Effect of Eicosapentaenoic Acid on Bone Changes due to Methylprednisolone in Rats
Author for correspondence: Maryam Rezaian, Department of Histology, Faculty of Veterinary Medicine, University of Tehran, P.O. Box: 14155-6453, Tehran, Iran (fax +98 21 66933222, e-mail firstname.lastname@example.org).
Abstract: The present study was conducted to investigate the effects of eicosapentaenoic acid on glucocorticoid-induced bone changes in rats, and to compare them with those of alendronate. Thirty six male Wistar rats, 2.5 months of age, were divided into six groups (n = 6 each) and treated with 0.9% NaCl (control), methylprednisolone 7 mg/kg, once a week subcutaneously, methylprednisolone + alendronate 20 µg/kg, twice a week subcutaneously and methylprednisolone + 80 or 160 or 320 mg/kg eicosapentaenoic acid, per day orally, for 6 weeks. At the end of the experiment, serum and urinary parameters of bone metabolism determined and bone histomorphometric analyses performed on cancellous bone of femoral epiphysis and metaphysis and cortical bone of tibial diaphysis. There were no significant differences in serum and urinary parameters among groups. Decrease of epiphyseal and metaphyseal trabecular width, epiphyseal bone area/tissue area and increase of epiphyseal trabecular separation observed in the methylprednisolone group compared to control. Alendronate restored all of these parameters except metaphyseal trabecular width, which increased significantly by eicosapentaenoic acid at the doses of 80 and 160 mg/kg. Effects of alendronate and 160 mg/kg eicosapentaenoic acid on bone area/tissue area, alendronate and eicosapentaenoic acid at the doses of 80 and 160 mg/kg on trabecular separation and alendronate and eicosapentaenoic acid at doses of 160 and 320 mg/kg on epiphyseal trabecular width were statistically similar. Methylprednisolone did not significantly change cortical bone parameters including cortical width and marrow area/cortical area. Eicosapentaenoic acid, especially, at the dose of 160 mg/kg exerts beneficial effects on methylprednisolone-induced bone changes in rats; these effects are similar or sometimes even better than alendronate.
Osteoporosis is probably one of the main limitations to long-term glucocorticoid therapy . It has been demonstrated that prolonged exposure to glucocorticoids at supraphysiological doses induces osteoporosis associated with an increased risk of bone fracture . Bisphosphonates, such as alendronate, can counteract the negative effects of glucocorticoids on bone ; but diet therapy and life style changes that minimize bone loss would be very helpful to decrease the necessity for drug therapy to prevent osteoporosis .
It is known that dietary sources of long chain n-3 fatty acids are essential for maintaining optimum health, since mammals can not synthesize fatty acids with double bonds post the Δ position . Eicosapentaenoic acid is the major long chain n-3 essential fatty acid present in fish oil . Although best known for their cardioprotective role, long chain polyunsaturated fatty acids and their metabolites also regulate bone metabolism and consequently may have a potential in the prevention and/or treatment of osteoporosis . Fish oil or n-3 essential fatty acid consumption has previously shown to protect against ovariectomy-induced bone loss in rats and mice [7–9]. Eicosapentaenoic acid-enriched diet has prevented the loss of bone weight and strength in ovariectomized rats fed with low calcium diet .
The purpose of the present study was to investigate the putative positive effects of eicosapentaenoic acid on bone, in the presence of detrimental bone changes due to methylprednisolone administration in rats and to compare these effects with those of common anti-osteoporosis drug, alendronate.
Materials and Methods
Animals. Thirty-six male Wistar rats, 2.5 months of age, with a body weight of 300 ± 36 g, purchased from Razi Serum and Vaccine Research Institute, Iran, and after a week of adaptation, randomly divided into six experimental groups (six animals each) and treated for 6 weeks as follows:
Group 1: 0.9% NaCl once a week S.C (control group).
Group 2: Methylprednisolone sodium succinate (Merck, France), 7 mg/kg once a week S.C.
Group 3: Methylprednisolone 7 mg/kg once a week S.C + alendronate sodium trihydrate (Arasto pharmaceutical and chemicals Inc., Iran) 20 µg/kg twice a week S.C.
Group 4: Methylprednisolone 7 mg/kg once a week S.C + eicosapentaenoic acid (S.L.A Pharma AG., Switzerland) 80 mg/kg daily by oral gavage.
Group 5: Methylprednisolone 7 mg/kg once a week S.C + eicosapentaenoic acid 160 mg/kg daily by oral gavage.
Group 6: Methylprednisolone 7 mg/kg once a week S.C + eicosapentaenoic acid 320 mg/kg daily by oral gavage.
Dose, dosage interval and duration of administration of methylprednisolone borrowed from studies performed by Wimalawansa et al. [11,12]. They induced bone changes in male Wistar rats by using methylprednisolone 7 mg/kg once a week S.C for 6 weeks; dosage regimen for treatment with alendronate was also chosen according to the study performed by Wimalawansa et al. .
During the experimental period, animals were weighed on a weekly basis and maintained on a 12-hr light/dark cycle at 20 ± 2° and allowed free access to tap water and a pelleted standard rat chow diet, provided by Razi Serum and Vaccine Research Institute, Iran.
Animals were treated ethically in compliance with the local regulations of University of Tehran, Faculty of Veterinary Medicine.
Sampling of urine blood and bones. On day 43, voiding urine samples collected from all animals and then the rats were anaesthetized with chloroform and blood samples obtained by cardiac puncture.
Sera harvested within an hour after sampling and serum and urine samples stored in –20° until use. After killing all animals under deep anaesthesia, left tibial and femoral bones were dissected for histomorphometric study.
Preparation of specimens for histomorphometric study. Left tibial and femoral bones were fixed in 4% formaldehyde solution and decalcified using Formic acid–Sodium Citrate method . From the tibial bone, transverse 5 µm cross-sections made perpendicularly to the long axis, such that fibula attaches into it. From the femoral bone, 5 µm longitudinal sections of the distal epiphysis and metaphysis were made in the median plate. All the sections were stained using Masson's trichrome method .
Histomorphometric parameters were determined by a digital photo microscope connected to a personal computer with Ziess axio vision LE software. Parameters measured in cancellous bone included epiphyseal and metaphyseal trabecular width, epiphyseal bone area/tissue area, epiphyseal trabecular separation and epiphyseal trabecular number.
The region of the cancellous bone marked for the measurements was the central zone of cancellous tissue, 2–3 mm below (epiphysis) or above (secondary spongiosa of metaphysis) the margins of the growth plate in the distal epiphysis and metaphysis of femoral bone. Marrow area/cortical area and cortical width were measured in the cross-sections of tibial diaphysis. All histomorphometric parameters determined two times by two independent persons (first author and Dr. S. Hamedi, a senior PhD student of Histology at our faculty, who was unaware of the grouping system of the experiment) and then we used the mean value of these measurements, so it seems that these parameters are measured unbiased.
The nomenclature of parameters is in compliance with ASBMR histomorphometry nomenclature committee .
Determination of serum and urine parameters. Serum Osteocalcin and C-Terminal Telopeptide of Collagen Type I (CTX) were measured using Rat-MID Osteocalcin ELISA and Rat-laps ELISA kits (Nordic bioscience diagnostics a/s, Denmark), respectively. Serum alkaline phosphatase was assayed by the photometric method provided by Pars Azmun Company, Ltd., Iran.
Serum and urine calcium, phosphorus and creatinine were determined by the cresophthalein complexon method, photometric method and Jaffe method, respectively. All reagents were prepared by Pars Azmun Company, Ltd., Iran.
Statistical analysis. All data were expressed as mean ± SD. Data comparisons among groups performed by one-way anova followed by Tukey's multiple comparison test. All statistical analyses were performed using the Sigma Stat 2 statistical package. A significance level of P < 0.05 was set for all comparisons.
During the first 4 weeks of the experiment, there was no significant difference in body weight among groups but at the end of 5th and 6th week, body weight of animals in the 160 mg/kg eicosapentaenoic acid-treated group was significantly lower than the methylprednisolone group (296 ± 32 g vs. 355 ± 32 g and P = 0.05 at the end of 5th week; and 291 ± 28 g vs. 360 ± 29 g and P = 0.009 at the end of 6th week).
Epiphyseal trabecular width, metaphyseal trabecular width and epiphyseal bone area/tissue area decreased (P < 0.001) and epiphyseal trabecular separation increased (P = 0.03) significantly in the methylprednisolone group compared to the control. Alendronate and eicosapentaenoic acid at the doses of 160 and 320 mg/kg significantly increased epiphyseal trabecular width in the methylprednisolone group (P < 0.001). This increase was statistically similar in the alendronate and eicosapentaenoic acid-treated groups.
Alendronate did not increase metaphyseal trabecular width significantly but the increase was significant in the 80 and 160 mg/kg eicosapentaenoic acid-treated groups with respect to the methylprednisolone group (P < 0.001), the effect of these two doses were statistically similar. Alendronate and eicosapentaenoic acid 160 mg/kg increased bone area/tissue area compared to the methylprednisolone group (P < 0.001) and their effects were similar. Alendronate and eicosapentaenoic acid at the doses of 80 and 160 mg/kg decreased trabecular separation in comparison with the methylprednisolone group (P = 0.02, P = 0.035 and P = 0.003, respectively). The decrease was statistically the same in all of these groups. There was no significant difference in trabecular number among the experimental groups. Methylprednisolone did not change cortical width significantly compared to the control, and there was no significant difference among the control group and alendronate- or eicosapentaenoic acid-treated groups. There was no significant change in marrow area/cortical area in the methylprednisolone-treated group with respect to the control group, or among the control group and alendronate- or eicosapentaenoic acid-treated groups. Histomorphometric data are shown in table 1.
Histomorphometric data presented as mean ± SD, n = 6 for each group. G1: control, G2: methylprednisolone (MP), G3: MP + alendronate, G4: MP + eicosapentaenoic acid (EPA) 80 mg/kg, G5: MP + EPA 160 mg/kg and G6: MP + EPA 320 mg/kg. Superscript letters are used to show groups with significant difference mentioned in each row. Ep.Tb.Wi: epiphyseal trabecular width, Mt.Tb.Wi: metaphyseal trabecular width, B.Ar/T.Ar: bone area/tissue area, Tb.N: trabecular number, Tb.Sp: trabecular separation, Ct.Wi: cortical width and Ma.Ar/Ct.Ar: marrow area/cortical area.
|Ep.Tb.Wi (µm)||a63.4 ± 2.3||b37.5 ± 2.12||c52.5 ± 2.61||d49.8 ± 2.08||e50.1 ± 2.03||f50.7 ± 1.76||a,b;a,c;a,d;a,e;a,f;b,|
e;b,f P < 0.001
P < 0.001;
a,e P = 0.042;
d,f P = 0.031;
c,d P = 0.005
a,b;b,c;b,e P < 0.001
|Mt.Tb.Wi (µm)||a38 ± 1.93||b33.5 ± 1.45||c35.2 ± 1.38||d38.8 ± 1.20||e40.9 ± 1.4||f35.8 ± 1.67|
|B.Ar/T.Ar (%)||a37.8 ± 2.5||b22.9 ± 1.7||c34.7 ± 4.5||d32.3 ± 2.4||e35.1 ± 3.1||f31.6 ± 2.5|
|Tb.N (1/mm)||6.1 ± 0.7||6.4 ± 1.6||6.8 ± 1.5||6.8 ± 1.6||7.3 ± 1.5||6.2 ± 0.5|
|Tb.Sp (µm)||a101 ± 10.2||b128 ± 38.3||c98 ± 19.6||d103 ± 22.3||e91 ± 19.6||f110 ± 10.8||a,b P = 0.03;|
b,c P = 0.02;
b,d P = 0.035;
b,e P = 0.003
|Ct.Wi (µm)||a337 ± 25.7||b347 ± 26.4||c350 ± 16.6||d333 ± 19.9||e310 ± 11.4||f347 ± 26||b,e P = 0.033;|
c,e P = 0.016
|Ma.Ar/Ct.Ar||a0.24 ± 0.02||b0.22 ± 0.02||c0.23 ± 0.01||d0.25 ± 0.01||e0.21 ± 0.02||f0.23 ± 0.01||b,d P = 0.026;|
d,e P = 0.001
Serum and urinary parameters.
There was no significant difference in serum CTX, osteocalcin, alkaline phosphatase, calcium, phosphorus and creatinine concentrations among groups (table 2). Urinary calcium/creatinine and phosphorus/creatinine ratios were statistically similar among experimental groups (table 3).
Serum parameters presented as mean ± SD, n = 6 for each group. G1: control, G2: methylprednisolone (MP), G3: MP + alendronate, G4: MP + eicosapentaenoic acid (EPA) 80 mg/kg, G5: MP + EPA 160 mg/kg and G6: MP + EPA 320 mg/kg. There was no significant difference in serum CTX, osteocalcin, calcium and phosphorus among groups P > 0.05.
|CTX (ng/ml)||16.55 ± 10.15||10.50 ± 5.98||12.85 ± 6.24||15.41 ± 5.27||14.46 ± 5.50||13.50 ± 5.74|
|Osteocalcin (ng/ml)||66.83 ± 61.21||54.07 ± 6.21||78 ± 56.86||70.35 ± 49.39||59.78 ± 21.21||82.13 ± 22.04|
|Alkaline phosphatase (mg/dl)||306.5 ± 51.8||320.8 ± 60.3||293.1 ± 48||259 ± 38.6||237.5 ± 31.43||257 ± 50.9|
|Calcium (mg/dl)||10.48 ± 2.42||10.51 ± 2.06||10.16 ± 1.64||7.61 ± 2.22||8.37 ± 1.63||9.14 ± 1.99|
|Phosphorus (mg/dl)||7.22 ± 1.16||7.68 ± 1.5||7.26 ± 1.27||8.43 ± 2.33||10.14 ± 1.53||8.32 ± 2.26|
|Creatinine (mg/dl)||0.68 ± 0.07||0.6 ± 0.06||0.68 ± 0.12||0.68 ± 0.1||0.63 ± 0.07||0.66 ± 0.06|
Urinary parameters presented as mean ± SD, n = 6 for each group. G1: control, G2: methylprednisolone (MP), G3: MP+alendronate, G4: MP + eicosapentaenoic acid (EPA) 80 mg/kg, G5: MP+ EPA 160 mg/kg and G6: MP+ EPA 320 mg/kg. There was no significant difference in calcium/creatinine and phosphorus/creatinine ratios among groups P > 0.05.
|Calcium/creatinine||0.04 ± 0.01||0.04 ± 0.02||0.03 ± 0.01||0.04 ± 0.01||0.03 ± 0.01||0.03 ± 0.0|
|Phosphorus/creatinine||1.65 ± 0.49||1.88 ± 0.59||1.63 ± 0.06|| 1.5 ± 0.57||1.87 ± 1.1||2.00 ± 1.6|
Epidemiological and longitudinal studies have reported a positive relationship between intake of n-3 family of long chain poly-unsaturated fatty acids and bone mineral density in postmenopausal women . In rats and mice, dietary supplementation with n-3 long chain polyunsaturated fatty acid-rich oils has improved maintenance of bone mass post-ovariectomy [7–9]. In addition, eicosapentaenoic acid-enriched diet has inhibited bone loss in ovariectomized rats fed with low calcium diet . However, there is no study conducted to demonstrate the effect of eicosapentaenoic acid on bone changes due to glucocorticoid administration. Glucocorticoid-induced osteopaenia and ovariectomy-induced osteopaenia are two contrasting types of osteopaenia. Histomorphometric studies performed by Nitta et al. demonstrated that the most characteristic consequence of methylprednisolone treatment in rats is a significant loss of trabecular bone resulting from thinning of trabecular bone with unchanged connectivity. Losses in the volume and thickness of trabecular bone in glucocorticoid-induced osteopaenia are due to acceleration of bone resorption (increase in Eroded Surface/Bone Surface) and depression of bone formation (Decrease in Osteoid Suface/Bone Surface). These characteristics are in a good contrast to those of ovariectomy-induced osteopaenia in which accelerated bone turnover in favour of bone resorption results in cut-but-non-thinned structure of the cancellous bone .
Regarding mechanism of action of glucocorticoids on bone, it is now apparent that their most important effect is decreasing bone formation. Glucocorticoids induce a 30% increase in osteoblast and osteocyte apoptosis and affect transcription of many of the genes responsible for synthesis of matrix constituents by osteoblasts such as type I collagen. Glucocorticoids also induce osteoclastogenesis and consequently increase bone resorption through enhancement of receptor activator of NF-κB (RANK) ligand (RANKL) production and inhibition of osteoprotegrin production. Because osteoprotegrin acts as a decoy receptor for RANKL, the final effect would be the overactivation of RANK which leads to an increase in bone resorption [17,18].
Bone loss due to estrogen deficiency is primarily as a result of increased osteoclast activity and accelerated bone resorption .
The aim of the present study was to determine whether eicosapentaenoic acid would exert beneficial effects on bone in glucocorticoid-treated rats and to compare the effects of eicosapentaenoic acid with those of the common anti-osteoporosis drug, alendronate.
It has been demonstrated that the key histological feature of glucocorticoid-induced cancellous bone loss is a reduction in the trabecular thickness, reflecting suppressed bone formation . In the present study, methylprednisolone induced cancellous bone loss that was manifested by a significant decrease in both epiphyseal and metaphyseal trabecular width and epiphyseal bone area/tissue area and an increase in epiphyseal trabecular separation. Alendronate restored all of these parameters except metaphyseal trabecular width, which increased significantly by eicosapentaenoic acid at the doses of 80 and 160 mg/kg. Effects of alendronate and 160 mg/kg eicosapentaenoic acid on bone area/tissue area, alendronate and eicosapentaenoic acid at the doses of 80 and 160 mg/kg on trabecular separation and alendronate and eicosapentaenoic acid at doses of 160 and 320 mg/kg on epiphyseal trabecular width were statistically similar. The effect of eicosapentaenoic acid on attenuating cancellous bone loss was found to be dose-dependent and the most effective dose was 160 mg/kg, except on increasing epiphyseal trabecular width that the effect of 320 mg/kg was only slightly better than 160 mg/kg.
It seems that glucocorticoid-induced cancellous osteopaenia in rats is a good responder to eicosapentaenoic acid. As noted previously, at the present study eicosapentaenoic acid with the dose of 160 mg/kg significantly attenuated cancellous bone loss due to glucocorticoid administration regarding histomorphometric parameters. In the study performed by Poulsen et al. (2007), diet supplemented with 0.5 g/kg body weight/day of eicosapentaenoic acid fed to ovariectomized rats for 16 weeks did not significantly increase bone mineral density compared to ovariectomized control rats . Considering different histomorphometric features and pathogenesis pathways of glucocorticoid-induced osteopaenia and ovariectomy-induced osteopaenia, as stated above, it is plausible that these two kinds of osteopaenia respond differently to the same agent. A fact that should be noted here is that these researchers did not perform bone histomorphometric study so the alterations in local bone formation and resorption and the modification of bone architecture remain uncertain.
Although glucocorticoid-induced osteoporosis is more evident in cancellous than in cortical bone , it has been also reported that glucocorticoid administration induces cortical osteopaenia in rats [21–23]. In the present study, methylprednisolone did not significantly change cortical bone parameters including cortical width and marrow area/cortical area of tibial diaphysis compared to control. Iwamoto et al. reported that treatment of rats with methylprednisolone for 4 weeks induced cancellous osteopaenia without significantly affecting cortical bones; however, continuation of the same regimen for 8 weeks resulted in cortical osteopaenia in addition to cancellous osteopaenia that was manifested by a decrease in the percent of cortical area and an increase in the percent of marrow area of tibial diaphysis . It seems that cortical osteopaenia due to glucocorticoid administration needs more time or total amount of drug than cancellous osteopaenia to be manifested.
Alendronate and eicosapentaenoic acid in three different doses did not exert any significant positive or negative effects on cortical bone parameters compared to control.
Serum bone resorption marker (CTX), bone formation markers (osteocalcin and alkaline phosphatase), calcium, phosphorus and creatinine concentrations were statistically similar among groups as well as urinary calcium/creatinine and phosphorus/creatinine ratios. In the study performed by Unoki, administration of prednisolone 2.5 mg/kg six times a week for 8 weeks to rats, although induced cancellous osteopaenia, did not significantly change urinary calcium and phosphorus compared to control group . In the study performed by Wang et al., administration of methylprednisolone at the doses of 0, 2.5, 5, 10 or 20 mg/kg to male rats daily for 4 weeks did not significantly change the values of serum calcium, phosphorus and resorption marker (Pyridinoline) among groups and osteocalcin and alkaline phosphatase decreased significantly only in 10 and 20 mg/kg methylprednisolone-treated groups .
This lack of significant difference in serum and urine biochemical markers in the present study may be due to a lack of statistical power to detect differences between groups or more probably due to relatively low cumulative dose of methylprednisolone used. Another explanation may be that by considering this fact that methylprednisol one sodium succinate is a short to medium acting glucocorticoid, there is a possibility that its effects on biochemical markers of bone metabolism have been vanished during the one week time course between the last dose of methylprednisol one and collecting of samples.
It seems that in the present study, histomorphometric parameters have been more appropriate indicators of detrimental changes of bone due to glucocorticoid administration in rats at least in cancellous bone.
In conclusion, eicosapentaenoic acid especially at the dose of 160 mg/kg exerts appreciable positive effects on detrimental bone changes due to glucocorticoid administration in rats. These effects are similar or even better (as seen with metaphyseal trabecular width) than a common anti-osteoporosis agent, alendronate.
Funding for this study was provided by University of Tehran, Faculty of Veterinary Medicine, Tehran, Iran, according to the research project number 7506006/6/3. Thanks are due to J. Slagel (S.L.A Pharma AG, Switzerland) for the generous donation of eicosapentaenoic acid. Authors are grateful to Dr. S. Hamedi for her assistance with measuring histomorphometric parameters.