European Journal of Nutrition

, Volume 53, Issue 5, pp 1155–1164

Pomegranate and its derivatives can improve bone health through decreased inflammation and oxidative stress in an animal model of postmenopausal osteoporosis

  • Mélanie Spilmont
  • Laurent Léotoing
  • Marie-Jeanne Davicco
  • Patrice Lebecque
  • Sylvie Mercier
  • Elisabeth Miot-Noirault
  • Paul Pilet
  • Laurent Rios
  • Yohann Wittrant
  • Véronique Coxam
Original Contribution

DOI: 10.1007/s00394-013-0615-6

Cite this article as:
Spilmont, M., Léotoing, L., Davicco, MJ. et al. Eur J Nutr (2014) 53: 1155. doi:10.1007/s00394-013-0615-6



Recently, nutritional and pharmaceutical benefits of pomegranate (PG) have raised a growing scientific interest. Since PG is endowed with anti-inflammatory and antioxidant activities, we hypothesized that it may have beneficial effects on osteoporosis.


We used ovariectomized (OVX) mice as a well-described model of postmenopausal osteoporosis to study the influence of PG consumption on bone health. Mice were divided into five groups as following: two control groups sham-operated and ovariectomized (OVX CT) mice fed a standard diet, versus three treated groups OVX mice given a modified diet from the AIN-93G diet, containing 5.7 % of PG lyophilized mashed totum (OVX PGt), or 9.6 % of PG fresh juice (OVX PGj) or 2.9 % of PG lyophilized mashed peel (OVX PGp).


As expected, ovariectomy was associated with a decreased femoral bone mineral density (BMD) and impaired bone micro-architecture parameters. Consumption of PGj, PGp, or PGt induced bone-sparing effects in those OVX mice, both on femoral BMD and bone micro-architecture parameters. In addition, PG (whatever the part) up-regulated osteoblast activity and decreased the expression of osteoclast markers, when compared to what was observed in OVX CT animals. Consistent with the data related to bone parameters, PG consumption elicited a lower expression of pro-inflammatory makers and of enzymes involved in ROS generation, whereas the expression of anti-inflammatory markers and anti-oxidant actors was enhanced.


These results indicate that all PG parts are effective in preventing the development of bone loss induced by ovariectomy in mice. Such an effect could be partially explained by an improved inflammatory and oxidative status.


Pomegranate Nutritional prevention Osteoporosis Animal model Inflammation Oxidative stress 


Pomegranate (Punica granatum L. Punicaceae), one of the oldest known edible fruits, represents a phytochemical reservoir with a high-potential medicinal value. This fruit, grown mainly in the Mediterranean region, has been used for centuries to treat many ailments such as parasitic and microbial infections, ulcers, diarrhea, hemorrhage, and dysentery [1, 2, 3]. This is why, over the past decade, the evaluation of nutritional and pharmaceutical benefits of pomegranate (PG) has raised a great scientific interest with more than 600 publications now available on the subject.

Pomegranate composition is very complex, and each part is built up with specific components. Indeed, the fruit can be divided in two parts: (1) the edible one, called the aril, comprising 78 % juice and 22 % seeds, which constitutes 52 % of the total fruit (w/w), and (2) the peel or pericarp, which is nonedible although it has been traditionally used in folk medicine [4]. A large array of phytochemicals has been identified in those two parts of PG, including polyphenolic compounds such as anthocyanins (cyaniding and delphinidin) in the juice and hydrolysable tannins (ellagic acid, punicalagin, gallagic acid) in the peel [4]. Actually, the main benefit of PG has been attributed to its unique polyphenols composition [5]. Indeed, PG polyphenols have been shown to exhibit high anti-oxidant and anti-inflammatory capacities interesting for the prevention of several age-related diseases [6]. In this light, the health benefits of PG consumption in preventing cancers [4] and cardiovascular diseases [7] have been widely focused. However, only a few studies have targeted the eventual benefice of PG consumption toward bone health [8, 9, 10], although in the present context of longer life expectancy, the prevalence of osteoporosis has been considerably increased. As a matter of fact, with more than 50 % of women and 20 % of men affected in the US population over 50-year old, osteoporosis, the most common bone disease, represents a major economic and public health issue [11], and this is true worldwide. This condition is characterized by a slow decline of bone mass and impaired micro-architecture leading to increased bone susceptibility to fracture [12, 13]. Pathology establishment involves disruption of formation/resorption balance through uncoupling osteoblasts and osteoclasts activities (Gallagher and Sai 2010; Kular, Tickner et al. 2012). Besides, the literature has linked inflammation and oxidative status establishment with bone alteration, osteoclast bone resorption being promoted while osteoblast-mediated formation is inhibited [14, 15].

Along with drugs ensuring bone homeostasis strategies, recent nutritional approaches have been developed and have revealed major interests for phenolic compounds such as flavonoids, anthocyanins, and tannins derived from vegetables and drinks (green tea, wine, etc.) [16, 17, 18]. In this context, we investigated the effect of PG consumption on bone status. So far, most of the studies targeting PG health benefits have been focused on juice or purified peel polyphenol extracts: ellagic acid [19], punicalagin [20], and anthocyanins [21]. In our hand, we investigated the in vivo (in a murine model of postmenopausal osteoporosis) potential benefits of PG, in the context of osteoporosis prevention through different nutritional approaches: (a) juice, (b) peel, and (c) whole fruit. The objective was to elucidate whether PG should be considered as one or as parts of new strategies to better contribute to bone health preservation.

Materials and methods


Pomegranates (the wonderful variety cultured in Israel) were purchased from POMONA (Clermont-Ferrand, France). Totum sample was prepared from the blended whole fruit (Philips HR2084, France) to obtain a homogenous mixture. Mashed peel was obtained following the same protocol as for the totum sample, separation of peel and arils being performed manually. Arils were squeezed, and juice was filtered through filter paper (Whatman filter paper Cat no.: 1001–090, Grade 1: 11 μm, GE healthcare). PG totum, juice, and peel used in the present study were analyzed for protein, free sugars, insoluble fiber, and polyphenols, see Table 1.
Table 1

Chemical composition of PG products


Chemical composition of PG totum, juice, and peel (g/100 g) of dry matter

Dry matter (g/100 g)

Free sugars

Insoluble fibers




20.787 ± 0.640

45.969 ± 1.557

24.767 ± 1.964

6.366 ± 0.476

0.678 ± 0.162


14.507 ± 0.995

92.141 ± 1.177

0.254 ± 0.030

2.926 ± 0.309

0.237 ± 0.195


22.680 ± 0.200

39.484 ± 2.390

30.003 ± 1.788

10.851 ± 1.016

0.808 ± 0.199

Values are presented as mean ± SD (means of 10 fruits, reproduced 3 times)

Totum and mashed peel were dried using a freeze drier (CHRIST, Gamma 1–20, Germany). The yield of the lyophilized powder of the PG totum was 21.2 and 25.4 % from PG peel. All samples were frozen at −20 °C, directly after preparation and until diet formulation.


All the animal procedures were approved by the institution’s animal welfare committee (Comité d’Ethique en Matière d’Expérimentation Animale Auvergne: CEMEAA) and were conducted in accordance with the European’s guidelines for the care and use of laboratory animals (2010-63UE). The animals were housed in the animal facilities of the Human Nutrition Unit at INRA Research Center ( (Agreement no. C6334514). Surgery was performed under anesthesia, and all efforts were made to minimize suffering.

Fifty-eight-week-old female C57BL/6J mice were purchased from JANVIER (St Berthevin, France). The animals were housed in a controlled environment (12:12 h light–dark cycle, at 20–22 °C, with 50–60 % relative humidity), 1 mice per plastic cage fed ad libitum and with free access to water. After an acclimatization period of 1 week, they were randomly divided into 5 groups (n = 10 per group). Two groups were surgically ovariectomized (OVX), while the other animals were sham-operated (SH), under anesthesia. The control mice (SH CT and OVX CT) were fed with a standard diet, modified from the AIN-93G powdered diet. PG groups were fed with the same standard diet, containing different amounts of PG parts. As far as the totum group is concerned (OVX PGt), 5.7 % of the PG lyophilized totum was added to the standard diet. With regards to the juice group (OVX PGj), 9.6 % of the PG fresh juice was added and 2.9 % of the PG lyophilized mashed peel was introduced into the diet for the peel group (OVX PGp). According to the human/mouse conversion principle [22], the experimental dose for PG juice (550 μL/mouse/day) in the present study was set equivalent to the corresponding nutritional dose of 250 mL for a 70-kg human subject. Then, the diet content of totum and peel from PG (nonedible parts) was calculated by considering the amount of fruit needed to achieve 550 μL of juice. In this light, daily quantity of PG given to the animals represented 1.4 g/mice/day of fresh totum (considering that juice = 37.5 % of the whole fruit) or 622 μg/mouse/day of fresh peel (based on the fact that peel = 42.4 % of the whole fruit). Diets were purchased from SAFE (Scientific Animal Food and Engineering, Augy, France).

Body weight was measured every 2 days during the study period. Besides, body composition was assessed at the beginning and at the end of the study, using a QMR EchoMRI-900™ system, without any anesthesia or sedation. After the 30-day treatment was completed, blood was withdrawn under anesthesia and centrifuged (3,000 rpm for 5 min at room temperature). Serum was frozen at −80 °C. Then, the mice were killed. Liver, spleen, and uterus were weighted. Femurs and tibias were harvested and stored at −80 °C prior to investigation.


Bone micro-architecture and bone mineral density analyses

After removing soft tissues, left femurs of the 14-week-old mice were placed in PBS buffer (Phosphate-Buffer Saline, Gibco) with 10 % formaldehyde at 4 °C for 1 week. Micro-architecture (secondary spongiosa) was analyzed using X-ray radiation micro-CT (SkyScan 1072). Pictures of 1,024 × 1,024 pixels were obtained at 37 kV and 215 μA. Bone morphological analysis was performed using an eXplore CT 120 scanner (GE Healthcare, Canada). Acquisitions were processed with X-ray tube settings at 100 kV and 50 mA.

Taqman low-density arrays (TLDA)

For each experimental group, four sets of extractions were performed with two tibias pooled in each. Frozen bones were ground in liquid nitrogen to obtain a fine powder. Then, total RNA was extracted from either bone powder using TRIzol reagent (Life Technology), following the manufacturer’s instructions. RNA was converted to cDNA using the high-capacity cDNA reverse transcription kit (Applied Biosystems). Resulting cDNA was used for TaqMan® low-density arrays (TLDAs) (Applied Biosystems 7900 HT real-time PCR system). Relative expression values were calculated using the comparative threshold cycle (2−ΔΔCT) according to Data Assist software (Applied Biosystems); 18 S, GAPDH, and actin served as housekeeping genes.

Statistical analysis

Results are expressed as means with their standard error (SEM). All the data were analyzed using XLSTAT (ExcelStat Pro software—Microsoft Office 2007) using a two-way analysis of variance (ANOVA) to test for difference among groups. If the result was found to be significant (p < 0.05), the fisher’s multiple comparison test was then used to determine specific difference between means.


Validation of the animal model

During the experimental period, mean body weight, as well as lean and fat mass, increased in all the experimental groups, which reflects the health of the animals (Fig. 1a). Furthermore, no significant difference was observed among groups, which means that neither castration (SH CT vs OVX CT) nor the different diets (OVX PG vs. OVX CT) have had an impact on weight change. Nevertheless, ovariectomy was associated with a significant decrease in uterine weight, in all the four OVX groups (Fig. 1b). This atrophy was not prevented by PG consumption (% change compared to the values measured in the SH CT group: −72.6 % in OVX CT mice, −80.1 % in OVX PGt mice, −78.1 % in OVX PGj mice, and −78.6 % in OVX PGp mice, p < 0.001). The fruit is thus devoid of any uterotrophic effect at this physiological level.
Fig. 1

Effect of ovariectomy on body weight, body composition, and uterine weight in mice under different diets. a Body weight and composition data. b Uterine weight data. Animals were fed a standard diet AIN-93 (SH and OVX control) or the same standard diet modified from AIN-93, containing 5.7 % of PG lyophilized mashed totum (OVX PGt), or 9.6 % of PG fresh juice (OVX PGj) or 2.9 % of PG lyophilized mashed peel (OVX PGp) for 30 days. n = 10 per group. Values are presented as mean ± SEM. ***p < 0.001 versus SH group

Consumption of pomegranate and its derivatives was associated with improvement of bone mineral density in OVX mice, the ultimate biomarker for bone health

The biological effect of ovariectomy and of the test systems has been studied was first investigated on bone morphological parameters, using micro-CT analysis. We thus confirmed the deleterious effect of ovariectomy on femoral bone mineral density (BMD), the animals in the OVX CT group having a lower BMD (p < 0.001), when compared to what was measured in the SH CT mice (−35.6 %) (Fig. 2a).
Fig. 2

Effect of consumption of PG or its derivatives on femoral bone parameters in OVX mice. a BMD analysis of left femurs: In sham-operated mice SH CT first bar, OVX mice given the control diet OVX CT second bar or a modified from AIN-93 containing 5.7 % of PG lyophilized mashed totum (OVX PGt third bar), or 9.6 % of PG fresh juice (OVX PGj fourth bar) or 2.9 % of PG lyophilized mashed peel (OVX PGp fifth bar). Each group contained 10 animals. Results are presented as mean ± SEM. **p < 0.01 and ***p < 0.001 versus OVX CT group. b, c Micro-CT analysis of left femurs in sham-operated (SH CT) or OVX animals given the control diet (OVX CT) or the diet, containing 5.7 % of PG lyophilized mashed totum (OVX PGt), or 9.6 % of PG fresh juice (OVX PGj) or 2.9 % of PG lyophilized mashed peel (OVX PGp) for 30 days (n = 6 per group). b Representative micro-CT images of the distal left femur for each group: SH CT, OVX CT, OVX PGt, OVX PGj, and OVX PGp. c Analysis of bone micro-architectural parameters. Values represent mean ± SEM. *p < 0.05 versus OVX CT group. BV/TV percent bone volume, TbN trabecular number, TbSP trabecular spacing, ConnDn connectivity density, Po tot total porosity, BS/TV bone surface density

Pomegranate consumption was able to counter such an osteopenic process induced by estrogen deficiency (% change in BMD values when compared to the OVX CT animals: PGt (+11.4 %), PGj (+27.1 %) (p < 0.001), and PGp (+16.2 %) (p < 0.05)).

Pomegranate was able to partially preserve bone micro-architecture in ovariectomized mice, but not its derivatives

Significant results were also observed on trabecular bone micro-architecture parameters of distal femur, as shown in Fig. 2b, c. As expected, ovariectomy significantly impaired bone micro-architecture, as shown by a decrease in trabecular number (TbN −25.0 %; p < 0.05), bone volume (BV/TV −23.6 %; p < 0.01), bone surface (BS/TV −23.9 %; p < 0.05), and connectivity (Conn Dn−19.7 %; p < 0.01) in the OVX CT, as compared with the SH CT one. Moreover, a significant increase in trabecular separation (TbSP +26.1 %; p < 0.001) and in total porosity (Po tot +3.5 %; p < 0.01) was observed as well. Consistently with BMD data, PG was endowed with bone-sparing properties as its consumption partially prevented (p < 0.01) all these micro-architectural deteriorations induced by OVX: BV/TV (+27.8 %), TbN (+26.7 %), Conn Dn (+24.6 %), BS/TV (+22.4 %), TbSP (−10.3 %), and Po tot (−3.0 %), when compared to what was measured in the OVX CT animals. With regards to PGj and PGp, only a trend toward improved biomarkers was observed, without reaching significant values.

Pomegranate was able to improve the expression profile of specific bone markers

To investigate the potential molecular mechanisms involved in the bone-sparing effect of PG, we performed transcriptomic analyses on bone tissue samples harvested from treated mice, using Taqman low-density arrays (TLDA). Due to the large amount of modulated genes, only results for the most important bone markers are presently given. Data from the OVX PG totum, juice, and peel groups were compared to those measured in the OVX CT animals. Consistently with data of bone phenotype, Fig. 3a indicates that ovariectomy (OVX CT) was associated with an up-regulation of two major osteoclastogenesis markers: the calcitonin receptor (CTR) for mature cells and the Fos for proliferating one (1.58-fold p < 0.001 and 1.61-fold, respectively, as compared to the SH CT group). In this context, PG consumption, regardless of the part of the fruit which was given, significantly counteracted this effect, as demonstrated by the decreased expression of CTR and Fos in all the groups fed the PG diet (PGt: 0.48-fold; PGj: 0.49-fold; PGp: 0.31-fold; p < 0.001 and PGt: 0.30-fold; PGj: 0.19-fold; PGp: 0.20-fold; p < 0.001, respectively, as compared to the OVX CT animals). Furthermore, the expression of the major integrin synthesized by mature and active osteoclasts (integrin β3; ITG β3) was significantly decreased in all the animals (PGt: 0.59-fold; PGj: 0.66-fold; PGp: 0.61-fold; p < 0.001). In the same way, the expression of key enzymes for osteoclast activity involved in collagen degradation such as metalloproteinase 2 (MMP2) was also significantly down-regulated in the OVX PG groups (PGt: 0.52-fold; PGj: 0.52-fold; p < 0.05; PGp: 0.48-fold; p < 0.001) as well as tartrate-resistant acid phosphatase (TRAP) expression was lowered, but only in the OVX PGp group (0.69-fold; p < 0.01), thus reflecting the inhibition of both osteoclast maturity and activity by the different PG studied parts.
Fig. 3

Effect of PG consumption on specific bone markers expression in femurs. a, b Expression profile analysis of bone markers in femurs from OVX mice raised either under control conditions (SH CT first bar, OVX CT second bar) or exposed to different parts of PG added in the diet : 5.7 % in the diet of PG lyophilized mashed totum (OVX PGt third bar) or 9.6 % of PG fresh juice (OVX PGj fourth bar), or 2.9 % of PG lyophilized mashed peel (OVX PGp fifth bar) for 30 days. Transcriptomic analysis of bone tissue mRNA levels was determined by Taqman low-density arrays. Results are presented as fold change compared to what was observed in the OVX CT group (fold change = 1) and expressed as mean ± SD (n = 8). *p < 0.05. a Osteoclastogenic genes: TRAP: tartrate-resistant acid phosphatase; CTR: calcitonin receptor; FOS; ITG b3: integrin β3; MMP2: metalloproteinase 2. b Osteogenic genes: OCN: osteocalcin; Lrp5; RunX2, Osterix; OPN: osteopontin

On another hand, while ovariectomy (OVX CT) decreased the expression of osteoblast markers (0.44-fold compared to SH CT group; p < 0.001), as shown in the Fig. 3b, TLDA analysis reveals that consumption of PG as the totum or as juice was associated with improved expression of a co-receptor implied in Wnt/β-catenin, one of major osteoclast signaling pathway: LRP5 (PGt: 1.86-fold, p < 0.001; PGj: 1.60-fold, p < 0.05 PGp: 1.19-fold compared with the OVX CT group). Nevertheless, the other major osteoblast differentiation markers that have presently been investigated were not modified by any of the studied diets.

Pomegranate improved bone inflammatory and oxidative status

To further conclude on the bone protective effect of PG linked to its consumption, we investigated the impact of PGt, PGj, and PGp on major inflammation and oxidation markers known to exacerbate bone resorption. Accordingly, a transcriptomic analysis of targeted inflammatory and oxidation markers was performed on bone tissue samples using TLDA, as previously described.

As expected, ovariectomy (OVX CT) enhanced the inflammatory context in bone environment when compared to sham-operated animals (SH CT), as demonstrated by the increased expression of pro-inflammatory markers such as CCL2 (4.76-fold, p < 0.001),while the expression of IL1-R2 and IL1-Rn, 2 anti-inflammatory mediators, was reduced (0.62-fold, p < 0.01 and 0.51-fold; p < 0.001, respectively) (Fig. 4a). In contrast, consumption of each part of PG, and of the entire fruit by OVX mice, was able to prevent the establishment of such an inflammatory status measured in the OVX CT animals given the control diet. This protection occurred mainly through a down-regulation of CCL2 (PGt: 0.16-fold; PGj: 0.22-fold; PGp: 0.20-fold; p < 0.001) and IL1-R1 (PGt: 0.68-fold; PGj: 0.68-fold; PGp: 0.56-fold; p < 0.001) and an up-regulation of IL1-R2 (PGt: 3.40-fold; PGj: 3.76-fold; PGp: 3.31-fold; p < 0.001) and IL1-Rn expression (PGt: 1.72-fold, p < 0.05; PGj: 1.89-fold; PGp: 1.77-fold; p < 0.001).
Fig. 4

Effect of PG consumption on inflammatory and oxidative status in femurs. a, b Expression profile analysis of inflammatory and oxidative stress markers in femurs harvested from SH or OVX mice raised either under control conditions (SH CT first bar, OVX CT second bar) or exposed to different parts of PG added to their diet: 5.7 % of PG lyophilized mashed totum (OVX PGt third bar), or 9.6 % of PG fresh juice (OVX PGj fourth bar) or 2.9 % of PG lyophilized mashed peel (OVX PGp fifth bar) for 30 days. Transcriptomic analysis of bone tissue mRNA levels was performed by Taqman low-density arrays. Results are presented as fold change compared to what was meadured in the OVX CT group (fold change = 1) and expressed as mean ± SD (n = 8). *p < 0.05. a Inflammation: CCL2 chemokine (C–C motif) ligand 2, INFγR1 interferon gamma receptor 1, IL1-R1 interleukine 1 receptor 1, IL1-R2 interleukine 1 receptor 2, IL1-Rn interleukine 1 receptor antagonist. b Oxidative stress: catalase; GPx glutathione peroxidase, GSR glutathione reductase, NOS2 nitric oxide synthase 2, NOX4 NADPH oxidase 4

On another hand, consistently, ovariectomy led to the establishment of an oxidative stress in mice bones as shown in the Fig. 4b by an increase in enzymes implied in ROS generation (nitric oxide synthase 2: NOS2 and NADPH oxidase 4: NOX4; 2.63-fold, p < 0.001; 1.37-fold p < 0.05, respectively, in the OVX CT animals compared to what was observed in the SH CT group), while expression of glutathione peroxidase: GPx and glutathione reductase: GSR (0.68-fold, p < 0.001; 0.41-fold p < 0.05, respectively, versus SH CT) was down-regulated. Interestingly, consumption of PG, whatever the part, was able to enhance those bone anti-oxidant defenses in OVX mice. Indeed, an up-regulation of those antioxidant enzymes expression [GPx (PGt: 1.72-fold, p < 0.05; PGj: 1.72-fold; PGp: 1.46-fold; p < 0.001) and GSR (PGt: 4.76-fold; PGj: 3.92-fold; PGp: 5.47-fold; p < 0.001)] was measured, while, in contrast, the expression of ROS producing enzymes was down-regulated: NOS2 (PGt: 0.24-fold; PGj: 0.23-fold; PGp: 0.27-fold; p < 0.001) and NOX4 (PGt: 0.23-fold; PGj: 0.20-fold; PGp: 0.16-fold; p < 0.001), further supporting the bone-sparing effect of PG.


The present study demonstrates for the first time that PG (as well as its derivatives (peel and juice)) consumption may contribute to bone health during aging. Actually, all PG parts were able to prevent bone loss in a well-characterized model of postmenopausal osteoporosis by modulating bone cells differentiation, as well as inflammatory and oxidative status in bone microenvironment.

First of all, to validate our experimental model, we assessed the effect of ovariecctomy on bone health. As a matter of fact, in comparison with what was observed in the SH CT group, the OVX control animals (i.e., fed the standard diet) exhibited decreased femoral BMD values and altered bone micro-architecture, as indicated by a lower number of trabeculae (TbN), decreased bone volume (BV/TV), bone surface (BS/TV), and connectivity density (ConnDn), in addition to a higher trabecular spacing (TbSp) and increased total porosity (Po tot) (Fig. 2). This is highly relevant to the human situation after menopause and consistent with the previous data [23, 24].

With regards to the dietary intervention with PG and its derivatives, the present study carried out in OVX mice showed that consumption of any of the presently studied part of PG elicited a protective effect on both bone mass and bone micro-architecture, the two parameters mainly altered in osteoporosis. Actually, our results evidenced a greater potential of PG juice and peel in preventing BMD loss, while when given as a totum (OVX PGt), PG better counteracted bone micro-architecture impairment (Fig. 2c). Accordingly, our data are consistent with those previously published by Mori-Okamoto et al. [25] showing that PG extracts from seed and juice exerted estrogen activities and improved bone parameters following ovariectomy. Nevertheless, in the present study and in contrast to the previous work, PG consumption (whatever the part) exhibited a bone sparing not only by modulating bone histological parameters but also through regulation of cellular processes, as well as by improving bone inflammatory and oxidative status.

Actually, PG is known for its large amount of phytochemicals and for the specific polyphenolic composition of its different parts, including anthocyanins (cyanidin, delphinidin, mainly in the juice) and hydrolyzable tannins in the peel such as ellagic acid, punicalagin, and gallagic acid [4]. Among the great variety of those chemicals, punicalagin and ellagic acid seem to be the two important components of PG [5]. It was raised that such a unique polyphenols composition could explain the potential of the PG to prevent various chronic diseases associated with aging, through its high anti-oxidant and anti-inflammatory capacities [3, 26, 27, 28]. In the present experiment, it appears that PG action could imply all its components: polyphenols (anthocyanins for the juice and ellagitannins for the peel), proteins, sugars, fibers, minerals, etc., although the relationship between the chemical constituents of PG and their pharmacologic effects is still not clear [6]. Accordingly, it appears crucial to investigate, like in the present study, the benefits of the whole fruit, considering the interactions between the various compounds in the PG matrix that could be of great importance for the final result [6].

Nevertheless, we demonstrated here that the juice or the peel exhibited a better action on BMD than the totum. This could be due to molecules common to both parts such as ellagic acid or punicalagin.

As a matter of fact, when considering the PG totum, those molecules could be less concentrated in the whole fruit, explaining that PG totum did not significantly impact BMD, although a trend was observed. As BMD mainly depends on hydroxyapatite content [29], such a result would suggest a greater calcium apposition rate in the animals given PG either under the peel-enriched diet or the juice, leading to a higher BMD. This is actually consistent with the recent published work by both [30] showing that a PG juice extract could increase calcium bone content during gestation in mice, even though a peel extract or a mixture of both were devoid of any significant effect.

Inversely, we demonstrated that the whole fruit had a greater impact on bone micro-architecture parameters than the juice or the peel alone. This could be explained by the molecules present in PG totum such as seed compounds (punicic acid for example) that we do not find in other parts of the fruit. As micro-architecture parameters mainly result from better bone remodeling and expansion of osteoblast pool (by proliferation, recruitment, and/or increased survival) [29], we could suggest that PG totum improved osteoblastogenesis, leading to an improved micro-architecture.

To further investigate cellular mechanism linked to such a bone rescue, the expression of the main markers for osteoblast and osteoclast differentiation and activity was analyzed. As expected, the osteopenia process induced by ovariectomy was associated with a down-regulation of osteoblast differentiation markers and an increase in osteoclast ones. With regards to present dietary interventions, we demonstrated that each part of PG was able to counteract such a bone loss in OVX mice (Fig. 3). Moreover, as inflammation and oxidative stress play a major role in bone remodeling by modulating bone resorption [15, 31], the expression of major markers of those processes was investigated in bone tissues (Fig. 4). In this light, our data dealing with the effect of ovariectomy control are consistent with previous publications [32]. Indeed, we confirmed that ovariectomy enhanced inflammation and oxidative stress in bones with unbalanced expression of pro- and anti-inflammatory and oxidative stress markers [32]. In this context, PG and its derivates were able to preserve this balance and consequently to prevent from excessive bone loss.

Nevertheless, we must stay aware that this study presents some limitations. Biomechanical and histomorphometry analyses could strengthen our data. With regards to the expression analysis of major bone, oxidative stress, and inflammation markers, proteomic investigation would allow to confirm what was observed on gene expression thanks to TLDA assessment.

In conclusion, those data suggest that PG may be used as an innovative alternative agent for the prevention of osteoporosis. However, further studies are now required to determine the detailed contribution of each nutrients, composing PG in mediating the observed bone-sparing effect, and to assess if those data, obtained in an animal model for the disease, can be extrapolated to the postmenopausal women. Nevertheless, further studies are required to fully determine the contribution of each PG parts properties on bone.


Greentech (GREENTECH SA, Saint-Beauzire, France) is kindly acknowledged for providing financial support for this study. The authors are as well grateful to (1) Paul Pilet for his assistance in collecting data of bone micro-architecture from micro-CT, (2) the people from the “Animal lab: Installation Experimentale de Nutrition” who provided every day cares to mice. This study was supported by INRA, UMR 1019, UNH, Clermont-Ferrand, France.

Conflict of interest

The authors have no conflict of interest to declare.

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Mélanie Spilmont
    • 1
    • 2
    • 3
    • 6
  • Laurent Léotoing
    • 1
    • 2
    • 3
  • Marie-Jeanne Davicco
    • 1
    • 2
    • 3
  • Patrice Lebecque
    • 1
    • 2
    • 3
  • Sylvie Mercier
    • 1
    • 2
    • 3
  • Elisabeth Miot-Noirault
    • 4
    • 5
  • Paul Pilet
    • 6
    • 7
  • Laurent Rios
    • 8
  • Yohann Wittrant
    • 1
    • 2
    • 3
    • 9
  • Véronique Coxam
    • 1
    • 2
    • 3
  1. 1.INRA, UMR 1019, UNH, CRNH AuvergneClermont-FerrandFrance
  2. 2.Equipe Alimentation, Squelette et MétabolismesClermont-FerrandFrance
  3. 3.Clermont Université, Université d’AuvergneUnité de Nutrition HumaineClermont-FerrandFrance
  4. 4.Clermont Université, Université d’AuvergneImagerie moléculaire et thérapie vectoriséeClermont-FerrandFrance
  5. 5.Inserm, U 990Clermont-FerrandFrance
  6. 6.Institut National de la Santé et de la Recherche Médicale, UMR S791Laboratoire d’Ingénierie Ostéo-Articulaire et DentaireNantesFrance
  7. 7.Pôle de Recherche et d’Enseignement Supérieur Université Nantes Angers Le Mans, Université de NantesUnité de Formation et de Recherche OdontologieNantesFrance
  8. 8.GREENTECH SA Biopôle Clermont-LimagneSaint-BeauzireFrance
  9. 9.INRA, Equipe ASM, UMR 1019Centre de Recherches INRA de Clermont-Ferrand/TheixClermont-FerrandFrance

Personalised recommendations