Urea Kinetics Varies in Jamaican Women and Men in Relation to Adiposity, Lean Body Mass and Protein Intake

Urea kinetics varies in Jamaican women and men in relation to adiposity, lean body mass and protein intake

SC Child1, MJ Soares2, M Reid2, C Persaud1, T Forrester2 and AA Jackson1

1Institute of Human Nutrition, University of Southampton, Bassett Crescent East, Southampton SO16 7PX; 2Tropical Metabolism Research Unit, University of the West Indies, Mona, Kingston 7, Jamaica

Objective: We have measured urea kinetics in normal adult men and women of different body composition to determine whether adiposity is associated with differences in the rate of urea production or endogenous urea hydrolysis. Design: Urea kinetics were determined from the excretion of [15N15N]urea in urine over a period of 48 h following a single oral dose of [15N15N]urea, in nine lean and nine obese women and in seven light and seven heavy males while they were consuming their habitual diets. Urinary 5-L-oxoproline was measured as an index of glycine metabolic status. Setting: The studies were carried out in the research ward of the Tropical Metabolism Research Unit, University of the West Indies. Results: Successful studies were completed in eight obese and ®ve lean women and in six heavy and ®ve light men. When compared with lean women, in obese women the rate of urea production and hydrolysis was signi®cantly greater and this difference could not be accounted for by the greater fat-free mass alone, and was in part associated directly with the increase in fat mass. The rate of urea production and hydrolysis was greater in heavy men than in light men, a difference which was attributed to an increase in dietary protein. In obese women and heavy men there was a signi®cantly higher rate of excretion of 5-L-oxoproline in urine when compared with lean women and lean men respectively. Conclusion: This paper highlights the dif®culty in identifying an appropriate reference with which to express results in people of different body composition. In obese women urea production and the hydrolysis of urea are increased, in part related to the increased fat-free mass, but also related to the increased fat mass itself. In obese women and men on high protein diets the greater rate of hydrolysis urea may be a re¯ection of an increased demand for the synthesis of non-essential amino acids, especially glycine. Sponsors: This work was carried out as a part of an elective study period of SCC, which was supported by the Wellcome Trust, Cow and Gate/Nutricia and Lederle. We acknowledge support from BBSRC and the Wessex Medical Trust. Descriptors: body composition; obesity; fat free mass; urea production; urea excretion; urea hydrolysis; 5-L- oxoproline

Introduction These observations have all been based upon urinary losses of urea or nitrogen and in the past there has been a There is wide variability among individuals in most aspects tendency to equate urinary urea excretion with the rate of of protein metabolism. Although it has been suggested that urea production and hence the rate of amino acid oxidation. this may be related to differences in lean body mass, there However, it is clear that under all normal circumstances at are few studies in which this has been explored directly least 25% of the urea produced is hydrolysed by the colonic (Waterlow, 1996). In addition to effects attributed to micro¯ora with the nitrogen being salvaged for further differences in lean body mass it has been suggested that metabolic interaction (Hibbert and Jackson, 1991), Figure there is a fundamental difference in nitrogen metabolism in 1. In circumstances when the body seeks to conserve obese individuals when compared with people with a nitrogen the extent of salvaging might be enhanced to normal body mass index (BMI) (Payne and Dugdale, 80% of urea production (Jackson, 1995; Steinbrecher et 1977). During fasting, nitrogen losses are reduced in the al, 1996). Therefore, the observation that obese individuals obese compared with lean individuals and obligatory nitro- conserve nitrogen more ef®ciently might be accounted for gen losses are also less. It has been suggested that these either by a reduced rate of amino acid oxidation, or an differences in external balance represent an innate differ- enhanced rate of salvage of urea nitrogen. ence in the propensity to deposit and mobilise lean or The weight gained by obese individuals represents a adipose tissue during growth or wasting, which is re¯ected change in body composition beyond simply an increase in in a difference in the rate at which protein or amino acids adipose tissue. There is also an increase in lean body mass, are oxidized (Henry, 1992). and therefore any changes in nitrogen metabolism might be a direct consequence of the change in lean body mass, Correspondence: Prof AA Jackson. Received 2 August 1996; revised 25 October 1996; accepted 2 November rather than being a feature of the increase in adipose tissue 1996 (Webster et al, 1984; Garrow and Webster, 1985). Urea kinetics and adiposity SC Child et al 108 study. The body weight, height and skinfold thicknesses were measured in all subjects by experienced observers using standard methods. Skinfold thicknesses were measured at the biceps, triceps, subscapular and suprailiac points on the non-dominant side using Harpenden calipers. Each skinfold was measured in triplicate and the mean used to determine fat mass (Durnin and Womersley, 1974). Fat free mass was calculated as the body weight minus the fat mass. An estimate of muscle mass was determined from the excretion of creatinine in urine over two 24 h periods, on the assumption that the excretion of 60 mg creatinine corresponded to 1 kg muscle. The non-muscle fat free mass was calculated as fat free mass minus muscle mass. The dietary intake at the time of the study was determined using a structured questionnaire to recall and describe all the food and drink consumed during the period. Further, an assessment of habitual intake was determined by conduct- ing two further 24 h recalls for each subject after they had completed the study. The dietary information obtained was analysed for energy and protein using a computer pro- gramme (Nutritionist II). Women were studied over days 6±11, or days 21±26 of their menstrual cycle (McClelland and Jackson, 1996). During the study, food was consumed in portions evenly balanced over the daytime period. Figure 1 Model of urea kinetics used in the present study, consisting of a Urea kinetics were measured using a single oral dose of single pool of urea with a single source of entry from endogenous urea 15 15 production and two of loss, to urinary excretion and colonic hydrolysis. [ N N]urea and the determination of the excretion of label in urine during the 48 h period following the dose We have used the urinary excretion of 5-L-oxoproline as (Jackson et al, 1993, Figure 1). At 0900 h on the ®rst day, a an index of the availability of glycine to the glutathione sample of urine was collected for the measurement of synthetic pathway, and hence as a marker for the adequacy baseline enrichment and as the start of the period of urine of glycine to meet overall demand (Jackson et al, 1987). collection. All urine passed over the following 48 h was The ability of the body to synthesise adequate amounts of collected, in two aliquots of 24 h each. The urine was non-essential amino acids appears to be linked to the collected into acidi®ed containers, 5 ml HCl 6 mol/L, and stored frozen until later analysis. An accurately known salvage of urea-nitrogen (Jackson, 1995). Glycine is 15 15 15 required in disproportionately large amounts for the synth- amount of [ N N]urea (99.8 atoms percent N, Bioquote esis of structural proteins, such as collagen, and haem. In Ltd, Ilkley, West Yorkshire, UK), of about 100 mg, was obese individuals, there is an increase in lean body mass made up in deionised water and taken orally as a single (Webster et al, 1984), with greater demand being placed on dose immediately after collection of the baseline sample of structural proteins and haem formation. urine. In this present study we have measured urea kinetics in a The concentration of ammonia nitrogen and urea nitro- group of women who were obese and compared the results gen in the urine were determined by the Berthelot method with women of normal weight. In order to understand the (Kaplan, 1965). Urinary 5-L-oxoproline was measured relative contribution of lean body mass, similar measure- enzymatically, following isolation by column chromatogra- ments were made in a group of young men who were body phy and hydrolysis to glutamic acid (Jackson et al, 1996). builders. As a normal part of their training regimen these Urea nitrogen was isolated for mass spectrometry using men consumed a diet high in protein and therefore it was short ion-exchange column chromatography (Jackson et al, possible to assess the effect of high protein intakes on urea 1980). Nitrogen gas was liberated from urea by reaction kinetics. with alkaline hyobromite. Nitrogen is released from urea in a monomolecular reaction and it is thus possible to determine the proportions of different isotopic species of urea Methods molecules by measuring the relative ion beams of nitrogen Ethical approval for the study was obtained from the ethical liberated with mass 28, 29 and 30 (Walser et al, 1954) committee of the Faculty of Medical Sciences and Uni- using a triple collector isotope ratio mass spectrometer versity Hospital of the West Indies. The subjects gave (Sira 10; VG Isogas, Cheshire, UK). The creatinine content informed written consent for their participation and were of each 24 h collection of urine was measured using a free to withdraw without prejudice at any stage. The study direct colorimetric method, as a check on the completeness was conducted in the Tropical Metabolism Research Unit of the urine collection. of the University of the West Indies. Urea kinetics were calculated using a model which Control subjects and obese women were recruited from presumes a single pool of urea which receives urea from the staff and students of the University and the body- endogenous production and from which urea moves either builders from a local health club. A medical history was to excretion in the urine or hydrolysis by the micro¯ora in taken from all subjects and each underwent a clinical the colon (Jackson et al, 1993). The dose of labelled urea is examination to exclude diabetes, hypertension and other absorbed in the stomach and mixes with the body pool, obvious systemic disease. Initially 18 women and 14 men Figure 1. It is assumed that the dose of labelled urea is were entered into the study. The subjects were admitted to handled in the same way as urea formed within the body the Tropical Metabolism Research Unit for the days of the and hence the proportion of the dose excreted should be the Urea kinetics and adiposity SC Child et al 109 same as the proportion of the urea produced in the body statistically signi®cant differences. The heavy men were which is excreted. Therefore, if the amount of urea excreted 8 cm taller than the light men, but this difference did not in the urine and the proportion of the dose of labelled urea reach statistical signi®cance. The heavy men were 25% excreted in the urine as [15N15N]urea are measured then the heavier, 19 kg, and their fat free mass was 14 kg greater rate of urea production can be calculated. The difference than the light men, statistically signi®cant differences. The between the urea produced and the urea excreted in the fat mass of the heavy men was not statistically signi®cantly urine is taken to be the urea hydrolysed in the bowel. different from the fat mass of the light men, but they did Although there is a diurnal variation in the size of the urea have a signi®cantly higher BMI, 25 compared with 21 kg/ pool, on a standard diet the urea pool is similar at the same m2. Light men and light women had similar weights and time each day for any individual (Quevado et al, 1994). In BMI, but differed in fat free mass and fat mass. Heavy men the present study urea excretion was measured for 24 and and heavy women had a similar fat free mass, but differed 48 h periods and it can be assumed that urea not excreted is in every other respect. The muscle mass and non-muscle fat hydrolysed in the bowel, not retained within the urea pool. free mass in heavy women was signi®cantly greater than in A proportion of the nitrogen derived from labelled urea will light women, by 30 and 60% respectively. In the heavy men be re-utilised for urea formation over the time course of the muscle mass was 30% greater than in light men, but there experiment and the extent of this can be determined by was no signi®cant difference in non-muscle fat free mass. measuring the amount of [15N14N]urea in urine. Under normal circumstances this is less than 15% of total urea Protein intake and urinary nitrogen production. However, where there is gastric infection with Helicobacter pylori the dose of [15N15N]urea might be Women. Both the light women and the heavy women had hydrolysed before it is absorbed and this gives rise to an similar intakes of protein in absolute terms, about 60±70 increase in the excretion of [15N14N]urea and the model no g/d. However, expressed in relation to body weight the light longer applies (Hibbert et al, 1992). women consumed signi®cantly more, 58%, than the heavy Comparisons between groups of data were sought using women and 28% more expressed relative to fat free mass. ANOVA and post hoc t-test for unpaired data. The data There was no statistically signi®cant difference in urea were also analysed using non-parameteric statistics with the excretion between the two groups of women, however same conclusions. Linear regression analysis was used to expressed, although the light women tended to excrete explore associations between variables. more when expressed as mgN/kg/d. Crude balance was calculated as nitrogen intake minus urinary urea nitrogen Results and there was no difference between the two groups of women. The study was completed satisfactorily in 13 women and 11 men. In three women and one man, there was an obvious Men. The heavy men ingested about 154 g protein/d, 65 g failure to collect all urine during the 48 h period. In two protein/d more than the light men, a statistically signi®cant women and two men the pattern of isotope excretion was difference which was maintained when expressed in rela- judged to be unsatisfactory and a high level of [15N14N]urea tion to body weight or fat free mass. There was a close in urine was suggestive of infection with Helicobacter linear relationship between intake and fat free mass pylori. Therefore, there were ®ve light females (LF) and (r 0.94, P < 0.001), Figure 2. Heavy men excreted eight heavy females (HF), and ®ve light males (LM) and almost twice as much urea-nitrogen as light men, and six heavy males (HM) for analysis. about one third more expressed in relation to body weight or fat free mass. For the heavy men, crude balance was Body composition nearly twice that of light men, a statistically signi®cant Table 1 shows that on average the women, age 30 years, difference. were older than the men, aged 23 y. Heavy women were on The excretion of 5-L-oxoproline was similar in light men average almost twice the weight of light women, 51 kg and light women, but for both heavy men and heavy heavier, and 6 cm taller, statistically signi®cant differences. women secretion was increased signi®cantly, by about The heavy women were all obese, with a BMI of 39 30%. These differences were no longer seen when 5-L- compared with 23 for the light women, a fat free mass oxoproline excretion was expressed relative to body which was 20 kg greater and a fat mass 30 kg greater, all weight.

Table 1 The age, height, weight, body mass index (BMI), body composition and urinary 5-L-oxoproline excretion in thirteen light and heavy females and eleven light and heavy males in whom urea kinetics were measured. Values are mean Æ s.d.

Female Male

LF HF P value LM HM P value ANOVA n 5 n 8 n 5 n 6 P value

Age, years 30 30 0.27 23 23 0.27 Height, m 1.64 0.05 1.70 0.05 0.04 1.78 0.04 1.86 0.09 0.118 0.0004 Weight, kg 62.7 5.9 113.8 17.8 0.003 66.9 7.0 85.8 4.5 0.006 0.0003 BMI, kg/m2 23.2 2.1 39.1 5.1 0.003 21.2 1.6 24.9 2.2 0.029 0.0016 Fat mass, kg 18.8 4.4 49.5 11.0 0.003 11.2 4.8 15.5 3.9 0.201 0.0018 Fat free mass, kg 43.9 4.4 64.3 7.4 0.003 55.7 3.4 70.2 6.3 0.006 0.0002 Muscle mass, kg 27.2 4.5 37.0 7.3 0.019 32.6 7.5 43.8 8.4 0.068 0.006 Non-muscle mass, kg 16.8 5.2 27.3 7.6 0.03 23.0 4.9 26.4 6.3 0.361 0.030 5-L-oxoproline, mmol/d 204 81 262 57 0.045 207 31 258 22 0.017 0.018 5-L-oxoproline, mmol/kg/d 3.2 1.1 2.4 0.7 0.057 3.1 0.4 3.0 0.3 0.855 0.185 Urea kinetics and adiposity SC Child et al 110

Figure 2 Urea kinetics were measured in thirteen women (light women, ^: heavy women, ^) and eleven men (light men, j, and heavy men, u) on their habitual diet. The relationship between fat mass and urea production is shown in Panel A, and between fat free mass and urea production in Panel B. The relations determined by linear regression analysis are shown in solid lines for women and broken lines for men.

Table 2 Dietary nitrogen intake (I), urea production (Pu), urinary urea excretion (Eu) and the rate of urea hydrolysis in the colon (H), and crude nitrogen balance (dietary nitrogen intake minus urinary urea nitrogen) were derived from measurements of urea kinetics in thirteen light and heavy females and eleven light and heavy males. Values are mean Æ s.d.

Female Male

LF HF P value LM HM P value ANOVA n 5 n 8 n 5 n 6 P value

g nitrogen/d Dietary nitrogen intake 9.8 2.7 11.6 3.1 0.340 14.3 1.2 24.6 8.8 0.010 0.001 Urea production 10.1 3.1 18.2 7.2 0.028 10.2 2.4 18.4 7.9 0.018 0.011 Urea excretion 6.3 1.9 8.5 2.0 0.079 5.6 0.9 9.8 2.9 0.018 0.015 Urea hydrolysis 3.8 1.8 9.7 7.4 0.057 4.4 2.1 8.6 5.8 0.100 0.055 Intake hydrolysis 13.6 3.2 21.2 8.4 0.079 18.7 2.7 33.1 14.1 0.018 0.003 Crude nitrogen balance 3.5 1.6 3.1 1.6 0.770 8.6 0.7 14.8 6.6 0.045 Pu/I, % 106 35 160 58 0.040 71 14 75 12 0.584 0.002 Pu/I H, % 73 11 86 6 0.040 56 6 54 5 0.465 0.000 Eu/I, % 64 14 75 10 0.107 40 4 41 9 0.715 0.001 H/Pu, % 37 12 48 19 0.242 42 1 44 13 0.855 0.675

Urea kinetics Inter-relationships The total nitrogen available to the body has been taken as protein intake urea hydrolysis (Langran et al, 1992). Urea production. Urea production was 80% greater in the Table 2 shows that although intake hydrolysis was heavy women than the light women, a statistically signi®- greater by 55% in the heavy women compared to light cant difference. Urea production was almost identical women this did not achieve statistical signi®cance, whereas between the two groups when expressed in relation to an increase of 77% in heavy men compared with light men body weight and similar expressed relative to fat free was statistically signi®cant. When expressed relative to mass. Urea production was 80% greater in heavy men body weight, there was no difference in the two groups than light men, a statistically signi®cant difference. of women. For men, the 36% increase relative to body Expressed in relation to body weight or fat free mass, weight in heavy men compared with light men failed to urea production was 40% greater in heavy men, but this achieve statistical signi®cance, whereas the 37% increase difference was not statistically signi®cant. Urea production relative to fat free mass did. However expressed, intake in gN/d was virtually identical in light men and light hydrolysis tended to be greater in men. women, and also in heavy men and heavy women. Table 3 shows that there were signi®cant differences between the groups when urea production was related to the Urea hydrolysis. In both males and females urea hydro- intake of nitrogen, or the intake hydrolysis, with the ratio lysis was about twice as great in the heavy compared with tending to be lower in men. Urea excretion as a proportion the light group, although in neither situation did it quite of intake was similarly lower in men. The proportion of achieve conventional levels of statistical signi®cance urea production which was hydrolysed did not vary across (P 0.057 for women and 0.068 for men). These differ- the different groups. These relationships were further ences were less evident when expressed relative to fat free explored using linear regression analysis. Table 4 and mass, and least when expressed relative to body weight. Figure 3, shows the results of bivariate comparisons. For Urea hydrolysis, gN/d, was similar in light men and light both men and women there was a highly signi®cant women, and also in heavy men and heavy women. correlation between nitrogen intake and urea excretion, Urea kinetics and adiposity SC Child et al 111 Table 3 Nitrogen intake, urea production, urinary urea excretion and the rate of urea hydrolysis in the colon, and crude nitrogen balance were derived from measurements of urea kinetics in thirteen light and heavy females and eleven light and heavy males, expressed relative to body weight or fat free mass (FFM). Values are mean Æ s.d

Female Male

LF HF LM HM ANOVA n 5 n 8 P value n 5 n 6 P value P value mg N/kg body weight/d Dietary nitrogen intake 158 48 100 17 0.028 215 21 285 84 0.068 0.001 Urea production 161 48 158 52 0.884 152 32 213 79 0.100 0.375 Urea excretion 100 32 75 14 0.057 86 14 115 29 0.068 0.042 Urea hydrolysis 60 27 83 57 0.770 65 29 99 62 0.314 0.591 Hydrolysis intake 218 54 183 56 0.380 280 36 381 140 0.144 0.007 mg N/kg FFM/d Dietary nitrogen intake 226 71 178 35 0.143 257 11 344 91 0.067 0.004 Urea production 233 81 282 11 0.464 182 40 256 35 0.067 0.292 Urea excretion 146 52 132 25 0.558 103 13 138 30 0.045 0.136 Urea hydrolysis 88 44 150 109 0.306 79 37 118 70 0.201 0.396 Hydrolysis intake 314 86 328 116 0.884 336 42 462 151 0.045 0.112

Table 4 Associations between aspects of body composition and variables determined when urea kinetics were measured in thirteen light and heavy females and eleven light and heavy males, by linear regression analysis

Females Males

Fat free mass Fat mass Fat free mass Fat mass

r P r P r P r P

Weight 0.97 < 0.0001 0.99 < 0.0001 0.91 0.0002 0.65 0.04 Nitrogen intake 0.48 0.098 0.47 0.109 0.91 0.0010 0.10 0.78 Urea production 0.65 0.017 0.82 0.001 0.89 0.0005 0.03 0.93 Urea excretion 0.53 0.061 0.62 0.024 0.91 0.0002 0.14 0.71 Urea hydrolysis 0.55 0.053 0.71 0.006 0.80 0.0060 0.04 0.93 Intake hydrolysis 0.65 0.016 0.78 0.018 0.88 0.0003 0.05 0.89

Urea production Urea excretion Urea production Urea excretion

Weight 0.76 0.002 0.60 0.032 0.72 0.012 0.72 0.012 Nitrogen intake 0.49 0.090 0.85 0.0002 0.94 < 0.0001 0.97 < 0.0001 Urea hydrolysis 0.94 < 0.0001 0.27 0.368 0.95 < 0.0001 0.87 0.0005 Intake hydrolysis 0.98 < 0.0001 0.58 0.036 0.98 < 0.0001 0.97 < 0.0001

Figure 3C. In men, nitrogen intake correlated with urea and urea hydrolysis (r2 0.71; F 22; P 0.001), with no production, but not in women, Figure 3A, however there further contribution from either fat free mass or fat mass. was a highly signi®cant relationship between inta- For women fat mass was highly statistically signi®cantly ke hydrolysis and production for both men and women, related to urea production (r2 0.67; F 23; P 0.0066) although the slope of the relationship was signi®cantly and urea hydrolysis (r2 0.50; F 11; P 0.0067), with different between the two groups, Figure 3B. For both no further contribution from either fat free mass or protein men and women there was a highly signi®cant relationship intake. For the entire group there was a signi®cant linear between urea production and urea hydrolysis, with an relationship between urinary 5-L-oxoproline and urea pro- almost identical relationship for each group. For all duction (r2 0.24; F 7; P 0.014). In women, this rela- females, urea production and urea hydrolysis were more tionship between 5-L-oxoproline and urea production was closely related to fat mass than fat free mass, Table 4, also seen, when the effect of fat mass had been taken into Figure 2. By contrast, in men, there was no relation account in a multiple regression analysis (r2 0.33; F 5; between production, excretion or hydrolysis with fat P 0.04). In men, urinary 5-L-oxoproline related to fat free mass, but a close correlation for each with fat free mass, mass, when urea production and protein intake were taken Table 4, Figure 2. into account in a multiple regression analysis (r2 0.33; The relative contributions of fat mass, fat free mass and F 5; P 0.04). protein intake to the variability in urea production and hydrolysis were explored by stepwise multiple linear Discussion regression in men and women. For the group as a whole there was a highly signi®cant, statistical relationship It is widely appreciated that metabolism is in¯uenced between fat free mass and urea production (r2 0.46; directly by a number of factors, amongst which gender, F 19; P 0.0003) and with urea hydrolysis (r2 0.32; weight, body composition (fat free mass and fat mass) and F 11; P 0.0042), with no further signi®cant contribu- dietary intake are of recognized importance. For energy tion being made by either fat mass or protein intake. In men intake and balance the relative effect of each variable has intake of protein was highly statistically signi®cantly been explored in some detail, but this is less so for protein related to urea production (r2 0.88; F 68; P 0.000) intake, nitrogen metabolism and dietary protein require- Urea kinetics and adiposity SC Child et al 112

Figure 3 Urea kinetics were measured in thirteen women (light women, ^: heavy women, ^) and eleven men (light men, j, and heavy men, u) on their habitual diet. The relationship between nitrogen intake and urea production is shown in Panel A; between nitrogen intake urea hydrolysis and urea production in Panel B; between nitrogen intake and urinary urea in Panel C; and between urea production and urea hydrolysis in Panel D. The relations determined by linear regression analysis are shown in solid lines for women and broken lines for men.

ments (Waterlow, 1996). Hence, the requirements for For this purpose urea kinetics were measured in men in protein have been variously expressed in relation to whom fat free mass varied over a similar range, but without weight, body composition or energy intake, although the any signi®cant increase in fat mass, Figure 2. We conclude justi®cation for selecting one over another is not good. The that there is a direct effect of fat free mass on urea kinetics, present work represents a ®rst attempt to identify some of which might, in part, account for differences between lean the main factors of importance in relation to urea kinetics and obese women. However, even when the likely in¯u- so that in future studies the magnitude of the effect of ence of fat free mass is taken into account there is an important variables can be explored individually in greater independent and signi®cant effect of fat mass itself on urea detail. Notwithstanding these considerations, the data have kinetics. Therefore, we conclude that in obese women there value in their own right and are of interest in identifying is an increase in the rate of urea production and in the rate speci®c interrelations of potential importance. of urea hydrolysis which is directly related to the increase It has been shown that the obligatory nitrogen loss and in fat mass, Table 4. fasting nitrogen loss in overweight and obese people is less In seeking a group in whom the effect of differences in than in lean individuals (Henry, 1992). We were interested fat free mass might be studied, we identi®ed a group of men to know the extent to which the reduced external losses who were heavy as a result of an increased muscle mass might be accounted for by an increase in urea nitrogen through regular weight training. However, as these men salvage rather than any change in urea production. The habitually consumed a high protein diet as part of their results show that in obese women urea kinetics are more training schedule a further variable of signi®cance was intense than in lean women with a greater rate of urea introduced. This did enable an exploration of the effect of production and of urea hydrolysis. However, it can not be long term ingestion of a diet high in protein on urea assumed that the differences in urea kinetics arise solely as kinetics, for which there is virtually no information in the a consequence of the increased fat mass as obese women literature. Compared with the men of normal weight, heavy also have a greater fat free mass, both muscle mass and men had a substantial increase in urea production, excretion non-muscle mass. In order to get some information on the and hydrolysis expressed as gN/d. However, this difference changes which might be attributed to fat mass, indepen- was reduced when expressed relative to either body weight dently of any effect of fat free mass, it was necessary to or fat free mass. It is dif®cult to differentiate the relative obtain some information on the changes in urea kinetics importance of changes in dietary protein and changes of fat which might be attributed to an increase in fat free mass. free mass in the men as the two covaried in the present Urea kinetics and adiposity SC Child et al 113

Figure 4 Urea kinetics have been measured using similar approaches in normal adult men and women in 10 studies in the literature. The rate of urea production is shown in relation to the protein intake for these studies (1) Jackson et al, 1984; (2) Hibbert et al, 1992; (3) Langran et al, 1992; (4) Danielsen and Jackson, 1992; (5) Forrester et al, 1994; (6) Bundy et al, 1993; (7) Meakins and Jackson, 1996; (8) McClelland and Jackson, 1996) and for adults in the present study, light males (j) and heavy males (u) and for light females (^) and heavy females (^) in the present study. Panel A; the relation between urea hydrolysis and protein intake, Panel B; the relation between urea hydrolysis as a percentage of urea production and protein intake, Panel C; for the same studies. The solid lines represents the relations determined by linear regression analysis. study. However, when the men and women are considered (Quevado et al, 1994). If the study had been for a duration together as a single population, there was a strong relation- less than 24 h then variations in urea pool size might have ship between urea production and fat free mass, Figure 2. introduced a considerable error in the results and would In the past we have shown that the method for measur- have had to be taken into account, but this is not a concern ing urea production is reproducible within individuals for studies which last for 24 or 48 h. It is interesting to note (Hibbert and Jackson, 1991; McClelland and Jackson, that for females urea excretion was higher for any level of 1996) and that different methods agree closely (Jackson protein intake, when compared with males. This difference et al, 1993) although amongst individuals there may be was also seen for urea production. In contrast the relation- wide differences, which the present work seeks to explore. ship between production and hydrolysis was identical for Interpretations in the present work are potentially con- females and males. Figure 2 shows that across the entire strained by the limitations in the methods which have range, there was a signi®cant relationship between fat mass been used: with measures of dietary intake and the com- and urea production in females, but not males, whereas the pleteness of the urine collection being the measurements relationship between fat free mass and urea production was most susceptible to error. Intake has been assessed on the similar in both sexes. basis of 24 h dietary records. There are well recognized Caution should be used in interpreting the data on crude problems with these approaches which are thought to be nitrogen balances. It has been a general, but confusing, exacerbated in the overweight and obese (Prentice et al, observation that as protein intake increases from modest to 1996). For these reasons we would be cautious about the high levels apparent nitrogen balance also increases with reliability of the protein intake. However, this caution the effect that impossibly high positive balances are should not be overstated. As shown in Figure 3C, in all obtained at high protein intakes. The present data might the subjects there was a consistent relationship between be of importance in this context, because urea production nitrogen intake and urinary urea. The subjects were under was constant over a wide range of intakes in association close supervision throughout, and creatinine was measured with enhanced urea salvage at high protein intakes. There- in the urine as a further check on their completeness. This fore, it follows that at high protein intakes nitrogen balance together with the linear relationship between intake and has to be achieved by the excretion of nitrogen in forms urinary losses suggests that any errors are unlikely to be other than as urea. We have no data to indicate other large. There is a diurnal variation in the urea pool, which is possible forms of nitrogen excretion, but faecal nitrogen more marked in individuals on higher protein intakes excretion might be of special importance. The appearance Urea kinetics and adiposity SC Child et al 114 of a difference in crude nitrogen balance between men and The results of the present study expand our understand- women in the present study may be real or related to the ing of factors which impact upon urea kinetics and nitrogen dif®culties in determining an accurate nitrogen balance metabolism in the body. We can conclude that urea kinetics with con®dence. are in¯uenced directly by gender, habitual protein intake, In Figure 4, a comparison is drawn between the results fat free mass and fat mass. The results underline the in the present study and others in which we have measured dif®culties in selecting an appropriate reference with urea kinetics in normal young men and women using the which to express results in individuals of varying body same methodologies (Jackson et al, 1984; Hibbert et al, composition. An attempt was made to differentiate the 1992; Langran et al, 1992; Danielsen and Jackson, 1992; relative in¯uence of the different factors through multiple Bundy et al, 1993; Forrester et al, 1994; Meakins and liner regression. Using this statistical technique, the major Jackson, 1996; McClelland and Jackson, 1996). It can be in¯uence upon urea production for the group as a whole seen that amongst the different studies protein intake varied was fat free mass, which also accounted best for the across a ®vefold range, but urea production was very variability in urea hydrolysis. However, in women, the similar, on average 180 mgN/kg/d, with a coef®cient of main factor which accounted for variability in urea produc- variation of only 12%. The results of the present study, for tion and hydrolysis was the fat mass, and in men the intake light and heavy men and light and heavy women, conform of protein. The results imply a fundamental difference in with the observations in other studies. Further, taken endogenous nitrogen metabolism in obese women. There is together the studies show a close inverse relationship the need to identify whether this is a consequence of between protein intake and the rate of urea hydrolysis, becoming obese, or whether it is an inherent feature with hydrolysis increasing by about 60 mgN/kg/d for a which might of itself contribute to the development of decrease in intake of 100 mgN/kg/d on average. Whereas obesity. the results for the heavy females and the heavy males ®t the general pattern for urea production, they tend not to for urea hydrolysis. The obese women hydrolyse less urea than expected from the general pattern and the heavy men References hydrolyse more urea than would be expected. Bundy R, Persaud C & Jackson AA (1993): Measurement of urea kinetics In the past, most of our studies on urea kinetics have with a single dose of [15N15N]-urea in free-living female vegetarians focused upon the extent to which an increase in the salvage on their habitual diet. Int. J. Food. Sci. Nutr. 44, 253±259. Danielsen M & Jackson AA (1992): Limits of adaptation to a diet low in of urea-nitrogen might represent a mechanism of adaptation protein in normal man: urea kinetics. Clin. Sci. 83, 103±108. when the body needs to economise on nitrogen. The present Durnin JVGA & Womersley J (1974): Body fat assessed from total body study is the ®rst in which we have explored directly the density and its estimation from skinfold thickness: measurements on effects of a high intake of protein. The ®nding that the rate 481 men and women aged from 16 to 72 years. Br. J. Nutr. 32, 77±97. of urea production on an intake of 300 mgN/kg/d is similar Forrester T, Badaloo AV, Persaud C & Jackson AA (1994): Urea production and salvage during pregnancy in normal Jamaican women. to that observed on intakes as low as 60 mgN/kg/d, Figure Am. J. Clin. Nutr. 60, 341±346. 4, demonstrates that protein intake cannot be the major Garrow JS & Webster J (1985): Quetelet's index (W/H2) as a measure of determinant of urea production. Moreover, at higher protein fatness. Int. J. Obesity. 9, 147±153. intake, urea hydrolysis is much greater than would have Harper AE, Benevenga NJ & Wohlhueter RM (1970): Effects of ingestion of disproportionate amounts of amino acids. Physiol. Rev. 50, 428± been predicted, absolutely and as a percentage of produc- 458. tion. We have shown that the movement of urea from the Henry CJK (1992). Quantitative relationships between protein and energy circulation to hydrolysis in the bowel can not be attributed metabolism: in¯uence of body composition. In: Scrimshaw NS, Schurch to a simple clearance from blood, but appears to be an B (eds). Protein-Energy Interactions. International Dietary Energy active, controlled process (Meakins and Jackson, 1996). If Consultancy Group: Lausanne, Switzerland, pp 191±200. Hibbert JM, Forrester T & Jackson AA (1992): Urea kinetics: comparison urea hydrolysis is purposive then we would expect the of oral and intravenous dose regimens. Eur. J. Clin. Nutr. 45, 405±409. increased hydrolysis on higher protein intakes to be ful- Hibbert JM & Jackson AA (1991): Variation in measures of urea kinetics ®lling a useful function. One possibility is the need to over four years in a single adult. Eur. J. Clin. Nutr. 45, 374±351. dispose of large quantities of essential amino acids, which Jackson AA, Golden MHN, Jahoor PF & Landman JP (1980): The isolation of urea-N and ammonia-N from biological samples for mass are potentially toxic. Methionine is the most toxic of all spectrometry. Anal. Biochem. 105, 14±17. amino acids and increased intakes are poorly tolerated Jackson AA, Picou D & Landman J (1984): The noninvasive measurement (Harper et al, 1970). Glycine is required to detoxify of urea kinetics in normal man by a constant infusion by 15N15N-urea. excess methionine and methionine loads potentially limit Hum. Nutr: Clin. Nutr. 38C, 339±354. the availability of glycine for other metabolic functions Jackson AA, Badaloo AV, Forrester T, Hibbert JM & Persaud C (1987): Urinary excretion of 5-oxoproline (pyroglutamic aciduria) as an index (Harper et al, 1970). The urinary excretion of 5-L-oxopro- of glycine insuf®ciency in normal man. Br. J. Nutr. 58, 207±214. line is a marker for the adequacy of glycine and the Jackson AA, Danielsen MS & Boyes S (1993): A non-invasive method for demonstration of a signi®cant increase in the urinary measuring urea kinetics with a single dose of [15N15N]urea in free-living excretion of 5-L-oxoproline on high protein diets would humans. J. Nutr. 123, 2129±2136. Jackson AA (1995): Salvage of urea nitrogen and protein requirements. ®t the suggestion that high protein intakes contain poten- Proc. Nutr. Soc. 54, 535±547. tially toxic levels of methionine (Jackson et al, 1987; Jackson AA, Persaud C, Meakins TS & Bundy R (1996): Urinary 5-L- Jackson et al, 1996). This raises the possibility that the oxoproline (pyroglutamic acid) excretion in normal adults increases on greater rate of urea hydrolysis is, in part, a response to the vegetarian or low protein diets. J. Nutr. 126, 2813±2822. Kaplan A (1965). Urea nitrogen and ammonia nitrogen. In: Meites S (ed). need to generate amino groups for enhanced de novo Standard Methods in Clinical Chemistry. Academic Press: New York, endogenous biosynthesis of conditionally essential amino pp 245±256. acids such as glycine (Jackson, 1995). The urinary excre- Langran M, Moran BJ, Murphy JL & Jackson AA (1992): Adaptation to a tion of 5-L-oxoproline was also increased signi®cantly in diet low in protein: effect of complex carbohydrate upon urea kinetics in the heavy females, suggesting that in obesity there is a normal man. Clin. Sci. 82, 191±198. McClelland ISM & Jackson AA (1996): Urea kinetics in healthy young constraint on the ability to synthesise glycine in amounts women: effects of menstrual cycle, contraceptive pill and protein intake. adequate for normal metabolism. Br. J. Nutr. 76, 199±209. Urea kinetics and adiposity SC Child et al 115 Meakins TS & Jackson AA (1996): Salvage of exogenous urea-nitrogen Steinbrecher HA, Grif®ths DM & Jackson AA (1996): Urea kinetics in enhances nitrogen balance in normal men consuming marginally normal breast-fed infants measured with primed/intermittent oral doses inadequate protein diets. Clin. Sci. 90, 215±225. of [15N,15N]urea. Acta. Paediatr. 85, 656±662. Payne PR & Dugdale AE (1977): A model for the prediction of energy Walser M, George J & Bodenlos LJ (1954): Altered proportions of balance and body weight. Ann. Human. Biol. 4, 525±535. isotopes of molecular nitrogen from biological samples for mass Prentice AM, Black AE, Coward WA & Cole TJ (1996): Energy expenditure spectrometry. J. Chem. Phys. 22, 1146. in overweight and obese adults in af¯uent societies: an analysis of 319 Waterlow JC (1996): The requirements of adult man for indispensable doubly-labelled water measurements. Eur. J. Clin. Nutr. 50, 93±97. amino acids. Eur. J. Clin. Nutr. 50, S151±S179. Quevado MR, Price GM, Halliday D, Pacy PJ, Millward DJ (1994): Webster JD, Hesp R & Garrow JS (1984): The composition of excess Nitrogen homeostasis in man: diurnal changes in nitrogen excretion, weight in obese women estimated by body density, total body leucine oxidation and whole body leucine kinetics during a reduction water and total body potassium. Hum. Nutr: Clin. Nutr. 38C, 299± from a high to a low moderate protein intake. Clin. Sci. 86, 185±193. 306.