Assessing the bone-healing potential of bone marrow mesenchymal stem cells in jawbone osteoporosis in albino rats

Background. Osteoporosis is one of the most common yet difficult to treat diseases. It affects millions of people and costs the health care systems billions worldwide. All of the available kinds of pharmacological treatment have multiple side effects, which is why a need for safer treatment options has emerged. Objectives. This study aimed to assess the bone-healing potential of bone marrow mesenchymal stem cells (BM-MSCs) in jawbone osteoporosis in Wistar albino rats.


Introduction
Osteoporosis is a "progressive skeletal systemic dis ease characterized by reduced bone mass and micro architectural deterioration, with a subsequent rise in bone fragility and fracture susceptibility". 1 It is the most com mon bone disease in humans, representing a major public health problem worldwide. 2 Osteoporosis causes the loss of bone mineral density (BMD) throughout the body, including the maxillary and mandibular bones. This reduced jawbone density leads to the increased porosity and accelerated resorption of the alveolar bone. Since the alveolar processes provide the teeth with support, their reduced bone density can have negative consequences on tooth stability. 3 Glucocorticoidinduced osteoporosis may happen in 30-50% of patients on glucocorticoid therapy. It is the main iatrogenic cause of secondary osteoporosis. Gluco corticoids increase the production of macrophage colony stimulating factor (MCSF) and receptor activator of nu clear factor kappabeta ligand (RANKL), and decrease the production of osteoprotegerin (OPG), subsequently increasing the number and activity of osteoclasts. 4 The pharmacological treatment of osteoporosis has been proven to have many side effects. Oral bisphosphonates, such as ibandronate, alendronate and risedronate, can cause up per gastrointestinal irritation and flulike symptoms. 5,6 Intra venous aminobisphosphonates, such as ibandronate and zoledronate, are potentially nephrotoxic, and can cause hypo calcemia and atrial fibrillation. They also pose an increased risk of heart failure, which is mostly due to the increased inci dence of hyperlipidemia, hypertension and peripheral artery disease among these patients. [5][6][7] Raloxifene can cause venous thromboembolism, whereas teriparatide has triggered con cerns about osteosarcoma in animal studies. 5 Mesenchymal stem cells (MSCs) obtained from bone mar row (BM) can be separated easily, have the ability to self renew, proliferate and differentiate into multilineages, retain the 'homing' feature, and are characterized by long storage without major loss of potency, thus making them the first op tion in repairing bone and treating related diseases. 8,9 Bone regeneration is a complex process involving the interrelation between adipogenic and osteogenic progeni tor cells, which are both derived from BM. 10 It has been confirmed that BMMSCs can be differentiated into os teoblasts and secrete many osteogenic factors after being cultured in vitro. They have also been proven to have great potential in bone and soft tissue repair in vivo. Moreover, they can migrate to the site of injury, creating an appro priate microenvironment for tissue repair. 11,12 In this context, this study aimed to assess the bone heal ing potential of BMMSCs in glucocorticoidinduced jaw bone osteoporosis by evaluating BMD as a primary out come in addition to the gene expression of RANKL and OPG, histopathological alterations, and the histomorpho metric analysis of the jawbones as secondary outcomes.

Material and methods
This experiment was conducted at the animal house at the Faculty of Medicine of Cairo University, Egypt, after obtaining the approval of the Institutional Animal Care and Use Committee (IACUC) (CUIIIF7317).

Isolation and culture of BM-MSCs
Bone marrow was flushed out of the tibias of male albino rats with an approximate weight of 100-120 g and age of 6 weeks by using phosphate buffered saline (PBS) (Invitrogen, Grand Island, USA) and centrifuged at 1,000 rpm for 5 min. A total of 35 mL of the flushed BM cells was layered over 15 mL FicollPaque™ (Invitrogen) and centrifuged at 400 × g for 35 min. The upper layer was discarded, leaving a mono nuclear cell (MNC) layer at the interphase. This MNC layer was collected, washed twice in PBS, and centrifuged at 200 × g for 10 min at 10°C. The isolated BMMSCs were cultured and propagated in 25milliliter culture flasks with RPMI1640 (Merck, Darmstadt, Germany) supplemented with 10% fetal bovine serum (FBS) (Thermo Fisher Scientific, Waltham, USA), 0.5% penicillin (Thermo Fisher Scientific) and streptomycin (Thermo Fisher Scientific). It was subse quently incubated at 5% CO 2 and 37°C until reaching 80-90% confluence within 14 days of culture. 13

Identification of BM-MSCs in the culture
Bone marrow MSCs were characterized in accordance with the International Society for Cellular Therapy guide lines 14 in terms of their morphology, adherence, fluores cenceactivated cell sorting (FACS) by assessing positivity for CD90 + , CD105 + and CD73 + , and negativity for CD14 − , CD34 − and CD45 − , and their capability to differentiate into osteoblasts, adipocytes and chondroblasts. The dif ferentiation of BMMSCs into ostoblasts was performed using a StemPro™ osteogenesis differentiation kit (Life Technologies, Carlsbad, USA); the cells were stained with the Alizarin Red S stain (SigmaAldrich, St. Louis, USA). Adipocyte differentiation was achieved with a StemPro™ adipogenesis differentiation kit (Life Technologies), and the cells were subsequently stained with the Oil Red O stain (SigmaAldrich). Chondroblast differentiation was performed using a StemPro™ chondrogenesis differen tiation kit (Life Technologies), after which the cells were stained with the Alcian Blue stain (SigmaAldrich).

Sample size
The sample size was calculated using the G*power software (https://www.psychologie.hhu.de/arbeitsgruppen/allgeme inepsychologieundarbeitspsychologie/gpower). In terms of the primary outcome (BMD) based on Uejima et al., the results revealed 124.32 ±4.88 mg/cm 2 for the control bones and 139.35 ±4.63 mg/cm 2 for the BMMSCsinjected bones 15 ; it was found that 3 rats per group with a total sam ple size of 12 rats was the appropriate sample size for the study (2 groups per each duration of 2 and 6 weeks). The power was 80% and the α error probability 0.05. The num ber was increased to 4 rats in each group to compensate for possible losses during the experiment for a total of 16 rats (4 × 4). The magnitude of the effect to be detected was es timated as mean and standard deviation (M ±SD) for the variable of interest, and was obtained from the scientific literature (Uejima et al. 15 ).

Experimental animals
A total of 16 healthy male Wistar albino rats (Rattus norvegicus) with an approximate weight of 150-200 g and age of 3-4 months were obtained from the animal house at the Faculty of Medicine of Cairo University. The animals were housed in a controlled, sterile environment (tempera ture 23 ±5°C and 12hour dark/light cycles). They had free access to a standard pellet diet and tap water ad libitum. They were maintained individually in stainless steel cages and kept under good ventilation. Each individual rat was considered an experimental unit within this study.

Study design
The animals were randomly divided using the Random Sequence Generator program (https://www.random.org/). Each animal was assigned a temporary random number. The rats were divided into 2 groups of 8 rats each (4 for each sacrifice date) as follows: -the osteoporotic group consisted of 8 glucocorticoid induced osteoporotic rats; these rats received no treat ment and served as a positive control group; -the BMMSCs group consisted of 8 glucocorticoid induced osteoporotic rats; these rats received a single intravenous injection of 50 million cultured BMMSCs in PBS through the tail vein. 16

Experimental procedure
The experiment was held at the animal house at the Faculty of Medicine of Cairo University. Four investiga tors were involved for each animal; the 1 st investigator was the only one aware of group allocation. During the experimental procedure, the 2 nd investigator was the only person to assess dosage administration for all animals. The 3 rd investigator assessed the outcomes in a masked fashion, without knowing the groups. The statistician performed data analysis and was unaware of group al location. All rats were intraperitoneally injected with 200 µg/100 g dexamethasone (Amriya Pharmaceuticals, Alexandria, Egypt) once daily for 30 days to induce os teoporosis. 17 Bone mineral density was measured at the jawbone to confirm that the model was successfully estab lished. Both the rats in the BMMSCs group and the rats in the untreated osteoporotic group were sacrificed 2 and 6 weeks after the introduction of treatment. The animals were sacrificed using an intraperitoneal overdose injec tion of an anesthetic mixture (ketamine/xylazine) (Sigma Aldrich). 18 Each lower jaw was dissected and processed for the assessment of the experimental outcomes.

Experimental outcomes
Dual-energy X-ray absorptiometry (DEXA) (primary outcome) The BMD of the mandibles was measured using dual energy Xray absorptiometry (DEXA). A Norland XR46 DXA scanner (Norland Corp., Fort Atkinson, USA), equipped with the appropriate software for bone assess ment, was utilized. The scan resolution was 0.5 × 0.5 mm and the scan speed was 60 mm/s. The analysis was carried out based on the image of the animal's jawbone on the screen, using a region of interest (ROI), which was de fined to include the bone area below the first molar. The results were displayed automatically in g/cm 2 . 19

Real-time polymerase chain reaction (RT-PCR) (secondary outcome)
Total RNA was extracted for further realtime poly merase chain reaction (RTPCR) analysis. RNA was iso lated with an RNeasy Micro Kit (Cat. No./ID: 74004; Qiagen, Hilden, Germany). A total of 20 ng of the isolated total RNA was used to develop cDNA by means of a high capacity cDNA Reverse Transcription Kit (Cat. No: 4368814; Applied Biosystems, Foster City, USA). The PCR analysis was performed using a SYBR ® Green Master Mix kit (Cat. No: 4344463; Applied Biosystems) with the StepOnePlus™ RealTime PCR System (Applied Biosystems), according to standard protocols. Briefly, the RTPCR thermal profile was programmed as follows: 10 min at 45°C for reverse transcription; 2 min at 98°C for the inactivation of reverse transcriptase (RT); and initial denaturation with 40 cycles of 10 s at 98°C, 10 s at 55°C and 30 s at 72°C for the am plification step. The relative quantity (RQ) values for each target gene were measured according to the calculation of ΔΔCt. The calculation of the RQ values for the studied genes was performed by 2 −ΔΔCt normalization to the house keeping gene GADPH. The primers for RANKL, OPG and GADPH are shown in Table 1.

Histopathological examination (secondary outcome)
The specimens were decalcified for 4 weeks, dehydrat ed in ascending grades of alcohol, cleared in xylol, and embedded in paraffin blocks. Serial sections of a thick ness of 5-6 µm were cut, mounted on glass slides, and stained with hematoxylin and eosin (H&E) for a routine histopathological examination.

Histomorphometric analysis (secondary outcome)
The area percentage of bone in the region below the first molar for each specimen was measured. The data was ob tained using a Leica Qwin 500 image analyzer computer system (Wetzlar, Germany). The area and area percentage of bone trabeculae were measured with an objective lens magnification of ×20 (a total magnification of ×200). Five fields were measured for each specimen. The bone area percentage was calculated in relation to a standard mea suring frame having an area of 118,476.6 μm 2 .

Statistical analysis
The data was coded and processed using the IBM SPSS Statistics for Windows software, v. 24.0 (IBM Corp., Armonk, USA). The data was presented as M ±SD. The Kolmogorov-Smirnov test showed that the data was normally distributed; thus, comparisons between the 2 groups were performed using the independent t test. A pvalue ≤0.05 was considered statistically significant.

Morphological characterization of BM-MSCs
The morphology of BMMSCs was observed under an inverted microscope (Invitrogen, Waltham, USA). Bone marrow MSCs were expanded successfully and adhered to culture flasks, showing a heterogenous population, displaying fibroblastlike morphology and forming a confluent monolayer at 14 days of culture ( Fig. 1(I)A). Bone marrow MSCs were differentiated into osteoblasts, stained with Alizarin Red S ( Fig. 1(І)B), adipocytes, stained with Oil Red O (Fig. 1(І)C), and chondroblasts, stained with Alcian Blue (Fig. 1(І)D).

DEXA results
A significant increase in BMD was observed in the BMMSCs group as compared to the osteoporotic group at the 1 st and 2 nd sacrifice times (p < 0.001) ( Table 2).

RT-PCR results
At the 1 st sacrifice time there was no significant difference between the osteoporotic and BMMSCs groups in terms of gene expression (p > 0.05). However, at the 2 nd sac rifice time there was a significant decrease in RANKL and a sig nificant increase in OPG gene expression in the BMMSCs group as compared to the osteoporotic group (p < 0.001) ( Table 2).

RANKL/OPG ratio in the study groups
At the 1 st sacrifice time there was no significant differ ence between the osteoporotic and BMMSCs groups in terms of RANKL/OPG ratio (p > 0.05). However, at the 2 nd sacrifice time there was a significant decrease in the RANKL/OPG ratio in the BMMSCs group as compared to the osteoporotic group (p < 0.001) ( Table 2).

Histopathological results
The histopathological examination of the alveolar bone in the osteoporotic group at the 1 st sacrifice time showed clear signs of osteoporosis, such as multiple osteoporotic cavities, scalloped resorptive pits and widened BM cavities surrounded by thin bone trabeculae. Bone marrow cavi ties showed areas of extravasated red blood cells (RBCs). Cellular degeneration was demonstrated by the diminished osteoblastic lining of BM cavities, the presence of shrunken osteocytes, empty lacunae, and bone areas free of osteo cytes. Multinucleated osteoclasts in Howship's lacunae and reversal lines were also observed (Fig. 2).
The histopathological examination of the alveolar bone in the BMMSCs group at the 1 st sacrifice time revealed some signs of healing, even though BM cavities were still wide. Bone lamellae showed a more homogenous and better organized architecture. Chronic inflammatory cell infil tration was detected. In addition, congested and dilated blood vessels were present. Some of widened BM cavities regained their osteoblastic lining, although they were still surrounded by thin bone trabeculae. A few osteoporotic and resorptive cavities were still observed. Areas of fatty degeneration were spotted where BM cavities contained multiple adipocytic cells. Regions of newly formed woven bone were noticed with disorganized fibers and osteo cytes that were rounder, larger and less regularly spaced than the ones in the lamellar bone. Areas of parallel fi bered bone were also found. It appeared as a transitional  4). BM-MSCs -bone marrow mesenchymal stem cells; BMD -bone mineral density; AU -arbitrary unit (relative expression); * statistically significant. bone tissue between the newly formed woven bone and the lamellar bone (Fig. 3). The histopathological examination of the alveolar bone in the osteoporotic group at the 2 nd sacrifice time pre sented augmented signs of osteoporosis. Marrow spaces exhibited diminished osteoblastic lining. Wide BM cavi ties revealed some fatty degeneration with the presence of multiple chronic inflammatory cells. Multiple osteopo rotic cavities and malorganized trabecular lamellae were detected. Osteocytes shrunken in their lacunae and some other empty lacunae were observed. Bone resting lines were also noticed (Fig. 4).
The histopathological examination of the alveolar bone in the BMMSCs group at the 2 nd sacrifice time displayed a great extent of bone healing. Bone marrow cavities returned to their normal size and regained their osteoblastic lining. Thick trabecular bone areas containing normal healthy osteocytes of normal size and shape, and resting in their lacunae were noticed. Bone resting lines surrounding marrow spaces were observed. Vascular spaces were noticed near the periodontal ligaments (PDLs). Some vascular spaces turned into second ary osteons surrounded by reversal lines. In addition, scal loped bone reversal lines were noticed, demonstrating the active bone formation and healing process (Fig. 5).

Discussion
The seriousness of osteoporosis lies in it being a totally asymptomatic illness that often remains undiagnosed un til it is manifested as a fracture. 2 All of the available kinds of pharmacological treatment are accompanied by mul tiple side effects, which is why a need for safer treatment options has emerged. 20 Hence, this study aimed to assess the bone healing potential of BMMSCs in glucocorticoid induced jawbone osteoporosis.
In the current work, osteoporosis was examined in the mandibular jawbones of adult rats. This is in accor dance with Jonasson and Rythén, who stated that the rate of bone turnover in the mandibular alveolar processes might be the fastest in the body; therefore, the initial signs of osteoporosis could be revealed there the earliest. 21 During the study period, BMD significantly deteriorat ed in the osteoporotic group. This finding is consistent with Cao et al., who reported that BMD significantly de creased in osteoporotic rats. 22 At the 2 nd sacrifice time, RANKL gene expression was significantly elevated in the osteoporotic group. This is in agreement with EghbaliFatourechi et al., who proved a significant upregulation of RANKL in osteoporotic postmenopausal women. 23 On the other hand, OPG gene expression was significantly decreased. This is in accor dance with a clinical study in which the delayed phase of osteoporosis caused OPG levels to decrease. 24 In the current investigation, the histopathological changes in the osteoporotic group became more notice able at the 2 nd sacrifice time. These observations resemble the findings of previous authors, who observed marked histopathological alterations in the mandibular alveolar bone spongiosa of osteoporotic rats. 17,25 Moreover, the great increase in BM fat content observed at the 2 nd sac rifice time was demonstrated by former studies, which reported that an increase in circulating glucocorticoids caused fatty tissue infiltration and the expansion of BM adipose tissue. 26,27 In this study, the histomorphometric analysis con firmed the histopathological architectural changes in the osteoporotic group. This coincides with the findings of Abuohashish et al., who determined that the morpho metric parameters of femur bones were significantly de creased in osteoporotic rats. 28 In the present work, after injecting the osteoporotic rats with BMMSCs, BMD significantly increased along the 2 sacrifice periods. This is quite similar to the observa tions of Hsiao et al., who confirmed that BMMSCs mi grated to bone after an intravenous injection, leading to an increased BMD and restored bone volume in an ovari ectomized mouse model. 29 This might be attributed to the ability of the transplanted MSCs to migrate to the site of injury and produce immunomodulatory cytokines and growth factors, helping local cells to recover and inducing the recruitment of new cells in the area. 8 In the BMMSCs group, RANKL gene expression sig nificantly decreased at the 2 nd sacrifice time as compared to the osteoporotic group. These results are supported by Li et al., who found BMMSCs to alleviate the symptoms of arthritis mainly by decreasing the levels of RANKL gene expression. 30 On the other hand, the level of OPG gene expression showed a significant increase. This coin cides with the results of Oshita et al., who reported that human MSCs produced OPG continuously, at both the mRNA and protein levels, which inhibited the osteoclas togenesis process. 31 In the current work, the histopathological findings in the BMMSCs group revealed gradual improvement between the 1 st and 2 nd sacrifice times. This is in agreement with Freitas et al., who demonstrated that the newly formed bone was observed 4 weeks after the injection of BMMSCs in rat calvarial defects. 32 This is also supported by Kim et al., who proved that MSCs reduced the progression of senile osteoporosis by sustaining osteocalcin levels in the circula tion, which resulted in improved bone microarchitecture. 33 Concerning the histomorphometric analysis, the BMMSCs group showed a significant progressive in crease in bone volume at the 2 nd sacrifice time. This is in agreement with de Melo Ocarino et al., who found that the greatest bone volume was achieved 2 months after inject ing the osteoporotic rats with differentiated BMMSCs. 34 Fig. 5. Photomicrographs of the BM-MSCs group at the 2 nd sacrifice time A -alveolar marrow spaces with a complete osteoblastic lining (bm), resting lines (arrows), vascular spaces (dotted arrows), secondary osteons (dotted circles), and normal periodontal ligaments (PDL) (×100 magnification); B -BM cavities with an osteoblastic lining (bm), osteocytes of normal size and shape (asterisks), vascular channels (dotted arrows), a resting line (arrow), reversal lines (arrow heads), and a normal periodontal ligament (PDL) (×200 magnification); C -normal osteocytes in their lacunae (asterisks) and bone reversal lines (arrows) denoting newly formed bone matrix (star) (×1,000 magnification).
The osteoporosis model should be further studied on animals with bone morphology closer to humans, as os teogenic healing in rats far exceeds that of a human. Also, more research techniques are needed to fully understand the healing mechanism of BMMSCs in the treatment of osteoporosis. Furthermore, more clinical trials should be conducted to determine the proper effective human dosage of BMMSCs in treating osteoporosis.

Conclusions
After assessing all the DEXA, RTPCR, histopathologi cal, and histomorphometric results, it was confirmed that BMMSCs had a positive effect on bone healing potential. It was also shown that the healing progress was achieved gradually along the experiment duration. Therefore, it can be concluded that BMMSCs could act as an effec tive treatment option for osteoporosis. However, further experiments with larger sample sizes are recommended to confirm these results.

Ethics approval and consent to participate
This experiment was conducted at the animal house at the Faculty of Medicine of Cairo University, Egypt, after obtaining the approval of the Institutional Animal Care and Use Committee (IACUC) (CUIIIF7317).

Data availability
All data generated and/or analyzed during this study is included in this published article.