INTRODUCTION

TOP ABSTRACT INTRODUCTION NITROGEN METABOLISM ANIMALS WITH A HIGH... SUMMARY REFERENCES

EVOLUTION CAN BE CONSIDERED a natural experimental process for selecting out animal design characteristics with a survival advantage. Some of these adaptations have occurred in response to conditions that could be considered analogous to chronic renal failure. For example, the vampire bat, which ingests an extremely high-protein diet, has adapted to a state of azotemia (41), whereas the American black bear can survive for months without intake or urination and yet have blood urea levels that actually fall during this period (1). How have these animals adapted to such unusual circumstances, and can we learn new approaches for the prevention/treatment of chronic renal failure from such natural models?

First, the general design features of mammals with respect to nitrogen metabolism and excretory functions of the kidney are briefly described. With these general characteristics as a reference state, what is known about the adaptation of several small mammals to an extremely high protein intake and about the bear's adaptation to months without intake and excretion is reviewed.

Although our understanding of these adaptations is limited, the existing data do suggest areas of inquiry that could result in new management strategies for chronic renal failure.

Design principles and quantitative functional interactions can be studied by the use of allometric scaling analysis (12). This technique deals with the effects of changes in body size (or scale) on biological functions and structures. Many morphological and physiological variables scale, relative to body size, according to allometric equations of the following general form: y = ax^b, where y is a biological variable, a is a proportionality coefficient, x is body mass, and b is the scaling exponent (59).

We would expect biological functions that are closely linked to scale to body mass in a similar fashion, i.e., to have similar scaling exponents. Furthermore, the ratio of biological functions with the same exponent would give a value that is independent of body size and, hence, a design feature of mammals in general (60, 68). It should be stressed, however, that these ratios do not necessarily allow any conclusions about cause and effect.

It is also obvious that design rules that apply to mammals varying in size from mouse to elephant will suppress individual species differences.

Perhaps the best-studied biological variable is resting metabolic rate (RMR), which has a scaling exponent of 0.75 (64, 76). About 20% of RMR can be accounted for by the energy expenditure for whole body protein turnover (73, 74), with muscle the principal tissue (73). In mammals studied at energy equilibrium, whole body protein turnover, which reflects the balance between synthetic and degradative processes, scales to body size, with an exponent of 0.74 (53). In addition, the turnover of tRNA and rRNA has been found to scale to body size, with exponents of 0.78 and 0.69, respectively (61).

Jackson and co-workers (17, 29, 33, 34) investigated urea production in normal adult men and women of different body composition consuming their habitual diets. Daily urea production per kilogram of body weight was relatively independent of daily protein intake over the range 0.5-1.4 g/kg body wt. However, there was a close relationship between urea production and RMR (33). In addition, with the use of these same data (17, 29, 33, 34), total daily urea production can be fitted to body mass by the following allometric equation: urea production (g N/day) = 0.52W^0.76, where W is body mass in kilograms. Although these data are for humans only and the weight range studied is limited, urea production over protein intakes of 0.5-1.4 g · kg¹ · day¹ scales to body size with an exponent similar to that of RMR. Finally, urinary nitrogen excretion (8) and glomerular filtration rate (GFR) (64) also scale to body size, with exponents of 0.72 and 0.77, respectively. The allometric equations for all these biological functions are given in Table 1. In summary, the scaling data describe close links between energy flux (RMR), protein and RNA turnover, urea production, and kidney excretory processes.

In addition, because the coupling between RMR and GFR is also observed in birds, reptiles, and fish (12, 65), this particular design feature has been well conserved from an evolutionary perspective.

	NITROGEN METABOLISM

TOP ABSTRACT INTRODUCTION NITROGEN METABOLISM ANIMALS WITH A HIGH... SUMMARY REFERENCES

For most amino acids, the initial catabolic step involves removal of the amino group by transdeamination with generation of ammonia. The carbon skeleton is further processed, and because the majority of amino acids are gluconeogenic, the end result is conversion to glucose. The subsequent fate of the ammonia depends on the type of animal. Mammals detoxify ammonia at a considerable energetic cost through the synthesis of urea, which is subsequently excreted.

A simplified model of nitrogen metabolism is illustrated in Fig. 1. The metabolic nitrogen pool consists mainly of free amino acids and ammonia. Amino acids enter the pool from ingested protein or from degradation of body protein. Catabolism of amino acids gives rise to urea. A portion of the urea appearing in the urea pool is excreted in the urine, and a portion is transferred to the (large) bowel, where it is hydrolyzed by bacterial urease. A part of this nitrogen (as ammonia) is resynthesized back into urea, whereas the remainder is available for other synthetic processes. The rate of urea production is equal to the rate of urea excretion by the kidney plus the rate of urea hydrolysis in the bowel.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Nitrogen metabolism (modified from Ref. 34).

The intestinal hydrolysis of urea by microflora appears to be a general mammalian feature (14, 24, 70, 79), and in fact the functional interactions between intestine, kidney, and urinary bladder are emerging as a sophisticated system for regulating nitrogen excretion/conservation. Specific urea transporters have been identified in the colon and kidney collecting duct (3, 54, 62, 66, 80). Urea transport processes (passive and active) can be induced in the collecting duct of rats fed a low-protein diet (32, 58, 66). Urea reabsorption from the mammalian urinary bladder has been observed in ground squirrels (24) and the bear (47).

The adaptive importance of renal urea transporters for purposes other than water conservation/osmoregulation is underscored by the observation that a similar transporter may exist in the freshwater trout (40). This fish demonstrates net tubular urea reabsorption against an apparent concentration gradient. Because, in freshwater fish, urea does not play any significant role in osmoregulation, the purpose of this reabsorptive process is unclear. Could the function of urea reabsorption in teleost fish be for nitrogen conservation?

A specific urea transport protein has also been described in mammalian liver (36), and it is believed that at least one function for this transporter is facilitation of the movement of urea, after its synthesis, out of hepatocytes and into the extracellular space. Menyhart and Grof (43) demonstrated that urea, even at concentrations as low as 8.3 mmol/l, produced significant inhibition of argininosuccinate lyase, one of the urea-ornithine cycle enzymes. Although ammonia is generated intramitochondrially in the liver, the final steps in urea synthesis, including the reaction catalyzed by argininosuccinate lyase, occur within the cytosol (15). Given the observation by Menyhart and Grof, the hepatic urea transporter could actually function as a regulator of ureagenesis by controlling the cytosolic concentration of urea. Klein et al. (36) also reported that the hepatic urea transport protein was significantly increased in abundance in rats made uremic by five-sixths nephrectomy. The physiological interpretation of this observation is not entirely clear. Klim et al. (37) described increased urea production from alanine or glutamine in isolated hepatocytes from uremic rats, whereas Cano et al. (16) reported decreased ureagenesis from these same substrates in isolated rat hepatocytes from their uremic animals (compared with controls in both cases). Even though the experimental designs differed, it is difficult to resolve the discrepancy between these results. If hepatic urea production is increased in chronic uremia, then upregulation of the urea transporter would be an appropriate adaptive response.

In humans, over the range of protein intakes in which daily urea production per kilogram of body weight is constant, there is an inverse relationship between intake and rate of urea hydrolysis (per kg body wt) and a positive relationship between intake and urea excretion per kilogram of body weight (17, 33).

Only limited urea kinetic studies have been done in other mammals (except the bear; see American black bear), but the results are consistent with the observations in humans. For example, in muskrats (14), the rate of urea hydrolysis was 67% higher in fall and winter, when forage protein levels are lowest, than in spring and summer. Herbivores such as the 13-lined ground squirrel and the Wyoming ground squirrel (24) have much higher levels of urea hydrolysis than the carnivorous badger (24) and the American marten (26). There is also good evidence in humans that a significant fraction of nitrogen released by intestinal hydrolysis can be used for synthetic processes other than the resynthesis of urea (21, 34, 38, 42, 50). That is to say, intestinal hydrolysis is not simply a "futile" process for cycling urea but should be considered a nitrogen salvage system contributing to "effective nitrogen intake" (33). The overall excretion/conservation of nitrogen involves the coordination of multiple processes such as GFR, urea reabsorption in the kidney collecting duct and urinary bladder, and transport of urea into the bowel.

In summary, mammals possess a set of biochemical/physiological processes that link energy flux, protein metabolism, and renal excretory functions. A component of the linkage involves urea transporters in the colon, kidney, liver, and probably urinary bladder, which regulate nitrogen excretion/conservation in response to changes in dietary protein intake.

The signals by which these interdependent processes communicate are unknown. There is circumstantial evidence (65) that ammonia might be a signal molecule mediating adjustments in GFR in response to changes in amino acid oxidation (i.e., urea production).

An additional constraint that relates to the capacity of structural elements in a set of physiological/biochemical processes needs to be introduced. A scaling analysis of oxygen supply during maximum performance in mammals has led to the principle of symmorphosis (30, 60). Symmorphosis postulates that the structural design of components comprising a system is matched quantitatively to functional demand (30). Applying this principle to the mammalian metabolic/renal set of functions, one would expect the magnitude of GFR (the scaling exponent of which is composed of the exponents for two structural parameters: glomerular number and capillary surface area) (64) to be appropriate for metabolic needs and not to contain a large excess capacity. This expectation will be addressed in ANIMALS WITH A HIGH PROTEIN INTAKE, in which we examine the features of animals adapted to a high protein intake.

	ANIMALS WITH A HIGH PROTEIN INTAKE

TOP ABSTRACT INTRODUCTION NITROGEN METABOLISM ANIMALS WITH A HIGH... SUMMARY REFERENCES

Can we establish a measure of how much GFR is appropriate for metabolic needs? Because metabolic demand relates to the rate of amino acid catabolism, whole body protein turnover or rate of urea production would be the best measure. Unfortunately, comparative animal data available for these functions are very limited. There are considerable data on urinary nitrogen excretion that could be used as an index of amino acid deamination/oxidation, because in mammals, urea nitrogen comprises the bulk of urinary nonprotein nitrogen (2). However, because a variable fraction of urea is transported into the bowel, where it is hydrolyzed, urinary nitrogen excretion will underestimate the extent of amino acid deamination. From a renal perspective, the net rate of urea (nitrogen) excretion will be a function of the rate of filtration less the rate of reabsorption in the collecting duct (32, 58, 66) and, possibly, the bladder (24, 47). Collecting duct urea reabsorption is increased in animals fed a low-protein diet (32, 58, 66). Because filtration is clearly the predominant process in the excretion of urea, and with the above qualifications in mind, the GFR-to-urinary nitrogen excretion ratio will be used as a crude global measure of the match between glomerular function and metabolic demand. For mammals, in general, what would be the expected value for this ratio; i.e., how much GFR does the "typical" mammal have for each gram of nitrogen excreted per day? There are several approaches to this question.

From the allometric equations in Table 1, for a mammal on a practically nitrogen-free but energy-complete diet, the value would be 32.7 (4.78/0.146) ml · min GFR¹ · g N excreted¹ · day¹. However, this value would probably be close to the upper limit in mammals. If, instead, we use the allometric equations for urea production and GFR (Table 1) and assume that 70% of urea nitrogen is excreted in the urine [30% of urea being hydrolyzed in the bowel (17)], then the ratio would be 13.1 (4.78/0.36) ml · min GFR¹ · g N excreted¹ · day¹. Finally, blood urea levels have been measured in many mammals (human, cow, dog, goat, guinea pig, horse, rabbit, rat, sheep, pig, and elephant) spanning a large size range and average 2-10 mmol/l (63, 67). This range of blood urea concentrations is probably related to the protein content of the diet. With the use of a GFR predicted by the allometric equation (Table 1) and the assumption that all urinary nitrogen is in the form of urea and urea clearance is ~50% of GFR (51), in a mammal of given weight, e.g., 70 kg, a blood urea concentration of 2 mmol/l would correspond to a ratio of 25 ml · min GFR¹ · g N excreted¹ · day¹ and a blood urea concentration of 10 mmol/l would correspond to a ratio of 5 ml · min GFR¹ · g N excreted¹ · day¹.

Thus the design feature of a typical mammal eating its usual diet would include a GFR of 5-25 ml · min¹ · g N excreted¹ · day¹ and a blood urea concentration of 2-10 mmol/l. If an animal increases its protein intake, the amount of urea nitrogen to be excreted would increase. The GFR-to-urinary nitrogen excretion ratio would fall unless GFR increased commensurate with the augmented nitrogen load. A fall in the ratio would be reflected by a rise in the blood urea concentration. Data in the literature indicate that the metabolic demands imposed by an increase in protein intake are not matched by an increase in GFR.

As reviewed by Brenner et al. (7), acute or chronic increases in protein intake can increase GFR 40-150% above baseline values. For example, GFR was ~70% higher in rats maintained on a 35% protein chow than in rats fed a 6% protein chow. Harbor seals can have a threefold increase in GFR several hours after a fish meal (28). However, GFR is not able to increase commensurate with metabolic demand. In dogs fed a meat meal (10 g/kg), GFR increased on average by 39% acutely, but plasma urea concentration increased an average of 67% acutely, indicating that the change in GFR was not sufficient to maintain a constant plasma urea level (48). The same observation has been made in adult men who were studied not acutely but over 2-wk periods after a change in protein intake (13). Urea clearance increased by 87% above baseline when protein intake was increased from the lowest level (zero) to the highest level (44 g/day). However, plasma urea concentration also rose by 165% above baseline, indicating that the change in GFR was not commensurate with the increase in urea synthesis. All these observations indicate a limited capacity of GFR to respond to increases in metabolic demand created by increases in protein intake and are consistent with the principle of symmorphosis. Because protein intake (and, hence, required nitrogen excretion rate) can be increased more than GFR, what is the lower limit of the ratio (GFR/urinary nitrogen excretion) that is tolerated in nature? In other words, from an evolution/natural selection perspective, how much mismatch between metabolic demand and glomerular function is allowed when this mismatch is created by a primary increase in metabolic demand?

There are several small mammals that naturally ingest a high-protein diet. The limited data available for these mammals in terms of nitrogen excretion and GFR are sufficient to allow us to estimate GFR-to-urinary nitrogen excretion ratios.

Marten and white-tailed prairie dog. Harlow and Buskirk (26) measured nitrogen metabolism in martens and white-tailed prairie dogs before and during a fast. Before fasting, the martens and prairie dogs were maintained on ad libitum food and water. Mean body mass was 1.2 and 1.15 kg for the prairie dog and marten, respectively. GFR, measured as creatinine clearance, was 5.06 and 2.46 ml/min in the marten and prairie dog, respectively. Harlow and Buskirk measured daily urine urea excretion, and the values before fasting were 65 and 15 mmol/day for the marten and prairie dog, respectively. If we assume that urinary nitrogen is entirely accounted for by urea nitrogen, the GFR-to-urinary nitrogen excretion ratios are 2.78 (5.06/1.82) and 5.86 (2.46/0.42) ml · min¹ · g N¹ · day¹ for the marten and prairie dog, respectively. These ratios are associated with blood urea concentrations before fasting of 16.2 and 12.1 mmol/l in the marten and prairie dog, respectively. Both of these animals do hydrolyze urea within the intestine (24, 26), but detailed urea kinetic measurements have not been made. About 5% of injected labeled urea is hydrolyzed in both animals.

Vampire bats. Vampire bats are nocturnal and forage once nightly, during which time the vampire bat engorges itself with blood until it is satiated. McFarland and Wimsatt (41) studied selected aspects of renal function in a colony of vampire bats fed bovine blood once a day. The rapid ingestion of blood results in a 30-40% increase in body weight, which can hinder the bat's flying ability. The vampire bat shows almost instantaneous micturition with feeding and can achieve peak urine flow rates of 4.0 ml · kg¹ · min¹ within 30 min after starting a feed. This osmotic diuresis is probably due to the rapid absorption of plasma fluids during feeding (with extracellular volume expansion) and is followed by a phase of water conservation. This phase is necessitated by the high insensible water loss and the fact that vampires do not utilize free water, even if it is available in their roosts. For example, 7 h after a meal, urine osmolality can reach 4,600 mosM, with urine urea concentrations of ~3,000-3,500 mmol/l.

1

1

1

Shrew. Shrews are small highly active mouselike mammals. Measurement of RMR in these animals is difficult because of their activity and excitability, and values have been obtained that are about two to three times that predicted by the mammalian allometric equation (10, 19, 39). Shrews have voracious appetites and consume ~70% of their body weight in food daily (18). The shrew feeds on a wide variety of common insects. The prey of the shrew is ~30% solid (39), and 50% of this solid portion is protein (10). Buckner (10) measured urinary nitrogen excretion rates in four species of shrew. The results for daily nitrogen excretion were 0.075 g for Suncus cinereus (3.6 g body wt), 0.140 g for Suncus arcticus (5.4 g body wt), 0.086 g for Microsorex hoyi (3.5 g body wt), and 0.130 g for Blarina brevicauda (20.1 g body wt).

American black bear. Unlike the vampire bat and shrew, which have adapted to a protein-rich diet, the American black bear has adapted to diametrically different conditions, which could be considered analogous to chronic renal failure. This mammal survives winter for up to 5 mo at near-normal body temperature without eating, drinking, urinating, or defecating (1). During this dormant period, RMR decreases by ~50% (44) and GFR by ~70% (9) compared with the active state. However, despite continual amino acid oxidation/deamination with urea production and complete reabsorption of urine by the bladder, the plasma urea concentration actually falls and lean body mass is maintained (4, 9, 47).

1

1

1

1

	SUMMARY

TOP ABSTRACT INTRODUCTION NITROGEN METABOLISM ANIMALS WITH A HIGH... SUMMARY REFERENCES

In mammals, RMR, whole body protein and RNA turnover, urea production, GFR, and urinary nitrogen excretion constitute a set of broad interdependent functions. The signals coordinating these functions are unknown, but ammonia may be one of the signal molecules. Additional features can be added to this framework. A variable fraction of produced urea is transported to the colon, where it undergoes hydrolysis. Some of the ammonia released is reformed back into urea, but most is available for other synthetic processes. Urea excretion by the kidney depends on GFR less urea reabsorption in the distal nephron and bladder. Because urea transporters have been identified in the colon and nephron collecting duct, it is quite plausible to consider GFR plus colonic and collecting duct transport rates as a coordinated system for regulating nitrogen excretion/conservation. For mammals in general, the capacity of processes such as GFR is approximately matched to RMR, whole body protein turnover, and rate of urea production. It is apparent, however, that this matching has limits. In mammals, such as the vampire bat and shrew, which have an exceedingly high protein intake, GFR is not commensurate with the large urea load to be excreted. The vampire bat and probably also the shrew are azotemic. Blood urea concentrations are 27-57 mmol/l in the vampire bat but, unfortunately, have not been measured in the shrew. How have these animals adapted to an azotemic internal environment that would include not only high urea levels but probably also high ammonia levels? Why do these mammals not develop progressive renal failure given their exposure to an intense, chronic hyperfiltration stimulus? The American black bear, on the other hand, has adapted to a prolonged period without intake or urine output. Despite continued amino acid catabolism with urea production, this mammal is able to completely salvage and reutilize urea nitrogen for protein synthesis. It is not known what signals initiate this metabolic adaptation during the bear's winter sleep or whether the bear possesses unique metabolic pathways not present in all mammals. Clearly, understanding the particular adaptations of the vampire bat, shrew, and bear to their unusual situations could lead to new interventions for the prevention and treatment of chronic renal disease. This approach underlies the principles of biomimicry. Biomimicry is the science that studies nature's models and then imitates or takes inspiration from these designs and processes to solve human problems. After 3.8 billion years of evolution, nature has learned what works, what is appropriate, and what lasts (6).