Fig 1.
a) Experimental setup for non-invasive loading of left C57BL/6J mouse tibia and the right nonloaded control tibia; b) sample preparation for imaging of the osteocyte lacuno-canalicular network (LCN) with confocal laser scanning microscopy (CLSM) (left to right: tibia extracted and dissected from the subject, sectioned along the sagittal plane, stained with rhodamine, and embedded in PMMA); c) single-slice CLSM image using two isolated signal channels (red: rhodamine; green: calcein) and their fused image (right) of a loaded posterior cortex at the mid-shaft region, showing the osteocyte LCN in the region of more mature bone (between the 1st set of calcein labels) and the osteocyte LCN in the regions of newly formed bone (between the 1st and 2nd set of calcein labels); d) 3D rendering of scanned osteocyte LCN in the volume and masking for different regions by using calcein double labels in 3D (orange: intercortical; blue: newly formed bone at the periosteal surface; pink: newly formed bone at the endosteal surface); e) an example of obtaining mathematical network description in a sub-volume with an osteocyte lacuna (beige almond-shaped object at the bottom left), by segmentation using adaptive thresholding to separate lacune and canaliculi, skeletonization and polynomial spline fitting to convert pixel-based data into a connectome network.
Fig 2.
Projected confocal laser scanning microscopy (CLSM) stacks to show two representative data sets of the cortices under the loaded condition (a) and the non-loaded condition (b).
Fig 3.
a) Box and whisker plot (indicating the median, 25% and 75% percentiles, and range of the variations in the sub-volumes) of canalicular densities (Can.Dn) from anterior and posterior cortices in the loaded (left) and non-loaded (right) tibiae from the three mice (M1-M3). b) Average canalicular densities (Can.Dn) evaluated in the newly formed regions (AE: Anterior Endosteal, PE: Posterior Endosteal; PP: Posterior Periosteal) and the pre-existing regions (Mid: Mid cortex). Note, we did not analyse the PP in the non-loaded tibia, nor the anterior posterior (AP) region in either the loaded or non-loaded tibia due to the limited amount of newly formed bone present.
Fig 4.
Colour maps of the canalicular density (Can.Dn) in the sagittal cross-sections of a) the anterior (partitioned into 3188 cubic sub-volumes in 3D with a side length of 5 μm) and b) the posterior (3714 cubic sub-volumes) cortices of the loaded tibia (representative example from mouse M1).
Dark red corresponds to regions of highest canalicular density and dark blue indicates regions of the lowest density, according to the colour bar on top. Black dashed lines are defined by the first set of calcein labels (inner), to separate the newly formed bone (semi-transparent pixels) to the pre-existing region (solid pixels).
Fig 5.
Loaded (blue) and non-loaded control (orange) spatial profiles of the canalicular density (Can.Dn) determined from thousands of sub-volumes, along the normalized length of the bone cortex from the endosteal to the periosteal region, with corresponding standard deviations in the shaded areas for all three mice (M1-M3).
Anterior Periosteal (AP), Anterior Endosteal (AE), Posterior Endosteal (PE) and Posterior Periosteal (PP) specify the orientations. Green dashed lines indicate the locations of calcein labels and, therefore, indicate the boundary to newly formed bone.
Fig 6.
Normalized frequency distributions of a) canalicular length and b) nodal degree of connectivity in a representative loaded mouse tibia. Green boxes highlight more than 70% of the canalicular lengths and more than 50% of the nodal connectivity from the rest in their distributions. Scatter plot of the percentages of long canaliculi (c, d) and tree-like (DoC = 3) nodes (e, f) as a function of the network density in the corresponding anterior (left) and posterior cortex (right), respectively. Blue circles correspond to a sub-volume in the pre-existing intercortical bone, fitted with a dashed line using linear regression. Red circles correspond to sub-volumes in the newly formed tissue. Solid lines are the result of a linear regression and should only serve as a guide to the eye.