. Stabilized fractures, in contrast, expressed no cartilage-specific genes and healed by intramembranous bone formation, with the abundant expression of bone-specific genes.
Genetic Diseases Completion of the Human Genome Project has greatly aided the identification of genes responsible for such disorders. It is particularly useful in conjunction with a genetic technique known as linkage analysis, which determines the position of (maps) a gene in relation to other known genes. Once linkage analysis has indicated the region of the chromosome to which the disease maps, it is possible to search the region for candidate genes with use of existing databases. Because DNA sequencing has become so rapid and affordable, suitable candidates can be examined by directly sequencing the relevant regions. Comparison between the sequences of unaffected and affected individuals permits identification of the precise mutations55. Furthermore, functional analysis can be carried out with a variety of techniques, particularly the production of genetically modified mice bearing the equivalent mutation. Although the murine mutant phenotype does not always match precisely the mutant phenotype accompanying the human disease, it is nevertheless frequently useful. For example, nearly all patients with achondroplasia have mutations in the FGF-receptor 356-58. Interestingly, this is one of the most mutable genes in the human genome. Four of the inactivating mutations found in humans have been generated in mice. Although the human and mouse phenotypes are not perfectly matched, all mice display dwarfism consistent with the human disease58.
Disease Mechanisms, with Osteoarthritis as an Example
New molecular techniques now permit novel approaches to the study of osteoarthritis. One such avenue is the use of chip technology and differential display, a related technique, to compare levels of mRNA expression and to identify differences in gene expression between chondrocytes recovered from normal joints and those from diseased joints (Fig. 4). Such studies are complicated by the heterogeneity of the disease, its insidious onset, and difficulties in obtaining human cartilage from individuals other than those with late-stage disease. Nevertheless, these methods have confirmed that there are numerous differences in gene expression between chondrocytes recovered from normal joints and those from joints affected by osteoarthritis (Fig. 4)64. Not all of the corresponding genes have been identified, but those that have include genes involved in inflammation, proliferation, and neoplasia. In related studies, the chondrocyte transcriptome has been found to reflect between 13,200 and 15,800 unique genes65. The best characterized 10,000 of these genes have been mapped, with detection of clusters of coordinately expressed genes in certain chromosomal locations65.
Studies such as those described above can be coupled with experiments on transgenic animals to carry out additional mechanistic investigations. For example, because of the possible importance of type-IX collagen in osteoarthritis, a Col9A1 knock-out mouse has been generated. Osteoarthritis develops in multiple joints in this mouse in the absence of chondrodysplasia66. Other models of osteoarthritis have been produced by, for example, introducing a truncated Col2A1 gene into mice67.
Molecular Therapies Recombinant Proteins With use of methods outlined earlier in this review, it is possible to produce very large amounts of a therapeutic protein for clinical use. To do this, cDNA encoding the protein of interest is isolated and is spliced into an expression plasmid. The plasmid is transferred into host cells, which then express the transgene of interest with high efficiency. In industrial facilities, large amounts of the protein, produced recombinantly in this fashion, can be recovered from the cell cultures, purified, and used therapeutically. In many cases, such as with BMP-2, the recombinant protein is generated in genetically modified bacteria. Other proteins, such as erythropoietin, require posttranslational modification that occurs only in eukaryotes and are produced in a relevant cell line, such as Chinese hamster ovary cells. Human insulin was the first recombinant protein to be used clinically. The use of recombinant proteins in orthopaedic practice began with erythropoietin in the 1990s. Erythropoietin is employed to promote the formation of erythrocytes and thus prevent and treat anemia without the need for blood transfusions. The recent availability of recombinant BMP-2 and BMP-7 as aids to fracture-healing and spine fusion is a good orthopaedic example of the power of molecular biology in translating a biological observation into a clinically useful product. As outlined in Table II, the finding in the mid-1960s that demineralized bone could induce ectopic, intramuscular bone formation initiated a chain of events culminating in the clinical application of recombinant BMPs in the late 1990s. Of note, the identification, cloning, and production of recombinant BMP-2 and 7 took approximately thirty years. Technology has improved to the point where, today, this process would take only a fraction of the time.
Therapeutic proteins need not necessarily be recombinant forms of naturally occurring molecules. Molecular technology permits enhancement of the product. For example, it is possible to engineer erythropoietin in a manner that increases the number of sugar molecules attached to it (glycosylation)68, which increases its physiological stability and permits less frequent dosing. Another example is the drug Enbrel (etanercept), which is being used with great success to treat rheumatoid arthritis. Enbrel is an engineered molecule containing two soluble tumor necrosis factor (TNF) receptors fused to part of an immunoglobulin molecule69; it does not exist as such in nature. Monoclonal antibodies constitute a rapidly growing area of protein therapeutics, diagnostics, and imaging substances70. Indeed, more than 200 antibodies, constituting 20% of all protein drugs, are presently being studied in clinical trials. Molecular techniques permit fully human monoclonal antibodies to be produced in murine cells, thereby obviating any immunological problems in the human host. Humira (adalimumab), an anti-TNF human monoclonal antibody used to treat rheumatoid arthritis, is a relevant example71.
Gene Therapy The repair and regeneration of bone, cartilage, ligaments, and other tissues of orthopaedic interest would be enhanced by the application of relevant growth factors. However, these growth factors are cleared quickly from the body. They need to be present for much longer periods to be effective. Solving this problem has proved surprisingly difficult and has greatly delayed the clinical application of growth factors. Data from preclinical studies strongly suggest that delivery of the relevant cDNA provides local, sustained, therapeutic synthesis of the transgene product. Moreover, healing of bone, cartilage, and ligaments has been enhanced in animal models with existing gene-transfer technology. Clinical trials of some applications, such as bone-healing, are on the horizon76. Additional orthopaedic opportunities for gene therapy are the treatment of chronic diseases such as arthritis77 and osteoporosis78,79. Currently, the treatment of chronic diseases such as rheumatoid arthritis with recombinant proteins requires regular injection or infusion, which is expensive, unpleasant, and accompanied by side effects. Gene transfer, in contrast, promises to generate these products endogenously in a sustained manner without the need for repeat doses. Two phase-I clinical trials of gene therapy for the treatment of rheumatoid arthritis have been successfully completed80,81, and the results from the first have recently been published82. In both trials, a cDNA encoding human interleukin-1 receptor antagonist (IL-1Ra) was transferred to the synovial linings of affected joints with use of a retrovirus as the vector. Subsequent, phase-II efficacy studies and a related phase-I study of the therapy for osteoarthritis83 are pending. A phase-I trial in which a recombinant virus encoding a TNF antagonist was injected into rheumatoid joints is just approaching completion. The application of antisense RNA molecules that inhibit the synthesis of TNF has shown efficacy in phase-II studies of patients with rheumatoid arthritis77. Vitravene (fomivirsen), which is applied locally to the eye to block cytomegalovirus infections, is the only antisense RNA drug given market clearance.
Small Molecules One such area is the identification of molecular targets. As noted earlier, components of intracellular signaling pathways, particularly protein kinases, are the subject of intense scrutiny and have led to the production of novel drugs for the treatment of cancer. Additional targets include receptors, transcription factors, and other molecules that regulate gene expression. In the future, in the place of BMP-2 protein therapy or gene therapy, patients may be able to swallow a pill whose active ingredient binds to the BMP-2 receptor and triggers a response that is identical to the one elicited by BMP-2 itself. Alternatively, it may be possible to develop orally active molecules that bind to the BMP-2 promoter region and transcriptionally activate BMP-2 synthesis. The possibilities are endless. Molecular modeling represents the second contribution of molecular biology to this process. Advances in x-ray crystallography84 and nuclear magnetic resonance spectroscopy85 permit the structures and conformations of biological and synthetic molecules to be determined with precision. They provide the opportunity to design small molecules that can bind to selected targets at specific sites in a predictable fashion. This eliminates much of the guesswork from drug development, resulting in the rapid creation of effective designer drugs.
Diagnosis and Prognosis As an alternative to using single genes and gene products in this way, use of gene chips and supporting software makes it possible to determine an entire expression repertoire. This will enable one to make differential diagnoses based on global gene-expression profiles. Such analyses will be particularly useful for the diagnosis of complex diseases that do not generate a single diagnostic molecular signal analogous to the EWS/FLI-1 fusion. Osteoarthritis, for example, is associated with altered expression of numerous cytokines, proteinases, matrix molecules, and so forth. Expression profiling may permit osteoarthritis to be diagnosed at a much earlier stage and with much more confidence than is presently possible, especially if the analysis can be performed on a blood sample rather than a cartilage biopsy specimen. Because gene expression is likely to change with time and with the progression of the disease, such methods can be used for staging and prognosis. As an example, patients with synovial sarcoma who have the SYT-SSX1 fusion have a worse prognosis than those who have the SYT-SSX2 fusion88. The National Institute of Arthritis and Musculoskeletal and Skin Diseases has launched an osteoarthritis initiative, with a major emphasis on developing markers that permit ready monitoring of osteoarthritis and its response to putative therapeutics. Both genomic and proteomic approaches are being used in this endeavor. Molecular profiling of this type may also permit accurate prediction of response to therapy. For example, patients with a breast cancer tumor that expresses estrogen receptor will respond to tamoxifen, whereas those with a tumor that does not express estrogen receptor will not89. Similarly, responses to the anti-cancer drug Iressa, mentioned earlier, depend on a specific mutation in the tumor cells90. Expression profiling of multiple genes is beginning to have an impact on the ability to determine prognosis. Recently, it has been shown that survival of patients with diffuse large-B-cell lymphoma can be predicted on the basis of the expression of six genes from a total of 10,000 candidates established in earlier studies91. If similar types of markers could be identified for common orthopaedic conditions, the impact would be huge. Predicting responses to expensive drugs such as Enbrel, for example, would have a major impact on the treatment of rheumatoid arthritis. Looking ahead, it might be possible to use molecular markers to determine whether a given nonunion would respond better to BMP-2, BMP-7, or some other osteogenic stimulus, or whether the chances of aseptic loosening of a joint replacement would be reduced by the selection of one prosthesis over another. Such advances would pave the way for comprehensive, personalized medicine based on the rapid genomic, proteomic, or other "-omic" analysis of a small volume of blood, urine, or other conveniently obtained bodily material.
Personalized Orthopaedic Care Many SNPs help to determine our susceptibility to different diseases, our responses to therapy, and other aspects of our individual medical makeup92. From a surgical perspective, they may indicate, for example, which patients are at risk for postoperative embolism, arthrofibrosis, or malignant hyperthermia. Polymorphisms may indicate which patients will have a nonunion following a certain type of fracture as well as whether osteoarthritis is likely to be progressive or not in a given individual. The potential for healing of a cartilage defect or for the development of adhesions after ligament surgery in a particular individual may also be determined by subtle genetic variations. Polymorphisms in a set of liver enzymes known as P450 explain why up to 10% of the population can take large doses of codeine without pain relief, while others are exquisitely sensitive. Pharmaceutical companies are currently engaged in an enormous effort to catalogue the repertoire of human SNPs and to determine how they influence responses to drugs and other forms of therapy; this has given birth to the new field called pharmacogenomics93. Matching individual polymorphisms to a menu of available treatment options will make it possible to tailor medical care so that individual patients receive personalized treatment best suited to their genotype. Such capabilities are expected to be available within the next few years. One controversial precursor to these developments is the recent clearance for marketing by the Food and Drug Administration (FDA) of a heart-failure drug specifically for African-Americans. This is the first time that the FDA has approved a drug for a specific racial group. Although it raises a plethora of sociopolitical issues, the decision has received the backing of the Association of Black Cardiologists.
Orthopaedic Practice of the Future Advances in molecular orthopaedics will also provide the physician with greater diagnostic, staging, prognostic, and treatment options that are individualized for each patient. For example, if the molecular profile of a patient indicates that he or she has early osteoarthritis with a low risk of progression, conservative treatment would perhaps be indicated. Conversely, prognosticators of rapid disease progression would indicate the need for aggressive therapy. A variety of different nonoperative interventions will be available, including orally active small molecules, protein therapy, gene therapy, and cell therapy. A review of the patient's SNPs, combined with gene-expression-array data derived with use of an osteoarthritis chip supplemented perhaps by a proteomic, transcriptomic, or other "-omic" analysis, will indicate which of these treatments is best suited for this individual. If the above agenda is successful, fewer patients will have disease progression to the stage requiring surgery, and molecular information will help operations to be more successful. The nature of the surgery will be based on molecular criteria. For example, these criteria will determine the most favorable type of prosthesis for an individual requiring total joint replacement. Following multiple trauma, molecular markers might indicate the best surgical approaches to fracture fixation as well as whether or not surgery needs to be supplemented with osteogenic growth factors delivered as proteins or genes or stimulated in some other manner such as with an osteogenic pill. Ultimately, the use of molecular orthopaedics will enable the orthopaedic surgeon to perform less surgery and pay greater attention to preventative and nonoperative interventions. The range of treatment options will increase dramatically; will include novel, biological therapies of various types; and will become individualized. These developments will bring dramatic changes to the practice of orthopaedics, in ways that will enhance the ability of the orthopaedist to prevent and treat diseases of the musculoskeletal system. Some of these capabilities should be in place before the end of the present Bone and Joint Decade.
NOTE: The authors thank Dr. Paul Robbins and Dr. Rudolf Fluckiger for reading earlier drafts of this paper.
|
Top
Abstract
Introduction
B (NF