No Bones About It

When examining bones and joints, toxicologic pathologists and those working on experimental models have the advantage of being able to examine sections from an equal number of age-matched control animals that are sectioned in the same plane as treated animals. Diagnostic pathologists have the advantage of input from clinicians who may have radiographs and, in some cases, other types of imaging data. Despite these advantages, bones and joints are often not favorite tissues, largely because of the tremendous heterogeneity in the appearance of normal features that can make lesions hard to distinguish. This heterogeneity is influenced by many variables, including but not limited to plane of section, differences in appearance of the various bones of the body and their associated structures (ligaments, tendons, menisci), age of the individual, and the fact that abnormal structures (eg, periarticular osteophytes) can assume a normal appearance through remodeling over time, thus requiring the pathologist to have expertise in the normal morphology (particularly the shape) of the affected site to appreciate changes due to chronic disease. Depending on the age of the animal, there may be the presence or absence of growth plates, remodeling in the cutback zone and endosteum, irregularities around ligament insertions and nutrient vessels, and a variable amount of new bone in the periosteum of the metaphysis. Whenever possible, plane of section should be standardized for comparison between affected and control tissues, particularly when histomorphometry measurements are taken. This is nicely exemplified in the article on animal models of osteoarthritis by McCoy ¹⁰ in this issue, in which an ideal, mid-coronal section of a murine stifle joint is included for orientation. Age of the individual can dramatically affect the healing potential of joint tissues, as evidenced by studies of osteochondrosis, a disease that develops only in immature individuals. Due to the remarkable healing properties of epiphyseal cartilage (discussed by Olstad et al ^14,15 in the studies on osteochondrosis), most lesions occurring in animals with this disease are subclinical and known to heal spontaneously. Largely because it is not possible to biopsy these sites and confirm their features in the human disease, subclinical lesions of osteochondrosis in humans often are misinterpreted by MD radiologists as “ossification variants” rather than evidence of disease. Conversely, lesions of osteoarthritis occurring in older individuals, in whom the articular cartilage undergoes chronic degenerative changes, remain static at best and usually progress (Figs. 1, 2).

Figure 1. Osteoarthritis, cynomolgus macaque, showing marked articular cartilage degeneration and fibrillation accompanied by subchondral sclerosis. Hematoxylin and eosin. Figure 2. Osteoarthritis, cynomolgus macaque, showing marked loss of toluidine blue staining, indicative of proteoglycan loss from the matrix in affected articular cartilage. Toluidine blue.

Preparing bones for histopathology can be problematic if careful handling is not followed throughout the process. This includes allotting sufficient time for fixation, collecting a sample of an appropriate thickness, rinsing to remove “bone dust,” and using an appropriate decalcifying agent in which the end point can be monitored to prevent over-decalcification. The processing schedule should allow good penetration of the solutions and impregnation of paraffin. At microtomy, using positively charged or coated slides ensures good adherence of the tissue. Finally, it is important to optimize the staining procedure to achieve crisp nuclear and cytoplasmic details. Most sections are routinely stained with hematoxylin and eosin, but broad horizons of color (with diagnostic significance) are available in bone and joint sections stained with safranin O (see Olstad et al, ¹⁴ Fig. 18) or toluidine blue, cationic dyes that stain proteoglycans and glycosaminoglycans in cartilage (Figs. 3–5). For those who keep their polarizing lens handy, examining the birefringence of collagen in bone is another scintillating reward and allows one to easily distinguish mature, lamellar bone from newly formed, woven bone (Figs. 6, 7).

Figure 3. Growth plate, normal rat. Intense red staining of glycosaminoglycans in cartilage of physis stained with safranin O. Figure 4. Tibia, rat. Intense red staining of glycosaminoglycans in articular cartilage stained with safranin O. Figure 5. Tibia, rat. Intense blue staining of glycosaminoglycans in articular cartilage stained with toluidine blue. Figure 6. Hypertrophic osteopathy, dog. Lamellar bone of the cortex (left) and woven bone in the periosteum (right). Hematoxylin and eosin (HE). (Image courtesy of Erik Olson) Figure 7. Hypertrophic osteopathy, dog. Lamellar and woven bone, viewed with polarized light. (Image courtesy of Erik Olson) Figure 8. Fluorochrome labeling with calcein. Cynomolgous macaque. Plastic-embedded section of bone stained with Villanueva and viewed with fluorescence microscopy. Osteoid stains orange, and the labels are visible as green lines where the calcein was deposited during active mineralization. (Image courtesy of Chris Jerome) Figure 9. Growth plate, rat. Movat’s pentachrome shows strong Alcian blue staining of cartilage in the hypertrophic zone. (Image courtesy of Kendall Frazier).

A few pathologists have access to specialized histology techniques providing undecalcified sections; interpreting these is a specialty area. Such sections allow additional opportunities to use fluorochromes, such as tetracyclines, to evaluate bone growth and remodeling (Fig. 8). And the glorious technicolor of Movat’s pentachrome (Fig. 9), a stain available for undecalcified sections, is a delightfully vibrant color experience for a weary pathologist’s eye. Although routine decalcification may render bone unsuitable for immunohistochemistry, this can sometimes be overcome by resourceful pathologists, as demonstrated by Muscatello et al ¹¹ in the study on Ellis-van Creveld syndrome in Grey Alpine cattle using decalcification in 10% EDTA and antigen retrieval procedures (see also Nagamine et al ¹² ). The intricacies of bone implant and device research are comprehensively reviewed in this issue by Wancket. ¹⁹ Since most pathologists, including most toxicologic pathologists, are unfamiliar with the types of testing conducted to evaluate implants and devices, the overview included in this article is a welcome inclusion. A practical insight into species differences in bone physiology helps to explain how these differences affect the choice of animal model for this type of research.

Interpretation of bone lesions is generally enhanced by correlation with radiographs or computed tomography as well as clinical and macroscopic observations. Conditions affecting the teeth and jaws can be particularly perplexing. Writing about the confusing array of ossifying oral cavity lesions in dogs that include ossifying fibroma, fibrous dysplasia, and osteomas, Soltero-Rivera et al ¹⁸ include radiography in their evaluation to provide a comprehensive evaluation of this topic. Similarly, Bell and Soukup ² describe clinical, radiological, and histological features of alveolar osteomyelitis in cats, another complicated oral disease. Smedley et al ¹⁷ provide histologic insight into what has been a puzzling and serious dental condition of older horses, equine odontoclastic tooth resorption and hypercementosis (EOTRH) syndrome. The remarkable diversity of histologic patterns and expression of cytoskeletal proteins in a large number of canine skeletal osteosarcomas of both the appendicular skeleton and the head are described by Nagamine et al. ¹² Finally, the articles by Olson and Shaw et al ¹³ on bone disease in marmosets and by Dittmer and Thompson ⁵ on the approach to investigating skeletal problems in livestock describe histologic lesions that alone are probably not specific enough to allow diagnosis. These articles clearly indicate the advantages of integrated analysis of clinical findings, imaging studies, gross and histopathology, and molecular investigation instead of a simple reliance on a single method of evaluation.

Similarly, understanding laminitis in horses, an age-old disease with several etiologies, has been greatly enhanced by Engiles et al, ⁶ who used micro–computed tomography (CT) and histology to measure and describe changes in the distal phalanx of affected horses. Karikoski et al ⁹ provide a detailed correlation of endocrinopathic laminitis with clinical, macroscopic, and microscopic findings. Another fine example is the article by Janes et al, ⁸ in which the cervical spine in young horses with and without spinal cord compression is systematically evaluated using magnetic resonance imaging (MRI), with histological confirmation of identified lesions. Although lesions were observed more frequently in affected horses, their generalized distribution and presence in unaffected horses support an underlying developmental cause for the lesions and clinical signs seen in this disease. This important information would not be possible without the combination of imaging and histological techniques. Additional examples are included in the Olstad et al ¹⁴ review of osteochondrosis, in which insights into disease pathogenesis have been gained through CT, micro-CT, and MRI studies, all of which offer a 3-dimensional perspective on the lesions occurring in this disease and will allow interspecies comparisons, including comparisons to the human disease. This article also provides insights on the importance of studying skeletal disease in the acute stages, as chronic lesions of diverse skeletal diseases (osteochondrosis, osteoarthritis, rheumatoid arthritis) can be closely similar, although the cause and pathogenesis of each may be completely different. This is particularly important in the study of animal models, in which experimental studies allow the availability of tissues in the early stages of disease that are usually unavailable for the corresponding human conditions. The article on subchondral bone cysts by Olstad et al ¹⁵ exemplifies this point by examining both naturally occurring and experimentally induced early lesions of osteochondrosis and clearly demonstrating, with specific examples, how subclinical lesions of osteochondrosis can result in either true cysts or pseudocysts. These insights are difficult, if not impossible, to gain by studying only end-stage tissues, which are the usual submissions in diagnostic cases in both veterinary and human medicine.

Many hormones, subcellular factors, and mediators are involved in the basic biology of bone growth. The roles of calcium, vitamin D, and parathyroid hormone are well known. The control of phosphorus by fibroblast growth factor 23 (FGF23) and proteins that regulate its activity is a less explored area. Hardcastle and Dittmer ⁷ have taken on the complex task of exploring and explaining the role this hormone has in the pathogenesis of diseases from rickets and osteomalacia to renal disease. The issue of determining the role of these minerals and hormones in naturally occurring diseases of bone can be difficult, as exemplified by the article by Olson and Shaw et al ¹³ that focuses on bone disease in common marmosets. The review of congenital skeletal abnormalities in livestock by Dittmer and Thompson ⁵ emphasizes the importance of good clinical and pathological descriptions and the similarities between congenital anomalies that can have different causes: genetic, nutritional, or teratogenic.

Examples of experimental approaches to evaluating models of bone and joint pathology include the review by Caplazi et al ⁴ on mouse models of rheumatoid arthritis by a multidisciplinary team of pathologists, immunologists, and imaging experts, which describes pathogenesis and potential drug targets along with examples of how different models may be used in pharmaceutical research. Another important area of research in bone pathology is the review by Simmons et al ¹⁶ on experimental methods that may result in a greater understanding of the bone metastatic process. Models that originate from such cases are of special interest for those who practice diagnostic pathology or conduct research on companion animal oncology.

The issue is rounded out by 2 articles by Bolon et al ³ and Adissu et al. ¹ Osteoprotegerin (OPG) is known mainly for its inhibition of osteoclastogenesis. In their article on degenerative joint disease (DJD) in OPG null mutant mice, Bolon et al provide evidence that OPG may have an essential role in maintenance of chondrocyte integrity. The interesting association of right ventricular epicardial fibrosis in mice with sternal segment dislocation described by Adissu et al is speculated to originate from trauma affecting the fourth intersternal joint, further emphasizing the importance of astute clinical observation and correlation with pathology.

We hope that this special issue will increase the enthusiasm for and improve the understanding of many conditions affecting the skeleton. There is so much to learn!