Abstract:
The Utilization of Ammonia-Nitrogen by the Sheep I. W. M C D ONALD * Summary Ammonia plays a central role in the nitrogen metabolism of the rumen; essentially, this is due, on the one hand, to its origin from the deamination of proteins and other nitrogenous substances and from the hydrolysis of urea, and on the other hand to its use by ruminal bacteria as a source of nitrogen for growth. Ammonia is absorbed from the rumen and converted in the liver to urea, which in part return' to the rumen in saliva and by s direct diffusion from the blood. These events and their implications in the nutrition of the sheep are reviewed, and attention is drawn to some deficiencies in present knowledge of this important aspect of ruminant metabolism. The fact that ammonia plays an important role in normal rumen metabolism was first recognised by McDonald (1948 a, and b), and has since been amply confirmed by other workers. The dynamic nature of events in the rumen and the heterogeneity of \ is contents make quantitative study extremely difficult and hence, although the major qualitative features are now clear, much work still remains to be done before the reactions within the rumen can be fully related to the nutrition and tissue metabolism of the ruminant. ORIGIN O F RUMEN AMMONIA The main source of rumen ammonia is fodder protein. The rumen contents are actively proteolytic (Schlottke, 1936; Sym, 1938; Pearson and Smith, 1943 ; Hoflund, Quin and Clark, 1948 ; Chalmers et. al., 1954; Warner, 1956; Annison, 1956)) yet amino-acids occur in the rumen in conspicuously low concentrations (el-Shazly, 1952 a; Lewis, 1955; Annison, 1956) ; the activity of rumen microbes in deaminating amino-acids (el-Shazly, 1952 b; Sirotnak et. al., 1953 ; Lewis, 1955; Warner, 1956) explains this finding and accounts fol the occurrence of ammonia. Deamination of proteins from a variety of sources has been demonstrated: casein, gelatin, ground-nut meal, herring meal, meadow hay, preserved grass (frozen, dried or ensiled) and fresh pasture (McDonald, 1952; Chalmers et al., 1954; Chalmers and Synge, 1954 ; Annison et. al., 1954; Johns, 1955; Head and Rook, 1955 ; Briggs, Hogan and Reid, 1957). The soluble proteins, e.g. in green leaves, peanut meal, or casein, are readily degraded in the rumen while highly insoluble proteins, e.g. zein, are very slowly attacked. The proteolytic activity is little influenced by diet, but the capacity of rumen bacteria to deaminate amino-acids is dependent on the presence in the diet of readily attacked protein (el-Shazly, 1952 b; Warner, 1956). Young growing plants contain significant amounts of aminoacids which contribute directly to the rumen ammonia. Some plants contain large amounts of nitrate, which is readily reduced, via nitrite, to ammonia (Sapiro et. al., 1949; Lewis, 1951). Other *Division of Animal Health and Production, C.S.I.R.O. Sheep Biology Laboratory, Prospect, New South Wales. 46 nitrogenous substances in plants (amides, amines, betaines, alkaloids, purines and nucleic acids) will presumably also be degraded by rumen organisms. Urea is rapidly hydrolysed in the rumen (Lenkeit and Becker, 1938). Urea is present in the sheep' saliva (McDonald, 1952; s Somers, 1957) and also reaches the rumen contents by diffusion from the blood (Simonnet, Le Bars and Molle, 1957; McDonald, unpublished data, 1957; Hogan, 1957)) and is thus an important source of rumen ammonia. THE CQNCENTRATION OF AMMONIA IN THE RUMEN The concentration of ammonia in the rumen depends on four major factors: rate of formation within the rumen, rate of passage to omasum, rate of absorption from rumen and rate of uptake by bacteria. At present it is not possible to give precise values for these rates in any given set of circumstances. (i) Rate of Ammonia Formation in the Rumen.-This will be influenced by feeding behaviour, e.g. pasture intake occurs over many hours, while stall-fed sheep may consume a day' ration s in an hour or so; this factor becomes very important in drought feeding when small supplements can be rapidly consumed. Most of the experimental work reported has been with animals fed once or twice daily and more work is required on sheep at pasture. A second factor is the ease of deamination of the nitrogenous compounds, especially proteins, in the feed. Thirdly, the level of ammonia is augmented by urea secreted in saliva and diffusing into the rumen from the blood. In mixed sheep saliva the urea concentration is approximately 60 per cent. of that in whole blood (McDonald, 1957) and the urea accounts for about 65-70 per cent. of the total-N (Somers, 1957) ; the existing data for the volume of saliva secreted by the sheep permit the calculation that some 0.5-I gm N can enter the rumen through the saliva (McDonald, 1948 b; Somers, 1957). No data are yet available to permit evaluation of the contribution from diffusion. (ii) Rate of Passage to Omasum.-Since fluid is continuously added to the rumen as saliva, and portions of the mixed contents are continuously being passed through the reticula-omasal orifice, the rate of passage of solutes out of the rumen is a logarithmic function of the amount present at any given time and is best expressed in terms of a biological half-life. In recent experiments on hay-fed sheep, the rate of passage of a soluble marker (polyethylene glycol-Sperber, Hyden and Ekman, 1953) varied considerably with an average half-life of 9 hours; for a sheep with rumen volume of 6 litres and rumen ammonia an average of 15 mg N/100 ml, this half-life would result in about 1.5 gm ammonia-N/day passing to the omasum. ( i i i ) R a t e of Absoription from the Rumen.-Absorption o f ammonia into the ruminal veins has been well established (McDonald, 1948 b; Bouckaert and Oyaert, 1952; Chalmers et. al., 1954) ; Lewis et al. (1957) and Hogan (1957) found that the rate of absorption is proportional to the concentration in the rumen and considered that absorption is by simple diffusion, although the rate of absorption may be influenced by pH or the concentration of other substances in the rumen. Lewis et al. (1957)) using Schambye' (1956) data s for portal blood flow in the conscious sheep, calculated that as much as half the daily nitrogen intake may be absorbed as ammonia from the rumen. Gray and Pilgrim (1956)) using N/lignin ratios, concluded that when a high quality lucerne hay was fed to sheep, some 60 per cent. of the fodder-N was absorbed from the rumen, presumably chiefly as ammonia. An alternative procedure is to compare the changes in concentration of ammonia and a soluble marker ; in an experiment with this technique, the absorption of ammonia, per hour, was found to be equivalent to the amount 47 contained in about 1 litre of rumen fluid; assuming a mea n con-centration of 15 mg ammonia-N/100 ml, this is equivalent to 3-4 gm N/day (McDonal d, unpublished data, 1957). These calculations, imperfect though they are, do indicate the importance of absorption of ammonia from the rumen. (iv) Rate of Uptake of Ammonia by Micro-organisms of the Rumen.-The present state of knowledge is very sketchy. The most serious deficiency is that we do not know the concentration of ammonia required for microbial growth, and. hence cannot determine, by direct analysis, whether the ammonia level in the rumen is adequate. It has been shown (McDonald, 1952; Annison et al., 1954) that when adequate starch is present, the ammonia concentration in the rumen may fall almost to zero -thi s suggests that bacteria may take up ammonia at very low concentrations, but gives no indication of the concentration at which N would be a limiting factor for bacterial growth. Again, it is not yet clear how the various carbohydrates compare in effectiveness as source of carbon and energy for bacterial growth when ammonia is the N- source. Dietetic experiments strongly suggest that cellulose is of little value in this direction but it is not known whether this is due to the slowness of cellulolysis in the rumen or whether, as seems more, probable, that the cellulolytic organisms have limited ability to use ammonia as a source of nitrogen for growth. Examination of rumen ammonia concentrations rather suggests that sugars (glucose, sucrose) are as effective in reducing ammonia as is starch, but the weight of dietetic evidence indicates that starch is the more useful compound. Lewis and McDonald (1957) found that xylan and levan reduced ammonia concentration as much as starch, but no dietetic evidence on their comparative value is available. TOXICITY OF AMMONIA McDonald (1948 b) found that the concentration of ammonia in peripheral blood of sheep was extremely low, and concluded that the ammonia absorbed from the rumen was wholly converted to urea by the liver. Lewis et al. (1957) found that changes in the rumen ammonia concentration of sheep fed various diets were paralleled by changes in portal blood ammonia concentration, but there was no increase in ammonia concentration in peripheral blood unless the rumen ammonia concentration was raised above 80 mg N/100 ml by adding ammonium acetate; similar results were obtained by Hogan (1957). It is clear therefore that an hepatic threshold is established by the rate at which the liver can remove ammonia from the portal blood; Head and Rook (1956) observed in cattle on spring pasture that ammonia absorption was sufficiently great to cause an increase in concentration in the peripheral blood; Johns (1955) did not record whether blood ammonia was elevated in grazing sheep with very high rumen concentrations. It has long been recognised that ammonium ion is toxic, and interest in this topic has increased in recent years (see review by Bessman, 1956). In the sheep the normal level of ammonia in peripheral blood is less than 2pg NHs-N/ml (McDonald, 1948 b; Repp et aZ., 1955; Lewis et al., 1957; Hogan, 1957). The symptoms of ammonia intoxication have been described by Dinning et at. (1948), Clark et al. (1951)) Repp et al. (1955) and Lewis et al. (1957) ; toxicity occurs when blood concentration rises above 8-12 ,ug NH3-N/ml (Repp et al., 1955; Lewis et ccl., 1957) Clark et al. (1951)) Kaishio et al. (1952) and Hale and King (1955) have doubted whether ammonia and urea poisoning are in fact due to the toxicity of ammonium ion to tissues; the pathogenesis of these syndromes has not been fully established, and the effects of ruminal stasis and alkalosis, occurring concomitantly with accumulation of ammonia in peripheral blood, are obscure. However, Payne (1955) 48 has found that ammonium hydroxide, chloride, acetate and carbamate all gave similar pharmacological effects when injected into dogs. From the evidence available it seems justifiable to conclude that any compound forming ammonia in the rumen will prove toxic if the rumen ammonia level is sufficiently raised so that the rate of ammonia absorption exceeds the conversion threshold of the liver and thus leads to a rise in peripheral blood ammonia to a level of more than 10 ,ug N/ml. N O N-P ROTEIN -N ITROGEN XN R UMINANT N UTRITION The capacity of N.P.N., especially urea, to replace part of the protein in ruminant diets has been the subject of hundreds of experiments (see reviews by Reid, 1953, and Hale, 1956). Although there is no doubt that N.P.N. can, through the use of ammonia-N for microbial growth in the rumen, provide protein to the ruminant, it is noteworthy that the limitations of this activity have not been ascertained. Emphasis has usually been placed on weight gain or milk production as the criterion of utilization of N.P.N.; however, under Australian conditions, this use of N.P.N. will probably find greater application in low levels of feeding under drought conditions; the work of Franklin, Briggs and McClymont (1955) leaves little doubt that this will prove an economic procedure. Urea is a normal constituent of the rumen contents and its subsequent fate and contribution to the animal' nitrogen metabolism s differs in no respect from that of ammonia derived from protein or other nitrogenous substances. Obviously then, a clear understanding of the normal nitrogen metabolism in the rumen will provide the background for prediction of the value of N.P.N. in various ruminant rations. It is probable that, provided other factors (e.g. palatability, toxicity) do not interfere, the value of any nonprotein-nitrogenous substance will depend on its conversion to ammonia in the rumen and on the suitability of the rumen contents as a medium for the growth of bacteria capable of using ammonia as a N-source. Recent experiments on sheep given urea with wheat or maizemeal indicate that the rate of hydrolysis of urea is extremely fast and that ammonia leaves the rumen in quantity before there is much fermentation of the starches; these observations suggest that the N taken up by bacterial growth under these conditions may be largely derived from urea returned to the rumen and not from that of the ration. Attention may be drawn to the relative amounts of protein-N and ammonia-N in the rumen. The rumen contents normally have an extremely high microbial population and this is reflected in a high protein content. Boyne et al. (1956) report analyses of rumen contents of sheep fed twice daily and receiving 34 gm N/day; at the period 12 hr. after feeding, the rumen-reticulum was found to contain approximately 28 gm N, i.e. over 80 per cent. of the daily intake; if one assumed a value of 20 mg ammonia-N/100 gm rumen contents in these sheep, the NH3-N would then represent only 4-5 per cent. of the total-N in the rumen. Comparable data are not available for sheep fed rations with relatively high proportions of N.P.N. but this example serves to indicate the need for information on the composition and changes in composition of the rumen contents in such cases. B IOLOGICAL V ALUE OF F ODDER P ROTEINS The complexities of nitrogen metabolism in the rumen have important implications in nutrition. In the non-ruminant, the biological value of a food protein is chiefly determined by its amino-acid composition. In the ruminant on the contrary, this would hold only 49 for that fraction of the food protein which escaped digestion in the rumen and passed unchanged into the abomasum where it could undergo peptic, and subsequently intestinal, digestion ; the fraction of portein degraded in the rumen would have value to the ruminant to the degree that it was converted into microbial protein. Assuming that the nitrogen metabolism of ruminant tissues is essentially the same as that of the non-ruminants, it is probable that only that nitrogen absorbed in the form of amino-acid assemblies can be utilized in the tissues; other forms of nitrogen absorbed would be valueless since they would merely contribute to an excess of N from non-essential amino-acids. Microbial activity will thus be profitable to the ruminant only if it leads to an increase in ' quality' or quantity of amino-acids available to the host. A consideration of the amino-acid composition of plant leaves suggests that no marked improvement in amino-acid composition would accrue by conversion to microbial protein (Weller, 1957) ; some seed proteins could be markedly improved in this way. It is probably safe to generalize as follows: on ordinary rations for maintenance or production, the greater the resistance of the protein to deamination in the rumen, the more effectively will it be utilized provided that subsequent digestion and absorption are not impaired. Such a situation has been strikingly demonstrated by Chalmers et aZ. (1954)) using casein as the main feed protein. When, however, the diet contains high proportions of starch a net gain of N can accrue due to the utilization of ammonia-N for microbial growth. It is therefore clear that, for the ruminants, biological value must be referred to the diet as a whole and not to the proteins of the diet alone. RUMEN A MMONIA AND BIDOD UREA C ONCENTRATION Since ammonia is absorbed from the rumen and converted by the liver into urea, it is obvious that this constitutes one factor influencing the concentration of urea in the blood. The passage of urea out of the blood stream into saliva and rumen similarly must play a role in this connection. Lewis (1957). postulated that the rumen ammonia level had a controlling influence on blood urea, since there was usually a close correlation between the blood urea concentration and the level of rumen ammonia at its maximum about 4 hr after feeding ; he suggested that this finding might form the basis of a supplementary test for assessing the value of protein in ruminant rations. The amount of urea-N in the blood and tissues is much in excess of the amount of ammonia-N in / the rumen (e.g. in a sheep of body weight 40 kg, rumen volume 5 litres, blood urea-N 15 mg/lOO ml, rumen NHs-N 12 mg/lOO ml, body water 60 per cent. B.W., then urea-N : NHS-N :: 6 : 1). This explains why blood urea concentration does not reflect the marked diurnal changes in rumen ammonia (Lewis, 1957). Little is known of the control of urea excretion rate by the ruminant kidney and thus it is not yet possible to calculate the direct loss of rumen ammonia via urea formation and urinary excretion. R EFERENCES 50 Briggs, P. K., Hogan, J. P., and Reid, R. L. (1957) .-Aust. J. agric. Res. 8: 674. Chalmers, M. I., Cuthbertson, D. P., and Synge, R. L. M. (1954).J. agric. Sci. 44: 254. Chalmers, M. I., and Synge, R. L. M. (1954) .-J. agric. Sci. 44: 263. Clark, R., Oyaert, W., and Quin, J. I. (1951) .-Onderstepoort J. vet. Sci. 25 : 73. Dinning, J. S., Briggs, H. M., Gallup, W. D., Orr, H. W., and Butler, R. (1948) .-Amer. J. Physiol. 153: 41. El-Shazly, K. (1952 a) .-Rio&em. J. 51.: 640. El-Shazly, K. (1952 b) .-Biochem. J. 51: 647. Franklin, M. C., Briggs, P. K., and McClymont, G. L. (1955) .-J. Aust. Inst. agric. Sci. 21: 216. Gray, F. V., and Pilgrim, A. F. (1956) .-Nature, Lond. 178: 94. Hale, W. H. (1956) .-J agric. Food Chem. 4: 948. Hale, W. H., and King, R. P. (1955) .-Proc. Sot. exp. BioZ., N.Y. 89: 112. Head, M. J., and Rook, J. A. F. (1955) .-Nature, Lond. 176: 262. Hoflund, S., Quin, J. I., and Clark, R. (1948).-Onderstepoort J. vet. Res. 23: 395. Hogan, J. P. (1957) , .-Thesis, University of Aberdeen. Johns, A. T. (1955). -N.Z. J. Sci. Tech. A37: 323. Kaishio, Y., Higaki, S., Harii, S., and Awai, Y. (1951) .-Bull. Nat. Inst. agric. Sci. (Japan) 2: 131. Lenkeit, W., and Becker, M. (1938) .--Z. Tiererniihr. FuttermitteZk. 1: 97. Lewis, D. (1951) .-Biochem. J. 48: 175. Lewis, D. (1955) .-Bet. J. Nutr. 9: 215. Lewis, D. (1957) .-J. agric. Sci. 43: 438. Lewis, D., Hill, K. J., and Annison, E. F. (1957) .-(In press). Lewis, D., and McDonald, I. W. (1957) .-( In press). McDonald, I. W. (1948 a) .-Thesis, Cambridge University. McDonald, I. W. (1948 b) .-Biochem. J. 42: 584. McDonald, I. W. (1952) .-Biochem. J. 51: 86. Payne, L. C. (1955). -Rep. Conf. Rumen Function, U.S.D.A. (Washington) . Pearson, R. M., and Smith, J. A. B. (1943) .-Biochem. J. 37: 153. Reid, J. T. (1953) .-J. Dairy Sci. 36: 955. Repp, W. W., Hale, W. H., Cheng, E. W., and Burroughs, W. (1955). -J. AnimaZ Sci. 14: 1 1 8 . Sapiro, M. L., Hoflund, S., Clark, R., and Quin, J. I. (1949) .Onderstepoort J. vet. Sci. 22: 357. Schambye, P. (1956) .-Nord. Vet.-Med. 8: 1001. Schlottke, E. (1936) .-S.B. naturf. Ges. Restock, Ser. 3, 6: 59. Simonnet, H. Le Bars, H., and Molle, J. (1957) .-C.R. Acad. Sci., Paris. 244: 943. Sirotnak, F. M., Doetsch, R. N., Brown, R. E., and Shaw, J. C. (1953) .-7. Dairy Sci. 36: 1117. Somers, M. (1957) .-Aust. vet. J. 33: 297. Sperber, I., Hyden, S., and Ekman, J. (1953).-KgZ. LantbruksHiigskoZ. Ann. 20: 337. Sym, E. A. (1938) .-Acta BioZ. exp., Varsovie. 12: 192. Warner, A. C. I. (1956) .-J. gen. Microbial. 14: 749. Weller, R. A. (1957) .-Aust. J. biol. Sci. 10: 384. 51