Identification and Characterisation of a Novel Antioxidant Activity for the BCAT1 Cxxc Motif: Implications for Myeloid Leukaemia Development

Recently the human cytosolic branched-chain amino transferase (BCAT1) has been implicated in the development of myeloid leukaemia. Previously studies identified a redox active CXXC motif for BCAT1 analogous to the mitochondrial isoform (BCAT2). Oxidation of this CXXC motif can regulate aminotransferase activity in both isoforms. Interestingly the reduction mid-point potential (Em) of the BCAT1 CXXC motif is around 80mV lower compared with BCAT2. This suggests that the BCAT1 CXXC motif is more reducing in nature i.e. has antioxidant capacity. Further evidence to support this notion comes from the observation that BCAT1 is significantly less sensitive to inactivation mediated by hydrogen peroxide (H2O2). This is an important observation, since H2O2 is a reactive oxygen species (ROS) implicated in myeloid leukaemia development. Here we provide the first evidence to show that the BCAT1 CXXC can metabolise H2O2. By using CXXC motif mutant constructs, we also evaluate the role of the BCAT1 CXXC motif in myeloid leukaemia cells.

To investigate the antioxidant capacity of the BCAT1 CXXC motif, cDNA (accession NM_001178094.1) was cloned into a pET28a vector for overexpression and purification in E.coli BL21(DE3) cells. Site directed mutagenesis of the CXXC motif subsequently followed creating three constructs: 1) BCAT1-WT (unmutated), 2) BCAT1-CXXS (C-terminal Cys → Ser mutant) and 3) BCAT1-SXXS (N/C-terminal Cys → Ser double mutant). Mutation of the BCAT1 CXXC motif was confirmed at the protein level by 5,5'-ditho-bis-(2-nitrobenzoic acid) titration. WT and CXXC motif mutant BCAT1 protein was added to 5mM H2O2 & monitored at λ240nm for the disappearance of H2O2. This was compared with 1U of catalase as a positive control. Our data show that 15μg of purified BCAT1-WT could metabolise H2O2 at a rate of 0.14±0.02 μmol/min. The BCAT1 CXXC motif mutants lacked capacity to do this. To verify our findings, BCAT1-WT treated with N-ethylmaleimide also lost the capacity to metabolise H2O2. This confirms that the antioxidant activity observed is Cys mediated. This novel finding for the BCAT1 CXXC motif may therefore be important in myeloid leukaemia development.

To evaluate the BCAT1 CXXC motif in myeloid leukaemia cells, WT and CXXS mutant BCAT1 was subcloned into a pLENTI-C-Myc-DDK lentiviral vector prior to transduction of U937 cells. Empty vector (EV) control transgenic U937 cells were also generated. Stable expression of BCAT1 was achieved and confirmed by western-blot and qPCR; BCAT1-WT (6.5±1.7 fold increase) & BCAT1-CXXS (5.1±2.6 fold increase). The BCAT1 and EV transgenic U937 cells were subjected to H2O2 mediated oxidative stress and monitored for cell viability after 24h. A significant difference in the LD50 between BCAT1-WT, BCAT1-CXXS and EV control cells was observed; 620±28mM, 306±37mM, 251±26mM respectively (p<0.05). This finding suggests that the BCAT1 CXXC provides protection against H2O2 mediated cell death. To corroborate these findings, the same cells treated with rotenone observed a similar pattern in LD50; 1.05mM, 0.06mM, 0.12mM respectively.

ROS are implicated in many cellular processes, including differentiation. Thus, we next asked whether the BCAT1 CXXC motif could suppress U937 cell differentiation mediated by phorbol 12-myristate 13-acetate (PMA). This compound was selected since ROS feature in PMA mediated differentiation. U937 cells were monitored for the expression of CD11b and CD36 following 48h PMA incubation. Our data show a significant reduction in the frequency of CD11b+/CD36+ U937 cells for BCAT1-WT (17.5±6.9%) compared with EV controls (54.0±10.5%, p<0.05). BCAT1-CXXS also showed a reduction (30.6±4.9%), but not to the same extent as BCAT1-WT. This finding was mirrored by the expression level of CD11b, where the MFI was significantly less for BCAT1-WT (88.0±16.3) compared with EV control (251.5±54.7, p<0.001) and BCAT1-CXXS (152.5±6.3, p<0.05). Analysis of intracellular ROS using 2′,7′-Dichlorodihydrofluorescein diacetate revealed a reduction in cellular ROS for BCAT1-WT cells (5929±13.8 MFI), compared with EV controls (8527±196 MFI) and BCAT1-CXXS (7879±18.4 MFI). This shows that the BCAT1 CXXC motif may control cellular ROS levels in myeloid leukaemia cells.

In summary, this study identifies a novel antioxidant role for the BCAT1 CXXC which confers protection against ROS mediated cell death and differentiation of myeloid leukaemia cells.