A fortuitous development was the discovery of the urea cycle just prior to the availability of 15
N and 13
C as metabolic tracers [6
]. Among the first applications of the isotopic method was the demonstration [7
] with a 13
probe that the carbon atom of urea derived from carbon dioxide (actually, HCO3−
is the source). The objective of most early isotopic (usually 15
N) studies of urea synthesis was not to measure flux through the urea cycle but to measure rates of total body protein synthesis [8
]. In the typical experiment a 15
N-labelled amino acid was administered to human subjects and the appearance of the label in urinary 15
N (most of which is comprised of [15
N]urea) was determined with isotope ratio-mass spectrometry [10
]. Whole body protein synthesis is then derived from the assumption that isotope retained in the body corresponds to that incorporated de novo
into body proteins [8
Technological advances enabled the use of stable isotopes to study flux through the urea cycle in youngsters with inborn errors of metabolism. Most early studies utilized isotope ratio-mass spectrometry as an analytic tool with which to measure isotopic enrichment. This approach had the advantage of exquisite sensitivity with regard to the detection of isotopic label, but it was far less sensitive with respect to the amount of material that is needed for analysis. In addition, isotope ratio-mass spectrometry requires the complete separation of the analyte of interest and the combustion of this analyze to a gaseous form (CO2 or N2) in order to determine the 13C/12C or 15N/14N isotopic ratio. The coupling of the mass spectrometer to a gas chromatograph facilitated pediatric investigations, since reproducible measurements now could be obtained even with relatively small samples sizes (often < 100 μl of plasma or serum). In addition, the great resolving power of gas chromatography, which can separate hundreds of metabolites in a single analysis, greatly amplifies the repertoire of compounds now susceptible to study.
A notable example of this approach is the study of Lee et al [11
], who measured urea turnover in a group of patients with urea cycle defects, including ornithine transcarbamylase deficiency (10 female/5 male), citrullinemia (5) and argininosuccinic aciduria (2). The experimental protocol involved the continuous infusion of [5-15
N]glutamine and [18
O]urea, which enabled total body turnover of the urea pool and of the amide-N of glutamine. Measurements of blood [15
N]urea defined the rate of nitrogen transfer from 5-N of glutamine to urea. They found that the ratio of [5-15
N]urea discriminated the control subjects from those with a urea cycle defect, with this parameter being much higher in patients with disease of neonatal onset.
This study involved a constant infusion of isotope over a relatively prolonged (7.5 hr) period. There are advantages to parenteral administration of tracer, which neutralizes the problem of variable absorption from the gastrointestinal tract. The constant infusion method also lends itself to a straightforward kinetic analysis. However, there are unavoidable problems attendant upon this method, which may be difficult to implement in children. A long infusion period is necessary to achieve steady-state in an individual with a urea cycle defect and consequent diminution of urea turnover. In addition, the relatively small vessels of infants and children commonly present a problem with regard to venous access. Children (as well as many adults) have found it difficult to summon the patience to comply with a long period of isotope infusion. The latter difficulty may be especially vexatious in a child (or adult) with a urea cycle defect, in whom impaired cognition and a relatively short attention span are common comorbidities.
A method based upon oral administration of isotope might prove more practicable to pediatric studies. To this end, we devised a protocol [12
] based upon the oral administration of 15
Cl. We utilized gas chromatography-mass spectrometry to monitor the appearance of label in [15
N]urea and [5-15
N]glutamine over a 4-hour time period following the test dose of tracer. Our study cohort for this X-linked disorder included 15 heterozygotes, of whom 6 were ostensibly asymptomatic, 8 were clearly symptomatic, and 1 was a severely affected individual with neonatal onset.
The results are shown in . We found essentially no difference with regard to the formation of [15N]urea (left panel) when we compared the response of the control group with that of asymptomatic carriers. In contrast, symptomatic heterozygotes converted significantly less 15N from ammonia to urea. A single, severely affected hemizygote was able to produce virtually no [15N]urea.
Figure 1 Left: Appearance of [15N]urea in blood after oral administration of 15NH4Cl (~ 0.37 mmol/kg; 98 atom % excess) to health controls (closed circles); asymptomatic heterozygotes with ornithine transcarbamylase deficiency (open circles); symptomatic heterozygotes (more ...)
Of note is the finding (, right) that the synthesis of [5-15
N]glutamine was greater even in asymptomatic carriers than in the control group. This observation indicates that ostensibly “asymptomatic” carriers are biochemically distinguishable from normal. In a subsequent study [14
] we explored this concept in greater detail by performing neuropsychological testing in 19 female heterozygotes for ornithine transcarbamylase deficiency. The carriers displayed normal IQ scores as well as strength in verbal intelligence and reading, but they manifested weakness in fine motor dexterity and lesser impairments in executive skills, nonverbal intelligence and mathematics. Interestingly, we found that the degree of residual urea synthetic capacity, as measured with the 15
Cl loading procedure (), often predicted cognitive measures. Thus, the isotopic method underscored the importance of maintaining good metabolic control, even in otherwise “asymptomatic” carriers. There was a significant positive correlation between individual 15
N incorporation into urea and performance in math (p = 0.0013), attention/executive (p = 0.001) and reading (p = 0.032). Indeed, the isotopic approach was more predictive of cognitive outcome than the allopurinol challenge test [14
No method is entirely free of disadvantages. Although the 15NH4Cl loading test proved highly discriminative, it did present problems: (a) The flavor of ammonium chloride is obnoxious and provoked gagging in several subjects. (b) It would be ethically impossible to administer an ammonium salt to an individual who already may have hyperammonemia. (c) Ammonia administration in such a clinical setting is experimentally unsound, since the expanded endogenous N pool would unduly dilute the isotopic tracer. (d) Finally, the method requires gas chromatography-mass spectrometry to analyze enrichment in [15N]urea and [5-15N]glutamine. As noted above, this technology is exquisitely sensitive with regard to sample size, but it is several orders of magnitude less sensitive with respect to the detection of isotopic enrichment. This becomes a confounding limitation in studying patients with severe urea cycle defects, in whom we anticipate very little incorporation of label into urea.
To this end, we endeavored to develop a method that would measure residual competency of the urea cycle by quantifying the synthesis of [13
C]urea from a 13
C-labeled precursor. The tracer we selected was [1-13
C]acetate, the carbon in which is incorporated into urea via the following reaction sequence:
In hepatic mitochondria acetate is very quickly oxidized to bicarbonate following condensation with oxaloacetate in the citrate synthetase reaction. Bicarbonate then enters the urea cycle after reacting with ammonium ion and ATP via carbamylphosphate synthetase. The 13C-carbamylphosphate ultimately becomes [13C]urea. We developed a novel method for determining isotopic abundance in [13C]urea. Since this approach deploys isotope ratio-mass spectrometry as an analytic tool, it has the advantage of being very sensitive, even when confronted with the challenge of detecting the relatively small amounts of [13C]urea that a severely affected patient typically produces.
An important indicator of the value of an isotopic method is whether it yields data that is of clinical utility in terms of devising novel therapies to improve the lives of affected patients. Our group has long been interested in N-carbamylglutamate, a synthetic glutamate derivative that appears to enhance flux through the urea cycle by stimulating the activity of carbamylphosphate synthetase (see above). The drug appears to act by mimicking the action of N-acetylglutamate, which is the naturally occurring obligatory effector of this reaction [15
]. We previously showed, using 15
Cl as tracer, that N-carbamylglutamate would virtually “cure” defective ureagenesis in patients with a congenital deficiency of N-carbamylglutamate synthetase [16
We tested [1-13C]acetate as a probe to see if this species would yield a comparable result. As shown in , when the isotope was given to a patient with a congenital deficiency of N-acetylglutamate synthetase, within minutes it was converted to 13CO2 (, panel A), a peak in which was attained by 30 minutes. This was true before (open circles) and after (closed circles) drug administration. Notwithstanding the robust production of 13CO2, the patient incorporated relatively little label into [13C]urea. This was reflected in a paucity of 13C either in the fraction of blood urea that was labeled (, panel B) or in the total concentration of blood [13C]urea (, panel C). The latter parameter represents the product of the fraction of blood urea labeled with 13C above baseline (, panel B) and the total blood urea concentration. It is evident from the data in that when this patient received treatment with N-carbamylglutamate for a period of 3 days, there was a near-complete restoration of ureagenesis. Indeed, the post-treatment appearance of label in [13C]urea is essentially indistinguishable from the control values (data not shown).
Figure 2 A patient with N-acetylglutamate synthetase deficiency was given an oral dose of [1-13C]acetate and samples subsequently were taken at the indicated times for analysis of 13C enrichment (atom % excess) in breath 13CO2 (Panel A), plasma [13C]urea (Panel (more ...)
In a subsequent study [17
] we extended this work to assess the efficacy of N-carbamylglutamate to improve metabolism in patients with propionic acidemia. Hyperammonemia is a common complication in this cohort. The precise mechanism is uncertain, but it probably entails compromise of flux through the carbamylphosphate synthase reaction. We therefore administered a short (3-day) course of N-carbamylglutamate to these patients and again utilized [1-13
C]acetate to quantify rates of urea production. We again were able to demonstrate a marked increase in [13
C]urea formation consequent to drug treatment.
C]acetate studies involve administration of a single oral dose of isotopic tracer, an approach we deployed in a “before-and-after” paradigm that entails study on 2 separate days, both prior to and following administration of N-carbamylglutamate. This paradigm could not disclose how quickly subjects respond to the medication – assuming that a response might occur within hours – since label in [13
C]urea does not return to baseline at the end of the 4-hour period of study (see ). We therefore developed a method that involves infusion of H13
over a 5-hour period with subsequent sampling of blood to detect 13
C in 13
C]urea. The rationale for this design is that by administering a priming dose and subsequent constant infusion of H13
we rapidly achieve a steady-state in this tracer since the body bicarbonate pool turns over very rapidly. Hepatocytes then draw upon this pool to form [13
We found in healthy controls [18
] that this approach rapidly (< 10 min) achieves a steady-state in 13
in expired air (). Label in plasma [13
C]urea conformed to a linear pattern (). At 90 minutes after the start of the H13
infusion we gave each subject a single oral dose of N-carbamylglutamate (, arrow). We then continued to monitor label in plasma [13
C]urea. In 5/6 subjects we noted an upward inflection of the line that described the rate of appearance of plasma [13
C]urea. These data indicate that N-carbamylglutamate augments ureagenesis even in healthy individuals and that it does so within 1–2 hours of drug administration.
Figure 3 Isotopic abundance (atom % excess) in 13CO2 in expired air in healthy adults who received an infusion of NaH13CO3 (0.089 mmol/kg/hr; 98 atom % excess) for a period of 300 minutes. The solid horizontal line denotes the mean isotopic abundance from 10 to (more ...)
Fig. 4 The concentration of [13C]urea in plasma during the constant infusion of NaH13CO3. The line represents linear regression analysis from 10 to 150 minutes. At 90 minutes each subject received an oral dose (50 mg/kg) of N-carbamylglutamate. The closed circles (more ...)