Direct Assessment of Rabbit Cortical Bone Basic Multicellular Unit Longitudinal Erosion Rate: A 4D Synchrotron-Based Approach

Abstract

Cortical bone remodeling is carried out by basic multicellular units (BMUs), which couple resorption to formation. Although fluorochrome labeling has facilitated study of BMU formative parameters since the 1960s, some resorptive parameters, including the longitudinal erosion rate (LER), have remained beyond reach of direct measurement. Indeed, our only insights into this spatiotemporal parameter of BMU behavior come from classical studies that indirectly inferred LER. Here, we demonstrate a 4D in vivo method to directly measure LER through in-line phase contrast synchrotron imaging. The tibias of rabbits (n = 15) dosed daily with parathyroid hormone were first imaged in vivo (synchrotron micro-CT; day 15) and then ex vivo 14 days later (conventional micro-CT; day 29). Mean LER assessed by landmarking the co-registered scans was 23.69 ± 1.73 μm/d. This novel approach holds great promise for the direct study of the spatiotemporal coordination of bone remodeling, its role in diseases such as osteoporosis, as well as related treatments. © 2022 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

Keywords: BASIC MULTICELLULAR UNIT; BONE MICRO-CT; CORTICAL POROSITY; LONGITUDINAL EROSION RATE; SYNCHROTRON RADIATION.

Fig. 1

Basic multicellular unit (BMU) morphology and remodeling states. Schematic view of the BMU in cortical bone (top) and its relation to porous structures (remodeling space and persisting vascular canal = orange) in the balanced, imbalanced, and uncoupled states (bottom).

Fig. 2

Assessment of basic multicellular unit (BMU) longitudinal erosion rate (LER) in cortical bone. Classical indirect method for inferring BMU LER from fluorochrome labeling of the closing cone (top) and the novel proposed direct method of measuring advancement of the cutting cone using in vivo imaging and image registration (bottom).

Fig. 3

Longitudinal erosion rate (LER) assessment based on synchrotron radiation (SR) micro‐CT and micro‐CT co‐registered scans. (Top left) Rabbit skeletal schematic displaying region of the distal tibia scanned (transparent, orange square). (Bottom left) The in vivo SR micro‐CT scan (translucent orange render) at time 1 (T1, day 15) is co‐registered within the ex vivo micro‐CT scan (translucent, green render) from time 2 (T2, day 29). (Center) Cortical porosity rendered alone revealing (A) classically shaped basic multicellular unit (BMU)‐related resorption spaces, (B) unchanged vascular canals, and (C) complex irregular BMU‐related resorption spaces. (Right) Zoomed‐in views (A1, B, C) from central panel and (A2) a schematic depiction of the direct method of LER assessment where LER = D/T2‐T1, where D denotes the distance between the tips of the cutting cones (in μm) and T2‐T1 denotes the time difference between the two scans (ie, 14 days). Scale bar = 1 mm.

Fig. 4

Representative 2D images depicting tibias from different animals scanned in vivo (day 15) at low (1 Gy, A), medium (2.5 Gy, B), and high (5 Gy, C) radiation dose. A roughly matching (eg, raw image before any registration) ex vivo image of the same animal depicted in (C) is provided in (D). Large basic multicellular unit (BMU)‐related resorption spaces are evident in all four images. (E) A representative image from one of the animals that could not be successfully registered because of the extensive intracortical remodeling as well as endosteal bone formation. Subtractive images for another specimen calculated after registration are presented in (F) and (G), representing cross‐sectional and longitudinal sections, respectively. Bone formation and resorption between the two scans are represented as bright and dark shades, respectively. Structures in common appear in shades of gray because of slight variations in gray intensity between the scans. Endosteal bone formation is evident in both views and an advancing BMU with new resorption at the cutting cone and trailing bone formation of the closing cone is present in (G).

Fig. 5

Direct assessment of basic multicellular unit (BMU) longitudinal erosion rate (LER) across dose groups. Individual LER measurements pooled across animals and groups (n = 186) were normally distributed (left) and mean LER data by group (right) are presented by box plots with the mean values from individual rabbits plotted as solid circles (n = 4 rabbits/group). One‐way ANOVA revealed no differences between radiation dose groups (α = 0.05).

Fig. 6

Comparison of parameters among the in vivo (synchrotron radiation [SR] micro‐CT—day 15) and ex vivo (micro‐CT—day 29) scans of the right and contralateral left tibias for different radiation dose groups (low dose [1 Gy]; medium dose [2.5 Gy]; high dose [5 Gy]). Data are presented by box plots (right in vivo [light gray]; right ex vivo [white]; left ex vivo [dark gray]) with the mean (or median for nonparametric tests) for each individual rabbit plotted as solid circles and outliers plotted as open circles. n = 5. Repeated measures ANOVA (**p < 0.05) comparing tibia scans across the radiation dose groups was employed for normal data. Wilcoxon signed‐rank test comparing tibia scans manually categorized by radiation dose group (*p < 0.05) was employed for non‐normal data. Significance level was α = 0.05. 3D measures (top row): Ct.Po = Cortical Porosity, Ca.Dm = Canal Diameter, and Ct.Th = Cortical Thickness. 2D measures (bottom row) Tt.ar = Total Area, Ct.Ar = Cortical Area, and Ma.Ar = Marrow Area.

Fig. 7

Histomorphometric analyses of transverse cortical bone sections comparing the previously in vivo imaged right (white) and contralateral left (gray) tibias. Data are presented as box plots with the mean (or median for nonparametric tests) for each individual rabbit plotted as solid circles and outliers plotted as open circles. n = 7 except for On.MAR and Ac.f measures, where dL.On were not detectable in one rabbit from each radiation dose group (low dose, 1 Gy [n = 4]; medium dose, 2.5 Gy [n = 4]; high dose, 5 Gy [n = 4]). Repeated measures ANOVA (**p < 0.05) comparing left and right tibias across the radiation dose groups was employed for normal data. Wilcoxon signed‐rank test comparing the right and left tibias manually categorized by radiation dose group (*p < 0.05) was employed for non‐normal data. Significance level was α = 0.05. For canal diameter (Ca.Dm), the plots reveal bimodal distributions of Ca.Dm sizes for all three groupings. The left peaks in the distributions reveal a large percentage of the porous canals at or near the resolution of the scans. The right peaks around 80 to 100 (ex vivo) and 120 (in vivo) μm represent the diameters of basic multicellular unit (BMU)‐related remodeling spaces. sL.On = Single‐Labeled Osteon; dL.On = Double‐Labeled Osteon; Rs.N = Resorption Cavities; a.Rm.Cr = Active Remodeling Centers; Ct.Ar = Cortical Area; W.Th = Wall Thickness; On.MAR = Osteonal Mineral Apposition Rate; Ac.f = Activation Frequency.