Serum Glutamate Dehydrogenase—Biomarker for Liver Cell Death or Mitochondrial Dysfunction?

We read, with great interest, a recent paper by Schomaker et al. (2013), who evaluated various enzymes, e.g., glutamate dehydrogenase (GLDH), malate dehydrogenase (MDH), purine nucleoside phosphorylase and paraxonase 1 as more specific alternatives to alanine aminotransferase (ALT) and aspartate aminotransferase for drug-induced liver injury. The results indicate that only GLDH and MDH are useful biomarkers of liver injury in humans. Of particular interest is the mitochondrial matrix enzyme GLDH. Because of its location within the cell, one would expect that mitochondrial damage would leak the enzyme into the cytosol from where it can be released into the plasma during cell necrosis. Based on this assumption, we have previously used GLDH not as a biomarker of cell death but as a mechanistic biomarker for mitochondrial dysfunction in acetaminophen (APAP) overdose patients (McGill et al., 2012a) and also in animals (McGill et al., 2012a,b). In a control experiment with mice, very high doses of furosemide caused extensive liver necrosis based on ALT release and histology but neither GLDH nor mitochondrial DNA (mtDNA) was appreciably increased in plasma (McGill et al., 2012a). In contrast, the well-documented mitochondrial damage after APAP (Jaeschke et al., 2012) caused substantial release of ALT, GLDH, and mtDNA in these animals (McGill et al., 2012a). Thus, based on these data, we concluded that the mechanism of APAP hepatotoxicity in humans involves mitochondrial dysfunction (McGill et al., 2012a). In addition, the onset of GLDH release is somewhat delayed compared with ALT release (McGill et al., 2012b), and we measured consistently lower GLDH activities in serum compared with ALT (McGill et al., 2012a,b). Hence, our data suggested that GLDH is more than a general cell death marker and may be a specific biomarker for mitochondrial dysfunction. There could be one caveat with measuring GLDH or mtDNA in serum or plasma of patients or animals. If cell contents are released from cells during necrosis, such as APAP-induced cell death, not only soluble cytosolic proteins but also intact mitochondria are released. If the blood is centrifuged at standard g-forces to remove blood cells, mitochondria remain in the serum and after freeze thawing the sample, mitochondrial GLDH, mtDNA, and MDH may be released and could then be measured together with enzymes liberated within the cell during mitochondrial damage. To avoid this confounding problem, we have centrifuged the freshly drawn blood at > 14,000 × g for 20min to pellet not only blood cells but also any intact mitochondria in the sample (McGill et al., 2012a). Thus, the measured GLDH activity and the mtDNA in our serum or plasma samples originated from damaged mitochondria during the pathophysiology of APAP-induced cell death. Schomaker et al. (2013), according to their description, only centrifuged the blood at 3000 × g for 10min, froze the serum, and thawed it later for enzyme analysis. Thus, there is the possibility that the authors measured both soluble GLDH and GLDH that was present in intact mitochondria released during necrosis. Although this methodological issue may lead to higher GLDH activities in serum of APAP overdose patients, it does not necessarily question the use of serum GLDH as biomarker for cell necrosis in APAP-induced liver injury because of the already well-established central role of mitochondrial damage in this pathophysiology in humans (McGill et al., 2012a) and in animals (Jaeschke et al., 2012). However, in samples from patients with liver injury where mitochondrial involvement is unknown, the interpretation of these data can be complicated. It certainly may be advisable to standardize sample collection and preparation in order to avoid comparing samples obtained under different conditions, which may include soluble enzyme with or without enzymes derived from intact mitochondria in serum.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.