CHARACTERIZATION OF BINDING AND ACTIVATION MECHANISMS OF BAK

Abstract

Background Cell death is required for the development and function of all multicellular organisms by eliminating damaged cells, thereby maintaining physiological homeostasis. Dysregulation of cell death contributes to the development of diseases such as cancer, autoimmune and neurodegenerative diseases. The onset of cell death involves the activation of BCL-2 family executioner proteins Bax and Bak by the BH3-only activators Bid, Bim or Puma. Executioner proteins Bak and Bax are both involved in the pore formation process on the mitochondrial membrane, however, recent evidence suggests that Bax and Bak are differentially activated. Understanding such mechanistic differences opens the possibilities for pharmaceutical interventions via Bax/Bak-mediated apoptosis due to their differential expression profiles across various diseased/healthy tissues. Methods FLIM (Fluorescence Lifetime Imaging Microscopy)-FRET (Förster resonance energy transfer) in live cells was done in BMK DKO and HCT116 DKO cell lines expressing the donor mCerulean3 or mTurquoise2 N-terminally fused to the protein-of-interest in cell lines stably expressing donor mCerulean3 (mC3)-fused to Bax or Bak without its carboxyl terminal, enabling survival of Bax/Bak expressing cells upon transient expression of the acceptor Venus-fused BH3-only activators. Transfected cells were imaged on the INO-FHS confocal microscope to obtain the fluorescence lifetime of mCerulean3.Fluorescent protein standards were used to convert intensity into concentration, and a Hill-slope fit of the binding curves was used to determine dissociation constants (KD) in live cells. The apparent KD values were obtained from the fitted Hill-equation (h=1) of the binding data after subtracting the fitted binding data of the negative collisional controls from each fitted construct's original binding data for each construct. If the final binding curves' BMAX y values (ΔѠ) were below the threshold of 0.04, the binding data were deemed insufficient (NSF). Unpaired student t-test with Welch's correction assuming Gaussian distribution with unequal standard deviation was used to determine statistical significance. Orthogonal in vitro binding assays using recombinant proteins were carried out to validate our FLIM-FRET findings. Activation of Bak by BH3-only proteins was tested using terbium-release liposome permeabilization assay in vitro and dox-inducible mC3-Bak cell death assay in live cells. Conclusion Here, we report that Bid is more efficient than Bim at activating Bak, with the BH3 region of Bid enabling high affinity binding to Bak. Further, effective Bak activation by Bim or Bid requires both binding of the BH3-motifs to the BH3-binding pocket of Bak and additional interactions between the activator proteins and the Bak-CTS. For Bim, the additional interaction with the Bak-CTS necessary for robust Bak activation is with the CTS of Bim.