Fd Chem. Toxic. Vol. 29, No. 6, pp. 387-390, 1991
0278-6915/91 $3.00+ 0.00 Copyright © 1991PergamonPress plc
Printed in Great Britain.All rights reserved
I N F L U E N C E OF DIETARY PROTEIN A N D GUT MICROFLORA ON E N D O G E N O U S SYNTHESIS OF NITRATE I N D U C E D BY BACTERIAL ENDOTOXIN IN THE RAT A. S. NIELSCH*'~',F. W. WARD*,M. E. COATES*:~, R. WALKER§ and I. R. ROWLANDII *Robens Institute, University of Surrey, §Department of Biochemistry, University of Surrey, Guildford, Surrey GU2 5XH and IIBIBRA, Woodmansterne Road, Carshalton, Surrey SM5 4DS, UK (Accepted 26 March 1991)
Abstract---Germ-free (GF) rats were maintained on a diet marginally adequate in protein, with and without a supplement of NH4C1. Their urinary excretion of total nitrogen, nitrate, urea and creatinine was measured before and for 4 days after injection of Escherichia coil endotoxin (lipopolysaccharide;LPS). Although more nitrogen was excreted by rats on the diet supplemented with NH4C1, nitrate excretion was increased to a similar extent in rats on both diets. This suggests that oxidation of ammonia released by deamination of amino acids is an unlikely pathway of nitrate synthesis. In a second experiment, nitrate excretion before and after injection of LPS was measured in GF and conventional (CV) rats given highor low-protein diets. Urinary 3-methylhistidine(3MH) was measured as an index of breakdown of tissue protein. In both environments, nitrate excretion was significantly greater, before and after LPS administration, by rats on the high-protein diet than by their counterparts on the low-protein diet, and was generally greater by GF than by CV rats. Since only small, non-significant rises in urinary 3MH were observed after LPS treatment, it was concluded that the bulk of the nitrogen required for nitrate synthesis in response to endotoxin is derived from dietary protein rather than from nitrogenous products of tissue breakdown.
excessive amount of protein, was measured before and after injection of LPS. In view of the reports that ammonia is a precursor of endogenous nitrate, a similar group of rats received the diet supplemented with 5 g NH4CI/kg. The reticuloendothelial system in G F rats is not exposed to microbial challenge, and so the synthesis of nitrate by immunostimulated macrophages might be expected to be less in the G F state. Nevertheless, in our experience nitrate excretion by G F rats is generally somewhat greater than that by their CV counterparts (Ward et al., 1989). For further evidence on this apparent anomaly, in the present paper, nitrate excretion by G F and CV rats given diets of high- (HP) or low-protein (LP) content was measured before and after injection of LPS. Endotoxin induces an inflammatory reaction with consequent breakdown of tissue proteins, the nitrogenous products of which could provide precursors for nitrate synthesis. To test this possibility, urinary excretion of 3-methylhistidine (3MH), an indicator of breakdown of skeletal muscle, was also measured to ascertain whether or not its pattern of excretion paralleled that of nitrate.
INTRODUCTION
The existence of a mammalian process for the endogenous synthesis of nitrate is now well established (Green et al., 1981; Tannenbaum et al., 1978; Witter et al., 1981). Our comparative studies in germ-free (GF) and conventional (CV) rats suggest that dietary protein is an important source of nitrogen for endogenous nitrate synthesis (Ward et al., 1989), and other workers have shown that ammonia, the ultimate endpoint of protein catabolism, is one of the precursors (Saul and Archer, 1984; Wagner et al., 1983). Since the demonstration by Wagner et al. (1983) that injection of rats with Escherichia coli endotoxin (lipopolysaccharide; LPS) greatly increased their urinary nitrate excretion, studies in vivo and in vitro have implicated macrophages in LPS-induced synthesis of nitrate (Stuehr and Marietta, 1985 and 1987). The first experiment described in the present paper explored the effects on endogenous nitrate synthesis of immunostimulation in the absence of gut microflora. Urinary excretion of nitrate by G F rats given a purified diet (C10), providing an adequate but not tPresent address: BP International Ltd, Surrey Research Park, Guildford, Surrey GU2 5YQ, UK. :~Present address: Department of Biochemistry, University of Surrey, Guildford, Surrey GU2 5XH, UK. Abbreviations: CV = conventional; GF = germ-free; HP = high protein; LP = low protein; LPS = lipopolysaccharide; 3MH = 3-methylhistidine.
MATERIALS AND METHODS
A n i m a l s a n d diets. G F and CV Lister Hooded rats of both sexes, bred at the University of Surrey, were used. The CV rats were housed in the open laboratory in standard rat cages with mesh floors. G F rats were
387
A . S . NIELSCI-i et al.
388 Table 1. Composition of the experimental diets Diet (g/kg) Ingredient
C 10
Lactalbumin Casein Maize starch Sucrose ~t-Cellulose Mineral mix* Vitamin mix* Choline chloride DL-Methionine
-100 652 100 50 35 8.66 1.34 3
LP 50 -898.6 --35 8.66 1.34 6.4
HP 200 -748.6 --35 8.66 1.34 6.4
*Ward and Coates (1987).
maintained in similar cages in plastic-film isolators, as described by Ward et al. (1986). From weaning until the beginning of the experiments, at about 6 wk of age, they received a commercial rodent diet of natural ingredients (GR3, Special Diet Services Ltd, Witham, Essex, UK) with a nitrate content of about 6 pg/g. The experimental diets were of purified ingredients and none contained more than 1.1 ~g nitrate/g. Their compositions are given in Table 1. All diets were sterilized by ),-radiation at 50 kGy. Autoclaved distilled water was supplied ad lib. At about 6 wk of age the rats were distributed between the experimental treatments and given the appropriate diets for 2 wk before urine collection was begun. Precise measurement of food intake could not be made inside the isolators, but final assessment of the amounts of food eaten by each group indicated little or no differences in food intake between G F rats and their CV counterparts. The animals were weighed at the end of the collection period and all parameters were expressed as per 100 g body weight. Analytical methods. Nitrate was measured by the chemiluminescence method of Waiters et al. (1987), as previously described by Ward et al. (1989). Total nitrogen was determined by a micro-Kjeldahl technique. Creatinine and urea nitrogen were measured colorimetrically with diagnostic kits nos 555 and 535, respectively (Sigma Chemical Co. Ltd, Poole, Dorset, UK). Urinary 3MH was determined as described by Jones et al. (1982). Since rats excrete a large proportion of their total 3MH in the N-acetylated form, the urines were hydrolysed before determination of total urinary 3MH. The urine was concentrated 3 × by rotary evaporation, an equal volume of concentrated HC1 was added and the mixture was auto-
claved for 2hr at 130°C. The hydrolysate was neutralized with 5 M-NaOH, derivatized with fluorescamine reagent (acetonitrile 160 mg/litre) and incubated at 80°C with 2 M-HCI for 1 hr to destroy fluorescamine derivatives of amino acids other than histidine and 3MH. The acid-stable fluorescent derivatives of histidine and 3MH were injected onto a column (15cm x 4.6mm Altex C18 ultrasphere; Beckman) and eluted with methanol-acetate buffer (48%:52%) at a flow rate of 1 ml/min. The eluates were quantified by means of a Gilson Spectra GIo fluorimeter, with input and output filters of < 390 nm and >460 nm, respectively. Experimentalprocedure. Experiment 1 explored the effects of LPS injection on the excretion of nitrate and other nitrogenous products by G F rats given a diet marginally adequate in protein, with or without a supplement of NH4C1. Groups of five female G F rats were maintained on a diet containing 100 g casein/kg, with or without addition of 5 g NH4CI/kg. Urine was collected in 1 M-HCI for 3 days, and the collections from each rat were pooled. The animals then received an ip dose of 100 pg LPS/kg body weight (E. coli serotype 0127:88, no. 3129, Sigma). Urine was collected daily for a further 4 days. The pooled 3-day collection prior to injection, and each individual daily collection thereafter, were analysed for total nitrogen, nitrate, urea nitrogen and creatinine. Experiment 2 investigated the effect of LPS injection on the excretion of nitrate and 3MH by G F and CV rats given diets with HP or LP content. Groups of two male and two female rats received diets containing either 50 or 200 g lactalbumin/kg for 2 wk, then urine was collected in 0.1 M-HC1 for 7 days, and the collections from each rat were pooled. The animals were injected ip with 100 #g LPS/100 g body weight and daily collections of urine were made for a further 4 days. The pooled initial sample and each daily post-injection sample were assayed for nitrate, creatinine and 3MH. Statistical analysis. For all parameters, an analysis of variance was done on the results of each daily collection, using the variance within groups as the estimate of error. Student's t-test was used to ascertain the significance of differences between the basal values and those on each day after administration of LPS. In experiment 2, the basal values for nitrate and 3MH differed between diets or environments, so the
Table 2. Effect of a dietary supplement of 0.5 g NH4CI/kg and of injection of endotoxin on daily urinary excretion of nitrogenous products by GF rats Hr after injection of endotoxin Parameter
Basal
24
48
72
96
26.7+0.8 35.8_+2.9*
32.8+4.6 42.4_+4.1
35.8_+ ll.l 40.6_+4.4
35.8+5.9 44.2_+4.2
31.3+2.0 38.5_+3.9
Total nitrogen (mg)
U A
Nitrate (/~g)
U A
Urea nitrogen (mg)
U A
17.9_+1.1 22.4_+3.2
24.1 _ + 4 . 7 28.7_+3.5
25.5_+5.6 27.1 _ + 4 . 0
26.8_+4.1 30.7_+3.9
21.9_+2.2 24.5_+3.2
Creatinine (mg)
U A
2.62 _+0.2 2.62_+0.2
2.79 _+0.3 2.72_+0.3
2.68 _+0.3 2.68_+0.3
3.03 _+ 0.2 3.03_+0.2
2.67 _+ 0.2 2.41 +0.1
159_+22.6 150_+ 1 0 . 2
649_+70.4++ 704_+82.7++
571 _+ 113.9t 504_+39.5++
187_+16.6 219_+14.8
U = unsupplemented diet A = diet supplemented with 5 g NH4C1/kg *Significantly different (Student's t-test) from corresponding U value at P < 0.05. tSignificantly different (Student's t-test) from basal value at P < 0.05. ++Significantly different (Student's t-test) from basal value at P < 0.01. Values (calculated as per 100 g body weight) are means + SEM of five rats.
167_+ 16.5 169_+ 13.4
Diet, gut flora and LPS-induced nitrate synthesis
389
Table 3. Effectsof injection of endotoxin on daily urinary excretion of nitrogenouscompounds by GF and CV rats given diets containing 50 (LP) or 200 (HP) g laetalbumin/kg Hr after injection of endotoxin Diet
Rat
Nitrate ~ g ) LP
HP Statistical comparison Statistical comparison 3 MH (pg) LP HP Statistical comparison Statistical comparison Creatinine (mg) LP HP
Basal
24
72
333 __.85 192_+26 647 _ 89 368_+37 P < 0.01 P < 0.01
148 _+26 116_+22 440 _+94 313_+11 NS P < 0.001
96
GF CV GF CV GF v. CV LP v. HP
149 -+ 10 146_+37 427 _+60 314_+47 NS P < 0.001
GF CV GF CV GF v. CV LP v. HP
34+_16 23 + 5 29 + 3 19_+4 P < 0.05 NS
29+11 43 _+14 32 _+6 29_+5 NS NS
52_+4 46 __.20 38 _+4 28_+6 NS NS
43_+8 29 + 9 27 _+4 31 _+7 NS NS
46_+12 27 _+9 28 _+4 23_+6 NS NS
2.91 2.24 2.71 2.28
2.45 2.86 1.98 1.52
2.80 2.76 2.79 2.26
2.53 2.45 2.78 2.70
3.12 2.23 2.64 2.03
GF CV GF CV
91 -+ 12 304_+39* 539 _+88 478_+108 NS P < 0.05
48
149 + 37 82_+6 530 + 73 300_+29 P < 0.05 P < 0.001
NS = not significant Values (calculated as per 100 g body weight)are means _+SEM of four rats, and those marked with asterisksdiffer significantly (Student's t-test) from the corresponding basal value (*P < 0.05). increase in excretion was considered a m o r e valid measure o f the effect o f LPS t h a n the actual a m o u n t excreted. The total a m o u n t excreted by each rat during the 4 days after LPS injection was calculated, and the increase over a similar period prior to injection was expressed as a percentage o f the basal value. RESULTS The G F animals in b o t h experiments remained u n c o n t a m i n a t e d throughout. The daily urinary excretions o f nitrogenous c o m p o u n d s , calculated as per 100 g b o d y weight, are recorded in Table 2. Creatinine excretion was similar in b o t h groups and remained fairly c o n s t a n t t h r o u g h o u t , indicating that the urinary collections were reliable 24-hr samples. As expected, total nitrogen excreted by rats given the diet supplemented with NH4C1 was significantly higher t h a n that excreted by the u n s u p p l e m e n t e d controls. This p a r a m e t e r remained higher, t h o u g h not significantly, after injection o f endotoxin. There was a small, but non-significant, increase in urea nitrogen Table 4. Increasesin nitrate and 3MH excretionby GF and CV rats, given diets containing 50 (LP) or 200 (HP) g lactalbumin/kg, during 4 days followingexposure to LPS, and expressedas a percentageof the amount excreted during four days prior to LPS injection 4 Days 4 Days Diet Rat beforeLPS after LPS % Increaser Nitrate Og/100 g body weight) LP GF 598 709 25 + 36 GV 582 657 38 _+43 HP GF* 1804 2259 27 +_ 15 CV 1256 1447 23 _+11 3MH (pg/100 g body weight) LP GF 136.3 169.2 32 _+27 CV 93.1 145.0 45 + 25 HP GF* 113.2 121.8 33 __.17 CV 77.5 111.2 52 -+ 21 *One rat died. ~'Values are means _+SEM of 4 rats. There were no significant differencesbetween GF and CV rats or LP and HP diets.
excretion by the supplemented rats. Nitrate excretion was n o t affected by the dietary supplement o f NH4Ci, but in b o t h groups there was a four-fold increase in nitrate excretion after injection o f LPS. Results o f the analyses o f daily urinary collections from G F and CV rats in experiment 2 are given in Table 3. Creatinine values again indicated satisfactory daily collections. Rats in b o t h environments given the H P diet excreted significantly m o r e nitrate than those given the LP diet, b o t h before and after administration o f LPS. Nitrate excretion o f G F rats was, in general, higher than that by their CV counterparts, although the increase was n o t always significant. After injection o f LPS, nitrate excretion rose in all groups (except for one anomalous value), but to a lesser extent t h a n in the previous experiment. The responses o f rats o n the H P diet were higher than those o n the LP diet, but followed a roughly similar pattern. Initial excretion o f 3 M H was higher by G F rats. Small, non-significant increases occurred after LPS injection, but no differences were observed between rats in the two environments or on the two diets. W h e n the increase in nitrate or 3 M H excretion after treatment with LPS was considered as a percentage o f the corresponding basal value (Table 4), there was an indication o f greater excretion o f 3 M H by CV c o m p a r e d with G F rats, but the difference did not reach statistical significance. DISCUSSION In confirmation o f our previous findings (Ward
et al., 1989), dietary protein concentration markedly influenced urinary excretion o f nitrate, with rats given the H P diet excreting two to three times m o r e nitrate t h a n those on the LP diet. This strengthens the suggestion that at least some endogenously synthesized nitrate is derived from dietary protein. One possible pathway for such synthesis is by hydrolysis o f the protein, d e a m i n a t i o n o f the a m i n o acids thus formed and oxidation o f the released a m m o n i a to
390
A.S. NIELSCHet al.
nitrate. The results of the first experiment, however, would argue against such a route, since a supplement of NH4C1 in diet marginally adequate in protein did not stimulate nitrate excretion, in spite of an increase in total urinary nitrogen. It is more likely that the synthesized nitrate is derived from oxidation of the amino group of arginine (Marietta, 1988), a process known to be stimulated by activation of macrophages (Iyengar et al., 1987). The results presented here are consistent with this hypothesis. Administration of bacterial endotoxin, LPS, caused an increase in nitrate excretion in all rats studied, although the magnitude of the increase was highly variable. Such variation in response between individuals and between experiments has been observed on previous occasions in both our laboratories. Although the LPS-induced increase in daily nitrate excretion was rarely significantly different from the basal value, the total nitrate excreted during the 4 days after LPS injection was always greater than in an equivalent period before LPS treatment (Table 4). It is pertinent to note that the response to LPS was greater in experiment 1, in which the diet contained casein, than in experiment 2, when lactalbumin was the protein source. Casein contains a higher proportion of arginine than lactalbumin (4.2 compared with 1.2g/16g protein nitrogen). The difference may also have been due to differences in antigenicity of the two proteins. As previously reported (Ward et aL, 1989), nitrate excretion by G F rats was usually greater than that by corresponding CV animals, but the magnitude of response to LPS exposure was similar in CV and G F rats (Table 4). The peak response in G F rats, whether on a HP or LP diet, occurred after 48 hr, but in CV rats the response to LPS peaked at 24 hr (Table 3). This may reflect differences in macrophage activation kinetics between rats with and without a microflora. However, since immunostimulation by LPS resulted in similar increases in nitrate synthesis in both environments, it can be assumed that the lower nitrate excretion by CV rats is due to microbial utilization or breakdown of nitrate. Urinary excretion of 3MH was not significantly affected by LPS treatment, implying that the dose administered was insufficient to cause severe tissue damage. It was, however, generally higher than the basal value in both CV and G F rats. From the cumulative 4-day excretion after LPS treatment, there was an indication that the CV rats on both diets had been more severely affected than their G F counterparts. Interestingly, the peak of 3MH occurred at a similar time to the peak of nitrate excretion, that is 24 hr after LPS exposure in CV rats and 48 hr in G F rats. Nevertheless, the much greater excretion of nitrate by rats on the HP diet was not paralleled by that of 3MH. It must be concluded, therefore, that although nitrogenous compounds released from damaged tissues may constitute a small proportion of the precursor(s) for nitrate synthesis, the bulk of the
nitrogen required appears to be derived from dietary protein. Acknowledgements--This work was supported by a grant from the Wellcome Trust, to whom we express our thanks. We are grateful to Professor D. J. Naismith and Dr P. Emery of King's College (Kensington Campus), London, for the use of equipment for the assay of 3MH, and to Miss V. Ronaasen for the care and management of the gnotobiotic rats. REFERENCES
Green L. C., Tannenbaum S. R. and Goldman P. (1981) Nitrate synthesis in the germfree and conventional rat. Science, New York 212, 56-58. Iyengar R., Stuehr D. J. and Marietta M. A. (1987) Macrophage synthesis of nitrite, nitrate and Nnitrosamines: precursors and role of respiratory burst. Proceedings of the National Academy of Sciences of the U.S.A. 84, 6369-6373. Jones A. D., Shorley D. and Hitchcock C. H. (1982) The determination of 3-methylhistidineand its application as an index for calculating meat content. Journal of the Association of Public Analysts 20, 89-94. Marletta M. A. (1988) Mammalian synthesis of nitrite, nitrate, nitric oxide and N-nitrosating agents. Chemical Research in Toxicology i, 249-257. Saul R. L. and Archer M. C. (1984) Oxidation of ammonia and hydroxylamine to nitrate in the rat and in vivo. Carcinogenesis 5, 77-8 i. Stuehr D. J. and Marietta M. A. (1985) Mammalian nitrate biosynthesis: mouse macrophages produce nitrite and nitrate in response to Escherichia coli lipopolysaccharide. Proceedings of the National Academy of Sciences of the U.S.A. 82, 7738-7742. Stuehr D. J. and Marietta M. A. (1987) Synthesis of nitrite and nitrate in murine macrophage cell lines. Cancer Research 47, 5590-5594. Tannenbaum S. R., Fett D., Young V. R., Sand P. C. and Bruce W. R. (1978) Nitrite and nitrate are formed by endogenous synthesis in the human intestine. Science, New York 2116, 1487-1489. Wagner D. A., Young V. R. and Tannenbaum S. R. (1983) Mammalian nitrate biosynthesis: incorporation of tSNH3 into nitrate is enhanced by endotoxin treatment. Proceedings of the National Academy of Sciences of the U.S.A. 80, 4518--4521. Waiters C. L., Gillatt P. N., Palmer R. C. and Smith P. L. R. (1987) A rapid method for the determination of nitrate and nitrite by chemiluminescence. Food Additives and Contaminants 4, 133-140. Ward F. W. and Coates M. E. (1987) Dietary fat and N-nitrosation in the rat. British Journal of Nutrition 58, 221-231. Ward F. W., Coates M. E. and Walker R. (1986) Nitrate reduction, gastro-intestinal pH and N-nitrosation in gnotobiotic and conventional rats. Food and Chemical Toxicology 24, 17-22. Ward F. W., Coates M. E. and Walker R. (1989) Influence of dietary protein and gut microflora on endogenous synthesis of nitrate and N-nitrosamines in the rat. Food and Chemical Toxicology 27~ 445-449. Witter J. P., Gatley S. J. and Balish E. (1981) Evaluation of nitrate synthesis by intestinal microorganisms in vivo. Science, New York 213, 449-450.
0278-6915/91 $3.00+ 0.00 Copyright © 1991PergamonPress plc
Printed in Great Britain.All rights reserved
I N F L U E N C E OF DIETARY PROTEIN A N D GUT MICROFLORA ON E N D O G E N O U S SYNTHESIS OF NITRATE I N D U C E D BY BACTERIAL ENDOTOXIN IN THE RAT A. S. NIELSCH*'~',F. W. WARD*,M. E. COATES*:~, R. WALKER§ and I. R. ROWLANDII *Robens Institute, University of Surrey, §Department of Biochemistry, University of Surrey, Guildford, Surrey GU2 5XH and IIBIBRA, Woodmansterne Road, Carshalton, Surrey SM5 4DS, UK (Accepted 26 March 1991)
Abstract---Germ-free (GF) rats were maintained on a diet marginally adequate in protein, with and without a supplement of NH4C1. Their urinary excretion of total nitrogen, nitrate, urea and creatinine was measured before and for 4 days after injection of Escherichia coil endotoxin (lipopolysaccharide;LPS). Although more nitrogen was excreted by rats on the diet supplemented with NH4C1, nitrate excretion was increased to a similar extent in rats on both diets. This suggests that oxidation of ammonia released by deamination of amino acids is an unlikely pathway of nitrate synthesis. In a second experiment, nitrate excretion before and after injection of LPS was measured in GF and conventional (CV) rats given highor low-protein diets. Urinary 3-methylhistidine(3MH) was measured as an index of breakdown of tissue protein. In both environments, nitrate excretion was significantly greater, before and after LPS administration, by rats on the high-protein diet than by their counterparts on the low-protein diet, and was generally greater by GF than by CV rats. Since only small, non-significant rises in urinary 3MH were observed after LPS treatment, it was concluded that the bulk of the nitrogen required for nitrate synthesis in response to endotoxin is derived from dietary protein rather than from nitrogenous products of tissue breakdown.
excessive amount of protein, was measured before and after injection of LPS. In view of the reports that ammonia is a precursor of endogenous nitrate, a similar group of rats received the diet supplemented with 5 g NH4CI/kg. The reticuloendothelial system in G F rats is not exposed to microbial challenge, and so the synthesis of nitrate by immunostimulated macrophages might be expected to be less in the G F state. Nevertheless, in our experience nitrate excretion by G F rats is generally somewhat greater than that by their CV counterparts (Ward et al., 1989). For further evidence on this apparent anomaly, in the present paper, nitrate excretion by G F and CV rats given diets of high- (HP) or low-protein (LP) content was measured before and after injection of LPS. Endotoxin induces an inflammatory reaction with consequent breakdown of tissue proteins, the nitrogenous products of which could provide precursors for nitrate synthesis. To test this possibility, urinary excretion of 3-methylhistidine (3MH), an indicator of breakdown of skeletal muscle, was also measured to ascertain whether or not its pattern of excretion paralleled that of nitrate.
INTRODUCTION
The existence of a mammalian process for the endogenous synthesis of nitrate is now well established (Green et al., 1981; Tannenbaum et al., 1978; Witter et al., 1981). Our comparative studies in germ-free (GF) and conventional (CV) rats suggest that dietary protein is an important source of nitrogen for endogenous nitrate synthesis (Ward et al., 1989), and other workers have shown that ammonia, the ultimate endpoint of protein catabolism, is one of the precursors (Saul and Archer, 1984; Wagner et al., 1983). Since the demonstration by Wagner et al. (1983) that injection of rats with Escherichia coli endotoxin (lipopolysaccharide; LPS) greatly increased their urinary nitrate excretion, studies in vivo and in vitro have implicated macrophages in LPS-induced synthesis of nitrate (Stuehr and Marietta, 1985 and 1987). The first experiment described in the present paper explored the effects on endogenous nitrate synthesis of immunostimulation in the absence of gut microflora. Urinary excretion of nitrate by G F rats given a purified diet (C10), providing an adequate but not tPresent address: BP International Ltd, Surrey Research Park, Guildford, Surrey GU2 5YQ, UK. :~Present address: Department of Biochemistry, University of Surrey, Guildford, Surrey GU2 5XH, UK. Abbreviations: CV = conventional; GF = germ-free; HP = high protein; LP = low protein; LPS = lipopolysaccharide; 3MH = 3-methylhistidine.
MATERIALS AND METHODS
A n i m a l s a n d diets. G F and CV Lister Hooded rats of both sexes, bred at the University of Surrey, were used. The CV rats were housed in the open laboratory in standard rat cages with mesh floors. G F rats were
387
A . S . NIELSCI-i et al.
388 Table 1. Composition of the experimental diets Diet (g/kg) Ingredient
C 10
Lactalbumin Casein Maize starch Sucrose ~t-Cellulose Mineral mix* Vitamin mix* Choline chloride DL-Methionine
-100 652 100 50 35 8.66 1.34 3
LP 50 -898.6 --35 8.66 1.34 6.4
HP 200 -748.6 --35 8.66 1.34 6.4
*Ward and Coates (1987).
maintained in similar cages in plastic-film isolators, as described by Ward et al. (1986). From weaning until the beginning of the experiments, at about 6 wk of age, they received a commercial rodent diet of natural ingredients (GR3, Special Diet Services Ltd, Witham, Essex, UK) with a nitrate content of about 6 pg/g. The experimental diets were of purified ingredients and none contained more than 1.1 ~g nitrate/g. Their compositions are given in Table 1. All diets were sterilized by ),-radiation at 50 kGy. Autoclaved distilled water was supplied ad lib. At about 6 wk of age the rats were distributed between the experimental treatments and given the appropriate diets for 2 wk before urine collection was begun. Precise measurement of food intake could not be made inside the isolators, but final assessment of the amounts of food eaten by each group indicated little or no differences in food intake between G F rats and their CV counterparts. The animals were weighed at the end of the collection period and all parameters were expressed as per 100 g body weight. Analytical methods. Nitrate was measured by the chemiluminescence method of Waiters et al. (1987), as previously described by Ward et al. (1989). Total nitrogen was determined by a micro-Kjeldahl technique. Creatinine and urea nitrogen were measured colorimetrically with diagnostic kits nos 555 and 535, respectively (Sigma Chemical Co. Ltd, Poole, Dorset, UK). Urinary 3MH was determined as described by Jones et al. (1982). Since rats excrete a large proportion of their total 3MH in the N-acetylated form, the urines were hydrolysed before determination of total urinary 3MH. The urine was concentrated 3 × by rotary evaporation, an equal volume of concentrated HC1 was added and the mixture was auto-
claved for 2hr at 130°C. The hydrolysate was neutralized with 5 M-NaOH, derivatized with fluorescamine reagent (acetonitrile 160 mg/litre) and incubated at 80°C with 2 M-HCI for 1 hr to destroy fluorescamine derivatives of amino acids other than histidine and 3MH. The acid-stable fluorescent derivatives of histidine and 3MH were injected onto a column (15cm x 4.6mm Altex C18 ultrasphere; Beckman) and eluted with methanol-acetate buffer (48%:52%) at a flow rate of 1 ml/min. The eluates were quantified by means of a Gilson Spectra GIo fluorimeter, with input and output filters of < 390 nm and >460 nm, respectively. Experimentalprocedure. Experiment 1 explored the effects of LPS injection on the excretion of nitrate and other nitrogenous products by G F rats given a diet marginally adequate in protein, with or without a supplement of NH4C1. Groups of five female G F rats were maintained on a diet containing 100 g casein/kg, with or without addition of 5 g NH4CI/kg. Urine was collected in 1 M-HCI for 3 days, and the collections from each rat were pooled. The animals then received an ip dose of 100 pg LPS/kg body weight (E. coli serotype 0127:88, no. 3129, Sigma). Urine was collected daily for a further 4 days. The pooled 3-day collection prior to injection, and each individual daily collection thereafter, were analysed for total nitrogen, nitrate, urea nitrogen and creatinine. Experiment 2 investigated the effect of LPS injection on the excretion of nitrate and 3MH by G F and CV rats given diets with HP or LP content. Groups of two male and two female rats received diets containing either 50 or 200 g lactalbumin/kg for 2 wk, then urine was collected in 0.1 M-HC1 for 7 days, and the collections from each rat were pooled. The animals were injected ip with 100 #g LPS/100 g body weight and daily collections of urine were made for a further 4 days. The pooled initial sample and each daily post-injection sample were assayed for nitrate, creatinine and 3MH. Statistical analysis. For all parameters, an analysis of variance was done on the results of each daily collection, using the variance within groups as the estimate of error. Student's t-test was used to ascertain the significance of differences between the basal values and those on each day after administration of LPS. In experiment 2, the basal values for nitrate and 3MH differed between diets or environments, so the
Table 2. Effect of a dietary supplement of 0.5 g NH4CI/kg and of injection of endotoxin on daily urinary excretion of nitrogenous products by GF rats Hr after injection of endotoxin Parameter
Basal
24
48
72
96
26.7+0.8 35.8_+2.9*
32.8+4.6 42.4_+4.1
35.8_+ ll.l 40.6_+4.4
35.8+5.9 44.2_+4.2
31.3+2.0 38.5_+3.9
Total nitrogen (mg)
U A
Nitrate (/~g)
U A
Urea nitrogen (mg)
U A
17.9_+1.1 22.4_+3.2
24.1 _ + 4 . 7 28.7_+3.5
25.5_+5.6 27.1 _ + 4 . 0
26.8_+4.1 30.7_+3.9
21.9_+2.2 24.5_+3.2
Creatinine (mg)
U A
2.62 _+0.2 2.62_+0.2
2.79 _+0.3 2.72_+0.3
2.68 _+0.3 2.68_+0.3
3.03 _+ 0.2 3.03_+0.2
2.67 _+ 0.2 2.41 +0.1
159_+22.6 150_+ 1 0 . 2
649_+70.4++ 704_+82.7++
571 _+ 113.9t 504_+39.5++
187_+16.6 219_+14.8
U = unsupplemented diet A = diet supplemented with 5 g NH4C1/kg *Significantly different (Student's t-test) from corresponding U value at P < 0.05. tSignificantly different (Student's t-test) from basal value at P < 0.05. ++Significantly different (Student's t-test) from basal value at P < 0.01. Values (calculated as per 100 g body weight) are means + SEM of five rats.
167_+ 16.5 169_+ 13.4
Diet, gut flora and LPS-induced nitrate synthesis
389
Table 3. Effectsof injection of endotoxin on daily urinary excretion of nitrogenouscompounds by GF and CV rats given diets containing 50 (LP) or 200 (HP) g laetalbumin/kg Hr after injection of endotoxin Diet
Rat
Nitrate ~ g ) LP
HP Statistical comparison Statistical comparison 3 MH (pg) LP HP Statistical comparison Statistical comparison Creatinine (mg) LP HP
Basal
24
72
333 __.85 192_+26 647 _ 89 368_+37 P < 0.01 P < 0.01
148 _+26 116_+22 440 _+94 313_+11 NS P < 0.001
96
GF CV GF CV GF v. CV LP v. HP
149 -+ 10 146_+37 427 _+60 314_+47 NS P < 0.001
GF CV GF CV GF v. CV LP v. HP
34+_16 23 + 5 29 + 3 19_+4 P < 0.05 NS
29+11 43 _+14 32 _+6 29_+5 NS NS
52_+4 46 __.20 38 _+4 28_+6 NS NS
43_+8 29 + 9 27 _+4 31 _+7 NS NS
46_+12 27 _+9 28 _+4 23_+6 NS NS
2.91 2.24 2.71 2.28
2.45 2.86 1.98 1.52
2.80 2.76 2.79 2.26
2.53 2.45 2.78 2.70
3.12 2.23 2.64 2.03
GF CV GF CV
91 -+ 12 304_+39* 539 _+88 478_+108 NS P < 0.05
48
149 + 37 82_+6 530 + 73 300_+29 P < 0.05 P < 0.001
NS = not significant Values (calculated as per 100 g body weight)are means _+SEM of four rats, and those marked with asterisksdiffer significantly (Student's t-test) from the corresponding basal value (*P < 0.05). increase in excretion was considered a m o r e valid measure o f the effect o f LPS t h a n the actual a m o u n t excreted. The total a m o u n t excreted by each rat during the 4 days after LPS injection was calculated, and the increase over a similar period prior to injection was expressed as a percentage o f the basal value. RESULTS The G F animals in b o t h experiments remained u n c o n t a m i n a t e d throughout. The daily urinary excretions o f nitrogenous c o m p o u n d s , calculated as per 100 g b o d y weight, are recorded in Table 2. Creatinine excretion was similar in b o t h groups and remained fairly c o n s t a n t t h r o u g h o u t , indicating that the urinary collections were reliable 24-hr samples. As expected, total nitrogen excreted by rats given the diet supplemented with NH4C1 was significantly higher t h a n that excreted by the u n s u p p l e m e n t e d controls. This p a r a m e t e r remained higher, t h o u g h not significantly, after injection o f endotoxin. There was a small, but non-significant, increase in urea nitrogen Table 4. Increasesin nitrate and 3MH excretionby GF and CV rats, given diets containing 50 (LP) or 200 (HP) g lactalbumin/kg, during 4 days followingexposure to LPS, and expressedas a percentageof the amount excreted during four days prior to LPS injection 4 Days 4 Days Diet Rat beforeLPS after LPS % Increaser Nitrate Og/100 g body weight) LP GF 598 709 25 + 36 GV 582 657 38 _+43 HP GF* 1804 2259 27 +_ 15 CV 1256 1447 23 _+11 3MH (pg/100 g body weight) LP GF 136.3 169.2 32 _+27 CV 93.1 145.0 45 + 25 HP GF* 113.2 121.8 33 __.17 CV 77.5 111.2 52 -+ 21 *One rat died. ~'Values are means _+SEM of 4 rats. There were no significant differencesbetween GF and CV rats or LP and HP diets.
excretion by the supplemented rats. Nitrate excretion was n o t affected by the dietary supplement o f NH4Ci, but in b o t h groups there was a four-fold increase in nitrate excretion after injection o f LPS. Results o f the analyses o f daily urinary collections from G F and CV rats in experiment 2 are given in Table 3. Creatinine values again indicated satisfactory daily collections. Rats in b o t h environments given the H P diet excreted significantly m o r e nitrate than those given the LP diet, b o t h before and after administration o f LPS. Nitrate excretion o f G F rats was, in general, higher than that by their CV counterparts, although the increase was n o t always significant. After injection o f LPS, nitrate excretion rose in all groups (except for one anomalous value), but to a lesser extent t h a n in the previous experiment. The responses o f rats o n the H P diet were higher than those o n the LP diet, but followed a roughly similar pattern. Initial excretion o f 3 M H was higher by G F rats. Small, non-significant increases occurred after LPS injection, but no differences were observed between rats in the two environments or on the two diets. W h e n the increase in nitrate or 3 M H excretion after treatment with LPS was considered as a percentage o f the corresponding basal value (Table 4), there was an indication o f greater excretion o f 3 M H by CV c o m p a r e d with G F rats, but the difference did not reach statistical significance. DISCUSSION In confirmation o f our previous findings (Ward
et al., 1989), dietary protein concentration markedly influenced urinary excretion o f nitrate, with rats given the H P diet excreting two to three times m o r e nitrate t h a n those on the LP diet. This strengthens the suggestion that at least some endogenously synthesized nitrate is derived from dietary protein. One possible pathway for such synthesis is by hydrolysis o f the protein, d e a m i n a t i o n o f the a m i n o acids thus formed and oxidation o f the released a m m o n i a to
390
A.S. NIELSCHet al.
nitrate. The results of the first experiment, however, would argue against such a route, since a supplement of NH4C1 in diet marginally adequate in protein did not stimulate nitrate excretion, in spite of an increase in total urinary nitrogen. It is more likely that the synthesized nitrate is derived from oxidation of the amino group of arginine (Marietta, 1988), a process known to be stimulated by activation of macrophages (Iyengar et al., 1987). The results presented here are consistent with this hypothesis. Administration of bacterial endotoxin, LPS, caused an increase in nitrate excretion in all rats studied, although the magnitude of the increase was highly variable. Such variation in response between individuals and between experiments has been observed on previous occasions in both our laboratories. Although the LPS-induced increase in daily nitrate excretion was rarely significantly different from the basal value, the total nitrate excreted during the 4 days after LPS injection was always greater than in an equivalent period before LPS treatment (Table 4). It is pertinent to note that the response to LPS was greater in experiment 1, in which the diet contained casein, than in experiment 2, when lactalbumin was the protein source. Casein contains a higher proportion of arginine than lactalbumin (4.2 compared with 1.2g/16g protein nitrogen). The difference may also have been due to differences in antigenicity of the two proteins. As previously reported (Ward et aL, 1989), nitrate excretion by G F rats was usually greater than that by corresponding CV animals, but the magnitude of response to LPS exposure was similar in CV and G F rats (Table 4). The peak response in G F rats, whether on a HP or LP diet, occurred after 48 hr, but in CV rats the response to LPS peaked at 24 hr (Table 3). This may reflect differences in macrophage activation kinetics between rats with and without a microflora. However, since immunostimulation by LPS resulted in similar increases in nitrate synthesis in both environments, it can be assumed that the lower nitrate excretion by CV rats is due to microbial utilization or breakdown of nitrate. Urinary excretion of 3MH was not significantly affected by LPS treatment, implying that the dose administered was insufficient to cause severe tissue damage. It was, however, generally higher than the basal value in both CV and G F rats. From the cumulative 4-day excretion after LPS treatment, there was an indication that the CV rats on both diets had been more severely affected than their G F counterparts. Interestingly, the peak of 3MH occurred at a similar time to the peak of nitrate excretion, that is 24 hr after LPS exposure in CV rats and 48 hr in G F rats. Nevertheless, the much greater excretion of nitrate by rats on the HP diet was not paralleled by that of 3MH. It must be concluded, therefore, that although nitrogenous compounds released from damaged tissues may constitute a small proportion of the precursor(s) for nitrate synthesis, the bulk of the
nitrogen required appears to be derived from dietary protein. Acknowledgements--This work was supported by a grant from the Wellcome Trust, to whom we express our thanks. We are grateful to Professor D. J. Naismith and Dr P. Emery of King's College (Kensington Campus), London, for the use of equipment for the assay of 3MH, and to Miss V. Ronaasen for the care and management of the gnotobiotic rats. REFERENCES
Green L. C., Tannenbaum S. R. and Goldman P. (1981) Nitrate synthesis in the germfree and conventional rat. Science, New York 212, 56-58. Iyengar R., Stuehr D. J. and Marietta M. A. (1987) Macrophage synthesis of nitrite, nitrate and Nnitrosamines: precursors and role of respiratory burst. Proceedings of the National Academy of Sciences of the U.S.A. 84, 6369-6373. Jones A. D., Shorley D. and Hitchcock C. H. (1982) The determination of 3-methylhistidineand its application as an index for calculating meat content. Journal of the Association of Public Analysts 20, 89-94. Marletta M. A. (1988) Mammalian synthesis of nitrite, nitrate, nitric oxide and N-nitrosating agents. Chemical Research in Toxicology i, 249-257. Saul R. L. and Archer M. C. (1984) Oxidation of ammonia and hydroxylamine to nitrate in the rat and in vivo. Carcinogenesis 5, 77-8 i. Stuehr D. J. and Marietta M. A. (1985) Mammalian nitrate biosynthesis: mouse macrophages produce nitrite and nitrate in response to Escherichia coli lipopolysaccharide. Proceedings of the National Academy of Sciences of the U.S.A. 82, 7738-7742. Stuehr D. J. and Marietta M. A. (1987) Synthesis of nitrite and nitrate in murine macrophage cell lines. Cancer Research 47, 5590-5594. Tannenbaum S. R., Fett D., Young V. R., Sand P. C. and Bruce W. R. (1978) Nitrite and nitrate are formed by endogenous synthesis in the human intestine. Science, New York 2116, 1487-1489. Wagner D. A., Young V. R. and Tannenbaum S. R. (1983) Mammalian nitrate biosynthesis: incorporation of tSNH3 into nitrate is enhanced by endotoxin treatment. Proceedings of the National Academy of Sciences of the U.S.A. 80, 4518--4521. Waiters C. L., Gillatt P. N., Palmer R. C. and Smith P. L. R. (1987) A rapid method for the determination of nitrate and nitrite by chemiluminescence. Food Additives and Contaminants 4, 133-140. Ward F. W. and Coates M. E. (1987) Dietary fat and N-nitrosation in the rat. British Journal of Nutrition 58, 221-231. Ward F. W., Coates M. E. and Walker R. (1986) Nitrate reduction, gastro-intestinal pH and N-nitrosation in gnotobiotic and conventional rats. Food and Chemical Toxicology 24, 17-22. Ward F. W., Coates M. E. and Walker R. (1989) Influence of dietary protein and gut microflora on endogenous synthesis of nitrate and N-nitrosamines in the rat. Food and Chemical Toxicology 27~ 445-449. Witter J. P., Gatley S. J. and Balish E. (1981) Evaluation of nitrate synthesis by intestinal microorganisms in vivo. Science, New York 213, 449-450.
