RAF kinases function as key intermediates within cellular signal transduction and are associated with critical cellular processes such as proliferation, cell survival and differentiation. Heterodimerization of BRAF and CRAF leads to activation of the Ras-Raf-MAPK pathway and signal transduction. Due to their vital cellular roles and oncogenic properties, RAF kinases have been extensively studied within the context of cancer.
Specific BRAF mutations that result in constitutive activation of downstream MEK/ERK signaling have been linked to a variety of cancers, and BRAF inhibitors that affect dimerization have demonstrated clinical relevance for cancer therapies.
GDC-0879, an ATP-competitive RAF inhibitor, was paradoxically found to induce CRAF-BRAF dimerization and activate downstream signaling.
Cell-based RAF kinase dimerization assays serve as a useful tool for understanding the function and mechanisms of action of compounds such as GDC-0879 and could identify additional compounds that can modulate this interaction.
One approach for cell-based protein interaction studies is bioluminescence resonance energy transfer (BRET), which uses a bioluminescent donor and fluorescent acceptor to monitor protein:protein interactions (PPI).
Specifically, the NanoBRET™ PPI assay uses NanoLuc® luciferase as the BRET energy donor and HaloTag® protein labeled with the fluorescent HaloTag® NanoBRET™ 618 Ligand (Cat.# G9801) as the energy acceptor to measure protein interaction in live cells.
The bright, blue-shifted NanoLuc® donor signal and red-shifted HaloTag® acceptor create an optimal pairing with minimal spectral overlap, increased signal and lower background compared to conventional BRET assays.
When designing a new NanoBRET™ assay, optimizing experimental conditions is important to achieve the best assay performance.
Tag orientation and position can influence signal output and protein function. We recommend that you optimize the NanoBRET™ assay by testing all possible donor and acceptor combinations. Proteins of interest should be tagged with NanoLuc® donor or HaloTag® acceptor at either the amino (N) or carboxy (C) terminus, resulting in four variants and eight potential donor/acceptor combinations.
Additionally, the ratios of vectors that are transfected should also be optimized to maximize dynamic range.
Here we describe the development of a NanoBRET™ PPI assay for monitoring the interaction between CRAF and BRAF proteins. We systematically screen all possible donor/acceptor combinations in the presence and absence of GDC-0879. We also optimize the donor-to-acceptor vector ratios for transfection and validate the interaction using a HaloTag® pull-down assay. We test the interaction assay in two cell lines, human embryonic kidney (HEK293) and the human colon carcinoma line (HCT116), which has a mutation in the KRAS gene and is often used as a cancer model and for inhibitor studies.
Together, these steps serve as an outline for NanoBRET™ PPI assay development.
Optimizing Tag Placement
To determine ideal placement of the BRET donor and acceptor, eight possible plasmid combinations were tested for the CRAF-BRAF protein interaction in HCT116 cells and HEK293 cells. Tag placement and experimental setup followed the protocol in the NanoBRET™ Protein:Protein Interaction Technical Manual #TM439.
Briefly, C- and N-terminal fusions of NanoLuc® donor and HaloTag® acceptor were placed on both BRAF and CRAF. HCT116 and HEK293 cells were co-transfected using FuGENE® HD Transfection Reagent (Cat.# E2311) with vectors using a 1:10 (1µg:10µg) donor-to-acceptor ratio.
Cells were replated into 96-well plates either containing or lacking (control) HaloTag® NanoBRET™ 618 Ligand. To induce the CRAF-BRAF interaction, a subset of cells were treated with 10µM GDC-0879 for 2 hours. NanoBRET™ NanoGlo® Substrate (Cat.# N1571) was added, and donor and acceptor signals were measured on the GloMax® Discover instrument (Cat.# GM3000) for both treated and untreated cells. NanoBRET™ signal was normalized to wells lacking HaloTag® NanoBRET™ 618 Ligand for the best assay window. The fold change in signal was determined between treated and untreated cells to determine the ideal tag placement.
CRAF-HaloTag® and BRAF-NanoLuc® vectors were co-transfected into HCT116 and HEK293 cells with donor-to-acceptor ratios of 1:1; 1:10; 1:100 and 1:1,000. As above in Optimizing Tag Placement, both treated (10µM of GDC-0879 compound) and untreated samples were measured on the GloMax® Discover instrument. Fold change was calculated between treated and untreated samples for each vector ratio.
GDC-0879 Dose-Response Curve
To measure induction of CRAF-BRAF interaction, CRAF-HaloTag®and BRAF-NanoLuc® vectors were co-transfected into HCT116 and HEK293 cells and treated with GDC-0879 compound to generate a dose-response curve. Briefly, cells were co-transfected with donor-to-acceptor ratios of 1:1 or 1:10. Cells were treated for 2 hours with GDC-0879 compound concentrations ranging from 0.001µM to 10µM. NanoBRET™ measurement and signal normalization was done as described previously.
HaloTag™ Pull-Down Assay
To validate the CRAF-BRAF protein interaction, we used a modified HaloTag® Mammalian Pull-Down assay (Cat.# G6509) (Figure 1), followed by bioluminescent detection (Steffen and Méndez-Johnson 2019). Briefly, HCT116 cells were co-transfected with the chosen HaloTag® and NanoLuc® fusions at a donor-to-acceptor ratio of 1:1. Cells were treated with GDC-0879 at concentrations of 0, 1µM or 10µM. HaloTag® protein fusions in cell lysates were incubated with HaloLink™ Resin (Cat.# G1912) and subsequently cleaved using ProTEV Plus (Cat.# V6101). The signal for copurified NanoLuc® luciferase fusion proteins was detected using Nano-Glo® Luciferase Assay (Cat.# N1110).
Figure 1. Modified HaloTag® Mammalian Pull-down System workflow. Cells are co-transfected with HaloTag® and NanoLuc® fusion constructs and lysed. The HaloTag® fusion protein is immobilized on HaloLink™ Resin and eluted along with interacting proteins by TEV cleavage. Copurified NanoLuc® luciferase fusion proteins are quantified.
To determine the ideal placement of NanoLuc® donor and HaloTag® acceptor tags, multiple vector combinations were assessed for signal dynamic range in the presence or absence of inhibitor compound. A total of four constructs were generated with C- and N-terminal fusions of NanoLuc® and HaloTag® proteins on both BRAF and CRAF, yielding eight possible vector combinations. Both HCT116 and HEK293 cell lines were transfected and a subset of cells were treated with GDC-0879 inhibitor compound. The highest NanoBRET™ signals were obtained for combinations marked by an arrow and varied by cell type (Figure 2). Importantly, vector combinations yielding the highest overall signal did not correspond to combinations with the largest fold change between treated and untreated samples. Because a specific CRAF-BRAF interaction occurs only with compound treatment, the NanoLuc® and HaloTag® fusions with the lowest basal level, the C-terminal-tagged proteins CRAF-HaloTag (CRAF-HT) and BRAF-NanoLuc (BRAF-NL), provided the largest fold change upon compound treatment and the best assay window. For this reason, we used CRAF-HT and BRAF-NL fusions for all further experiments.
Figure 2. Tag placement optimization. NanoBRET™ ratio and fold change obtained with eight plasmid combinations. CRAF and BRAF tagged with N- or C-terminal NanoLuc® (NL) or HaloTag® (HT) proteins were co-transfected into HCT116 cells (Panel A) or HEK293 cells (Panel B) with 1:10 donor-to-acceptor DNA ratio and treated with or without 10µM of GDC-0879 compound (n = 4).
To minimize the amount of unbound donor and maximize dynamic range for the NanoBRET™ signal, we optimized the ratio of donor-to-acceptor vectors for the chosen combination during co-transfection. We used both HCT116 and HEK293 cell lines and tested donor-to-acceptor vector ratios of 1:1, 1:10, 1:100 and 1:1,000. Once again, a portion of cells were treated with GDC-0879 to induce the interaction. The highest fold change was obtained for the donor-to-acceptor ratio of 1:1 on both cell lines. Both the 1:1 and 1:10 ratios were further analyzed (Figure 3).
Figure 3. Vector ratio optimization. NanoBRET™ ratio and fold change were obtained with different donor-to-acceptor ratios using CRAF-HaloTag® and BRAF-NanoLuc® plasmids. HCT116 cells (Panel A) or HEK293 cells (Panel B) were co-transfected with 1:1, 1:10, 1:100 or 1:1,000 donor-to-acceptor DNA ratios and treated with or without 10µM GDC-0879 (n = 4).
To investigate the effect of GDC-0879 dose on CRAF and BRAF heterodimerization, various concentrations of compound were measured in NanoBRET™ experiments. HCT116 and HEK293 cell lines with donor-to-acceptor ratios of 1:1 and 1:10 were analyzed. Figure 4 shows the dose-response curve for both vector ratios. Confirming previous experiments, a greater dynamic range was obtained with the 1:1 vector ratios, and a similar response was observed with both cell lines. These data demonstrate the need for both tag and vector ratio optimization for the most effective NanoBRET™ assay. Furthermore, once optimized, the NanoBRET™ PPI Assay can be used as a highly quantitative measure of protein interaction that responds as expected to compound modulators.
Figure 4. GDC-0879 dose-response curve. Fold change in NanoBRET™ CRAF-BRAF interaction signal obtained at various concentrations of GDC-0879 compound. HEK293 cells (Panel A) or HCT116 cells (Panel B) were co-transfected with 1:1 or 1:10 donor-to-acceptor DNA ratios and treated with GDC-0879 at the indicated concentrations (n = 4).
Finally, to confirm the CRAF-BRAF interaction we took advantage of another use for the HaloTag® and NanoLuc® proteins and used a modified HaloTag® Mammalian Pull-Down assay followed by bioluminescent detection of the NanoLuc® fusion. HCT116 cells were treated with various concentrations of inhibitor compound to stimulate dimerization. After protein pull-down and cleavage, the Nano-Glo® Luciferase Assay was used to detect any BRAF-NL that copurified with CRAF-HT. Luciferase signal was compared to the HaloTag® control vector. In the absence of GDC-0879, we observed approximately a tenfold enrichment of BRAF compared to the HaloTag® vector controls (Figure 5). We observed a significant increase in the fold enrichment of BRAF following treatment with increasing concentrations of GDC-0879. These data confirm that the energy transfer being measured in the NanoBRET™ assay was due to a physical interaction between BRAF-NL and CRAF-HT fusion proteins.
Figure 5. HaloTag® pulldown of BRAF-NL. Fold enrichment of NanoLuc® luciferase activity in HCT116 cells co-expressing CRAF-HT with BRAF-NL in the presence of 0, 1µM or 10µM GDC-0879 compound (n = 3).
Here we present an outline for NanoBRET™ protein:protein interaction assay development using the highly studied CRAF-BRAF interaction. We demonstrate the importance of tag placement and vector co-transfection ratio optimization in the presence of an inducer or inhibitor. Notably, the same CRAF-HT and BRAF-NL fusions displayed the highest fold change in both HTC116 and HEK293 cells. Further, we show stimulation of heterodimerization in a GDC-0879 dose-dependent manner. An additional unique aspect of the NanoBRET™ technology is that the HaloTag® and NanoLuc® fusions have a broader functionality and can be used to perform a modified HaloTag® pull-down assay followed by NanoLuc® quantitation to validate the CRAF-BRAF interaction and demonstrate that the results observed with energy transfer indicate physical interaction between the two proteins.