Severity of hyperammonemic encephalopathy correlates with brain ammonia level and saturation of glutamine synthetase in vivo.

Correlation among in vivo glutamine synthetase (GS) activity, brain ammonia and glutamine concentrations, and severity of encephalopathy was examined in hyperammonemic rats to obtain quantitative information on the capacity of GS to control these metabolites implicated in the etiology of hepatic encephalopathy. Awake rats were observed for neurobehavioral impairments after ammonium acetate infusion to attain a steady-state blood ammonia concentration of 0.9 (group A) or 1.3 mumol/g (group B). As encephalopathy progressed from grade III to IV, brain ammonia concentration increased from 1.9 to 3.3 mumol/g and then decreased to 1.3 mumol/g on recovery to grade III. In contrast, brain glutamine concentration was 26, 23, and 21 mumol/g, respectively. NH(4+)-infused rats pretreated with L-methionine DL-sulfoximine reached grade IV when brain ammonia and glutamine concentrations were 3.0 and 5.5 mumol/g, respectively; severity of encephalopathy correlates with brain ammonia, but not glutamine. In vivo GS activity, measured by NMR, was 6.8 +/- 0.7 mumol/h/g for group A and 6.2 +/- 0.6 mumol/h/g for group B. Hence, the in vivo activity, shown previously to increase with blood ammonia over a range of 0.4-0.64 mumol/g, approaches saturation at blood ammonia > 0.9 mumol/g. This is likely to be the major cause of the observed accumulation of brain ammonia and the onset of grade IV encephalopathy.

Dependence of in vivo glutamine synthetase activity on ammonia concentration in rat brain studied by 1H - 15N heteronuclear multiple-quantum coherence-transfer NMR.

The dependence of the in vivo rate of glutamine synthesis on the substrate ammonia concentration was studied in rat brain by 1H-15N heteronuclear multiple-quantum coherence-transfer NMR in combination with biochemical techniques. In vivo rates were measured at various steady-state blood and brain ammonia concentrations within the ranges 0.4-0.55 mumol/g and 0.86-0.98 mumol/g respectively, after low-rate intravenous 15NH4+ infusion (isotope chase). The rate of glutamine synthesis at steady state was determined from the change in brain [5-15N]glutamine levels during isotope chase, observed selectively through the amide proton by NMR, and 15N enrichments of brain glutamine and of blood and brain ammonia measured byN gas chromatography-MS. The in vivo rate (v) was 3.3-4.5 mumol/h per g of brain at blood ammonia concentrations (s) of 0.40-0.55 mumol/g. A linear increase of 1/v with 1/s permitted estimation of the in vivo glutamine synthetase (GS) activity at a physiological blood ammonia concentration to be 0.4-2.1 mumol/h per g. The observed ammonia-dependence strongly suggests that, under physiological conditions, in vivo GS activity is kinetically limited by sub-optimal in situ concentrations of ammonia as well as glutamate and ATP. Comparison of the observed in vivo GS activity with the reported in vivo rates of glutaminase and of gamma-aminobutyrate (GABA) synthesis suggests that, under mildly hyperammonaemic conditions, glutamine is synthesized at a sufficiently high rate to serve as a precursor of GABA, but glutaminase-catalysed hydrolysis of glutamine is too slow to be the sole provider of glutamate used for GABA synthesis.