Dogmas and controversies in the handling of nitrogenous wastes: Excretion of nitrogenous wastes in human subjects

Protein intake is essential for survival. Its oxidation yields two major nitrogen-containing waste products, urea and ammonium(NH₄⁺), which must be removed by renal excretion. While noting the specific physiology that influences their individual excretion rates, integrative aspects should also be considered, i.e. the need to ensure that these excretions do not alter the composition of the urine and thereby cause urinary solutes to precipitate within the urinary tract. Urea is not an effective urine osmole; accordingly, in this paper urea excretion is analyzed in the context of controlling the minimum urine flow rate. In addition to considering the physiology of acid balance(Kamel et al., 2002),NH₄⁺ excretion is analyzed in the context of the need to maintain a urine pH of ∼6.0 to minimize the risk of kidney stone formation(Coe and Parks, 2000).

Two aspects are explored in this section: first, we comment on the urine flow rate needed to minimize the risk of forming kidney stones; second, we consider how the minimum urine flow rate may be higher when urea is excreted.

To illustrate this point, the rise in the ion product for calcium ×oxalate, the constituents of the commonest form of kidney stones in humans(Coe and Parks, 2000), are calculated at a 24 h average urine flow rate (1.2 ml min^–1),at the usual mean overnight urine flow rate of 0.6 ml min^–1(Halperin et al., 2002), and at a lower urine flow rate of 0.3 ml min^–1. For the purpose of this calculation, we shall assume that the excretion rates for calcium and oxalate remain constant (Table 1). The calcium × oxalate ion product is 4- and 16-fold higher at the 0.6 and 0.3 ml min^–1 flow rates than at the 1.2 ml min^–1 rate. The major danger of precipitation occurs when the urine flow rate decreases from 0.6 to 0.3 ml min^–1.

Of great interest, vasopressin also causes the insertion of urea transporters into the luminal membrane of the inner medullary collecting duct(MCD) (Sands, 1999). In rats consuming their usual diet, the MCD is sufficiently permeable to urea that its concentrations are similar in the urine and in the renal papilla(Gowrishankar et al., 1998)(Table 2). Hence urea is not an effective osmole in the urine and it does not control the urine flow rate in this setting.

When the urine is electrolyte-poor, its flow rate could decline sufficiently to cause a precipitate to form because of the low rate of excretion of effective osmoles (Equation 1, Table 1). An example of this conundrum is a human in a hot environment who sweats and ingests little salt or water. A deficit of Na⁺, Cl^– and water will develop and the urine will contain very few of the usual effective osmoles(Na⁺ and Cl^–).

Ketonuria of prolonged fasting seems to waste useful energy. Moreover, to achieve acid–base balance, these ketoacid anions must be excreted with NH₄⁺, adding to the apparent disadvantage in their excretion because the source of the nitrogen would be from lean body mass(Halperin et al., 1989). Kamel et al. (1998) postulated that there was an advantage in excreting NH₄⁺ plus ketoacid anions instead of urea during prolonged fasting, when viewed as a physiological adaptation to allow for the excretion of urine with a flow rate in a safer range (∼0.5 ml min^–1). In more detail, when 2 mmoles of NH₄⁺ are excreted along with 2 mmoles of ketoacid anions, this results in acid–base balance; however, the urine will contain four effective milliosmoles containing the same two nitrogens that would be present in one millimole of the ineffective urine osmole, urea. While useful in the prolonged fasted state, this mechanism does not provide a means to ensure a safe minimum urine flow rate in fed mammals.

The only non-electrolyte osmole that is excreted in a sufficient quantity in a fed mammal is urea, but urea does not seem to be an effective urine osmole when vasopressin acts in rats consuming the usual amount of electrolytes (Table 2).

To gain insights into how to resolve this conundrum, the concentration of urea was measured in the urine and in the interstitial compartment of the papilla in rats consuming a low-electrolyte diet while they received vasopressin. The concentration of urea was significantly higher in the urine than in the papillary interstitial compartment(Table 2). This implies that urea is not freely permeable in the inner MCD when the urine is electrolyte-poor. Accordingly, we speculated that under this circumstance,urea could become an effective osmole in the inner MCD despite the continuing presence of vasopressin (Fig. 1). In fact, the change in urea permeability need only be modest to ensure the needed small rise in the urine flow rate.

In summary, to allow for water conservation, urea is not an effective osmole when vasopressin acts in fed mammals(Gamble et al., 1934). Nevertheless, when the urine is electrolyte-poor, there is a risk of precipitate formation in the urinary tract if urea does not become an effective osmole to ensure a safe minimum urine flow rate. The mechanism that could permit this physiological alteration in urea reabsorption is a subject of ongoing studies.

The excretion of NH₄⁺ is examined from two similar perspectives. The first focuses on its primary function, allowing the body to achieve acid–base balance when the kidney must eliminate a H⁺load. The second perspective is the requirement to maintain a urine pH of∼6.0 to minimize the risk of uric acid precipitation(Fig. 2).

The excretion of NH₄⁺ can be thought of as having two components. First, NH₄⁺ is synthesized in the proximal convoluted tubule (PCT) along with HCO₃^–; its precursor is the electroneutral amino acid, glutamine(Halperin et al., 1989). The selection of this glutamine pathway of metabolism in PCT cells depends on having a low intracellular fluid (ICF) pH. Generation of this new HCO₃^– is not complete until NH₄⁺ is made into an end-product of metabolism;NH₄⁺ must be excreted in the urine otherwise it would be returned to the liver and converted to urea + H⁺.

Second, for high rates of secretion of NH₄⁺, ammonia(NH₃) must diffuse across the renal medullary interstitial compartment to the MCD. In this process, NH₄⁺ are reabsorbed in the thick ascending limb of the loop of Henle, replacing K⁺ on the Na⁺, K⁺, 2-Cl^–cotransporter (NKCC). This generates the `single effect' for recycling of NH₄⁺ in the loop of Henle and the build-up of the high concentration of NH₃ in the medullary interstitial compartment to facilitate its diffusion into the lumen of the MCD (Fig. 3 ). To complete the transfer of NH₃ to the final urine as NH₄⁺, the pH of the luminal fluid in this nephron segment should be less than 6.0 (Knepper et al., 1989); this low urine pH is caused by H⁺secretion in the MCD, primarily by the H⁺-ATPase.

A low urine pH is needed to achieve high rates of excretion of NH₄⁺. Hence the price to pay to defend acid–base balance would be to accept a risk of precipitation of uric acid in the terminal nephron. This led us to reconsider the traditional view of the physiology of the excretion of NH₄⁺ and the role of its medullary shunt pathway.

As judged from the data in subjects fed NH₄Cl on a chronic basis(Madison and Seldin, 1958; Simpson, 1971) and during the ketoacidosis of prolonged fasting (Kamel et al., 1998), a low urine pH is not needed to augment the excretion of NH₄⁺ because maximum excretion rates occur when the urine pH is ∼6.0.

We address this question under three headings: theoretical considerations,an experiment using loop diuretics in rats to inhibit the reabsorption of NH₄⁺ in the loop of Henle, and data from rats with chronic metabolic acidosis that permit a quantitative analysis of the contribution of the medullary interstitial shunt pathway to NH₄⁺ excretion.

Diffusion is a slow process, which requires a high concentration of the substance that will diffuse and the absence of a barrier for its diffusion. Although NH₃ is the species of NH₄⁺/NH₃ transported across the basolateral membrane out of cells of the medullary thick ascending limb of the loop of Henle (Kikeri et al., 1989),owing to the interstitial fluid pH and the pK for NH₄⁺,the interstitial concentration of NH₃ is 1/100 that of NH₄⁺. Of greater importance, there are barriers for the diffusion of NH₃ in the renal medullary interstitial compartment,i.e. the lipid component of cell membranes of the MCD.

Another way to gain insights into the importance of a pathway is to inhibit it. Because NH₄⁺ is reabsorbed in the loop of Henle, a loop diuretic should markedly reduce medullary NH₄⁺shunting. Following action of the loop diuretic, ethacrynic acid, rather than reducing the rate of excretion of NH₄⁺ as one would expect if this NH₃ shunt pathway were critical for NH₄⁺ excretion, the rate of NH₄⁺excretion rose significantly (Vasuvattakul et al., 1993). Because the urine pH fell in the experiments employing ethacrynic acid, we used the loop diuretic, furosemide, which has a minor carbonic anhydrase effect that prevents the fall in the urine pH. Furosemide led to a marked rise in the rate of excretion of NH₄⁺, but in these experiments, this rise could not be attributed to a fall in the urine pH (Table 4). Therefore this medullary shunt pathway lowers rather than raises the rate of NH₄⁺ excretion. Hence another hypothesis is needed to identify a possible function of this shunt pathway.

Sajo et al. (1981) measured NH₄⁺ excretion in rats with chronic metabolic acidosis using a microcatheterization technique to sample fluid from the junction of the cortical and medullary collecting ducts. They found that approximately 75%of the NH₄⁺ excreted was added prior to entry into the MCD (Fig. 3). Therefore a high rate of shunting of NH₄⁺ from the loop of Henle to the MCD is not required to achieve a high rate of excretion of NH₄⁺ in the rat.

The major function of the medullary NH₃ shunt pathway is not to achieve high rates of excretion of NH₄⁺, but possibly to prevent a large fall in the urine pH. This can be accomplished by having a robust luminal NH₄⁺/H⁺ ion exchanger to remove H⁺ secreted by the MCD during chronic metabolic acidosis(Verlander et al., 2003; Weiner and Verlander, 2003). In this hypothesis, distal H⁺ secretion will provide the driving force for the transport of NH₄⁺ into the lumen of the MCD and hence this process functions as an adjuster of the urine pH.

The overall process is described in more detail in Fig. 4. It begins with the reabsorption of NH₄⁺ from the loop of Henle, which adds NH₃ to the medullary interstitial compartment (the H⁺ to convert it to NH₄⁺ are added at site 4 in Fig. 4). Recycling of NH₄⁺ in the loop of Henle raises the concentration of NH₄⁺ in the medullary interstitium(Knepper et al., 1989) (site 1, Fig. 4). NH₄⁺ can diffuse rapidly enough through the renal medullary interstitial compartment because its concentration is high. NH₄⁺ crosses both lipid-containing cell membranes of the MCD via two different Rh-glycoproteins that function as NH₄⁺/H⁺ exchangers, one on the basolateral and another on the luminal membrane of these cells (sites 3 and 4 in Fig. 4)(Verlander et al., 2003; Weiner and Verlander, 2003). The combination of H⁺ exit from, and NH₄⁺entry into, the lumen of the MCD could adjust the urine pH upward (towards 6.0) by removing luminal H⁺ despite continuing H⁺secretion by the H⁺-ATPase. The net result is a final urine pH that is ∼6.0 and a somewhat higher rate of NH₄⁺excretion.

In summary, we suggest that NH₄⁺ plays a direct role in both the traditional physiology (generates new HCO₃^–) and in the integrative physiology (adjusts the urine pH to ∼6.0). For the latter, NH₄⁺ diffuses across the medullary interstitial compartment and that there is a H⁺-linked counter-transport system with NH₄⁺across both polar membranes of MCD cells. Moreover, the driving force for this process is the secretion of H⁺ by the MCD, because the concentration of NH₄⁺ in the urine is higher than in the papillary interstitial compartment (Table 2). Thus, the activity of the NH₄⁺/H⁺ exchanger in the luminal membrane might act as an `adjuster' for the final urine pH to minimize the risk of uric acid stone formation (Fig. 4).

Our goal was to consider the renal excretion of the major nitrogenous waste products, urea and NH₄⁺, from the perspectives of their specific physiology and of integrative physiology, i.e. the need to ensure that the composition of the urine would minimize the risk of forming kidney stones. For urea, a system was needed to make it become an effective urine osmole in electrolyte-poor urine and thereby avoid a dangerously low urine flow rate. A special consideration for the excretion of NH₄⁺ was to ensure that the urine pH was close to 6.0 without sacrificing acid–base balance. During chronic metabolic acidosis, the high rate of excretion of NH₄⁺ needs to be achieved while maintaining a urine pH of 6.0. The medullary shunt process for NH₃, which requires a low luminal pH, does not appear to be important for NH₄⁺ excretion. Rather, this shunt pathway utilizes the diffusion of NH₄⁺ plus two NH₄⁺/H⁺ exchangers in the MCD, and it can prevent a large fall in the urine pH due to distal H⁺secretion.

This research was supported by a grant from Canadian Institutes of Health Research – MT-15485.