Spatial control of perilacunar canalicular remodeling during lactation

Osteocytes locally remodel their surrounding tissue through perilacunar canalicular remodeling (PLR). During lactation, osteocytes remove minerals to satisfy the metabolic demand, resulting in increased lacunar volume, quantifiable with synchrotron X-ray radiation micro-tomography (SRµCT). Although the effects of lactation on PLR are well-studied, it remains unclear whether PLR occurs uniformly throughout the bone and what mechanisms prevent PLR from undermining bone quality. We used SRµCT imaging to conduct an in-depth spatial analysis of the impact of lactation and osteocyte-intrinsic MMP13 deletion on PLR in murine bone. We found larger lacunae undergoing PLR are located near canals in the mid-cortex or endosteum. We show lactation-induced hypomineralization occurs 14 µm away from lacunar edges, past a hypermineralized barrier. Our findings reveal that osteocyte-intrinsic MMP13 is crucial for lactation-induced PLR near lacunae in the mid-cortex but not for whole-bone resorption. This research highlights the spatial control of PLR on mineral distribution during lactation.


• Standard Micro-computed Tomography Measurements
• Synchrotron Radiation Micro-computed Tomography Measurements

• Syris Phase Simulation
Quantitative RT-qPCR Gene expression data were obtained from harvested humeri bone from 16-week-old mice, cleaned of any muscle and periosteum.The epiphysis was removed, and the bone marrow was removed via centrifugation.Humeri bones were then snap-frozen in liquid nitrogen and homogenized with an Omni homogenizer in Qiazol Lysis Reagent (Qiagen, Cat # 79306) as previously described [1,2,3].RNA isolation was performed using the RNAeasy Mini Kit (Qiagen, Cat # 74106) following the manufacturer's instructions.Quantification of the extracted RNA was obtained using a Nanodrop spectrophotometer (Thermo Fisher Scientific), and 3ug of RNA was reversed-transcribed using an iScript cDNA synthesis kit (BioRad, Cat # 1708891) following the manufacturer's instruction.For reverse transcription quantitative real-time PCR (RT-qPCR) reactions, 60ng of cDNA from n=6-8 mice/group was performed using the primer sequences in Table S1.Technical replicates were used in each sample (n=6-8 mice/group).Analysis of RT-qPCR was quantified using the comparative C(T) method [4,5], also known as the DDCt method, to obtain the fold change and normalized to the housekeeping gene, beta-actin.Table S1: Primer sequences for RT-qPCR analysis

Immunofluorescence
Right femurs of 16-week-old mice were dissected, fixed in 10% neutral buffered formalin (NBF), decalcified in 10% disodium and tetrasodium EDTA, followed by serial ethanol dehydration and paraffin-embedded.Paraffin sections (6µm) in the axial plane across the mid-cortical region of the femur were stained with primary antibody for MMP13 (1:20, ab39012) with Alexa Fluor 594 Tyramide SuperBoost Kit (Invitrogen, Cat #: B40925) following the manufacturer's instructions.Briefly, paraffin sections were deparaffinized, rehydrated, and incubated in Uni-Trieve (Innovex Biosciences, Cat #: NB325) at 65°C for 30 mins, blocked with 10% goat serum for 1hr at room temperature, and incubated with primary antibody for MMP13 or Rabbit IgG (ab172730) as a negative control for overnight at room temperature.Secondary goat anti-rabbit IgG (provided in the Alexa Fluor 594 Tyramide SuperBoost Kit) was incubated for 1-2 hrs at room temperature and detected using Alexa Fluor Tyramide working solution for 10 mins at room temperature.Slides were mounted with Prolong Gold antifade Mountant with DAPI (Invitrogen, Cat #: P36935) and imaged using Leica DM2500.Four regions of each 2-4 sections were imaged at 40X from every n=2-3 mice/group.

Synchrotron Radiation Micro-computed Tomography Measurements
To further examine bone resorption due to lactation, we measured the endocortical perimeter (Ec.Pm) and the periosteal perimeter (Ps.Pm) for each bone using Dragonlfy's Bone Analysis Tool.Due to lactation, we observed a 16% increase Ec.Pm in the control (WT) group (p=0.18) and a significant 31% increase in Ec.Pm in the MMP13 ocy-/-group.No significant changes were detected in the Ps.Pm as a result of lactation.These results confirm that cortical bone resorption during lactation occurs at the bone's endocortical region [6].

Syris Phase Simulation
During the reconstruction of the synchrotron X-ray radiation micro-computed tomography (SRµCT) images, we utilized phase retrieval to retrieve the absorption component of the image and minimize the impact of phase at material interfaces.The effectiveness of the phase retrieval was confirmed through phase simulation using Syris, the synchrotron radiation imaging simulation repository [7] (Figure S4).To simulate the phase contribution in bone, we first took a reconstructed SRµCT image and binarized it using simple thresholding.Using the binarized image, we obtained image projections using Tomopy [8].Phase was simulated in these projections using Syris.Before phase propagation, each projection was converted to a static body that would mimic the material properties of hydroxyapatite using Syris.These projections with a phase contribution were then reconstructed using Tomopy.Images were reconstructed with and without phase retrieval to analyze the effectiveness of phase retrieval in reducing the phase component of the image.Paganin phase retrieval was performed using the phaseCT repository from Forien et al. [9,10].The absorption (β) and phase (δ) coefficients for hydroxyapatite were used for this phase retrieval.The β and δ values used were 1.48130805E-08 and 2.01842317E-06, respectively.We found that Syris closely modeled the real phase contribution in bone and that the Paganin phase retrieval implemented by Forien et al. [9,10] effectively minimized this phase contribution.Due to these results, we performed localized mineralization measurements in the bone surrounding the lacunae.

Figure S3 :
Figure S3: Measurements of cortical tibia bone surface perimeter using SRµCT images.a) Measurements of the endocortical perimeter (Ec.Pm).b) Measurements of the periosteal perimeter (Ps.Pm).

Figure S4 :
Figure S4: Phase simulation results using Syris.a) Mineral profiles computed using a 3D distance map away from the lacunae for the original image, the simulated phase image, and the simulated phase retrieval image.b) Representative image slices showing the original image, the simulated phase image, and the simulated phase retrieval image.