- Email: [email protected]
Whey protein suppresses the osteoclast-mediated bone resorption and osteoclast cell formation
PII : S0958-6946(97)00073-3
Int. Dairy Journal 7 (1997) 821—825 ( 1998 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0958-6946/98/$19.00#0.00
Whey Protein Suppresses the Osteoclast-mediated Bone Resorption and Osteoclast Cell Formation Yukihiro Takadaa *, Naomichi Kobayashia, Hiroaki Matsuyamaa, Ken Katoa, Junichi Yamamuraa, Masatoshi Yahiroa, Masayoshi Kumegawab and Seiichiro Aoea a Snow Brand Milk Product Co. Ltd., Nutritional Science Laboratory, 1-1-2 Minamidai, Kawagoe, Saitama 350-11, Japan b Department of Oral Anatomy, Meikai University School of Dentistry, 1-1 Keyakidai, Sakado, Saitama 350-02, Japan (Received 20 February 1997; accepted 19 October 1997) ABSTRACT Effects of whey protein on bone resorption and osteoclastic cell formation were evaluated. In the pre-existed and newly formed osteoclast, in bone resorption methods using an unfractionated bone cell culturing system, whey protein suppressed the area of pits formed by osteoclasts. In the osteoclastic cell formation method using the hemopoietic blast cell culturing system, whey protein also suppressed osteoclastic cell formation. These activities were resistant to heat when the protein was treated at 75—90°C for 10 min. Heat-treated whey protein was first fractionated on a Mono S column, and the active fraction (basic protein fraction) was then applied to Superose 12. The molecular weights of the active components were approximately 23 000 and 10 000 Da, as determined by gel filtration. The inner solution of an everted gut-sac incubated in a solution of intact basic protein (BP), pepsin-digested BP or pepsin/pancreatin-digested BP also suppressed osteoclast-mediated bone resorption. Thus, these active components can possibly be absorbed or transported by the intestines. These results showed that whey protein contains an active component that suppresses osteoclast-mediated bone resorption and osteoclastic cell formation. ( 1998 Elsevier Science Ltd. All rights reserved
INTRODUCTION
using cultures of unfractionated bone cells on dentine slices (Takada et al., 1992). Using this culturing system, it is possible to elucidate the effects of osteotropic factors, reflecting the net effects of direct and indirect actions of factors. On the other hand, Kurihara et al. (1989) developed an osteoclastic cell formation system from hematopoietic blast cells that were generated from the spleen cells of 5-fluorouracil-treated mice. In this study, we examined the effect of whey protein on osteoclast-mediated bone resorption using pre-existing and newly formed osteoclast bone resorption methods. The effect of whey protein on osteoclastic cell formation using the osteoclastic cell formation method from hemopoietic blast cells was also examined. We characterized the active component by ion-exchange chromatography and gel filtration chromatography and by using rat everted gut-sacs (Martin and Deluca, 1969). We also examined the permeability of the gut to the protein.
Milk is well known to be a safe food, beneficial to human health, that can be taken for a long time. Whey protein originates from milk and is obtained from whey by ultrafiltration, reverse osmosis, chromatography and dialysis to remove lactose and other components (Glover, 1971; Hiddink et al., 1978; Kessler, 1981; Skudder, 1985). Whey is obtained by adding an acid or rennet to milk and removing the formed coagulate. Whey protein is usually produced as a by-product of cheese or casein manufacturing. One-hundred mL of cow’s milk contains 0.55 g of whey protein and the major proteins are a-lactalbumin (20%), b-lactoglobulin (58%), immunoglobulin (13%) and serum albumin (7%) (Ito, 1991). Milk has a functional role in growth of newborn animals, and so milk whey protein may possibly have components that affect bone metabolism. Recently, we found that milk whey protein was effective in increasing bone strength and the content of collagen-specific amino acids such as a hydroxyproline in ovariectomized (OVX) rats (Takada et al., 1993). We also found that whey protein is effective for osteoblast cell proliferation and differentiation (Takada et al., 1996). Bone resorption is important for the metabolism of bone formation. Osteoclastic bone resorption is regulated at two different stages; the formative stage and the activation stage of osteoclasts (Mundy and Roodman, 1987). We established a biological system for evaluating bone resorption, osteoclast recruitment and activation
MATERIALS AND METHODS Materials Alpha-minimum essential medium (a-MEM) was purchased from Flow Laboratories (McLean, VA, USA) and fetal bovine serum (FBS) was obtained from Gibco (Grand Island, NY, USA). Whey protein was prepared from a whey protein concentrate (WPC) by dialysis. The WPC was purchased from New Zealand Dairy Board (Wellington, New Zealand). Mono S and Superose 12 columns were obtained from Pharmacia (Sweden).
* Corresponding author. 821
822
Y. Takada et al.
Suckling ICR and 6-week-old BDF1 mice were purchased from the Shizuoka Laboratory Animal Center (Shizuoka, Japan). 1,25(OH) D was obtained from Dr. 2 3 S. Ishizuka (Teijin Biomedical Research Institute, Tokyo, Japan). Murine recombinant interleukin-3 (IL-3) was provided by Dr. T. Sudo (Toray Industries Inc., Kanagawa, Japan). Human recombinant IL-6 and murine recombinant granulocytemacrophage-colony stimulating factor (GM-CSF) were kindly provided by Ajinomoto Co. Ltd. (Kanagawa, Japan) and Sumitomo Pharmaceutical Co. Ltd. (Osaka, Japan), respectively. Assay of pre-existing and newly formed osteoclast-mediated bone resorption using unfractionated bone cells Unfractionated bone cells were prepared from the femoral bones of 13-day-old mice (Takada et al., 1992). The bones were dissected free of soft tissues and transferred to a 50 mL test-tube. For preparation, the tube was vigorously agitated for 0.5 min with a test-tube mixer (Yamato, Tokyo, Japan) in 10 mL of a-MEM, and the large fragments were allowed to sediment for 3 min. The number of living cells was counted with a hemocytometer and cytospun cells were stained for tartrate-resistant acid phosphatase (TRAP) at 37°C for 5 min and for alkaline phosphatase to determine osteoclasts and osteoblasts, respectively (Takada et al., 1992). Dentine slices (150 km thick), prepared with a diamond saw, were punched out as circles 6 mm in diameter, sonicated and sterilized in 75% alcohol. After the slices had been washed with a-MEM, each was placed into a well of a 96-well plate. For assaying the effects of whey protein on bone resorption by pre-existing osteoclasts, 200 kL of a-MEM supplemented with 5% FBS plus 3]105 cells were added to each well containing a dentine slice, and the cells were then incubated at 37°C in a CO 2 incubator (5% CO : 95% air) for 2 h. Then, the slices 2 were transferred to fresh medium containing whey protein, and were incubated for the indicated number of days. The medium was changed every 2 days. To assay bone resorption by newly formed osteoclasts, we precultured the cells in whey protein-free medium for 6 days before transferring them to medium containing whey protein. After culturing for the 4 days, the cells were brushed off the dentine using a motorized hand-held brush (Yoshida Seiko Co. Ltd., Tokyo, Japan) for 10 s, and the dentine slices were then stained with acid hematoxylin (Sigma) for 3 min in the 96-well plate. The total area of pits was measured by an image analyzer (PIASLA555, PIAS Co. Ltd., Tokyo, Japan) to detect possible osteoclastic bone resorption. The pit area was determined from an image taken by a video camera attached to a light microscope (objective lens, ]2; TV camera magnification, ]14). An image of the pits alone was obtained by filtering and by using noise erasure treatments. After treatment into two values (black and white), the total surface area of the remaining image was measured as the pit area. The number of TRAP-positive cells was also determined. Assay of osteoclastic cell formation from splenic blast cells Six-week-old female BDF1 mice were administered 5-fluorouracil (Hoffman Laroche Co., Basal, Switzer-
land) at a dosage of 150 mg kg~1 through a tail vein. Splenic blast cells were prepared as described previously (Kurihara et al., 1989). Spleen cells from the treated mice were cultured in 35 mm non-tissue culture dishes (Coster, Cambridge, MA) in a-MEM containing 1.2% methylcellulose (Aldrich Chemical Co., Milwaukee, WI), 50 U mL~1 IL-3, 30 ng mL~1 IL-6, 1% deionized BSA (fraction V; Sigma) and 30% FBS. Blast cell colonies in the cultures were lifted from the dishes and the cell number was counted with a hemocytometer. Fifteen microliters of medium containing 1000 blast cells, 5% FBS and 100 IU mL~1 GM-CSF were spotted into each well of a 48-well microplate (Coster, Cambridge, MA). After incubation for 6 h, 250 kL of the same medium containing 10~8 M 1,25(OH) D was added to the cells 2 3 and incubation was continued for 7 days. The cells were either left untreated or treated with whey protein. After 4 days of treatment, the cells were stained for TRAP. The number of TRAP-positive MNCs (multi nuclei cells) with three or more nuclei was measured. Chromatography Whey protein (50 g) was dissolved in 1 L of de-ionized water. After adjusting to pH 4.0, heat treating at 80°C for 10 min and centrifuging for 30 min at 5000 g, the supernatant was lyophilized. The heat-treated whey protein fraction was dissolved in 5 mM sodium phosphate buffer (pH 7.0), assayed chromatographically on a Mono S column (10/10) and eluted with a gradient of 0—1 M NaCl in the same buffer. Then, the basic protein (BP) fraction of bone resorption inhibitory activity was analyzed chromatographically on a Superose 12 gel chromatography (0.5 M NaCl in 5 mM sodium phosphate buffer). Molecular weight was determined using a molecular weight determination kit (Pharmacia). Transport study using an everted gut-sac of the rat small intestine Rat everted gut-sac was prepared from a SpragueDawley rat (Charles River, Japan) intestine (Martin and Deluca, 1969). The first 10 cm of the intestine from the stomach was dissected free and the tissue was immediately rinsed with Ringer’s buffer (150 mM NaCl, 4 mM KCl, 3 mM CaCl , pH 7.4). The intestine was everted in a man2 ner such that the distal end of the segment remained tied to the everting rod. It was then blotted, trimmed to a length of 5.5 cm. The everted gut-sac was filled with 0.5 mL of Ringer’s buffer supplemented with 0.5% BSA for keeping same osmotic pressure against the outside solution using 1 mL of syringe (Terumo, Tokyo) fitted with a blunt needle. The everted gut-sac was tied shut on both ends of intestine. The preparation of the everted gut-sac was completed within 3 min. The everted gut-sac was then placed and incubated in 300 mL Erlenmeyer flask containing 100 mL of a 0.5% solution of intact BP, 0.5% pepsin-digested BP, or pepsin/pancreatin-digested BP in Ringer’s buffer for 1 h at 37°C. Oxygen was continuously bubbled through the incubation buffer throughout the experiment. The incubation period of 1 h was chosen on the basis of experiments showing that there was little further transport after 90 min (Martin and Deluca, 1969). In experiments in which the inside fluid of the sac was changed, the sac was removed from the flask and most of the fluid allowed to drip off using kimtowel.
Whey protein suppresses the osteoclast-mediated bone resorption and osteoclast cell formation
823
The inner solution of everted gut-sac was collected with a syringe fitted with a needle and tested for bone resorption inhibitory activity using unfractionated bone cells. RESULTS The number of living unfractionated bone cells and percent of TRAP-positive multinucleate cells obtained from ten mice (five replicate experiments) were 1.1—1.5]108 cells and 0.10—0.20%, respectively. Bone marrow cells were obtained from a bone cut across its epiphysis by flushing out the cells with a syringe (needle gauge, 23G; 2 mL of a-MEM). The number of living bone marrow cells and percent of TRAP-positive multinucleate cells obtained from ten mice (five replicate experiments) were 3.2—4.0]107 cells and 0.0005%—0.001%, respectively. The number and percentage of osteoclastcontaining unfractionated bone cells was much higher than that of the osteoclast-containing bone marrow cells. The whey protein suppressed the area of pits formed by osteoclasts and reached a maximum at 1 mg mL~1, with the pit area being 2.4-fold smaller than the control value (Fig. 1A). The osteoclast bone resorption inhibitory activity of the whey protein was not affected by heating the protein at 75—95°C for 10 min (Fig. 2). We also examined the effect of whey protein on bone resorption using the newly formed osteoclast bone resorption evaluation method. When the whey protein was added to the culture at day 7, at which time the cultures contained only a few TRAP-positive cells (Takada et al., 1992), the whey protein inhibited the area of pits formed by osteoclasts depending on the dose and the effect reached a maximum at 1 mg mL~1, with the pit area being 2.1-fold smaller than the control value (Fig. 1B). The whey protein also inhibited the MNCs formation dose-dependently (Fig. 3). The inhibitory activity of whey protein on MNCs formation was not affected by heat treatment of the protein at 75—95°C for 10 min (data not shown). There was one major peak with apparent bone resorption inhibitory activity using unfractionated bone cells in a BP fraction (Fig. 4). The active fraction was then applied to a Superose 12 column. The molecular weights of
Fig. 2. Effect of heat treatment of whey protein on pre-existing osteoclast bone resorption using unfractionated bone cells. Results are expressed as the mean$SD. *Significantly different from the control group (P(0.05).
Fig. 3. Effect of whey protein on osteoclast formation from splenic blast cells using 5-fluorouracil-treated spleen cells. Results are expressed as the mean$SD. *Significantly different from the control group (P(0.05).
Fig. 1. Effect of whey protein on pre-existing osteoclast and newly formed osteoclast bone resorption using unfractionated bone cells. (A) Pre-existing osteoclast bone resorption, (B) newly formed osteoclast bone resorption. Results are expressed as the mean$SD. *Significantly different from the control group (P(0.05).
824
Y. Takada et al.
Fig. 4. Mono S cation ion-exchange chromatographic-pattern. Thick line represents the absorbance at 280 nm (A280); dotted line (— — —) represents the NaCl gradient; thin line (——) represents the fraction assay of pre-existed osteoclast bone resorption; dark box (d) (fraction 13—15) represents the active fraction.
Fig. 5. Gel filtration chromatographic pattern of whey protein. Thick line (——) represents the absorbance at 280 nm (A280); thin line (——) represents the fraction assay of pre-existed osteoclast bone resorption. Arrow represents the active fraction.
these active components were approximately 23 000 and 10 000 Da (Fig. 5). The inner solutions of the intact BPtreated, pepsin-digested BP-treated and pepsin/pancreatin digested BP-treated everted gut-sac also had bone resorption inhibitory activity using unfractionated bone cells (Fig. 6). DISCUSSION In our previous work, we established a biological system for evaluating bone resorption, osteoclast recruitment and activation using cultures of unfractionated bone cells on dentine slices (Takada et al., 1992). In this culturing system, it was possible to elucidate the effects of osteotropic factors such as 1,25(OH) D , hPTH and cCT 2 3 as reflecting the net effects of direct and indirect actions of factors as they occur in intact bones. Moreover, since unfractionated bone cells cultured for 6 days contain few osteoclasts, only bone resorption by newly formed osteoclasts can be evaluated (Takada et al., 1992). We first
used pre-existed osteoclast bone resorption and newly formed osteoclast formation. In the pre-existing and newly formed osteoclast bone resorption methods, the whey protein suppressed the area of pits formed by osteoclasts based on dose. Whey protein had direct or indirect effects on pre-existed osteoclasts. Whey protein also had direct or indirect effects on newly formed osteoclasts or new osteoclast formations. The whey protein was non-toxic as it increased the stromal cell number (data not shown). Next, we examined the effect of osteoclast formation with the osteoclastic cell (TRAP-positive MNCs) formation method with hemopoietic blast cells (Kurihara et al., 1989). In the TRAP-positive MNCs formation method, the osteoclastic cells formed were multinucleate and possessed several properties characteristic of osteoclasts, such as high levels of expression of TRAP and calcitonin receptors. The cells also resorbed the bone, as evidenced by their co-culturing with living bone rudiments. In this culturing system, no stromal cells are present and the cell population remains the same, so it is possible to evaluate osteoclastic cell formation from splenic blast
Whey protein suppresses the osteoclast-mediated bone resorption and osteoclast cell formation
825
Fig. 6. Effects of the inner solutions of an everted gut-sacs incubated in solutions of native or proteinase digested whey protein on pre-existing osteoclast bone resorption using unfractionated bone cells. Results are expressed as the mean$SD. *Significantly different from the control group (P(0.05).
cells precisely. Whey protein effect was also dose-dependent. These results suggest that whey protein inhibits osteoclast precursor differentiation. Osteoclast-mediated bone resorption activity was attributed to a basic protein (BP) fraction according to the results of Mono S chromatography. Most milk proteins such as casein, a-lactalbumin and b-lactoglobulin have acidic isoelectric points, but a few are basic. The molecular weights of the active components were comparatively low, approximately 23 000 and 10 000 Da, as estimated by the results of Superose 12 chromatography. Previously, we found that orally administered whey protein was effective in increasing bone strength and bone proteins such as collagen in OVX rats (Takada et al., 1993), which suggests that the active components are absorbed by the small intestine. Therefore, we examined the possible transport of the active components by using everted gut-sacs made from rat small intestine. In this system (Martin and Deluca, 1969), glucose and lactoferrin were detected in the inner solution of the everted gut-sac, but lactose and dextran were not. After incubating everted gut-sacs in solutions of intact BP, pepsin-digested BP or pepsin/pancreatin-digested BP, the inner solution was withdrawn and it also suppressed osteoclast-mediated bone resorption. These results suggested that the active component could be absorbed or transported by the intestines. These findings suggest that whey protein contains active components that suppress osteoclast-mediated bone resorption and osteoclastic cell formation and that these active components can probably be absorbed or transported by the intestines. To our knowledge, the data presented here are the first to indicate that whey protein affects bone resorption. It is necessary to study the significance of existing mechanisms concerning the effect of whey protein on changes in bone structure and bone metabolism in detail.
REFERENCES Glover, F. A. (1971) Concentration of milk by ultrafiltration and reverse osmosis. Journal of Dairy Research 38, 373—379. Hiddink, J., de Bore, R. and Romijn, D. J. (1978) Removal of milk salts during ultrafiltration of whey and buttermilk. Netherlands Milk and Dairy Journal 32, 80—93. Ito, T. (1991) Science of breast milk. New Food Industry 33(3), 73—80. Kessler, H. G. (1981) Ultrafiltration-processing variables. In Food Engineering and Dairy ¹echnology, ed. A. K. Freising, pp. 86—100. Federal Republic, Germany. Kurihara, N., Suda, T., Miura, Y., Nakamuchi, K., Kodama, H., Hiura, K., Hakeda and Y. and Kumegawa, M. (1989) Generation of osteoclasts from hematopoietic cells. Blood 74, 1295—1302. Martin, D. L. and Deluca, H. F. (1969) Influence of sodium on calcium transport by the rat small intestine. American Journal of Physiology 216, 1351—1359. Mundy, G. R. and Roodman, G. D. (1987) Osteoclast ontogeny and function. Bone and Mineral Research 5, 209—280. Skudder, P. J. (1985) Evaluation of a porous silica-based ion-exchange medium for the production of protein fractions from rennet- and acid-whey. Journal of Dairy Research 52, 167—168. Takada, Y., Yahiro, M. and Nakajima, I. (1993) Effect of milk components on calcium absorption and bone metabolism. In Characterization of Milk component and health, eds K. Yamauchi, T. Imamura and T. Morita, pp. 171—185. Kouseikan, Tokyo. Takada, Y., Aoe, S. and Kumegawa, M. (1996) Whey protein stimulates cell proliferation and differentiation in osteoblastic MC3T3-E1 cells. Biochemical and Biophysical Research Communication 223, 445—449. Takada, Y., Kusuda, M., Hiura, K., Sato, T., Mochizuki, H., Nagao, Y., Tomura, M., Yahiro, M., Hakeda, Y., Kawashima, H. and Kumegawa, M. (1992) A simple method to assess osteoclast-mediated bone resorption using unfractionated bone cells. Bone and Mineral Research 17, 347—359.
Int. Dairy Journal 7 (1997) 821—825 ( 1998 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0958-6946/98/$19.00#0.00
Whey Protein Suppresses the Osteoclast-mediated Bone Resorption and Osteoclast Cell Formation Yukihiro Takadaa *, Naomichi Kobayashia, Hiroaki Matsuyamaa, Ken Katoa, Junichi Yamamuraa, Masatoshi Yahiroa, Masayoshi Kumegawab and Seiichiro Aoea a Snow Brand Milk Product Co. Ltd., Nutritional Science Laboratory, 1-1-2 Minamidai, Kawagoe, Saitama 350-11, Japan b Department of Oral Anatomy, Meikai University School of Dentistry, 1-1 Keyakidai, Sakado, Saitama 350-02, Japan (Received 20 February 1997; accepted 19 October 1997) ABSTRACT Effects of whey protein on bone resorption and osteoclastic cell formation were evaluated. In the pre-existed and newly formed osteoclast, in bone resorption methods using an unfractionated bone cell culturing system, whey protein suppressed the area of pits formed by osteoclasts. In the osteoclastic cell formation method using the hemopoietic blast cell culturing system, whey protein also suppressed osteoclastic cell formation. These activities were resistant to heat when the protein was treated at 75—90°C for 10 min. Heat-treated whey protein was first fractionated on a Mono S column, and the active fraction (basic protein fraction) was then applied to Superose 12. The molecular weights of the active components were approximately 23 000 and 10 000 Da, as determined by gel filtration. The inner solution of an everted gut-sac incubated in a solution of intact basic protein (BP), pepsin-digested BP or pepsin/pancreatin-digested BP also suppressed osteoclast-mediated bone resorption. Thus, these active components can possibly be absorbed or transported by the intestines. These results showed that whey protein contains an active component that suppresses osteoclast-mediated bone resorption and osteoclastic cell formation. ( 1998 Elsevier Science Ltd. All rights reserved
INTRODUCTION
using cultures of unfractionated bone cells on dentine slices (Takada et al., 1992). Using this culturing system, it is possible to elucidate the effects of osteotropic factors, reflecting the net effects of direct and indirect actions of factors. On the other hand, Kurihara et al. (1989) developed an osteoclastic cell formation system from hematopoietic blast cells that were generated from the spleen cells of 5-fluorouracil-treated mice. In this study, we examined the effect of whey protein on osteoclast-mediated bone resorption using pre-existing and newly formed osteoclast bone resorption methods. The effect of whey protein on osteoclastic cell formation using the osteoclastic cell formation method from hemopoietic blast cells was also examined. We characterized the active component by ion-exchange chromatography and gel filtration chromatography and by using rat everted gut-sacs (Martin and Deluca, 1969). We also examined the permeability of the gut to the protein.
Milk is well known to be a safe food, beneficial to human health, that can be taken for a long time. Whey protein originates from milk and is obtained from whey by ultrafiltration, reverse osmosis, chromatography and dialysis to remove lactose and other components (Glover, 1971; Hiddink et al., 1978; Kessler, 1981; Skudder, 1985). Whey is obtained by adding an acid or rennet to milk and removing the formed coagulate. Whey protein is usually produced as a by-product of cheese or casein manufacturing. One-hundred mL of cow’s milk contains 0.55 g of whey protein and the major proteins are a-lactalbumin (20%), b-lactoglobulin (58%), immunoglobulin (13%) and serum albumin (7%) (Ito, 1991). Milk has a functional role in growth of newborn animals, and so milk whey protein may possibly have components that affect bone metabolism. Recently, we found that milk whey protein was effective in increasing bone strength and the content of collagen-specific amino acids such as a hydroxyproline in ovariectomized (OVX) rats (Takada et al., 1993). We also found that whey protein is effective for osteoblast cell proliferation and differentiation (Takada et al., 1996). Bone resorption is important for the metabolism of bone formation. Osteoclastic bone resorption is regulated at two different stages; the formative stage and the activation stage of osteoclasts (Mundy and Roodman, 1987). We established a biological system for evaluating bone resorption, osteoclast recruitment and activation
MATERIALS AND METHODS Materials Alpha-minimum essential medium (a-MEM) was purchased from Flow Laboratories (McLean, VA, USA) and fetal bovine serum (FBS) was obtained from Gibco (Grand Island, NY, USA). Whey protein was prepared from a whey protein concentrate (WPC) by dialysis. The WPC was purchased from New Zealand Dairy Board (Wellington, New Zealand). Mono S and Superose 12 columns were obtained from Pharmacia (Sweden).
* Corresponding author. 821
822
Y. Takada et al.
Suckling ICR and 6-week-old BDF1 mice were purchased from the Shizuoka Laboratory Animal Center (Shizuoka, Japan). 1,25(OH) D was obtained from Dr. 2 3 S. Ishizuka (Teijin Biomedical Research Institute, Tokyo, Japan). Murine recombinant interleukin-3 (IL-3) was provided by Dr. T. Sudo (Toray Industries Inc., Kanagawa, Japan). Human recombinant IL-6 and murine recombinant granulocytemacrophage-colony stimulating factor (GM-CSF) were kindly provided by Ajinomoto Co. Ltd. (Kanagawa, Japan) and Sumitomo Pharmaceutical Co. Ltd. (Osaka, Japan), respectively. Assay of pre-existing and newly formed osteoclast-mediated bone resorption using unfractionated bone cells Unfractionated bone cells were prepared from the femoral bones of 13-day-old mice (Takada et al., 1992). The bones were dissected free of soft tissues and transferred to a 50 mL test-tube. For preparation, the tube was vigorously agitated for 0.5 min with a test-tube mixer (Yamato, Tokyo, Japan) in 10 mL of a-MEM, and the large fragments were allowed to sediment for 3 min. The number of living cells was counted with a hemocytometer and cytospun cells were stained for tartrate-resistant acid phosphatase (TRAP) at 37°C for 5 min and for alkaline phosphatase to determine osteoclasts and osteoblasts, respectively (Takada et al., 1992). Dentine slices (150 km thick), prepared with a diamond saw, were punched out as circles 6 mm in diameter, sonicated and sterilized in 75% alcohol. After the slices had been washed with a-MEM, each was placed into a well of a 96-well plate. For assaying the effects of whey protein on bone resorption by pre-existing osteoclasts, 200 kL of a-MEM supplemented with 5% FBS plus 3]105 cells were added to each well containing a dentine slice, and the cells were then incubated at 37°C in a CO 2 incubator (5% CO : 95% air) for 2 h. Then, the slices 2 were transferred to fresh medium containing whey protein, and were incubated for the indicated number of days. The medium was changed every 2 days. To assay bone resorption by newly formed osteoclasts, we precultured the cells in whey protein-free medium for 6 days before transferring them to medium containing whey protein. After culturing for the 4 days, the cells were brushed off the dentine using a motorized hand-held brush (Yoshida Seiko Co. Ltd., Tokyo, Japan) for 10 s, and the dentine slices were then stained with acid hematoxylin (Sigma) for 3 min in the 96-well plate. The total area of pits was measured by an image analyzer (PIASLA555, PIAS Co. Ltd., Tokyo, Japan) to detect possible osteoclastic bone resorption. The pit area was determined from an image taken by a video camera attached to a light microscope (objective lens, ]2; TV camera magnification, ]14). An image of the pits alone was obtained by filtering and by using noise erasure treatments. After treatment into two values (black and white), the total surface area of the remaining image was measured as the pit area. The number of TRAP-positive cells was also determined. Assay of osteoclastic cell formation from splenic blast cells Six-week-old female BDF1 mice were administered 5-fluorouracil (Hoffman Laroche Co., Basal, Switzer-
land) at a dosage of 150 mg kg~1 through a tail vein. Splenic blast cells were prepared as described previously (Kurihara et al., 1989). Spleen cells from the treated mice were cultured in 35 mm non-tissue culture dishes (Coster, Cambridge, MA) in a-MEM containing 1.2% methylcellulose (Aldrich Chemical Co., Milwaukee, WI), 50 U mL~1 IL-3, 30 ng mL~1 IL-6, 1% deionized BSA (fraction V; Sigma) and 30% FBS. Blast cell colonies in the cultures were lifted from the dishes and the cell number was counted with a hemocytometer. Fifteen microliters of medium containing 1000 blast cells, 5% FBS and 100 IU mL~1 GM-CSF were spotted into each well of a 48-well microplate (Coster, Cambridge, MA). After incubation for 6 h, 250 kL of the same medium containing 10~8 M 1,25(OH) D was added to the cells 2 3 and incubation was continued for 7 days. The cells were either left untreated or treated with whey protein. After 4 days of treatment, the cells were stained for TRAP. The number of TRAP-positive MNCs (multi nuclei cells) with three or more nuclei was measured. Chromatography Whey protein (50 g) was dissolved in 1 L of de-ionized water. After adjusting to pH 4.0, heat treating at 80°C for 10 min and centrifuging for 30 min at 5000 g, the supernatant was lyophilized. The heat-treated whey protein fraction was dissolved in 5 mM sodium phosphate buffer (pH 7.0), assayed chromatographically on a Mono S column (10/10) and eluted with a gradient of 0—1 M NaCl in the same buffer. Then, the basic protein (BP) fraction of bone resorption inhibitory activity was analyzed chromatographically on a Superose 12 gel chromatography (0.5 M NaCl in 5 mM sodium phosphate buffer). Molecular weight was determined using a molecular weight determination kit (Pharmacia). Transport study using an everted gut-sac of the rat small intestine Rat everted gut-sac was prepared from a SpragueDawley rat (Charles River, Japan) intestine (Martin and Deluca, 1969). The first 10 cm of the intestine from the stomach was dissected free and the tissue was immediately rinsed with Ringer’s buffer (150 mM NaCl, 4 mM KCl, 3 mM CaCl , pH 7.4). The intestine was everted in a man2 ner such that the distal end of the segment remained tied to the everting rod. It was then blotted, trimmed to a length of 5.5 cm. The everted gut-sac was filled with 0.5 mL of Ringer’s buffer supplemented with 0.5% BSA for keeping same osmotic pressure against the outside solution using 1 mL of syringe (Terumo, Tokyo) fitted with a blunt needle. The everted gut-sac was tied shut on both ends of intestine. The preparation of the everted gut-sac was completed within 3 min. The everted gut-sac was then placed and incubated in 300 mL Erlenmeyer flask containing 100 mL of a 0.5% solution of intact BP, 0.5% pepsin-digested BP, or pepsin/pancreatin-digested BP in Ringer’s buffer for 1 h at 37°C. Oxygen was continuously bubbled through the incubation buffer throughout the experiment. The incubation period of 1 h was chosen on the basis of experiments showing that there was little further transport after 90 min (Martin and Deluca, 1969). In experiments in which the inside fluid of the sac was changed, the sac was removed from the flask and most of the fluid allowed to drip off using kimtowel.
Whey protein suppresses the osteoclast-mediated bone resorption and osteoclast cell formation
823
The inner solution of everted gut-sac was collected with a syringe fitted with a needle and tested for bone resorption inhibitory activity using unfractionated bone cells. RESULTS The number of living unfractionated bone cells and percent of TRAP-positive multinucleate cells obtained from ten mice (five replicate experiments) were 1.1—1.5]108 cells and 0.10—0.20%, respectively. Bone marrow cells were obtained from a bone cut across its epiphysis by flushing out the cells with a syringe (needle gauge, 23G; 2 mL of a-MEM). The number of living bone marrow cells and percent of TRAP-positive multinucleate cells obtained from ten mice (five replicate experiments) were 3.2—4.0]107 cells and 0.0005%—0.001%, respectively. The number and percentage of osteoclastcontaining unfractionated bone cells was much higher than that of the osteoclast-containing bone marrow cells. The whey protein suppressed the area of pits formed by osteoclasts and reached a maximum at 1 mg mL~1, with the pit area being 2.4-fold smaller than the control value (Fig. 1A). The osteoclast bone resorption inhibitory activity of the whey protein was not affected by heating the protein at 75—95°C for 10 min (Fig. 2). We also examined the effect of whey protein on bone resorption using the newly formed osteoclast bone resorption evaluation method. When the whey protein was added to the culture at day 7, at which time the cultures contained only a few TRAP-positive cells (Takada et al., 1992), the whey protein inhibited the area of pits formed by osteoclasts depending on the dose and the effect reached a maximum at 1 mg mL~1, with the pit area being 2.1-fold smaller than the control value (Fig. 1B). The whey protein also inhibited the MNCs formation dose-dependently (Fig. 3). The inhibitory activity of whey protein on MNCs formation was not affected by heat treatment of the protein at 75—95°C for 10 min (data not shown). There was one major peak with apparent bone resorption inhibitory activity using unfractionated bone cells in a BP fraction (Fig. 4). The active fraction was then applied to a Superose 12 column. The molecular weights of
Fig. 2. Effect of heat treatment of whey protein on pre-existing osteoclast bone resorption using unfractionated bone cells. Results are expressed as the mean$SD. *Significantly different from the control group (P(0.05).
Fig. 3. Effect of whey protein on osteoclast formation from splenic blast cells using 5-fluorouracil-treated spleen cells. Results are expressed as the mean$SD. *Significantly different from the control group (P(0.05).
Fig. 1. Effect of whey protein on pre-existing osteoclast and newly formed osteoclast bone resorption using unfractionated bone cells. (A) Pre-existing osteoclast bone resorption, (B) newly formed osteoclast bone resorption. Results are expressed as the mean$SD. *Significantly different from the control group (P(0.05).
824
Y. Takada et al.
Fig. 4. Mono S cation ion-exchange chromatographic-pattern. Thick line represents the absorbance at 280 nm (A280); dotted line (— — —) represents the NaCl gradient; thin line (——) represents the fraction assay of pre-existed osteoclast bone resorption; dark box (d) (fraction 13—15) represents the active fraction.
Fig. 5. Gel filtration chromatographic pattern of whey protein. Thick line (——) represents the absorbance at 280 nm (A280); thin line (——) represents the fraction assay of pre-existed osteoclast bone resorption. Arrow represents the active fraction.
these active components were approximately 23 000 and 10 000 Da (Fig. 5). The inner solutions of the intact BPtreated, pepsin-digested BP-treated and pepsin/pancreatin digested BP-treated everted gut-sac also had bone resorption inhibitory activity using unfractionated bone cells (Fig. 6). DISCUSSION In our previous work, we established a biological system for evaluating bone resorption, osteoclast recruitment and activation using cultures of unfractionated bone cells on dentine slices (Takada et al., 1992). In this culturing system, it was possible to elucidate the effects of osteotropic factors such as 1,25(OH) D , hPTH and cCT 2 3 as reflecting the net effects of direct and indirect actions of factors as they occur in intact bones. Moreover, since unfractionated bone cells cultured for 6 days contain few osteoclasts, only bone resorption by newly formed osteoclasts can be evaluated (Takada et al., 1992). We first
used pre-existed osteoclast bone resorption and newly formed osteoclast formation. In the pre-existing and newly formed osteoclast bone resorption methods, the whey protein suppressed the area of pits formed by osteoclasts based on dose. Whey protein had direct or indirect effects on pre-existed osteoclasts. Whey protein also had direct or indirect effects on newly formed osteoclasts or new osteoclast formations. The whey protein was non-toxic as it increased the stromal cell number (data not shown). Next, we examined the effect of osteoclast formation with the osteoclastic cell (TRAP-positive MNCs) formation method with hemopoietic blast cells (Kurihara et al., 1989). In the TRAP-positive MNCs formation method, the osteoclastic cells formed were multinucleate and possessed several properties characteristic of osteoclasts, such as high levels of expression of TRAP and calcitonin receptors. The cells also resorbed the bone, as evidenced by their co-culturing with living bone rudiments. In this culturing system, no stromal cells are present and the cell population remains the same, so it is possible to evaluate osteoclastic cell formation from splenic blast
Whey protein suppresses the osteoclast-mediated bone resorption and osteoclast cell formation
825
Fig. 6. Effects of the inner solutions of an everted gut-sacs incubated in solutions of native or proteinase digested whey protein on pre-existing osteoclast bone resorption using unfractionated bone cells. Results are expressed as the mean$SD. *Significantly different from the control group (P(0.05).
cells precisely. Whey protein effect was also dose-dependent. These results suggest that whey protein inhibits osteoclast precursor differentiation. Osteoclast-mediated bone resorption activity was attributed to a basic protein (BP) fraction according to the results of Mono S chromatography. Most milk proteins such as casein, a-lactalbumin and b-lactoglobulin have acidic isoelectric points, but a few are basic. The molecular weights of the active components were comparatively low, approximately 23 000 and 10 000 Da, as estimated by the results of Superose 12 chromatography. Previously, we found that orally administered whey protein was effective in increasing bone strength and bone proteins such as collagen in OVX rats (Takada et al., 1993), which suggests that the active components are absorbed by the small intestine. Therefore, we examined the possible transport of the active components by using everted gut-sacs made from rat small intestine. In this system (Martin and Deluca, 1969), glucose and lactoferrin were detected in the inner solution of the everted gut-sac, but lactose and dextran were not. After incubating everted gut-sacs in solutions of intact BP, pepsin-digested BP or pepsin/pancreatin-digested BP, the inner solution was withdrawn and it also suppressed osteoclast-mediated bone resorption. These results suggested that the active component could be absorbed or transported by the intestines. These findings suggest that whey protein contains active components that suppress osteoclast-mediated bone resorption and osteoclastic cell formation and that these active components can probably be absorbed or transported by the intestines. To our knowledge, the data presented here are the first to indicate that whey protein affects bone resorption. It is necessary to study the significance of existing mechanisms concerning the effect of whey protein on changes in bone structure and bone metabolism in detail.
REFERENCES Glover, F. A. (1971) Concentration of milk by ultrafiltration and reverse osmosis. Journal of Dairy Research 38, 373—379. Hiddink, J., de Bore, R. and Romijn, D. J. (1978) Removal of milk salts during ultrafiltration of whey and buttermilk. Netherlands Milk and Dairy Journal 32, 80—93. Ito, T. (1991) Science of breast milk. New Food Industry 33(3), 73—80. Kessler, H. G. (1981) Ultrafiltration-processing variables. In Food Engineering and Dairy ¹echnology, ed. A. K. Freising, pp. 86—100. Federal Republic, Germany. Kurihara, N., Suda, T., Miura, Y., Nakamuchi, K., Kodama, H., Hiura, K., Hakeda and Y. and Kumegawa, M. (1989) Generation of osteoclasts from hematopoietic cells. Blood 74, 1295—1302. Martin, D. L. and Deluca, H. F. (1969) Influence of sodium on calcium transport by the rat small intestine. American Journal of Physiology 216, 1351—1359. Mundy, G. R. and Roodman, G. D. (1987) Osteoclast ontogeny and function. Bone and Mineral Research 5, 209—280. Skudder, P. J. (1985) Evaluation of a porous silica-based ion-exchange medium for the production of protein fractions from rennet- and acid-whey. Journal of Dairy Research 52, 167—168. Takada, Y., Yahiro, M. and Nakajima, I. (1993) Effect of milk components on calcium absorption and bone metabolism. In Characterization of Milk component and health, eds K. Yamauchi, T. Imamura and T. Morita, pp. 171—185. Kouseikan, Tokyo. Takada, Y., Aoe, S. and Kumegawa, M. (1996) Whey protein stimulates cell proliferation and differentiation in osteoblastic MC3T3-E1 cells. Biochemical and Biophysical Research Communication 223, 445—449. Takada, Y., Kusuda, M., Hiura, K., Sato, T., Mochizuki, H., Nagao, Y., Tomura, M., Yahiro, M., Hakeda, Y., Kawashima, H. and Kumegawa, M. (1992) A simple method to assess osteoclast-mediated bone resorption using unfractionated bone cells. Bone and Mineral Research 17, 347—359.