Correspondence to: Seth W. Donahue, Dept. of Mechanical Engineering, Colorado State University, Flint Animal Cancer Center, 300 West Drake Road, Fort Collins, Colorado. Fax: 970-297-1254. E-mail: firstname.lastname@example.org
Intracortical bone remodeling—bone resorption by osteoclasts followed by bone formation by osteoblasts—produces secondary osteons, which contain a pore known as a Haversian canal (Sedlin and Frost, 1963). The periphery of secondary osteons are well defined by the histological feature known as the cement line (Schaffler et al., 1987). Intracortical bone remodeling is persistent throughout life, leading to age-related increases in osteon population density (OPD) in many mammals including humans (Kim et al., 2007; Han et al., 2009; Villa and Lynnerup, 2010), macaques (Havill, 2004), and chimpanzees (Mulhern and Ubelaker, 2009). Thus, there is an age-related increase in Haversian canal density, which contributes to age-related increases in intracortical porosity. Unbalanced bone resorption and formation also contribute to aging-induced increases in intracortical porosity. Intracortical porosity increases with age in many species including humans (Wang and Ni, 2003; Thomas et al., 2005), dogs (Frank et al., 2002), horses (Hawkins and Stover, 1997), and chickens (Srinivasan et al., 2000). Disuse also increases intracortical porosity by unbalancing bone resorption and formation (Gross and Rubin, 1995; Garber et al., 2000; Li et al., 2005). In contrast, hibernating bears are a naturally occurring model for the prevention of both age-related and disuse osteoporoses (Donahue et al., 2006; McGee et al., 2007a, 2007b, 2008; McGee-Lawrence et al., 2009). Black bears spend 0.25–7 months in hibernation annually depending on climate and food availability, yet bears in Michigan and Utah that hibernate for 5–6 months annually experience no losses in bone geometrical or mechanical properties (McGee et al., 2007b, 2008; McGee-Lawrence et al., 2009). Additionally, intracortical porosity significantly decreases with age in bears from Utah and Michigan (McGee et al., 2007a,b, 2008; McGee-Lawrence et al., 2009). The age-related increase in human intracortical porosity is strongly negatively correlated with bone strength (McCalden et al., 1993), and increased porosity may contribute to fracture risk (Patsch et al., 2013). Thus, improving our understanding of the unique bone metabolism in hibernating bears will increase our ability to develop treatments for age- and disuse-related osteoporoses in humans.
Hibernation is characterized by metabolic suppression for the conservation of metabolic energy during the winter when food is scarce. Basal metabolic rates in hibernating bears are only 25% of summer levels (Toien et al., 2011). Similarly, the activation frequency of intracortical bone remodeling in bears during hibernation is only 25% of summer levels (McGee et al., 2008), suggesting a link between energy conservation and skeletal preservation during hibernation. Uniquely, hibernating bears are able to maintain balanced bone resorption and formation during disuse (McGee et al., 2008). Surprisingly, intracortical porosity was lower in hibernating bears than in summer bears, despite the hibernating bears being physically inactive for 4 months (McGee et al., 2008). The suppressed intracortical bone remodeling and balanced resorption/formation during hibernation likely contribute to the age-related decrease in intracortical porosity in bears (McGee-Lawrence et al., 2009). To further study age-related changes in bone properties in hibernating bears, we evaluated femurs from bears from various latitudes with various hibernation durations. We hypothesized that bears with longer hibernation durations would have smaller age-related increases in OPD and greater age-related decreases in intracortical porosity than bears with short hibernation durations.
Forty-seven black bear femurs from our archives were used for this study. All samples were from male bears (16 from Utah, 11 from West Virginia, and 20 from Florida). Ages of the bears were determined using dental cementum annuli (Bourque et al., 1978). Ages of the bears ranged from 1 to 19 years for Utah bears, 1–13 years for West Virginia bears, and 1–9 years for Florida bears. Utah bears died in May, September, and October. Florida bears died in January (2), February (3), July (1), August (3), October (2), November (7), and December (2), and West Virginia bears died in May and June. Soft tissue was removed from the bones, which were subsequently stored at −20°C. Hibernation length is correlated to food availability and weather patterns (Schooley et al., 1994); bears in southern latitudes enter hibernation dens later and emerge earlier (Oli et al., 1997). In Florida, where food is abundant throughout most of the year, black bears hibernate for about 1–4 weeks, if at all. In West Virginia, black bears hibernate about 3–4 months annually, and in Utah hibernation typically lasts 5–6 months.
The midshaft of the femur was defined as one-half of total femur length. This midpoint was thin-sectioned using a diamond saw (Isomet 1000, Buehler, Lake Bluff, IL) to obtain cross-sections that were imaged with a digital camera (SPOT Insight QE, Diagnostic Instruments, Sterling Heights, MI). Cortical cross-sectional area was determined using a digitized section image and image analysis software (Scion Image, Frederick, MD). A 15-mm segment located distal to the midpoint of the diaphysis was removed and fixed in 70% ethanol for 48 hr. Bone segments were then embedded in methylmethacrylate. Two sections from each femur were prepared using a diamond saw and ground to a thickness of 70–90 μm. One section was stained in 1% basic fuchsin stain for 30 sec in increasing ethanol solutions (70%–100% ethanol). Sections were then mounted onto slides and imaged at 40× magnification using a light microscope and SPOT Insight QE digital camera (Diagnostic Instruments, Sterling Heights, MI). Microstructure was analyzed using BIOQUANT OSTEO software (Nashville, TN) to determine porosity. Porosity was defined as ratio of void space to bone area. Void spaces were defined as all porous spaces and included Haversian canals, Volkmann's canals, remodeling cavities, and primary bone vascular cavities, but excluded osteocyte lacunae and canaliculi.
The sections adjacent to those used for porosity measurements were stained in toluidine blue to permit visualization of cement lines. The sections were etched in 2% formic acid for 3 min, rinsed in distilled water, placed in 70% ethanol for 15 min, and stained in the toluidine blue solution for 5 min. The solution consisted of 1 g of toluidine blue powder and 1 g of sodium tetraborate decahydrate per 100 mL of distilled water. The procedures were conducted on an orbital shaker plate to maintain uniform staining and staining intensity. Sections were then placed in increasing ethanol concentrations in 30 sec intervals before being mounted on slides for analysis. This staining procedure allows visualization of cement lines that define secondary osteons (Fig. 1). Resorbing and refilling cavities were not included in OPD calculations. Sections were imaged at 40× magnification and the number of secondary osteons was quantified. OPD was determined by summing total secondary osteons (defined as including intact Haversian canals) and secondary osteon fragments (containing a partial or no Haversian canal) and dividing by cortical cross-sectional area. Resorption and refilling cavities were quantified separately and normalized by cortical cross-sectional area to determine resorption and refilling cavity densities. A cavity was classified as a refilling cavity if there was an intact cement line and an osteoid seam. A cavity was classified as a resorption cavity if it was not surrounded by a cement line. Haversian canal area and osteonal area were quantified for osteons sampled from the entire cross-section. Images were collected from the periosteal and endosteal regions of each anatomical octant (i.e., 16 images for each cross-section) at 100× magnification.
A 10-mm section located proximal to the midpoint of the diaphysis was removed and cleaned of marrow. Each segment was rehydrated for 4 hr in 0.15 M saline and weighed to determine wet mass. Sections were then dried in a furnace for 24 hr at 100°C, after which dry mass was determined. Dried sections were ashed at 600°C for 48 hr to remove organic matrix. Ash fraction, a measure of mineral content, was calculated as the ratio of ash mass/dry mass of the bone.
Analysis of Covariance was performed in MATLAB (MathWorks, Natick, MA) to test for the equality of slopes and compare each dependent variable with location (i.e., state) and age as covariates. Significance was determined based on an alpha level of 0.05.
Histological Features of Black Bear Femoral Cortical Bone
Similar to many large rapidly growing quadrupeds, black bear cortical bone consists of a mixture of primary and secondary bone (Fig. 2). Resorption cavities are eventually refilled by osteoblasts to produce secondary osteons. Osteon density is typically greatest in the older bone near the endosteal surface.
OPD and Remodeling Cavity Densities
Cortical bone cross-sectional area was greater (P = 0.0041) in Florida bear femurs than Utah and West Virginia bear femurs. Cortical cross-sectional area increased (P = 0.0001) with age in bears from all three states, however, the rate of the age-related increase was not different (P = 0.2156) between the three states. West Virginia black bears had greater OPD (P = 0.006) than those from Utah (Figs. 3 and 4). There was no difference (P = 0.1278) in OPD between West Virginia and Florida bears, or between Florida and Utah bears (P = 0.2296). OPD increased (P < 0.0032) with age in Florida and West Virginia samples, but not in Utah black bear samples (Fig. 2). There was no difference (P = 0.742) in the slopes of OPD vs. age between Florida and West Virginia bears, but OPD increased at a faster rate (P = 0.0474) in West Virginia bears than in Utah bears, and OPD tended (P = 0.0862) to increase at a faster rate in Florida bears than in Utah bears. There were no age-related (P = 0.27) or latitude-related differences (P = 0.32) in refilling cavity density. There were also no age-related (P = 0.09) or latitude-related (P = 0.55) differences in resorbing cavity density. There was no age-related difference (P = 0.29) in osteonal area, but there was a trend (P = 0.09) for osteons to be largest Florida bears followed by West Virginia bears, then Utah bears. There was no latitude-related (P = 0.24) difference in Haversian canal area; however, Haversian canal area tended to increase in Florida bears (P = 0.055) but not West Virginia (P = 0.71) or Utah (P = 0.28) bears.
Both Florida (P = 0.0001) and West Virginia (P = 0.0006) samples demonstrated lower intracortical porosity compared to samples from Utah; however, intracortical porosity in bear bones from Florida and West Virginia were not different from each other (P = 0.6525) (Fig. 5). Porosity decreased (P = 0.0007) with age in bear bones from all three states; the rate of the age-related decrease in intracortical porosity were not different (P = 0.3343) between samples from the three states.
Utah bear bones had greater (P = 0.0005) mineral content than Florida bear bones. Utah bears tended (P = 0.0604) to have greater ash fraction than those from West Virginia, and West Virginia bears tended (P = 0.0955) to have greater ash fraction than Florida bears. In bones from all three latitudes, ash fraction increased (P < 0.0001) with age (Fig. 6). The slope of ash fraction vs. age tended to be steeper in Florida samples than in Utah (P = 0.0559) or West Virginia samples (P = 0.0688), but no difference (P = 0.5957) was shown in the slopes between Utah and West Virginia samples.
Bears greatly reduce intracortical bone remodeling during hibernation, which causes intracortical porosity to be lower and mineralization to be higher in hibernation than in summer (McGee et al., 2008). As fewer new osteons are formed during the hibernation season, bears with longer hibernation periods were expected to show smaller age-related increases in OPD, greater age-related decreases in intracortical porosity, and greater age-related increases in mineralization as compared to bears with shorter or no annual hibernation periods. The goal of the current study was to compare age-related changes in OPD, intracortical porosity, and mineralization in femurs from bears from three different latitudes, which undergo varying hibernation durations. These findings increase our understanding of the unique bone metabolism that occurs in bears during hibernation and have implications for more broadly understanding the disuse- and age-related effects on bone in various species. Ultimately, the biological mechanisms that prevent disuse- and age-related osteoporoses in bears may contribute to the development of new therapies for treating osteoporosis. We found longer hibernating bears did indeed demonstrate lower OPD and higher cortical bone mineralization than bears with shorter hibernation durations, but we surprisingly found longer hibernating bears had higher intracortical porosity. However, most notably, we found bears from all three states showed age-related decreases in intracortical porosity, indicating that regardless of hibernation duration, black bears do not show the disuse- or age-related increases in intracortical porosity which is typical of other animals. (Srinivasan et al., 2000; Frank et al., 2002; Wang and Ni, 2003; Thomas et al., 2005).
We previously showed that hibernation does not have deleterious effects on bone properties in bears from northern latitudes (McGee et al., 2008; McGee-Lawrence et al., 2009). In the current study, we did not directly evaluate the effects of hibernation on bone properties in bears from the three different latitudes. However, given that bears from Utah do not show differences in bone properties before and after hibernation (McGee-Lawrence et al., 2009), it seems unlikely that bears with even shorter hibernation durations (i.e., those from West Virginia and Florida) would experience bone loss during hibernation. This is supported by the finding that serum markers of bone turnover in hibernating bears in Virginia show decreased turnover with balanced bone formation and resorption (Bradford et al., 2010). Bears in Florida hibernate for approximately 1–4 weeks annually, if at all. It is unknown if or how long the Florida bears used for this study hibernated. Male bears in Florida are often large enough and have sufficient fat stores that they do not hibernate at all. Surprisingly, the Florida bears showed the same rate of age-related decrease in intracortical porosity as bears from the other two states. This was surprising as we expected the suppression of intracortical bone remodeling during hibernation to have greater effects on age-related decreases in porosity in longer hibernating bears by reducing the rate of Haversian canal accumulation with age.
However, in addition to Haversian canals, resorption and refilling cavities, Volkmann's canals, and vascular porosities in primary bone contribute to the overall intracortical porosity. There were no differences in resorbing and refilling cavity densities between the bears from different geographical locations. There was also no difference in the size of osteons in the bones from the three different latitudes, which suggests there was no difference in resorption cavity size over the bears' lifetime. There were also no differences in the size of Haversian canals in bear bones from the three different states. It is possible that Utah bears have greater vascular porosities in primary bone and/or they have a more extensive Volkmann's canal network. Another potential confounding factor is that the bones in this study are from bears that died at different times of year, although all bears died while physically active (i.e., not while hibernating). There are some seasonal changes in bone remodeling in nonhibernating animals (Hill et al., 2007). Thus, another possible explanation for why porosity is higher in Utah bears, even though OPD was lower, is due to seasonal differences in bone remodeling at the time of death. Notwithstanding the unclear relationship between OPD and porosity, intracortical porosity decreases with age and is relatively low (less than 5%) in skeletally mature bears from all three states, even in Utah bears, as compared to the high cortical bone porosities of aging humans and other mammals like macaques. Importantly, age-related decreases in intracortical porosity of male bears show the opposite behavior (Fig. 5) of the age-related increases in human males, which increase from approximately 5% in the second decade of life to values approaching 20% in the seventh decade of life (Bousson et al., 2001).
This study provides further support that hibernating bears prevent age-related increases in intracortical porosity. We previously attributed age-related decreases in porosity (McGee-Lawrence et al., 2009) to the suppressed intracortical remodeling that occurs during hibernation (McGee et al., 2008). However, the current study shows age-related decreases in intracortical porosity regardless of hibernation duration. The age-related decreases in intracortical porosity and age-related increase in mineralization contribute to the age-related increase in bone strength in bears (McGee et al., 2007a), which is remarkable considering bears are physically inactive for as long as 6 months annually. Thus, the biological mechanisms that maintain balanced bone formation and resorption during the physical inactivity that occurs during hibernation and throughout life should continue to be studied as they may contribute to the development of new therapies for osteoporosis.
The authors gratefully acknowledge Dr. David Burr for toluidine staining advice, and Mike Orlando, Dr. Mike Vaughan, and Dr. Hal Black for providing black bear femurs from Florida, West Virginia, and Utah, respectively.