To isolate pharmacologically safe compounds that can induce apoptosis of tumor cells, leaves of an aromatic plant ( Zanthoxylum schinifolium ), which are widely used as a food flavor and herbal medicine in Korea and Japan, were sequentially extracted by organic solvents. An apoptogenic ingredient in the methylene chloride extract was further purified by silica gel column chromatography and identified as auraptene (AUR). The IC ₅₀ value of AUR against Jurkat T cells was 16.5 μg/ml. After the treatment of Jurkat T cells with AUR, the endoplasmic reticulum (ER) stress-mediated activation of caspase-12 and -8 and subsequent apoptotic events including c-Jun N-terminal kinase (JNK) activation, cleavage of FLICE inhibitory protein and Bid, mitochondrial cytochrome c release, activation of caspase-9 and -3, degradation of poly (ADP-ribose) polymerase and apoptotic DNA fragmentation were induced in a dose-dependent manner. The cytotoxicity of AUR was not blocked by the anti-Fas neutralizing antibody ZB-4. The AUR-induced cytotoxicity and apoptotic events were abrogated by ectopic over-expression of Bcl-xL or addition of the pan-caspase inhibitor z-VAD-fmk. The individual or simultaneous addition of the m-calpain inhibitor (E64d), JNK inhibitor (SP600125) and mitochondrial permeability transition pore inhibitor (CsA) failed to prevent apoptotic events including caspase-8 activation and Bid cleavage, unless the caspase-8 inhibitor (z-IETD-fmk) was combined, whereas AUR-induced caspase-12 activation was sustained even in the concomitant presence of z-IETD-fmk. These results demonstrated that the apoptotic effect of AUR on Jurkat T cells was exerted by the ER stress-mediated activation of caspase-8, and the subsequent induction of mitochondria-dependent or -independent activation of caspase cascade, which could be suppressed by Bcl-xL.

In the present study, we first report two forms of human phosphoserine aminotransferase (PSAT) cDNA (HsPSATα and HsPSATβ). HsPSATα has a predicted open reading frame comprising 324 amino acids, encoding a 35.2 kDa protein (PSATα), whereas HsPSATβ consists of an open reading frame comprising 370 amino acids that encodes a 40 kDa protein (PSATβ). PSATα is identical with PSATβ, except that it lacks 46 amino acids between Val290 and Ser337 of PSATβ, which is encoded by the entire exon 8 (138 bp). Both PSATα and PSATβ can functionally rescue the deletion mutation of the Saccharomyces cerevisiae counterpart. Reverse transcriptase–PCR analysis revealed that the expression of PSATβ mRNA was more dominant when compared with PSATα mRNA in all human cell lines tested. PSATβ was easily detected in proportion to the level of mRNA; however, PSATα was detected only in K562 and HepG2 cells as a very faint band. The relative enzyme activity of glutathione S-transferase (GST)–PSATβ expressed in Escherichia coli appeared to be 6.8 times higher than that of GST–PSATα. PSAT mRNA was expressed at high levels (approx. 2.2 kb) in the brain, liver, kidney and pancreas, and very weakly expressed in the thymus, prostate, testis and colon. In U937 cells, the levels of PSAT mRNA and protein appeared to be up-regulated to support proliferation. Accumulation of PSAT mRNA reached a maximum in the S-phase of Jurkat T-cells. These results demonstrate that although two isoforms of human PSAT can be produced by alternative splicing, PSATβ rather than PSATα is the physiologically functional enzyme required for the phosphorylated pathway, and indicate that the human PSAT gene is regulated depending on tissue specificity as well as cellular proliferation status with a maximum level expression in the S-phase.

The cytoprotective mechanism of l-serine against oxidative stress-mediated neuronal apoptosis was investigated in mouse hippocampal neuronal HT22 cells. Treatment with the reactive oxygen species (ROS) inducer 2,3-dimethoxy-1,4-naphthoquinone (DMNQ) increased cytosolic and mitochondrial ROS and apoptosis, without necrosis, in HT22 cells. ROS-mediated apoptosis was accompanied by the induction of the endoplasmic reticulum (ER) stress-mediated apoptotic pathway, involving CHOP/GADD153 upregulation, JNK and p38 MAPK activation, and caspase-12 and caspase-8 activation, and subsequent induction of the mitochondrial apoptotic pathway through BAK and BAX activation, mitochondrial membrane potential (Δψm) loss, caspase-9 and caspase-3 activation, PARP cleavage, and nucleosomal DNA fragmentation. However, the DMNQ-caused ROS elevation and ER stress- and mitochondrial damage-induced apoptotic events were dose-dependently suppressed by co-treatment with l-serine (7.5-20 mM). Although DMNQ reduced both the intracellular glutathione (GSH) level and the ratios of reduced GSH to oxidized GSH (GSSG), the reduction was restored by co-treatment with l-serine. Co-treatment with GSH or N-acetylcysteine also blocked DMNQ-caused ROS elevation and apoptosis; however, co-treatment with the GSH synthesis inhibitor buthionine sulfoximine significantly promoted ROS-mediated apoptosis and counteracted the protection by l-serine. In HT22 cells, DMNQ treatment appeared to tilt the mitochondrial fusion-fission balance toward fission by down-regulating the levels of profusion proteins (MFN1/2 and OPA1) and inhibitory phosphorylation of profission protein DRP1 at Ser-637, resulting in mitochondrial fragmentation. These DMNQ-caused alterations were prevented by l-serine. A comparison of mitochondrial energetic function between DMNQ- and DMNQ/l-serine-treated HT22 cells showed that the DMNQ-caused impairment of the mitochondrial energy generation capacity was restored by l-serine. These results demonstrate that l-serine can protect neuronal cells against oxidative stress-mediated apoptotic cell death by contributing to intracellular antioxidant GSH synthesis and maintaining the mitochondrial fusion-fission balance.

Oxidative stress and mitochondrial dysfunction play critical roles in neurodegenerative diseases. Glutathione (GSH), a key brain antioxidant, helps to neutralize reactive oxygen species (ROS) and maintain redox balance. We investigated the effectiveness of L-cysteine (L-Cys) in preventing apoptosis induced by the ROS generator 2,3-dimethoxy-1,4-naphthoquinone (DMNQ) in mouse hippocampal neuronal HT22 cells, as well as alleviating memory and cognitive impairments caused by the GSH synthesis inhibitor L-buthionine sulfoximine (BSO) in mice. DMNQ-induced apoptotic events in HT22 cells, including elevated cytosolic and mitochondrial ROS levels, DNA fragmentation, endoplasmic reticulum stress, and mitochondrial damage-mediated apoptotic pathways were dose-dependently abrogated by L-Cys (0.5-2 mM). The reduced intracellular GSH level, caused by DMNQ treatment, was restored by L-Cys cotreatment. Although L-Cys did not significantly restore GSH in the presence of BSO, it prevented DMNQ-induced ROS elevation, mitochondrial damage, and apoptosis. Furthermore, compared to N-acetylcysteine and GSH, L-Cys had higher 2,2-diphenyl-1-picrylhydrazyl and 2,2-azino-bis-3-ethylbenzothiazoline-6-sulphonic acid radical-scavenging activity. L-Cys also restored mitochondrial respiration capacity in DMNQ-treated HT22 cells by reversing mitochondrial fission-fusion dynamic balance. BSO administration (500 mg/kg/day) in mice led to neuronal deficits, including memory and cognitive impairments, which were effectively mitigated by oral L-Cys (15 or 30 mg/kg/day). L-Cys also reduced BSO-induced ROS levels in the mice hippocampus and cortex. These findings suggest that even though it does not contribute to intracellular GSH synthesis, exogenous L-Cys protects neuronal cells against oxidative stress-induced mitochondrial damage and apoptosis, by acting as a ROS scavenger, which is beneficial in ameliorating neurocognitive deficits caused by oxidative stress.