What to Expect (from the NanoBRET® Assay) When You're Expecting (a Protein:Protein Interaction)

The short answer is, you don’t—at least not initially. NanoBRET requires that one of your proteins gets tagged with NanoLuc® luciferase while the other protein is tagged with HaloTag® protein. But which protein should receive which tag? And should the tag be placed on the N-terminus or the C-terminus of each protein? If structural information is known, it may seem logical to place a tag at a specific terminus. But these are questions that must be answered empirically. Adding a tag to a protein can affect its ability to fold properly as well as its expression level. It’s also possible that the tag itself could alter or prevent the very interaction you are trying to detect. Therefore, it is recommended to construct a series of eight NanoBRET® expression vectors incorporating all possible combinations of protein, tag, and terminus.

Full-length Nanos is composed of three distinct regions: N-terminal (N), Zinc Finger (Z) and C-terminal (C). The ZC regions have been shown to be crucial for repression, while the N region only makes a minor contribution. Due to the difficulty in expressing full-length Nanos and Pum, electrophoretic mobility shift assays demonstrating complex formation between Pum and Nanos has been performed with only Nanos ZC and the RNA-binding domain of Pum. So, we chose to use those protein domains in developing our assay.

In our example, we created eight clones of the Drosophila Pum RNA Binding Domain (Pum-RBD) and the Nanos zinc-finger and C-terminal domains (Nanos-ZC) by appending NanoLuc® luciferase (NL or NLuc) donor and HaloTag® (HT) acceptor tags to Pum-RBD and Nanos-ZC using the NanoLuc® NanoBRET® PPI Flexi® Starter System. Pum-RBD and Nanos-ZC were first cloned into the pFN21A (N-terminal HaloTag®) vector and then additional clones with the proper tags at each orientation were transferred from the pFN21A vectors using Flexi® cloning. Before proceeding to the assay, each clone was sequenced to ensure that the correct DNA sequence was present in-frame with no mutations introduced. Clones were also created for full-length Nanos. Results with the Nanos-ZC were found to be similar to full-length Nanos, so we show only data with Nanos-ZC here. 

It’s easy—simply pair up the eight possible tagged combinations of your proteins of interest and use NanoBRET to determine which of the eight pairs resulted in the best energy transfer. In our example, the eight combinations of Pum-RBD and Nanos-ZC are shown below.

1. A transfection mixture for each vector combination consisting of 1µg of the HaloTag® vector and 0.1µg of the NanoLuc® vector (NanoLuc:HaloTag ratio at 1:10) was added to attached HEK293 cells in a 6-well plate, and proteins were expressed for approximately 20 hours at 37°C and 5% CO2.
2. Transfected HEK293 cells were collected and divided into two pools. For experimental samples, 1μl of 0.1mM HaloTag® NanoBRET® 618 Ligand was added per milliliter of cells (100nM final concentration). For no-acceptor controls, 1μl of DMSO was added per milliliter of cells (0.1% DMSO final concentration).
3. One hundred microliters of each pool of the cells was re-plated into multiple wells of a 96-well plate and incubated at 37°C and 5% CO2 overnight for ~20 hours.
4. Twenty-five microliters of diluted NanoBRET® Nano-Glo® Substrate was added to cells and donor emission (460nm) and acceptor emission (618nm) was measured using the GloMax® Discover plate reader. 

To calculate NanoBRET® ratios, the acceptor emission value for each pair was divided by the donor emission value, and these raw values were multiplied by 1,000 to convert to milliBRET units (mBU). The mean corrected mBU was then calculated by subtracting the mean mBU of the no ligand control from the mean mBU of the experimental sample. In order to determine assay consistency, a statistic known as a
Z´ factor was calculated comparing the mean and standard deviation values of the experimental samples and the no ligand control. The closer the calculated Z´ factor is to 1, the higher the quality of the assay, with an assay value between 0.5–1 considered to be in the acceptable range for signifying low variability. 

The NanoBRET® assay can be further optimized using the best PPI combination for optimal donor and acceptor DNA ratio to minimize unbound donor and maximize dynamic range by titration of the NLuc donor. In other words, we want to increase the sensitivity of the assay. For our Pum and Nanos example, the NanoLuc® vector (C-terminal tagged Nanos-ZC) was diluted 1:2, 1:5, 1:10, 1:100, and 1:1000. Transfection mixtures consisting of 1μg of HaloTag® vector (C-terminal tagged Pum-RBD) and titrated NanoLuc® vector were created. As the amount of NLuc vector was decreased, the NanoBRET® ratios increased (Figure 4). Raw donor and acceptor values also decreased with decreasing amounts of Nluc vector as expected. The highest NanoBRET® ratio was seen with the 1:1000 dilution, however, assay variability was high. Thus, the 1:100 NLuc to HT DNA ratio was chosen for further testing as it showed a high NanoBRET® ratio with high assay consistency based on Z´ factor calculations.

1. Weidmann et al. (2016). Drosophila Nanos acts as a molecular clamp that modulates the RNA-binding and repression activities of Pumilio. eLife 5:e17096.