Diets Higher in Dairy Foods and Dietary Protein Support Bone Health ...

ORIGINAL E n d o c r i n e

ARTICLE R e s e a r c h

Diets Higher in Dairy Foods and Dietary Protein Support Bone Health during Diet- and Exercise-Induced Weight Loss in Overweight and Obese Premenopausal Women Andrea R. Josse, Stephanie A. Atkinson, Mark A. Tarnopolsky, and Stuart M. Phillips Exercise Metabolism Research Group, Department of Kinesiology (A.R.J., S.M.P.), and Departments of Pediatrics (S.A.A.) and Pediatrics and Medicine (M.A.T.), McMaster University, Hamilton, Ontario, Canada L8S 4K1

Context: Consolidation and maintenance of peak bone mass in young adulthood may be compromised by inactivity, low dietary calcium, and diet-induced weight loss. Objective: We aimed to determine whether higher intakes of dairy foods, dietary calcium, and protein during diet- and exercise-induced weight loss affected markers of bone health. Participants: Participants included premenopausal overweight and obese women. Design/Intervention: Ninety participants were randomized into three groups (n ⫽ 30 per group): high protein and high dairy (HPHD), adequate protein and medium dairy (APMD), and adequate protein and low dairy (APLD), differing in dietary protein (30, 15, or 15% of energy, respectively), dairy foods (15, 7.5, or ⬍2% of energy from protein, respectively), and dietary calcium (⬃1600, ⬃1000, or ⬍500 mg/d, respectively). Outcome Measures: Serum and urine bone turnover biomarkers, serum osteoprotegerin (OPG), receptor activator of nuclear factor-␬B ligand (RANKL), PTH, 25-hydroxyvitamin D, leptin, and adiponectin measured at 0 and 16 wk. Results: All groups lost equivalent body weight (P ⬍ 0.05). N-telopeptide, C-telopeptide (CTX), urinary deoxypyridinoline, and osteocalcin increased in APLD (P ⬍ 0.01), whereas in HPHD, osteocalcin and procollagen 1 amino-terminal propeptide (P1NP) increased (P ⬍ 0.05), and all resorption markers remained unchanged. P1NP to CTX and OPG to RANKL ratios increased in HPHD (P ⬍ 0.005), and P1NP to CTX ratio decreased in APLD (P ⬍ 0.05). PTH decreased in HPHD and APMD vs. APLD (P ⬍ 0.005), and 25-hydroxyvitamin D increased in HPHD (P ⬍ 0.05), remained unchanged in APMD, and decreased in APLD (P ⬍ 0.05). Leptin decreased and adiponectin increased in APMD and HPHD only (P ⬍ 0.001). Conclusions: Hypoenergetic diets higher in dairy foods, dietary calcium, and protein with daily exercise, favorably affected important bone health biomarkers vs. diets with less of these bonesupporting nutrients. (J Clin Endocrinol Metab 97: 251–260, 2012)

ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2012 by The Endocrine Society doi: 10.1210/jc.2011-2165 Received July 27, 2011. Accepted October 6, 2011. First Published Online November 2, 2011

Abbreviations: APLD, Adequate protein and low dairy; APMD, adequate protein and medium dairy; BMC, bone mineral content; BMD, bone mineral density; BSAP, bone-specific alkaline phosphatase; CTX, C-telopeptide; CV, coefficient of variation; 25(OH)D, 25-hydroxyvitamin D; DXA, dual-energy x-ray absorptiometry; HPHD, high protein and high dairy; IDEAL, Improving Diet, Exercise and Lifestyle; NTX, N-telopeptide; OC, osteocalcin; OPG, osteoprotegerin; P1NP, procollagen 1 amino-terminal propeptide; RANKL, receptor activator of nuclear factor-␬B ligand; uDPD, urinary deoxypyridinoline.

J Clin Endocrinol Metab, January 2012, 97(1):251–260

jcem.endojournals.org

251

252

Josse et al.

Dairy, Weight Loss, and Bone Health in Young Women

he majority of adult bone mass is accrued by the end of adolescence with final consolidation between the second and third decade of life (1, 2). Efforts to maximize the accretion of bone by early adulthood, and its maintenance thereafter, are important because attaining and maintaining a higher peak bone mass may delay the decline in bone health with age and lessen the disease burden of osteoporosis (3, 4). Although genetics plays a large role in determining an individual’s peak bone mass, lifestyle factors such as nutrition and exercise can also make a considerable contribution to its accrual and maintenance (2). Dairy foods provide about 70% of the dietary calcium in the diets of Canadians (4) and Americans (5, 6). It has been suggested that without the consumption of dairy foods, it would be difficult to meet the current dietary recommendations for several essential nutrients including calcium, potassium, magnesium, vitamin D (if fortified), and certain B vitamins (6, 7). Dairy foods are also good sources of high quality protein (8). With respect to bone health, calcium and protein contribute to the structural integrity and strength of bone by influencing bone mineralization (via the formation of hydroxyapatite crystals) and collagen formation, respectively (2, 7). Higher intakes of dietary calcium and vitamin D reduce circulating PTH concentrations, which also positively affects bone mass and reduces rates of bone turnover (2). Although a higher body weight is associated with greater bone mass (9), weight loss through energy restriction can adversely affect bone health (10, 11). The relationship between body mass and bone likely reflects the established positive effect that mechanical loading has on bone (10, 12) and, conversely, the negative effect that prolonged inactivity/unloading has on bone mass (13). Nonetheless, recent clinical trials have demonstrated that the reductions in bone mass sometimes observed with energy restriction can be offset with increased consumption of dietary protein (emphasizing dairy), increased dietary calcium intake, and exercise (10, 14, 15). Strategies to maintain or improve bone health during weight loss should include increased intake of bone-supporting nutrients such as protein, calcium, and vitamin D as well as weight-bearing exercise to stimulate bone. Although the individual effects of dairy, calcium, protein (14 –17), and exercise (10, 12, 18) on bone during weight loss have been studied in premenopausal women, no trial has combined all these aforementioned strategies together in one study to support bone health. Therefore, we conducted a controlled randomized weight loss intervention trial, the Improving Diet Exercise and Lifestyle (IDEAL) for Women Study (19), which was designed to achieve weight loss with a high ratio of fat to lean (muscle) mass loss and be supportive of bone health. We employed mod-

T


est dietary energy restriction (⬃500 kcal/d) and daily exercise (⬃250 kcal/d) including aerobic and resistance training with varied intakes of protein and dairy foods. We hypothesized that during 16 wk of energy restriction with daily exercise, higher dairy, higher protein, and dietary calcium consumed at or above recommended daily intakes (20 –22) would provide adequate levels of bone-supporting nutrients to positively affect markers of bone health in overweight or obese, premenopausal, young women. Furthermore, we hypothesized that body weight and fat loss would affect circulating levels of leptin and adiponectin and that the changes in adipokine levels relate to changes in several measured bone markers.

Subjects and Methods Participants The IDEAL for Women study was approved by the research ethics board of the Hamilton Health Sciences and conformed to the most recent Canadian government tri-council funding policy statement on the use of human subjects in research (available at www.pre.ethics.gc.ca/eng/index/) (23). All participants were premenopausal, overweight or obese women (body mass index between 27 and 40 kg/m2) between the ages of 19 and 45y. Other general inclusion criteria included participants who were otherwise healthy (as assessed through a standard medical screening questionnaire), habitually low dairy consumers (no more than one serving of dairy per day or ⬍600 mg calcium/d), generally sedentary, regularly menstruating, not pregnant or nursing, and not taking vitamin or mineral supplements. Additional details on participant recruitment procedures and participant flow have been outlined elsewhere (19).

Study protocol The IDEAL for Women study was a randomized, controlled, parallel intervention trial where participants were randomly assigned to one of three groups: high protein and high dairy (HPHD), adequate protein and medium dairy (APMD), and adequate protein and low dairy (APLD). This report focuses on the bone-specific outcomes assessed in the IDEAL for Women study including dietary calcium, protein, and vitamin D; serum 25hydroxyvitamin D [25(OH)D), PTH, osteoprotegerin (OPG), receptor activator of nuclear factor-␬B ligand (RANKL), leptin, and adiponectin; serum and urinary biomarkers of bone turnover; and bone mineral density (BMD) and bone mineral content (BMC) obtained by dual-energy x-ray absorptiometry (DXA). Results for body composition, anthropometry, strength, fitness, blood lipids, and inflammatory markers have been previously reported (19).

Diets and energy restriction Maintenance energy requirements were calculated per participant using the Mifflin-St. Jeor equation (24) with a sedentary activity factor. Once determined, it was reduced by 500 kcal and used as the participant’s new maximum energy intake level. The three intervention groups differed in the amount and type of protein consumed: the APLD group maintained stable baseline


TABLE 1.


253

Baseline characteristics of participants in the IDEAL for Women Study

Variable BMI (kg/m2) Height (m) Age (yr) Body weight (kg) BMD (g/cm2) BMC (g) PTH (pmol/liter) 25(OH)D (nmol/liter) OC (␮g/liter) BSAP (␮g/liter) P1NP (␮g/liter) NTX (nmol/liter BCE) CTX (nmol/liter BCE) uDPD (nM/mM Cr) OPG (ng/liter) RANKL (ng/liter) Leptin (mg/liter) Adiponectin (mg/liter)

APLD 31.5 ⫾ 0.6 163 ⫾ 1 28 ⫾ 1 84.0 ⫾ 2.1 1.17 ⫾ 0.02 2333 ⫾ 54 5.0 ⫾ 0.3 48.4 ⫾ 3.2 7.8 ⫾ 0.4 16.3 ⫾ 1.0 55.2 ⫾ 1.8 11.5 ⫾ 0.3 0.74 ⫾ 0.01 5.3 ⫾ 0.07 94.7 ⫾ 1.6 5.9 ⫾ 0.3 10.3 ⫾ 0.3 9.2 ⫾ 0.2

APMD 31.8 ⫾ 0.6 164 ⫾ 1 26 ⫾ 1 85.3 ⫾ 2.1 1.13 ⫾ 0.02 2259 ⫾ 47 4.7 ⫾ 0.2 49.3 ⫾ 3.1 7.7 ⫾ 0.4 14.5 ⫾ 0.7 52.9 ⫾ 1.7 10.5 ⫾ 0.3 0.75 ⫾ 0.01 5.3 ⫾ 0.07 94.1 ⫾ 1.5 6.1 ⫾ 0.3 10.5 ⫾ 0.3 8.9 ⫾ 0.3

HPHD 31.4 ⫾ 0.6 166 ⫾ 1 30 ⫾ 1 87.1 ⫾ 2.1 1.17 ⫾ 0.02 2427 ⫾ 57 5.4 ⫾ 0.3 51.0 ⫾ 4.0 7.5 ⫾ 0.5 15.7 ⫾ 1.0 49.2 ⫾ 2.0 10.8 ⫾ 0.3 0.77 ⫾ 0.01 5.4 ⫾ 0.07 97.5 ⫾ 1.5 6.0 ⫾ 0.3 10.2 ⫾ 0.4 8.8 ⫾ 0.2

P value (ANOVA) 0.88 0.12 0.38 0.55 0.17 0.084 0.22 0.88 0.90 0.36 0.11 0.087 0.26 0.70 0.27 0.90 0.80 0.49

Values are means ⫾ SE; n ⫽ 90 (30 per group). BCE, Bone collagen equivalents; Cr, creatinine; BMI, body mass index.

dairy intakes of zero to one serving per day and consumed 15% of their daily energy from nondairy sources of high quality protein (meat, eggs, fish, chicken, pork, legumes, soy, and wheat); the APMD group had three to four servings of dairy per day and consumed 15% of their daily energy from high quality protein (7.5% of energy as protein from dairy); and the HPHD group had six to seven servings of dairy per day and consumed 30% of their daily energy from high quality protein (15% of energy as protein from dairy). All participants received individualized diet counseling by study dietitians and research nutritionists on a biweekly basis to ensure dietary compliance. Additional details relating to the diets and nutritional counseling have been reported elsewhere (19).

Study foods Two groups (APMD and HPHD) received all dairy products needed to control their daily dairy and calcium intakes (donated by Agropur Dairy Cooperative, Longueuil, Quebec, Canada). Please refer to Table 1 in our earlier publication for the nutritional breakdown of the study foods (19).

Exercise training Participants all underwent the same exercise protocol. They engaged in various modes of aerobic exercise everyday, 5 d/wk with us and 2 d/wk on their own on weekends. In addition, participants engaged in a progressive, individualized resistance training regimen 2 d/wk (upper body, lower body split) with either a personal trainer or kinesiologist. Additional details regarding the exercise program are outlined elsewhere (19).

Dual-energy x-ray absorptiometry DXA scans (QDR-4500A; Hologic Inc., Waltham, MA; software version 12.31) were carried out at the McMaster University Medical Centre. Participants underwent DXA scans at wk 0 and 16 to determine whole-body BMD and BMC. For each scan, participants wore a standard hospital gown or the same loose clothing of their own and were scanned at the same time of day. They were also asked to follow a similar daily routine on the scan days (i.e. controlling their daily exercise and meal times).

Blood and urine samples Blood samples were obtained before and after the intervention between 0630 and 1000 h after an overnight fast of 10 –12 h into tubes containing no additives (serum). They were then processed (centrifuged and aliquoted) in our laboratory within 1 h of collection and stored in ⫺20 C freezers for later analysis.

TABLE 2. Dietary intakes from participants in the IDEAL for Women study Dietary variable Energy intake (kcal/d)d Energy restriction achieved (kcal/d)d Protein %/d g/d g/kg 䡠 d Fat (%/d) Carbohydrate (%/d) Dietary fiber (g/d) Calcium (mg/d) Vitamin D (␮g/d)e Vitamin C (mg/d) Vitamin A (mg RAE/d) Iron (mg/d)

APLD 1320 ⫾ 40a

APMD HPHD a,b 1430 ⫾ 42 1500 ⫾ 36b

⫺498 ⫾ 41

⫺477 ⫾ 42

⫺435 ⫾ 37

18 ⫾ 1b 28 ⫾ 1c 16 ⫾ 1a a b 55 ⫾ 1 66 ⫾ 2 108 ⫾ 3c 0.72 ⫾ 0.02a 0.84 ⫾ 0.02b 1.33 ⫾ 0.04c 28 ⫾ 1a 24 ⫾ 1b 31 ⫾ 1c 56 ⫾ 1a 58 ⫾ 1a 41 ⫾ 1b 21 ⫾ 1 299 ⫾ 22a 0.7 ⫾ 0.1a

18 ⫾ 1 16 ⫾ 1 1200 ⫾ 19b 1840 ⫾ 13c 9.8 ⫾ 0.3b 13.2 ⫾ 0.2c

119 ⫾ 19a 79 ⫾ 6a,b 76 ⫾ 6b 1.71 ⫾ 0.22a 1.85 ⫾ 0.16a 2.55 ⫾ 0.18b 10 ⫾ 0.6a

15 ⫾ 0.5b

11 ⫾ 0.6a

Values are means ⫾ SE; n ⫽ 90 (30 per group). Data were taken from poststudy 7-d food records. Table is adapted from Ref. 19. RAE, Retinol activity equivalents. a– c Means in a row with superscripts without a common letter differ, P ⬍ 0.05. d

1 kcal ⫽ 4.18 kJ.

e

1 ␮g ⫽ 40 IU.

254

Josse et al.


Once the study was completed, frozen serum samples were taken to the Core Laboratory at the McMaster University Medical Centre for analysis of 25(OH)D by RIA [DiaSorin Canada Inc., Mississauga, Ontario, Canada; coefficient of variation (CV) of 14%] and intact PTH by Immulite 2500 autoanalyzer (Siemens Healthcare Diagnostics, Deerfield, IL; CV of 4%) CV of 4%. Osteocalcin (OC) by RIA (Nichols Institute Diagnostics, San Juan Capistrano, CA; CV of 6%), bone-specific alkaline phosphatase (BSAP) by RIA (Quidel, San Diego, CA; CV of 4%), procollagen type 1 N-terminal propeptide (P1NP) by ELISA (antibodies-online Inc. Atlanta, GA; CV of 4%), C-telopeptide (CTX) by ELISA (Nordic Bioscience, Herlev, Denmark; CV of 1%), N-telopeptide (NTX) by ELISA (Ostex International, Inc., Seattle, WA; CV of 4%), OPG by ELISA (Biomedica-Gruppe, Wien, Austria; CV of 1%), RANKL by ELISA (BiomedicaGruppe; CV of 6%), leptin by ELISA (Millipore, Etobicoke, Ontario, Canada; CV of 5%), and adiponectin by ELISA (Millipore; CV of 3%) were batch analyzed upon study completion. Fasting spot urine samples were obtained at the same time as the blood samples, and aliquots were stored at ⫺20 C for later analysis. Upon study completion, frozen urine samples were analyzed for urinary deoxypyridinoline (uDPD) by ELISA (Pyrilinks-D; Metra Biosystems, Mountain View, CA; CV of 3%) and corrected for urinary creatinine (Cayman Chemical Co., Ann Arbor, MI). All investigators and laboratory technicians were unaware of the participants’ group assignment during analysis.


Statistics Statistical analyses were performed using SPSS version 18.0 (SPSS, Chicago, IL). Statistical significance was set at P ⬍ 0.05, and all data are expressed as means ⫾ SE. Differences between groups in all baseline variables were compared by univariate ANOVA. Analyses of variables over time were performed using a two-way, repeated-measures ANOVA with time (before and after) as the within-subject factor, and group (APLD, APMD, and HPHD) as the between-subject factor. Significant F ratios were further analyzed, and differences were isolated with Tukey’s post hoc test. Univariate ANOVA with Tukey’s post hoc tests were carried out to compare changes from baseline between groups. Pearson correlation coefficients were also calculated for several variables.

Results Of the 90 women randomized, nine dropped out after wk 8 for reasons unrelated to the study: six in the APLD group and three in the HPHD group. Body composition results demonstrated that, despite equivalent weight loss across all groups (pooled mean change over time, ⫺4.3 ⫾ 0.7 kg, P ⬍ 0.05), the HPHD group experienced greater total fat loss, visceral fat loss, and lean (muscle) mass gains (19). Table 1 shows the baseline values for all measured variables. None of the baseline values were statistically different between any of the groups. For baseline dietary intake data, please refer to the other publication (19). Dietary intakes Dietary protein consumption increased in the HPHD group, and dietary calcium and vitamin D intake increased in the HPHD and APMD groups and decreased in the APLD group over 16 wk (Table 2). Dual-energy x-ray absorptiometry We observed no significant changes in any group over 16 wk in BMD (g/ cm2: APLD, ⫺0.02 ⫾ 0.01; APMD, 0.01 ⫾ 0.001; HPHD, 0.00 ⫾ 0.004) or BMC (g: APLD, ⫺8.16 ⫾ 7.4; APMD, 9.9 ⫾ 7.0; HPHD, ⫺1.9 ⫾ 7.3).

FIG. 1. Serum 25(OH)D and PTH levels before and after intervention. A, Serum 25(OH)D. *, P ⬍ 0.05; **, P ⬍ 0.005 vs. HPHD at same time point; means not bearing the same letter within a group are significantly different from each other, P ⬍ 0.05. B, ⌬Serum 25(OH)D. *, P ⱕ 0.001. C, Serum PTH. *, P ⬍ 0.001 vs. APLD at same time point; means not bearing the same letter within a group are significantly different from each other, P ⬍ 0.05. D, ⌬Serum PTH. *, P ⬍ 0.005.

25-Hydroxyvitamin D 25(OH)D (Fig. 1, A and B) increased significantly in the HPHD group, whereas the APMD group remained unchanged and the APLD group showed a significant reduction over 16 wk. For both the APMD and APLD groups, serum 25(OH)D at wk 16 was



255

APMD group (Fig. 2, A and B). BSAP showed no change over time both within and between groups (Fig. 2, E and F). P1NP increased significantly in both the HPHD and APMD groups with no change in the APLD group (Fig. 2, C and D). Markers of bone resorption: NTX, CTX, and uDPD NTX and CTX increased significantly in the APLD group, and these increases were significantly greater than those observed in both the APMD and HPHD groups (Fig. 3, A–D). The HPHD group showed no significant change in any serum resorption marker, and NTX was significantly increased in the APMD group. uDPD did not change over time in the HPHD or APMD groups, but increased significantly in the APLD group (Fig. 3, E and F). P1NP to CTX ratio Expressedasthenetchange(afterminus before), the APLD group (⫺9.0 ⫾ 4.1 ␮g/ nmol) was significantly lower than both the HPHD (21.6 ⫾ 4.1 ␮g/nmol, P ⬍ 0.001) and APMD (11.8 ⫾ 3.8 ␮g/ nmol, P ⬍ 0.001) groups. There was no significant difference between the HPHD and APMD groups. OPG and RANKL OPG increased significantly in both the APMD and HPHD groups with no change in the APLD group over 16 wk (Fig. 4, A and B). The change in OPG in the HPHD group was significantly greater than the APLD group (P ⬍ 0.005). RANKL was significantly decreased in the HPHD group, and this decrease was significantly greater than the slight increase observed in the APLD group (Fig. 4, C and D). The OPG to RANKL ratio increased significantly over 16 wk only in the HPHD group, and at wk 16, the ratio in the HPHD group was significantly greater than both the APMD and APLD groups (Fig. 4E).

FIG. 2. Serum markers of bone formation before and after intervention. A, Serum OC. *, P ⱕ 0.001 vs. APLD at same time point; means not bearing the same letter within a group are significantly different from each other, P ⬍ 0.05. B, ⌬Serum OC. **, P ⬍ 0.005; *, P ⬍ 0.05. C, Serum P1NP. *, P ⬍ 0.05; **, P ⬍ 0.005 vs. APLD at same time point; means not bearing the same letter within a group are significantly different from each other, P ⬍ 0.001. D, ⌬Serum P1NP. *, P ⬍ 0.005. E, Serum BSAP (no statistical differences). F, ⌬Serum BSAP (no statistical differences).

significantly lower than the HPHD group (P ⬍ 0.05; Fig. 1A). Intact PTH Both the APMD and HPHD groups showed significant reductions in PTH (Fig. 1, C and D) over time with a greater reduction in the HPHD group, whereas the APLD group showed a significant increase over 16 wk. Serum PTH at wk 16 in the APLD group was significantly higher than the other two groups (P ⬍ 0.001) (Fig. 1C). Markers of bone formation: OC, BSAP, and P1NP Serum OC increased significantly over time in the HPHD and APLD groups and remained unchanged in the

Leptin and adiponectin Leptin decreased significantly in the APMD and HPHD groups over 16 wk with no significant change in the APLD group (Fig. 5, A and B). At wk 16, the HPHD group had

256

Josse et al.



both change and wk 16 RANKL (r ⫽ ⫺0.28; P ⫽ 0.011 and r⫽ ⫺0.22; P ⫽ 0.048, respectively).

Discussion Our results demonstrate that consumption of dairy foods and higher protein resulted in improved markers of bone health and calcium metabolism in overweight and obese young women over 16 wk of diet- and exercise-induced weight loss. Both the APMD and HPHD groups achieved dietary calcium and vitamin D intakes that were above the old dietary reference intakes (on which the study was modeled) (20). However, by wk 16, serum 25(OH)D rose significantly only in the HPHD group to a mean value in excess of 50 nmol/liter, which is suggested to be adequate for bone health (21). In contrast, the APLD group showed a small but significant decline in 25(OH)D over the 16 wk. We observed a number of findings indicative of positive changes in bone health in the HPHD group and, for the most part, also in the APMD group. The changes observed spanned a wide variety of bone-related processes including bone matrix turnover, bone colFIG. 3. Serum markers of bone resorption before and after intervention. A, Serum NTX. lagen turnover, and osteoclast differentia*, P ⬍ 0.001 vs. APLD at same time point; †, P ⬍ 0.01 vs. APMD at same time point; means tion. These processes all changed in a not bearing the same letter within a group are significantly different from each other, P ⬍ manner that suggests that higher dairy 0.01. B, ⌬Serum NTX. *, P ⬍ 0.05; **, P ⬍ 0.001. C, Serum CTX. *, P ⬍ 0.005 vs. APLD at same time point; means not bearing the same letter within a group are significantly different food and calcium consumption during from each other, P ⬍ 0.001. D, ⌬Serum CTX. *, P ⬍ 0.005. E, uDPD. Means not bearing the diet- and exercise-induced weight loss same letter within a group are significantly different from each other, P ⬍ 0.05. F, ⌬uDPD (no augments new bone formation. In genstatistical differences). eral, changes in bone biomarkers in the APLD group were in the opposite disignificantly lower leptin levels than the APLD group (P ⬍ rection and were indicative of increases in bone resorp0.05). Adiponectin increased in the APMD and HPHD tion, reduced bone collagen formation, and increased osgroups over 16 wk with no significant change in the APLD teoclast activity. With respect to the adipokines, leptin group (Fig. 5, C and D). At wk 16, the HPHD group had decreased and adiponectin increased significantly in the significantly higher adiponectin levels than the APMD APMD and HPHD groups only with no change in the (P ⬍ 0.01) and APLD (P ⬍ 0.005) groups. APLD group. This may have an impact on bone metaboAdiponectin levels at wk 16 were correlated with changes in total body fat (r ⫽ ⫺0.22; P ⫽ 0.048), visceral lism because leptin has been shown to inhibit bone refat volume (r ⫽ ⫺0.41; P ⫽ 0.010), P1NP (r ⫽ 0.29; P ⫽ modeling (25) and adiponectin to promote it (26). In ad0.008), and CTX (r ⫽ ⫺0.31; P ⫽ 0.005). In addition, dition, P1NP and CTX (biomarkers of type 1 collagen leptin levels at wk 16 correlated with both change and wk synthesis and degradation, respectively) both correlated 16 OPG (r ⫽ ⫺0.26; P ⫽ 0.022 and r ⫽ ⫺0.31; P ⫽ 0.005, with adiponectin levels, further indicating a role for the respectively), and changes in adiponectin correlated with adipokines in the regulation of bone turnover. Thus, in



257

and/or consuming a diet higher in dairy-source protein during weight loss minimizes bone turnover in overweight women. In our study, the loading of resistance exercise may have acted to offset the negative effect of energy restriction on bone by providing an osteogenic stimulus (10). However, because all women participated in the same exercise program and different rates of bone resorption were still evident, we surmise that even with daily weight-bearing exercise, including loaded contractions twice weekly, the intake of higher dairy, calcium, and vitamin D in the HPHD and APMD groups offered greater protection against bone resorption and potential bone loss usually seen with weight loss. Consumption of adequate dietary calcium in both the APMD and HPHD groups may have contributed to the significant reductions observed in PTH over 16 wk (Fig. 1). We have shown this previously in young women who underwent resistance training while consuming and additional 1 liter of fat-free milk/d for 12 wk (28). In the HPHD group, higher intakes of calcium and vitamin D and lower circulating PTH may have also helped maintain stable levels of remodeling by reducing the reFIG. 4. OPG, RANKL, and OPG to RANKL ratio before and after intervention. A, Serum OPG. sorptive effect on bone (2, 29). In con*, P ⬍ 0.05; **, P ⬍ 0.001 vs. APLD at same time point; †, P ⬍ 0.01 vs. APMD at same time trast, in the APLD group, the low dipoint; means not bearing the same letter within a group are significantly different from each etary calcium and vitamin D intakes other, P ⬍ 0.05. B, ⌬Serum OPG. *, P ⬍ 0.005. C, Serum RANKL. *, P ⬍ 0.005 vs. APLD at same time point; means not bearing the same letter within a group are significantly different and the increase in PTH likely contribfrom each other, P ⬍ 0.05. D, ⌬Serum RANKL. *, P ⬍ 0.05. E, OPG to RANKL ratio. *, P ⬍ uted to elevations in markers of remod0.001 vs. APLD at same time point; †, P ⬍ 0.01 vs. APMD at same time point; means not eling favoring resorption. Enhanced bearing the same letter within a group are significantly different from each other, P ⬍ 0.005. bone remodeling, particularly in those with suboptimal intakes of dietary calaccordance with our initial hypotheses, we demonstrated cium and vitamin D, could adversely affect the density and that diets adequate or higher in bone-supporting nutrients strength of bone, increasing the risk for osteoporosis (3). primarily from dairy foods positively affected markers of Adequate provision of high quality dietary protein is of bone turnover favoring bone formation, decreased circulating PTH, increased 25(OH)D, potentially reduced os- central importance to bone strength and health given its teoclast formation and function via changes in OPG and role as the primary structural component of bone (30). RANKL, and fostered appropriate adipokine responses to Positive relationships between higher protein intakes and promote bone formation and inhibit resorption in young BMD, BMC, reduced bone resorption, and reduced fracwomen who lost weight through a hypoenergetic diet- and ture risk have been observed in different populations (2, 30, 31), and a recent meta-analysis supports a positive exercise-induced weight loss program. Energy restriction and weight loss stimulate bone re- association between dairy-based protein intake and bone sorption (9, 27). However, in some cases (14, 16, 18), but health (32). In addition, lower protein consumption has not all (15, 17), maintaining adequate calcium intake been shown to influence PTH regulation by causing sec-

258

Josse et al.



formation and correlates well with other formation markers in healthy individuals, increases in OC could reflect greater rates of turnover or remodeling as opposed to greater rates of bone formation or bone mass accretion, especially when resorption markers are also increased (35). Several mechanisms may explain the increase in turnover rates favoring resorption seen in the APLD group with the greatest effect possibly being attributed to low dietary calcium and vitamin D intakes and subsequently higher PTH levels. We recognize that a large inherent biological variability exists in the measurement of bone turnover biomarkers (36), and due to the duration of our intervention, our biomarker results did not translate into DXA-measured bone mass changes. Nevertheless, the multiple biomarkers we measured showed quite consistent results FIG. 5. Leptin and adiponectin before and after intervention. A, Serum leptin. *, P ⬍ 0.05 vs. demonstrating a positive effect of consumAPLD at same time point; means not bearing the same letter within a group are significantly different from each other, P ⬍ 0.001. B, ⌬Serum leptin (no statistical differences). C, Serum ing dairy foods on bone remodeling and adiponectin. *, P ⬍ 0.005 vs. APLD at same time point; †, P ⬍ 0.01 vs. APMD at same time bone health in young women during point; means not bearing the same letter within a group are significantly different from each weight loss. other, P ⬍ 0.001. D, ⌬Serum adiponectin. **, P ⬍ 0.001; *, P ⬍ 0.05. To our knowledge, this is the first study to assess the effect of energy reondary hyperparathyroidism (33). Kerstetter et al. (33) striction and exercise in premenopausal women on the demonstrated that despite the maintenance of adequate OPG-RANKL signaling system with and without the prodietary calcium and vitamin D intakes, after just 4 d on low vision of dairy. Increased OPG (and decreased RANKL) protein (⬍0.7 g/kg 䡠 d), PTH concentrations increased, and particularly an increased OPG to RANKL ratio is whereas with higher protein (⬃1.0 g/kg 䡠 d), PTH reprotective for bone (via inhibition of osteoclastogenesis) mained unchanged. In our study, the APLD group con(37), which is what we observed in the HPHD group (Fig. sumed protein at a lower level (0.72 g/kg 䡠 d, which 4). In addition, we also observed significant negative coramounted to 55 ⫾ 7 g/d and approximately 16% of their relations between OPG and leptin and RANKL and adidaily energy), along with inadequate intakes of calcium and vitamin D. Both of these factors provide further ex- ponectin. In support of these relationships, previous replanation for the significant increase in PTH seen in this search has demonstrated that leptin and OPG have group. On the other hand, the HPHD group showed the correlated negatively in populations of healthy adult largest decrease in serum PTH, possibly due to their women (38) and obese children (having lower OPG and greater consumption of protein (1.33 g/kg 䡠 d; 108 ⫾ 18 higher leptin) (39) and that leptin-deficient mice have ing/d; 28% of daily energy) as well as greater intakes of creased bone mass (23). Furthermore, adiponectin has been shown to inhibit RANKL-induced osteoclastogendietary calcium and vitamin D. Higher rates of bone remodeling favoring resorption esis in vitro (26). In summary, we have demonstrated that the consumphave been associated with increased severity of osteoporosis and greater fracture risk in older persons (34). In the tion of diets higher in protein with an emphasis on dairy APLD group, the rise in bone turnover markers would foods during a diet- and exercise-induced energy deficit in favor resorption as indicated by significant increases in overweight and obese premenopausal women positively NTX, CTX, and uDPD (Fig. 3) and a significant decrease affected markers of bone turnover and adipokine levels, as in the ratio of P1NP to CTX over 16 wk. Of note, OC also well as markers of calcium, vitamin D status, and bone rose significantly, but this could be for many reasons. Al- metabolism. Moreover, diets with protein intakes at or though OC is generally categorized as a marker of bone around the current recommended dietary allowance (22)


with at least 1000 mg/d calcium, i.e. the APMD group, with protein and calcium intakes of 0.84 g/kg 䡠 d and 1200 mg/d, respectively, also offered favorable bone benefit compared with diets with no or low consumption of dairy foods. Thus, our data provide a good rationale to recommend consumption of dairy foods to aid in high quality weight loss (19) and the promotion of bone health in young women who are at the age when achieving and maintaining peak bone mass is of great importance (2).

Acknowledgments We thank T. Prior and the other members of the Exercise Metabolism Research Group (EMRG). We also sincerely thank the research assistants, dietitians/nutritionists, and thesis students who helped run the study: P. Kocsis, J. Wood, L. Wright, G. Zubic, S. French, K. Theeuwen, J. Clark, A. Paashuis, R. Robinson, K. Booker, J. Jackson, H. Zoschke, L. Bellamy, S. Losier, M. Labouesse, K. Howarth, D. Gallo, H. Robertshaw, and M. Marcinow. We appreciate the many hours put in by all our volunteers. Finally, we thank our study participants for their time, constant effort, and compliance with the study protocol. Address all correspondence and requests for reprints to: Stuart M. Phillips, Ph.D., McMaster University, 1280 Main Street West, Hamilton, Ontario, Canada L8S 4K1. E-mail: [email protected]. This work was supported by Dairy Farmers of Canada, The U.S. Dairy Research Institute, and the Canadian Institutes of Health Research. This study was registered at clincaltrials.gov as NCT00710398. A.R.J., S.A.A., M.A.T., and S.M.P. designed the research project. A.R.J. and S.M.P. conducted the research. A.R.J. and S.M.P. analyzed the data and conducted the statistical analyses. A.R.J., S.A.A., M.A.T., and S.M.P. helped write the final manuscript, and all approved the final content. Disclosure Summary: All authors state that they have no conflicts of interest.

References 1. Prentice A, Schoenmakers I, Laskey MA, de Bono S, Ginty F, Goldberg GR 2006 Nutrition and bone growth and development. Proc Nutr Soc 65:348 –360 2. Heaney RP 2009 Dairy and bone health. J Am Coll Nutr 28(Suppl 1):82S–90S 3. Heaney RP 2000 Calcium, dairy products and osteoporosis. J Am Coll Nutr 19:83S–99S 4. Poliquin S, Joseph L, Gray-Donald K 2009 Calcium and vitamin D intakes in an adult Canadian population. Can J Diet Pract Res 70: 21–27 5. 2009 The role of dairy products in diet. Introduction. J Am Coll Nutr 28(Suppl 1):69S–72S 6. Gao X, Wilde PE, Lichtenstein AH, Tucker KL 2006 Meeting adequate intake for dietary calcium without dairy foods in adolescents aged 9 to 18 years (National Health and Nutrition Examination Survey 2001–2002). J Am Diet Assoc 106:1759 –1765


259

7. Weaver CM 2010 Role of dairy beverages in the diet. Physiol Behav 100:63– 66 8. Heaney RP, Layman DK 2008 Amount and type of protein influences bone health. Am J Clin Nutr 87:1567S–1570S 9. Shapses SA, Riedt CS 2006 Bone, body weight, and weight reduction: what are the concerns? J Nutr 136:1453–1456 10. Villareal DT, Fontana L, Weiss EP, Racette SB, Steger-May K, Schechtman KB, Klein S, Holloszy JO 2006 Bone mineral density response to caloric restriction-induced weight loss or exercise-induced weight loss: a randomized controlled trial. Arch Intern Med 166:2502–2510 11. Hinton PS, LeCheminant JD, Smith BK, Rector RS, Donnelly JE 2009 Weight loss-induced alterations in serum markers of bone turnover persist during weight maintenance in obese men and women. J Am Coll Nutr 28:565–573 12. Rector RS, Loethen J, Ruebel M, Thomas TR, Hinton PS 2009 Serum markers of bone turnover are increased by modest weight loss with or without weight-bearing exercise in overweight premenopausal women. Appl Physiol Nutr Metab 34:933–941 13. Giangregorio L, Blimkie CJ 2002 Skeletal adaptations to alterations in weight-bearing activity: a comparison of models of disuse osteoporosis. Sports Med 32:459 – 476 14. Bowen J, Noakes M, Clifton PM 2004 A high dairy protein, highcalcium diet minimizes bone turnover in overweight adults during weight loss. J Nutr 134:568 –573 15. Wagner G, Kindrick S, Hertzler S, DiSilvestro RA 2007 Effects of various forms of calcium on body weight and bone turnover markers in women participating in a weight loss program. J Am Coll Nutr 26:456 – 461 16. Thorpe MP, Jacobson EH, Layman DK, He X, Kris-Etherton PM, Evans EM 2008 A diet high in protein, dairy, and calcium attenuates bone loss over twelve months of weight loss and maintenance relative to a conventional high-carbohydrate diet in adults. J Nutr 138:1096 –1100 17. Shapses SA, Von Thun NL, Heymsfield SB, Ricci TA, Ospina M, Pierson RN Jr, Stahl T 2001 Bone turnover and density in obese premenopausal women during moderate weight loss and calcium supplementation. J Bone Miner Res 16:1329 –1336 18. Redman LM, Rood J, Anton SD, Champagne C, Smith SR, Ravussin E 2008 Calorie restriction and bone health in young, overweight individuals. Arch Intern Med 168:1859 –1866 19. Josse AR, Atkinson SA, Tarnopolsky MA, Phillips SM 2011 Increased consumption of dairy foods and protein during diet- and exercise-induced weight loss promotes fat mass loss and lean mass gain in overweight and obese premenopausal women. J Nutr 141: 1626 –1634 20. Institute of Medicine 1997 Dietary reference intakes for calcium, phosphorus, magnesium, vitamin D, and fluoride. Washington, DC: National Academy Press 21. Ross AC, Manson JE, Abrams SA, Aloia JF, Brannon PM, Clinton SK, Durazo-Arvizu RA, Gallagher JC, Gallo RL, Jones G, Kovacs CS, Mayne ST, Rosen CJ, Shapses SA 2011 The 2011 report on dietary reference intakes for calcium and vitamin D from the Institute of Medicine: what clinicians need to know. J Clin Endocrinol Metab 96:53–58 22. Institute of Medicine 2005 Dietary reference intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein, and amino acids. Washington, DC: National Academy Press 23. Ducy P, Amling M, Takeda S, Priemel M, Schilling AF, Beil FT, Shen J, Vinson C, Rueger JM, Karsenty G 2000 Leptin inhibits bone formation through a hypothalamic relay: a central control of bone mass. Cell 100:197–207 24. Frankenfield D, Roth-Yousey L, Compher C 2005 Comparison of predictive equations for resting metabolic rate in healthy nonobese and obese adults: a systematic review. J Am Diet Assoc 105:775–789 25. Karsenty G, Oury F 2010 The central regulation of bone mass, the first link between bone remodeling and energy metabolism. J Clin Endocrinol Metab 95:4795– 4801

260

Josse et al.


26. Oshima K, Nampei A, Matsuda M, Iwaki M, Fukuhara A, Hashimoto J, Yoshikawa H, Shimomura I 2005 Adiponectin increases bone mass by suppressing osteoclast and activating osteoblast. Biochem Biophys Res Commun 331:520 –526 27. Bacon L, Stern JS, Keim NL, Van Loan MD 2004 Low bone mass in premenopausal chronic dieting obese women. Eur J Clin Nutr 58:966 –971 28. Josse AR, Tang JE, Tarnopolsky MA, Phillips SM 2010 Body composition and strength changes in women with milk and resistance exercise. Med Sci Sports Exerc 42:1122–1130 29. Holick MF 2008 Vitamin D: a D-Lightful health perspective. Nutr Rev 66:S182–S194 30. Rizzoli R, Bonjour JP 2004 Dietary protein and bone health. J Bone Miner Res 19:527–531 31. Feskanich D, Willett WC, Stampfer MJ, Colditz GA 1996 Protein consumption and bone fractures in women. Am J Epidemiol 143: 472– 479 32. Darling AL, Millward DJ, Torgerson DJ, Hewitt CE, Lanham-New SA 2009 Dietary protein and bone health: a systematic review and meta-analysis. Am J Clin Nutr 90:1674 –1692 33. Kerstetter JE, Svastisalee CM, Caseria DM, Mitnick ME, Insogna

34.

35. 36.

37. 38.

39.


KL 2000 A threshold for low-protein-diet-induced elevations in parathyroid hormone. Am J Clin Nutr 72:168 –173 Riggs BL, Melton 3rd LJ 2002 Bone turnover matters: the raloxifene treatment paradox of dramatic decreases in vertebral fractures without commensurate increases in bone density. J Bone Miner Res 17: 11–14 Polak-Jonkisz D, Zwolinska D 1998 Osteocalcin as a biochemical marker of bone turnover. Nephrology 4:339 –346 Civitelli R, Armamento-Villareal R, Napoli N 2009 Bone turnover markers: understanding their value in clinical trials and clinical practice. Osteoporos Int 20:843– 851 Boyce BF, Xing L 2007 Biology of RANK, RANKL, and osteoprotegerin. Arthritis Res Ther 9(Suppl 1):S1 Tenta R, Panagiotakos DB, Fragopoulou E, Nomikos T, Pitsavos C, Chrysohoou C, Antonopoulou S, Stefanadis C 2010 Osteoprotegerin and nuclear factor-kappaB ligand are associated with leptin and adiponectin levels, in apparently healthy women. J Musculoskelet Neuronal Interact 10:174 –179 Dimitri P, Wales JK, Bishop N 2011 Adipokines, bone-derived factors and bone turnover in obese children; evidence for altered fatbone signalling resulting in reduced bone mass. Bone 48:189 –196

Save the Date for Clinical Endocrinology Update, September 13-15, 2012, Miami, Florida. www.endo-society.org/CEU