Glutamate Dehydrogenase Is Important for Ammonia Fixation and Amino Acid Homeostasis in Brain During Hyperammonemia

Abstract

Impaired liver function may lead to hyperammonemia and risk for hepatic encephalopathy. In brain, detoxification of ammonia is mediated mainly by glutamine synthetase (GS) in astrocytes. This requires a continuous de novo synthesis of glutamate, likely involving the action of both pyruvate carboxylase (PC) and glutamate dehydrogenase (GDH). An increased PC activity upon ammonia exposure and the importance of PC activity for glutamine synthesis has previously been demonstrated while the importance of GDH for generation of glutamate as precursor for glutamine synthesis has received little attention. We therefore investigated the functional importance of GDH for brain metabolism during hyperammonemia. To this end, brain slices were acutely isolated from transgenic CNS-specific GDH null or litter mate control mice and incubated in aCSF containing [U-¹³C]glucose in the absence or presence of 1 or 5 mM ammonia. In another set of experiments, brain slices were incubated in aCSF containing 1 or 5 mM ¹⁵N-labeled NH₄Cl and 5 mM unlabeled glucose. Tissue extracts were analyzed for isotopic labeling in metabolites and for total amounts of amino acids. As a novel finding, we reveal a central importance of GDH function for cerebral ammonia fixation and as a prerequisite for de novo synthesis of glutamate and glutamine during hyperammonemia. Moreover, we demonstrated an important role of the concerted action of GDH and alanine aminotransferase in hyperammonemia; the products alanine and α-ketoglutarate serve as an ammonia sink and as a substrate for ammonia fixation via GDH, respectively. The role of this mechanism in human hyperammonemic states remains to be studied.

Keywords: alanine; brain; glutamate; glutamate dehydogenase; glutamine; hyperammonemia; pyruvate carboxylase (PC).

FIGURE 1

The percent ¹³C labeling in metabolites in hippocampal slices from Glud1^lox/lox (filled bars) and CNS-Glud1^–/– (open bars) mice following metabolism of [U-¹³C]glucose via subsequent turns in the TCA cycle. Percent of alanine and lactate being M + 3 labeled (A) and the percent of citrate M + 5, citrate M + 6, glutamate M + 4, glutamate M + 5, glutamine M + 4, glutamine M + 5, GABA M + 4, and aspartate M + 4 (B). Hippocampal slices from Glud1^lox/lox and CNS-Glud1^–/– mice were incubated in the presence of [U-¹³C]glucose for 60 min as detailed in Materials and methods. The enrichment was determined from GC-MS analysis of slice extracts as detailed in Materials and methods. Results are averages ± SEM (n = 10), and the asterisk indicates a statistically significant difference between Glud1^lox/lox and CNS-Glud1^–/– mice (P < 0.05). Cit, Citrate; Glu, Glutamate; Gln, Glutamine; GABA, γ-aminobutyric acid; Asp, Aspartate.

FIGURE 2

The percent ¹⁵N labeling in metabolites in cortical slices from Glud1^lox/lox (filled bars) and CNS-Glud1^–/– mice (open bars) mice following incubation in the presence of 1 or 5 mM ¹⁵NH₄Cl for 60 min as detailed in Materials and methods. The percent mono-labeled glutamate (A), aspartate (B), alanine (C), GABA (D) as well as the percent of glutamine being mono- or double labeled (E) was determined from GC-MS analysis of slice extracts as detailed in Materials and methods. Results are averages ± SEM (n = 5–8), and the asterisk indicates a statistically significant difference between Glud1^lox/lox and CNS-Glud1^–/– mice (P < 0.05).

FIGURE 3

Hippocampal slices from Glud1^lox/lox mice were incubated in medium containing 5 mM [U-¹³C]glucose and 0 mM (control; filled bars), 1 mM (hatched bars), or 5 mM (open bars) NH₄Cl. The effect of ammonia treatment on the amounts (nmol/mg protein) of alanine, glutamine, GABA, glutamate, and aspartate (A) and the percent ¹³C labeling originating from [U-¹³C]glucose in alanine M + 3 and lactate M + 3 (B) and in citrate M + 5, citrate M + 6, glutamate M + 4, glutamate M + 5, glutamine M + 4, glutamine M + 5, GABA M + 4, and aspartate M + 4 (C). The hippocampal slices from Glud1^lox/lox mice were incubated in the presence of [U-¹³C]glucose for 60 min as detailed in Materials and methods. The amounts of metabolites were determined employing HPLC while the percent enrichment was determined from GC-MS analysis of slice extracts as detailed in Materials and methods. Results are averages ± SEM (n = 6–10), and the asterisk indicates a statistically significant difference between hippocampal slices from Glud1^lox/lox mice incubated in the absence and presence of 1 or 5 mM NH₄Cl (P < 0.05). Ala, Alanine; Lac, Lactate; Cit, Citrate; Glu, Glutamate; Gln, Glutamine; GABA, γ-aminobutyric acid; Asp, Aspartate.

FIGURE 4

Hippocampal slices from CNS-Glud1^–/– mice were incubated in medium containing 5 mM [U-¹³C]glucose and 0 mM (control; filled bars), 1 mM (hatched bars), or 5 mM (open bars) NH₄Cl. The effect of ammonia treatment on the amounts (nmol/mg protein) of alanine, glutamine, GABA, glutamate, and aspartate (A) and the percent ¹³C labeling originating from [U-¹³C]glucose in alanine M + 3 and lactate M + 3 (B) and in citrate M + 5, citrate M + 6, glutamate M + 4, glutamate M + 5, glutamine M + 4, glutamine M + 5, GABA M + 4, and aspartate M + 4 (C). The hippocampal slices from CNS-Glud1^–/– mice were incubated in the presence of [U-¹³C]glucose for 60 min as detailed in Materials and methods. The amounts of metabolites were determined employing HPLC while the percent enrichment was determined from GC-MS analysis of slice extracts as detailed in Materials and methods. Results are averages ± SEM (n = 6–10), and the asterisk indicates a statistically significant difference from hippocampal slices from CNS-Glud1^–/– mice incubated in the absence and presence of 1 or 5 mM NH₄Cl while a number sign indicates a statistically significant difference between hippocampal slices from CNS-Glud1^–/– mice incubated in 1 and 5 mM NH₄Cl (P < 0.05). Ala, Alanine; Lac, Lactate; Cit, Citrate; Glu, Glutamate; Gln, Glutamine; GABA, γ-aminobutyric acid; Asp, Aspartate.

FIGURE 5

Cartoon demonstrating the metabolic pathways involved in ammonia detoxification in brain slices from Glud1^lox/lox mice (A) and from CNS-Glud1^–/– mice (B). (A): In brain slices from Glud1^lox/lox mice glucose enters the TCA cycle via PDH and PC for de novo synthesis of glutamate (1), GDH is important for ammonia fixation and de novo synthesis of glutamate (2), glutamate is used as precursor for glutamine synthesis in astrocytes (3), and The concerted action of GDH and ALAT provides a mechanism for continuous fixation and disposal of ammonia as alanine is released from the brain (4). (B): In brain slices from CNS-Glud1^–/– mice, the lack of GDH impairs the possibility for de novo synthesis of glutamate (1), which results in less glucose entering the TCA via PC (2), and the direction of ALAT is reversed in order to sustain glutamate synthesis that can serve as precursor for glutamine synthesis (3). In addition, neurotransmitter glutamate cleared from the synapse by astrocytes is used for glutamine synthesis (4).