Dried plum mitigates spinal cord injury‐induced bone loss in mice

Abstract Spinal cord injury (SCI) is accompanied by rapid loss of bone and increased risk of low impact fractures. Current pharmacological treatment approaches have proven to be relatively ineffective in preventing or treating bone loss after SCI. Dietary supplementation with dried plum (DP) has been shown to have dramatic effects on bone in various other disease models. In this study, we tested the efficacy of DP in preventing bone loss after SCI and restoring bone that has already been lost in response to SCI. Male C57BL/6J mice (3‐month‐old) underwent SCI and were fed a diet containing 25% DP by weight or a control diet for up to 4 weeks to assess whether DP can prevent bone loss. To determine whether DP could restore bone already lost due to SCI, mice were put on a control diet for 2 weeks (to allow bone loss) and then shifted to a DP supplemented diet for an additional 2 weeks. The skeletal responses to SCI and dietary supplementation with DP were assessed using microCT analysis, bone histomorphometry and strength testing. Dietary supplementation with DP completely prevented the loss of bone and bone strength induced by SCI in acutely injured mice. DP also could restore a fraction of the bone lost and attenuate the loss of bone strength after SCI. These results suggest that dietary supplementation with DP or factors derived from DP may prove to be an effective treatment for the loss of bone in patients with SCI.


| INTRODUCTION
Patients with spinal cord injuries (SCI) rapidly lose bone. 1,2 Bone formation rates can decrease as fast as 1% per week and sublesional bone mineral density (BMD) can decrease by as much as 40% in SCI patients. 3,4 Bone loss after SCI leads to increased risk of low impact fractures and significantly increases the morbidity and mortality of SCI patients. 5 It has been reported that as high as 52% of SCI patients are diagnosed as osteoporotic within 12 months of injury. 6 Fractures most often involve the tibia or fibula, but upper extremity fractures also occur, most commonly in those with higher cord lesions. Falls from a wheelchair and transfers are the most common causes of fracture, although fractures can also result from low-impact activities, such as performing range-of-motion activities. 3 Fracture healing is Xuhui Liu and Mengyao Liu made equal contributions to this work. more challenging for SCI patients compared to healthy patients. It is reported that as high as 83% of long bone fractures in SCI patients require operative management. 7 Pharmacological treatments for the loss of bone in SCI patients have been relatively ineffective. Vitamin D supplementation to restore vitamin D levels in individuals with vitamin D deficiency is commonly used but has not been effective in preventing and restoring the bone loss associated with SCI. 8 Despite some success, the effects of the bisphosphonates have been inconsistent. A recent meta-analysis containing 19 studies involving 364 patients showed that acute SCI participants treated with bisphosphonate therapy demonstrated a trend toward less bone loss. 9 Bauman et al. found that zoledronic acid did not prevent BMD loss at the knee in SCI patients. 10 Denosumab, a human monoclonal antibody to RANKL, has been shown to decrease bone turnover and a recent study showed that Denosumab decreases bone turnover markers in patients with recent SCI. 11 However, its long-term effect in preventing bone loss and restoring bone after SCI remain unknown. Teriparatide (recombinant PTH 1-34) has also been studied but the effects in patients with SCI are unclear. Teriparatide combined with robotically assisted gait training was evaluated in 12 chronically injured nonambulatory subjects with low bone mass.
Trabecular thickness increased at the knee at 3 months but not 6 or 12 months. BMD measurements after treatment at the spine and hips were not statistically significant. 12 In a recent larger clinical trial in patients with chronic SCI, Teriparatide treatment resulted in a significant increase in spine BMD at 1 year and improvements to the hip at 2 years. 13 Although some of these studies are promising, issues with patient compliance, adverse side effects, cost and long-term efficacy remain.
The use of specific foods and nutritional supplements has gained increasing attention as alternate treatment approaches. We have shown that dietary supplementation with dried plum (DP) can have dramatic effects on bone. [14][15][16][17][18] Dried plum diets can increase cancellous bone volume in adult and aged mice by 65% and 33%, respectively. Dried plum also improves bone strength and prevents bone loss and restores bone already lost due to estrogen deficiency in rats. 19 Although limited, studies in humans also suggest that dietary DP can increase BMD. 20 To determine whether dietary supplementation with DP can prevent the loss of bone induced by SCI, and restore bone that has already been lost in response to SCI, we fed mice a diet containing 25% DP in a mouse model of SCI.

| Animals
Male C57BL/6J mice (3 months of age) were obtained from the Jackson Laboratory (Sacramento, CA). The animals were housed in air-filtered, humidity-and temperature-controlled rooms with equal 12 hours light-12 hours dark cycles and fed a standard mouse diet before the study began. All mice went through a one-week acclimatization period after arriving at our facility before the experiments were conducted. In total, 53 mice were used in this study (including 45 with SCI and 8 without SCI). The detailed animal groups and numbers are listed in Table 1. The animal protocol for the study was in accordance with the NIH Guide for the Care and Use of Laboratory Animals and approved by the Animal Care and Use Committee at the Veterans Affairs Medical Center, San Francisco.

| Spinal Cord Injury
Mice underwent a dorsal midthoracic laminectomy followed with spinal cord contusion using a modified Allen weight-drop method as described previously. 21 The injury was induced by dropping a 35 g stainless steel rod onto the exposed spinal cord at the T10 level with a penetrating depth of 1.8 mm from a height of 50 mm, generating a complete paraplegia. Animals were under anesthesia with 1-4% isoflurane in oxygen inhalation during all procedures. In total, 45 mice underwent spinal cord contusion procedure. Six of them were excluded from the study due to incomplete SCI (attrition rate = 14.3%).
Thirty nine of the mice with complete SCI were used in this study.
Bupivacaine (0.25%, 0.1 mL) was injected to the incision site before surgery and Buprenorphine (0.1 mg/kg) was injected twice a day after surgery for the first two postoperational days.

| Plum Diet
DP was received as a kind gift from the California Dried Plum Board.
Mice were fed either a control diet (AIN-93 M) or a diet containing 25% DP by diet weight (AIN-93 M control diet containing 25% DP, w/w) for 2 or 4 weeks. The DP diet was selected based on its effectiveness to alter bone turnover in previous studies. 14,16 Control and experimental diets were formulated in a pellet form and contained an equal amount of energy, protein, fat, carbohydrate, calcium, phosphorus and other nutrients. Details of the diets are published elsewhere. 14 The mice have access to the food at all time in the cage.

| Animal protocols
Our objectives were to determine: 1) whether dietary supplementation with DP can prevent the loss of bone induced by SCI (prevention experiment) and 2) whether the bone lost after SCI can be replaced by switching to a DP diet (recovery experiment). To accomplish our first objective mice were divided into two groups (control diet and DP diet, N = 7 in control diet group and N = 8 in DP diet group) and fed their respective diets beginning immediately after SCI. in vivo microCT scanning was performed three times, at base line (immediately before SCI) and 2 and 4 weeks after SCI. Body weight was measured at the time of scanning. To accomplish our second objective we divided mice into 4 groups: group 1 was euthanized at baseline without SCI, group 2 underwent SCI and was fed the control diet for 2 weeks, group 3 underwent SCI and was fed the control diet for 4 weeks and group 4 underwent SCI and was fed the control diet for 2 weeks (to allow bone loss) and then switched to the DP diet for an additional 2 weeks.
To test our hypothesis, we compared the mice switched to the DP diet (group 4) to the control diet groups at 2 weeks (group 2) and 4 weeks (group 3). Comparing group 2 to group 4 allowed us to determine whether we could restore bone that had been lost. Comparing

| Mechanical Testing
After the frozen bones were thawed, three-point bending was used to assess bone strength in the femoral diaphysis using a Bose ElectroForce 3220 (Bose, Corp. Framingham, Massachusetts). The methods have been described previously. 22 The span between the left overhang and right overhang was 7.65 mm. The Maximum Load (N) and stiffness (N/mm) were recorded using the manufacturers' software.

| Bone Histomorphometry
The distal end of the left femur was processed undecalcified for quantitative bone histomorphometry as described previously. 23  One section per sample was quantified.

| Statistical analysis
Data are reported as mean ± SD with n = 7-8 mice per experimental group. A two-way ANOVA and a Holm-Sidak post hoc test were used in the prevention studies to compare groups. A Dunnet's test was used in the recovery studies to compare the switched diet group to T A B L E 1 Animal groups and numbers in this study

| RESULTS
Dietary supplementation with DP completely prevented the loss of bone and bone strength induced by SCI (Figures 1, 2, 3). MicroCT images at baseline, 2 and 4 weeks after SCI in mice fed either the DP diet or the control diet are shown in Figure 1. Mice fed the control diet lost 53% of their cancellous bone volume by 2 weeks after SCI.
Four weeks after SCI, the loss was 71%. No change in cancellous bone volume compared to baseline was detected in the mice fed the DP diet at either 2 or 4 weeks after SCI. Trabecular thickness and number and connectivity density decreased in the control diet mice but not in the DP diet mice. The SMI increased in the control mice but did not To determine whether DP could restore bone after it had already been lost from SCI, we compared mice after 2 and 4 weeks on the control diet to mice fed the control diet for 2 weeks and then switched to the DP diet for 2 more weeks. MicroCT images and quantitative analyses are shown in Figure 3. Switching to the DP diet at 2 weeks after SCI restored a fraction of the bone lost. Trabecular number, but not thickness was also restored in part. Switching to the DP diet also attenuated the changes in SMI, tissue density and con- In the prevention experiment, all mice lost weight following SCI.
The body weights of mice in the control diet group were 26.3 ± 1.5 g (baseline), 23.0 ± 1.9 g (2 weeks after SCI) and 21.5 ± 2.5 g (4 weeks after SCI). The body weights of mice in the plum diet group were 26.1 ± 1.7 g (baseline), 22.2 ± 1.8 g (2 weeks after SCI) and 20.8 ± 1.6 g (4 weeks after SCI). Two-way ANOVA analysis showed a significant effect of time (P < .01), but not diet. In the recovery experiment, the body weights of mice in each group were: 24.9 ± 1.2 g (baseline), 19.9 ± 1.2 g (2 weeks control diet), 22.2 ± 1.2 g (4 weeks control diet) and 22.9 ± 1.7 g (2 weeks control diet +2 weeks plum diet). Dunnet's test showed a significant difference in body weights (P < .01) between mice fed the control diet for 2 weeks and mice fed the control diet for 2 weeks and then switched to the DP diet. There was no difference in body weights between mice fed the control diet for 4 weeks and mice in the switched diet group.

| DISCUSSION
The results of our studies suggest that dietary supplementation with DP can completely prevent the loss of bone induced by SCI in a mouse model. The results show that the loss of bone following acute SCI is associated with trabecular thinning and a lower trabecular number. In DP fed mice, trabecular thickness and number were preserved. Preservation of bone mass and strength is critical in the acutely injured patient. In chronically injured patients who have already lost bone the goal is to prevent further loss or restore the bone lost.
F I G U R E 3 Top: Typical uCT images (cross section views in the upper row and sagittal section views in the lower row) of the distal femoral metaphysis at baseline and in mice fed a control diet or diet containing 25% DP for 2 or 4 weeks after SCI, and a group of mice fed the control diet for 2 weeks and then switched to the DP diet for 2 more weeks. DP restored a fraction of the bone lost and attenuated the changes in microarchitecture induced by SCI. Bottom: Quantification of diaphysis of femurs results. Cancellous bone volume (BV/TV) and micro-architecture (Tb.Th, Tb.N, Conn.D, SMI) from μCT analysis in the distal femoral metaphysis in mice fed the control diet for 2 or 4 weeks and mice fed the control diet for 2 weeks and then switched to the DP diet for 2 more weeks (mean with error bar for SD. n = 8 in each group, a p < .05 compared to mice on the control diet for 4 weeks and, b P < .05 compared to mice on the control diet for 2 weeks) Dietary supplementation with DP beginning 2 weeks after SCI and after a deficit of bone had been established (−40%), restored a fraction of the lost bone. This was associated with an increase in trabecular number but not trabecular thickness suggesting that the DP diet is increasing trabecular bridging but not adding new bone to trabecular surfaces. Importantly a DP diet can prevent bone loss when treatment is started after a bone deficit has occurred and may even be able to restore some lost bone but complete restoration is unlikely. Longer DP treatment periods, however, may promote further bone recovery.
Although DP supplemented diets can protect against bone lost following SCI, the mechanisms involved are poorly understood. Our previous findings suggest that in mature mice the primary cellular effector for the increase in bone volume is a decrease in osteoclastic number and bone resorption. 13 Consistent with this idea, DP has been shown to alter allocation of cells in the hematopoietic lineage and inhibit recruitment of osteoclast precursors into the osteoclast lineage. 24 It appears that the gain in bone occurs because of bone formation and resorption where the decrease in resorption is greater than the decrease in formation.
All mice in the prevention experiment lost weight, a finding consistent with previous experiments, but there was no difference in the amount of weight loss between control and DP fed mice. These observations suggest that the preservation of bone by DP supplementation was not a consequence of preservation of weight. Furthermore, body weights in the recovery experiment were not different between control diet mice at 4 weeks and mice switched to the DP diet. These results suggest that the fractional recovery of bone was not a consequence of an increase in body weight.
The bioactive factors in DP have not been identified. Previous studies suggest that the effects of DP on bone may be mediated through DP-specific polyphenols, although other nonphenolic compounds are likely to also be involved. 18,25 In DP fed mice the serum level of several polyphenols, including quercetin, increases and quercetin has been shown to decrease bone loss in both diabetic rats 26 F I G U R E 4 Typical histomorphology images. A, A low magnification overview of the femur indicating the region of interest. B-E, TRAP staining images (counter stained with toluidine blue) and, F-I, fluorochrome images of the distal femoral metaphysis at baseline (B, F) and in mice fed a control diet for 2 (C, G) or 4 weeks after SCI (D, H), and a group of mice fed the control diet for 2 weeks and then switched to the DP diet for 2 more weeks (E, I). Osteoclast surface was indicated with arrows and ovariectomized mice. 27 Curcumin, another polyphenol has been reported in both preclinical 28 31 Disruption of the microbiome, a process known as dysbiosis, after SCI has been considered as a disease-modifying factor 32 and fecal transplant prevented gut dysbiosis and anxiety-like behavior after spinal cord injury in rats. 33 However, the role of dysbiosis in SCI-induced osteopenia remains unclear.
Dried plum contains a complex of prebiotic factors and changes in the microbiome have been linked to changes in mineral metabolism. [34][35][36][37] Intestinal calcium bioavailability has also been shown to be associated with changes in short chain fatty acids (SCFA), 36 although dietary consumption of plum has been reported to have no effect on SCFA. 38 Other factors reported to affect bone, such as vitamin K (menaquinone), are produced in the intestine and synthesis is altered by the composition of the intestinal microbiota. 39,40 Collectively, these observations suggest that the effects of DP on bone in SCI patients may be mediated through changes in the gut microbiome.
In conclusion, dietary supplementation with DP can prevent bone loss in an acute model of SCI, and restore a fraction, but not all, of the bone that has already been lost in mice following SCI. A clinical trial treating SCI patients with DP is currently underway. If shown effective, dietary supplementation with DP or drugs derived from DP may prove to be an effective treatment for the loss of bone induced by SCI.  with three-point bending in mice of baseline, fed a control diet for 2 or 4 weeks, and mice fed with control diet for 2 weeks and then switched to the DP diet for 2 more weeks after SCI (mean with error bar for SD, n = 8 in each group, a P < .05 compared to mice on the control diet for 4 weeks)