Source: YALE UNIVERSITY submitted to NRP
DIETARY PROTEIN AFFECTS CALCIUM AND BONE METABOLISM
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
0185794
Grant No.
00-35200-9579
Cumulative Award Amt.
$140,000.00
Proposal No.
2000-01094
Multistate No.
(N/A)
Project Start Date
Sep 1, 2000
Project End Date
Aug 31, 2004
Grant Year
2000
Program Code
[(N/A)]- (N/A)
Recipient Organization
YALE UNIVERSITY
105 WALL ST
NEW HAVEN,CT 06511-6614
Performing Department
INTERNAL MEDICINE
Non Technical Summary
Osteoporosis is a serious disorder affecting millions of Americans. Despite this, we know very little about how nutrients affect bone health. In previous work, we have observed that modest dietary protein restriction in adult women impairs intestinal calcium absorption and leads to an abnormal rise in parathyroid hormone secretion, so called secondary hyperparathyroidism. Further we have found that when women consume 80 mg/d of calcium and the RDA for protein they develop secondary hyperparathyroidism. Since these findings were published, the RDA for calcium has increased from 800 mg to 1,000 mg in young women. These results have led us to formulate the following two specific aims for the current proposal. 1. We will feed women graded levels of calcium while they consume the RDA for protein (0.8 g protein/kg) to determine the optimal level of calcium required at this level of protein intake. 2. We will directly compare the effects of omnivore to vegan protein sources on calcium and bone metabolism in women. The experimental diets used in our previous studies contained both animal and vegetable proteins. Will the same changes observed with these mixed protein sources occur when vegetable/soy is the sole source of dietary protein. These studies will refine current recommendations about appropriate levels of dietary protein and calcium intake as well as sources of protein needed to ensure bone health. Our goal of both of these studies is to be able to recommend appropriate levels of dietary protein and calcium and the type of protein needed to optimize mineral metabolism and
Animal Health Component
100%
Research Effort Categories
Basic
(N/A)
Applied
100%
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
70260101010100%
Goals / Objectives
1. Determine the level of dietary calcium required to optimize calcium metabolism when the RDA for protein is consumed. Our initial finding, that secondary hyperparathyroidism (due to reduced intestinal calcium absorption) occurred in women ingesting the RDA for protein, was made while the subjects consumed 800 mg/d of calcium (the RDA for calcium at the time the study was done). Recently, the RDA for calcium was increased to 1000 mg/d in young women. We wish to determine if this higher level of calcium intake will abrogate the rise in parathyroid hormone levels observed during moderately low protein intakes. 2. Directly compare the effects of omnivore to plant-based protein sources on calcium and bone metabolism in women. Our experimental diets previously contained both animal and plant-based proteins (omnivore). Little is known about plant-based proteins. We define plant proteins as all non-animal sources of protein, including grains and cereals, fruits, soy, and vegetables. Because of health concerns and religious beliefs, there are many individuals who exclude meat completely or nearly completely from their diet. We wish to determine if a diet based solely on plant protein based solely on plant protein induces the same changes as those observed in women consuming omnivore diets.
Project Methods
Protocol #1 (addressing objective #1). Calcitropic hormone response to graded levels of dietary calcium. We wish to determine if increasing the level of dietary calcium will prevent the abnormalities in calcium metabolism observed when the RDA for protein (0.8 g/kg) is consumed. The plan is to study 12 young women (age 20-40 years) using a 4-day experimental model. A 2-week adjustment period will precede each experimental period during which all subjects will consume the RDA for protein (0.8 g protein/kg). Subjects will then consume one of 4 levels of calcium (800, 1000, 1200, or 1500 mg) during the 4 day experimental period. The 4 experimental diets will be randomized and each subject will receive all four levels of calcium. We will collect blood and urine samples on days 0 and 4 of each experimental period to measure markers of bone turnover and calcium regulating hormones. The study aim is to determine the level of dietary calcium required to optimize calcium metabolism when the RDA for protein is consumed. Protocol #2. (addressing objective #2). Comparison of the effects of animal and plant protein sources in women. The aim of this study is to examine the effects of animal versus plant-based protein on bone and calcium metabolism. We intend to study 12 healthy women who will complete the following 4 experimental 4-day interventions: 1. omnivore low protein (0.7 g protein/kg) 2. vegan low protein (0.7 g protein/kg) 3. omnivore high protein (2.1 g protein/kg) 4. vegan high protein (2.1 g protein/kg) After a 2-week adjustment period, subjects will receive one of four experimental diets in random order. This 3-week cycle will be repeated three more times until the subject has received all 4 experimental diets. During days 0 and 4 of each experimental period, we will measure markers of bone turnover and calcium regulating hormones. The metabolites that we will measure include markers of bone formation (osteocalcin), markers of bone resorption (urine and serum N-telopeptide cross links), serum and urine calcium, phosphorus, sodium, creatinine; and calcium regulating hormones (PTH, 1,25 (OH)2 vitamin D, nephrogenous cyclic AMP). We plan to study non-Hispanic Caucasian or Asian women because these two groups are at highest risk for osteoporosis and these groups were used to generate preliminary data.

Progress 09/01/00 to 08/31/04

Outputs
The 2 Specific Aims are to: 1. Directly compare the effects of omnivore to plant-based protein sources on Ca and bone metabolism in women. 2. Determine the level of dietary Ca required to optimize Ca metabolism (ameliorate the secondary hyperparathyroidism) when the RDA for protein is consumed. Both studies addressing the 2 aims are completed and will be described below. Study Aim #1. The Vegan Study: Our nutrition intervention studies typically employ omnivorous protein sources (mixed animal and vegetable proteins), typical of what most Americans consume. We are interested in determining how the source of dietary protein influences Ca and skeletal homeostasis. Therefore, we designed a study to assess Ca homeostasis in women consuming vegetable-based protein (mostly soy) and compared it to omnivore proteins (our standard diet). We used low isoflavone soy foods in the experimental diets to eliminate the potential contributions of this estrogen-like compound. The research protocol was similar to those used routinely in our laboratory and consists of 4 experimental interventions each lasting 4 days. 1) omnivore low protein (0.7 g protein/kg); 2) vegan low protein (0.7 g protein/kg); 3) omnivore high protein (2.1 g protein/kg); 4) vegan high protein (2.1 g protein/kg). Twenty healthy Caucasian women (12 premenopausal and 8 postmenopausal) completed every intervention in random order. There were no age differences between the young and the older women, so their data were pooled. As expected, there were no differences in baseline measurements of any of the variables (data not shown). The figure shows the important variables at Day 4. During the omnivore diets, urinary Ca increased as dietary protein increased (low omni versus high omni). The high protein diet-induced hypercalciuria was also seen during the vegan diet, but the rise in urine calcium was blunted in comparison to the omnivore diet. At day 4, serum parathyroid hormone (PTH) was significantly elevated during the low protein diet, in comparison to the high protein diet. We know from our other studies that the rise in serum PTH in response to a protein- restricted diet is due to a decrease in intestinal Ca absorption. Interesting, the secondary hyperparathyroidism was more pronounced during the low protein vegan diet, in comparison to the low protein omnivore diet. Low Protein High Protein: Omnivore 20.7 % 24.2 %; Soy 17.6 % 18.7 %. These data demonstrate that vegetable-based protein diets are changing Ca homeostasis, but we were uncertain whether the change reflected a change in skeletal remodeling, intestinal calcium absorption or both. To answer this question, we studied Ca absorption (using dual stable Ca isotopes) in 6 women consuming the 4 interventions outlined above. In both circumstances, intestinal Ca absorption, as directly measured with isotopes, was lower during the soy diet in comparison to the omnivore diet. This change did not reach statistical difference. The decrease in intestinal calcium absorption during the low protein soy diet explains the more significant rise in PTH on the low protein soy vs. low protein omnivore diet. The decrease in intestinal calcium absorption during the high protein soy diet explains the blunted hypercalciuric response.

Impacts
Americans are not consuming enough calcium to support bone health and this is particularly true in women. The growing gap between the recommended intake for calcium and the actual intake is a major public health concern that will contribute to osteoporosis. Our study shows that the current RDA for calcium in young adult women (1000 mg) may still not be enough to support normal calcium homeostasis when the RDA for protein is consumed. In fact, an intake of 1200 mg calcium may be more beneficial based on our short-term study. We hope that our study will provide valuable data to inform the newest recommendations for dietary calcium.

Publications

  • Affenito S, Lambert-Lagace, L, Kerstetter JE, Denmark-Wahnefried W, Position of the American Dietetic Association: Nutrition and Women's Health. J Am Diet Assoc. 2004;104:1-18.
  • Kerstetter JE, O'Brien KO, Insogna KL. High protein diets, calcium economy, and bone health. Top Clin Nutr 2004; 19:56-66.
  • Kerstetter JE, O'Brien KO, Insogna KL. Dietary protein, calcium metabolism and skeletal homeostasis revisited. Am J Clin Nutr, 2003;584S-592S.
  • Kerstetter JE, O'Brien KO, Insogna KL. Low protein intake: the impact on calcium and bone homeostasis in humans. J Nutr. 2003;133:855S-861S.
  • Kerstetter JE, O'Brien KO, Insogna, KL Dietary protein and intestinal calcium absorption, Letter to the Editor, Am J Clin Nutr. 2001;73:990-991.
  • Kerstetter JE, Insogna, KL. Dietary Protein, Calcium and Bone Metabolism: Time to Revisit. Vegetarian Nutrition Update, Vegetarian Nutrition Dietetic Practice Group, a practice group of the American Dietetic Association, Spring 2001, Volume X, Number 3, 1-12.
  • Ilich, JZ, Kerstetter, JE. Nutrition and bone health revisited: A story beyond calcium. J Am Col Nutr. 2000;19: 715-737.
  • Kerstetter JE, Svastisalee CM, Mitnick ME, Caseria DM, Insogna KL. A threshold for low-protein-diet-induced elevations in parathyroid hormone. Am J Clin Nutr. 2000;72:168-173.
  • Kerstetter JE, Looker AC, Insogna KL. Low protein intake and low bone density (Letter to the Editor). Calcified Tissue International. 2000;66:313.
  • Ilich JZ, Kerstetter JE. Nutrition and Bone Health. In: Management of Fractures in Severely Osteoporotic Bone: Orthopaedic and Pharmacological Strategies. Ed, Obrant K., Springer-Verlag, London, 2000:362-382.
  • Kerstetter JE, O'Brien KO, Insogna KL. Low protein intake: Its impact on calcium and bone homeostasis in humans. J Nutrition, In Press, 2003.
  • Kerstetter JE, O'Brien KO, Insogna KL. Dietary protein, calcium metabolism and skeletal homeostasis revisited. Am J Clin Nutr, In Press, 2003.
  • Kerstetter, JE, O'Brien KO, Insogna, KL Dietary protein and intestinal calcium absorption, Letter to the Editor, Am J Clin Nutr, 2001,73:990-991.
  • Kerstetter, JE, Insogna, KL. Dietary Protein, Calcium and Bone Metabolism: Time to Revisit. Vegetarian Nutrition Update, Vegetarian Nutrition Dietetic Practice Group, a practice group of the American Dietetic Association, Spring 2001, Volume 10, Number 3, 1-12.
  • Raphael RH, Kerstetter JE, Svastisalee CM, O'Brien KO, Wall DE, Mitnick ME, Caseria DM, Insogna KL. Comparative effects of omnivore and vegan proteins on calcium homeostasis. Am Soc Bone Min Res, 2002:17:S470.
  • Raphael RH, Kerstetter JE, Svastisalee CM, O'Brien KO, Wall DE, Mitnick ME, Caseria DM, Insogna KL. Calcium homeostasis in vegan protein diets compared to omnivore diets. J Am Diet Association, 2002;102:A-77.
  • Kerstetter JE, Svastisalee CM, Raphael RH, Wall DE, Mitnick ME, Caseria DM, Insogna KL. Vegan protein impacts calcium metabolism differently than omnivorous proteins. 4th International Congress of Vegetarian Nutrition, Loma Linda University, San Bernardino, CA, 2002:52.
  • Kerstetter JE, Svastisalee CM, Mitnick ME, Caseria DM, Insogna KL. Dietary protein-induced changes in mineral metabolism are not influenced by age or sex. Am Soc Bone Min Res 2000:S423.


Progress 10/01/01 to 09/30/02

Outputs
The 2 Specific Aims are to: 1. Directly compare the effects of omnivore to plant-based protein sources on Ca and bone metabolism in women. 2. Determine the level of dietary Ca required to optimize Ca metabolism (ameliorate the secondary hyperparathyroidism) when the RDA for protein is consumed. Both studies addressing the 2 aims are completed and will be described below. Study Aim #1. The Vegan Study: Our nutrition intervention studies typically employ omnivorous protein sources (mixed animal and vegetable proteins), typical of what most Americans consume. We are interested in determining how the source of dietary protein influences Ca and skeletal homeostasis. Therefore, we designed a study to assess Ca homeostasis in women consuming vegetable-based protein (mostly soy) and compared it to omnivore proteins (our standard diet). We used low isoflavone soy foods in the experimental diets to eliminate the potential contributions of this estrogen-like compound. The research protocol was similar to those used routinely in our laboratory and consists of 4 experimental interventions each lasting 4 days. 1) omnivore low protein (0.7 g protein/kg); 2) vegan low protein (0.7 g protein/kg); 3) omnivore high protein (2.1 g protein/kg); 4) vegan high protein (2.1 g protein/kg) bone health. In addition, our data are important for the soy industry. If soy foods do indeed decrease intestinal calcium absorption (as we are currently confirming this fining in a larger group of women), it would support the practice of fortifying processed soy foods with calcium. Study #2. The Calcium Dose Response Study: Our previous work showed that within 4 days of a low protein and calcium diet (0.8 g/kg and 800 mg, respectively), healthy young women develop secondary hyperparathyroidism (2 HPT) due to a decrease in intestinal calcium absorption. The latest Recommended Dietary Allowance (RDA) for protein remains at 0.8 g protein/kg, while the recommendation for calcium was increased from 800 to 1000 mg. Therefore, the purpose of this study was to determine if the latest recommendation for calcium will ameliorate the 2 HPT observed when healthy young women consume the RDA for protein. We hope to determine the optimal intake of calcium needed to support normal calcium metabolism when 0.8 g protein/kg is consumed. The study was randomized, double blind and placebo-controlled consisting of 4 cycles. Each cycle included a 2-week adjustment diet followed by a 4 d experimental diet containing the RDA for protein and one of 4 levels of calcium (800, 1000, 1200 and 1500 mg). Ten healthy young women received all 4 levels of dietary calcium, in random order. Blood and urine were sampled at the beginning and end of each experimental period. Mean serum parathyroid hormone (PTH) levels were elevated by day 4 (above the normal range 20 mmol) when the diets contained 800 and 1000 mg calcium, but not when the diets contained 1200 and 1500 mg calcium (NS). There was a progressive increase in Day 4 mean 24-h urinary calcium excretion with each increase in dietary calcium; the increment between the 1000 mg and 1200 mg calcium interventions was significant (p = 0.025). When the average Day 4 urine calcium on the 2 low calcium diets (800 and 1000 mg) was compared using a paired t-test to the average on the 2 high calcium diets (1200 and 1500 mg), the mean value was significantly higher during the 2 high calcium diets (115-12 versus 155-8 mg, p < 0.01). These data suggest that healthy young women consuming the RDA for protein and calcium (1000 mg) for 4 days demonstrate acute perturbations in calcium homeostasis. At least 1200 mg of dietary calcium is needed to normalize serum PTH when the RDA for protein is consumed.

Impacts
Americans are not consuming enough calcium to support bone health and this is particularly true in women. The growing gap between the recommended intake for calcium and the actual intake is a major public health concern that will contribute to osteoporosis. Our study shows that the current RDA for calcium in young adult women (1000 mg) may still not be enough to support normal calcium homeostasis when the RDA for protein is consumed. In fact, an intake of 1200 mg calcium may be more beneficial based on our short-term study. We hope that our study will provide valuable data to inform the newest recommendations for dietary calcium.

Publications

  • Kerstetter JE, O'Brien KO, Insogna KL. Dietary protein, calcium metabolism and skeletal homeostasis revisited. Am J Clin Nutr, 2003;584S-592S.
  • Kerstetter JE, O'Brien KO, Insogna KL. Low protein intake: the impact on calcium and bone homeostasis in humans. J Nutr. 2003;133:855S-861S.
  • Kerstetter JE, O'Brien KO, Insogna, KL Dietary protein and intestinal calcium absorption, Letter to the Editor, Am J Clin Nutr. 2001;73:990-991.
  • Kerstetter JE, Insogna, KL. Dietary Protein, Calcium and Bone Metabolism: Time to Revisit. Vegetarian Nutrition Update, Vegetarian Nutrition Dietetic Practice Group, a practice group of the American Dietetic Association, Spring 2001, Volume X, Number 3, 1-12.
  • Ilich, JZ, Kerstetter, JE. Nutrition and bone health revisited: A story beyond calcium. J Am Col Nutr. 2000;19: 715-737.
  • Kerstetter JE, Svastisalee CM, Mitnick ME, Caseria DM, Insogna KL. A threshold for low-protein-diet-induced elevations in parathyroid hormone. Am J Clin Nutr. 2000;72:168-173.
  • Kerstetter JE, Looker AC, Insogna KL. Low protein intake and low bone density (Letter to the Editor). Calcified Tissue International. 2000;66:313.
  • Ilich JZ, Kerstetter JE. Nutrition and Bone Health. In: Management of Fractures in Severely Osteoporotic Bone: Orthopaedic and Pharmacological Strategies. Ed, Obrant K., Springer-Verlag, London, 2000:362-382.
  • Kerstetter JE, O'Brien KO, Insogna KL. Low protein intake: Its impact on calcium and bone homeostasis in humans. J Nutrition, In Press, 2003.
  • Kerstetter JE, O'Brien KO, Insogna KL. Dietary protein, calcium metabolism and skeletal homeostasis revisited. Am J Clin Nutr, In Press, 2003.
  • Kerstetter, JE, O'Brien KO, Insogna, KL Dietary protein and intestinal calcium absorption, Letter to the Editor, Am J Clin Nutr, 2001,73:990-991.
  • Kerstetter, JE, Insogna, KL. Dietary Protein, Calcium and Bone Metabolism: Time to Revisit. Vegetarian Nutrition Update, Vegetarian Nutrition Dietetic Practice Group, a practice group of the American Dietetic Association, Spring 2001, Volume 10, Number 3, 1-12.
  • Raphael RH, Kerstetter JE, Svastisalee CM, O'Brien KO, Wall DE, Mitnick ME, Caseria DM, Insogna KL. Comparative effects of omnivore and vegan proteins on calcium homeostasis. Am Soc Bone Min Res, 2002:17:S470.
  • Raphael RH, Kerstetter JE, Svastisalee CM, O'Brien KO, Wall DE, Mitnick ME, Caseria DM, Insogna KL. Calcium homeostasis in vegan protein diets compared to omnivore diets. J Am Diet Association, 2002;102:A-77.
  • Kerstetter JE, Svastisalee CM, Raphael RH, Wall DE, Mitnick ME, Caseria DM, Insogna KL. Vegan protein impacts calcium metabolism differently than omnivorous proteins. 4th International Congress of Vegetarian Nutrition, Loma Linda University, San Bernardino, CA, 2002:52.
  • Kerstetter JE, Svastisalee CM, Mitnick ME, Caseria DM, Insogna KL. Dietary protein-induced changes in mineral metabolism are not influenced by age or sex. Am Soc Bone Min Res 2000:S423.
  • Affenito S, Lambert-Lagace, L, Kerstetter JE, Denmark-Wahnefried W, Position of the American Dietetic Association: Nutrition and Women's Health. J Am Diet Assoc. 2004;104:1-18.
  • Kerstetter JE, O'Brien KO, Insogna KL. High protein diets, calcium economy, and bone health. Top Clin Nutr 2004; 19:56-66.


Progress 10/01/00 to 09/30/01

Outputs
The 2 Specific Aims are to: 1. Directly compare the effects of omnivore to plant-based protein sources on Ca and bone metabolism in women. 2. Determine the level of dietary Ca required to optimize Ca metabolism (ameliorate the secondary hyperparathyroidism) when the RDA for protein is consumed. The study addressing the first aim is approximately 75% completed and will be described below. The study addressing the second aim has just begun and will not be presented. For the second study, we are currently analyzing 3 women out of the planned 12 subjects and the intervention remains blinded. Our intervention studies typically employ omnivorous protein sources (mixed animal and vegetable proteins), typical of what most Americans consume. We are interested in learning how the source of dietary protein influences Ca and skeletal homeostasis. Therefore, we designed a study to assess a Ca homeostasis in women consuming vegetable-based protein (mostly soy) and compared it to omnivore proteins (our standard diet). We used low isoflavone soy foods in the experimental diets to eliminate the potential contributions of this estrogen-like compound. The research protocol is similar to that used routinely in our laboratory and consists of 4 experimental interventions each lasting 4 days: 1) omnivore low protein (0.7 g protein/kg), 2) vegan low protein (0.7 protein/kg), 3) omnivore high protein (2.1 g protein/kg), 4) vegan high protein (2.1 g protein/kg). Sixteen healthy Caucasian women (8 premenopausal and 8 postmenopausal) completed every intervention in random order; data from 13 of the subjects are presented in the Figure. As expected, during the omnivore diets, urinary Ca increased as dietary protein increased and serum parathyroid hormone (PTH) rose significantly during the low protein diet (due to a decrease in intestinal Ca absorption). There were significant differences in urinary Ca between the 2 different sources of protein. Under high protein conditions, urinary Ca excretion was dramatically lower during the vegan diets when compared to the omnivorous diets (P<0.002). During the low protein diets, mean urinary Ca was generally lower on the vegan diet, although the difference did not reach statistical significance. However, the secondary hyperparathyroidism was clearly more exaggerated during the low protein vegan diet versus the low protein omnivore diet in every subject (P<0.0001). These data clearly demonstrate that vegetable-based protein diets are changing Ca homeostasis in a manner quantitatively different than omnivorous diets. To begin to ascertain whether these quantitative differences are due to differential effects in bone, intestine or both we have begun to undertake calcium kinetic analyses. We have thus far studied 1 subject as she consumed all 4 diets. Her intestinal Ca absorption decreased by 9% (absolute value) between the low omni and the low vegan diets; and by 6% between the high omni and the high vegan diets. Bone balance was considerably more negative on the low protein vegan versus the low protein omnivore diet. Overall, the low protein vegan diet appears to decrease intestinal Ca absorption to a greater extent (1 subject)

Impacts
It has been argued that the amino acid composition of vegetable proteins might preserve bone by reducing the amount of diet-derived fixed acid that requires buffering in bone. Thus, we expected soy proteins to improve Ca economy and bone health. However, our acute studies suggest quite the opposite; that the incorporation of soy protein into the diet decreases intestinal Ca absorption and blunts urinary Ca excretion to a greater extent than does a comparable omnivorous diet. This results in more pronounced secondary hyperparathyroidism when dietary protein is low, which may translate to more bone loss. These are important findings given the large number of adults that are incorporating soy into their diets with the hope of improving bone health. Our data suggest that calcium economy is substantially altered during ingestion of a vegan diet and that further studies are needed on how best to correct these abnormalities. Until such information is available it may be premature to assume that all vegan diets are beneficial to skeletal health.

Publications

  • KERSTETTER, JE, O'BRIEN, KO, INSOGNA, KL. Dietary protein and intestinal calcium absorption, Letter to the Editor, AM J Clin Nutr, 2001, 73:990-991.
  • KERSTETTER, JE, INSOGNA, KL. Dietary protein, calcium and bone metabolism: time to revisit. Vegetarian Nutrition Update, Vegetarian Nutrition Dietetic Practice Group, a practice group of the American Dietetic Association, Spring 2001, Volume X, Number 3, 1-12.
  • ILICH, JZ, KERSTETTER, JE. Nutrition and bone health revisited: a story beyond calcium. J Am Col Nutr. 2000; 19: 715-737.
  • KERSTETTER JE, SVASTISALEE CM, MITNICK ME, CASERIA DM, INSOGNA KL. A threshold for low-protein-diet-induced elevations in parathyroid hormone. AM J Clin Nutr. 2000; 72:168-173.
  • KERSTETTER JE, LOOKER AC, INSOGNA KL. Low protein intake and low bone density (Letter to the Editor). Calcified Tissue International. 2000; 66:313.
  • ILICH JZ, KERSTETTER, JE. Nutrition and bone health. In: Management of fractures in severely osteoporotic bone: orthopaedic and pharmacological strategies., Obrant K. ed, Springer-Verlag, London, 2000:362-382.