Novel 2,5-hexanedione analogues. Substituent-induced control of the protein cross-linking potential and oxidation susceptibility of the resulting primary amine-derived pyrroles

Comparative covalent protein binding of 2,5-hexanedione and 3-acetyl-2,5-hexanedione in the rat

2,5-Hexanedione (HD) is the metabolite implicated in n-hexane neurotoxicity. This gamma-diketone reacts with protein lysine amines to form 2,5-dimethylpyrrole adducts. Pyrrole adduction of neurofilaments (NF) and/or other axonal proteins was proposed as a critical step in the neuropathy. While pyrrole adduction is widely accepted as necessary, subsequent pyrrole oxidation, which may result in protein cross-linking, was alternatively postulated as the critical mechanistic step. Previous studies have indicated that 3-acetyl-2,5-HD (AcHD), an analogue that forms pyrroles that do not oxidize, was not neurotoxic in rats. However, relative levels of pyrrole adduction of NF or other axonal proteins were not reported. In the present study, groups of 6 male Wistar rats were given saline, [1,6-(14)C]-HD (3 mmol/kg/d), or [5-(14)C]-AcHD (0.1 mmol/kg/d), i.p. for 21 d. HD- and AcHD-treated rats lost 10% and gained 14% body weight, respectively, compared to a 22% gain for control rats. At termination, HD- and AcHD-treated rats exhibited mean scores of 3.5 and 1.4, respectively, for hindlimb weakness (0-5 scale). Incorporation of radiolabel from HD was 27.8 +/- 3.9, 13.9 +/- 2.6, and 7.8 +/- 0.6 nmol/mg in plasma protein, purified globin, and axonal cytoskeletal proteins, respectively, compared to 0.6 +/- 0.1, 1.6 +/- 0.5, and 1.0 +/- 0.1 for AcHD. Binding of HD to the NF-L, -M, and -H subunit proteins from treated animals was 4-, 24-, and 13-fold higher, respectively, that that of AcHD, indicating differing stoichiometry and patterns of NF adduction for the two diketones. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) analysis of globin and NF proteins did not demonstrate protein cross-linking for either diketone at the dose levels and time period examined. These results indicate that that the lack of neurotoxicity previously reported for AcHD may reflect differences in adduct levels at critical axonal target sites rather than an inability to form cross-linking adducts. Based on these data, further studies are required to fully assess the neurotoxic potency of AcHD and other non-cross-linking analogues as compared to HD.

2,5-Hexanedione-induced changes in the neurofilament subunit pools of rat peripheral nerve

Axon atrophy is the principle morphological feature of the peripheral neuropathy induced by 2,5-hexanedione (HD). Axon caliber is determined by a stationary neurofilamentous cytoskeleton that is maintained through dynamic interactions with mobile neurofilament (NF) subunits. To determine the effects of HD on the stationary and mobile NF pools, groups of rats were exposed to HD at dosing schedules (175 mg/kg x 101 days or 400 mg/kg x 26 days) that produced moderate levels of neurological deficits and, as assessed by previous studies, prevalent axon atrophy in peripheral nerve. Sciatic and tibial nerves from HD-intoxicated rats and their age-matched controls were triton-extracted and separated by differential centrifugation into a high-speed pellet (P1) of NF polymer and a corresponding supernatant fraction (S1), which presumably contained mobile monomer. Cytoskeletal proteins (NF-L, NF-M, NF-H and beta-tubulin) in each fraction were determined by immunoblot analysis. Results show that regardless of HD dose-rate, triton-soluble NF subunits in the supernatant fractions were significantly reduced, whereas triton-insoluble proteins in the corresponding pellets were inconsistently affected. Beta-tubulin also exhibited inconsistent fractional changes, while abnormal higher molecular weight NF proteins were detected primarily in the triton-insoluble fraction. Studies with antibodies directed against phosphorylated (RT97) and non-phosphorylated (SMI32) epitopes on NF-H did not reveal major changes in subunit phosphorylation. These results suggest that HD intoxication is primarily associated with depletion of soluble NF proteins, which could produce axon atrophy through disruption of cytoskeletal turnover and maintenance.

Phosphatidylethanolamine modification by oxidative stress product 4,5(E)-epoxy-2(E)-heptenal

Phosphatidylethanolamine and ethanolamine were incubated with 4,5(E)-epoxy-2(E)-heptenal to analyze the changes produced in amino phospholipids as a consequence of their reaction with this lipid oxidation product. The reactions, which were followed by GC/MS or HPLC/MS, produced in a first step the corresponding imines, which later evolved into two main products. These products, which were isolated and characterized by (1)H and (13)C NMR and mass spectrometry, resulted in the corresponding phosphatidylethanolpyrroles (PEPs) and phosphatidylethanol-2-(1-hydroxypropyl)pyrroles (PEHPs). PEPs were relatively stable, but PEHPs polymerized spontaneously. This polymerization, which was followed in 4,5(E)-epoxy-2(E)-heptenal/ethanolamine incubations with the detection of the corresponding dimers and trimers, was related to the brown color and fluorescence development produced. In addition, when hydroxyalkylpyrroles were incubated in the presence of single pyrroles, mixed polymers of the two pyrroles were obtained, and these polymers were produced in a higher extent than those derived exclusively from PEHP. These results confirm that lipid oxidation products are able to react with amino phospholipids analogously to protein amino groups, and therefore, both amino phospholipids and proteins may compete for lipid oxidation products. Although these results do not demonstrate the formation of pyrrolyzed phospholipids in vivo, they suggest that analogously to pyrrolyzed proteins, these modified phospholipids are likely to be produced during the phospholipid peroxidation process. In addition, these results also provide new bases for assessing the role of phospholipid modification in certain diseases, pathophysiological conditions, and aging.

1,2-diacetylbenzene, the neurotoxic metabolite of a chromogenic aromatic solvent, induces proximal axonopathy

Several widely used aromatic hydrocarbon solvents reportedly induce blue-green discoloration of tissues and urine in animals and humans. The chomophore has been proposed to result from a ninhydrin-like reaction with amino groups in proteins. The present study examines the neurotoxic property of 1,2-diacetylbenzene (1,2-DAB), the active metabolite of the chromogenic and neurotoxic aromatic solvent 1,2-diethylbenzene. Rats treated with 1,2-DAB, but not with the nonchromogenic isomer 1,3-DAB or with ninhydrin developed blue discoloration of internal organs, including the brain and spinal cord. Only 1,2-DAB induced limb weakness associated with nerve fiber changes, which were most prominent in spinal cord and spinal roots. Changes began with the formation of proximal, neurofilament-filled axonal swellings of the type seen after treatment with 3,4-dimethyl-2,5-hexanedione, a potent derivative of the active metabolite of the neurotoxic aliphatic hydrocarbon solvents n-hexane and methyl n-butyl ketone. These compounds are metabolized to a gamma-diketone that forms pyrroles with target proteins, such as neurofilament proteins. A comparable mechanism is considered for 1,2-DAB, an aromatic gamma-diketone.