Looking to perform a non-radioactive EMSA to study protein binding to ssDNA. Before doing so, I’m looking to optimize detection.

The two detection series are:

1) biotinylated oligo, strep-HRP incubation (1:5000 O/N 4C in the dark), 1’ rabbit anti-strep pAb (1:3000 O/N 4C), 2’ goat anti-rabbit-HRP (1:3000 RT 2hra in the dark) —-> ECL (Pierce Super Signal Pico Plus ECL) —> x-ray film development

OR

2) biotinylated oligo, streptavidin-AP (1:3000 O/N 4C) > NBT/BCIP (37C at pH 9 until development) —> image

To start, our lab purchased a Mirus Label-IT biotin labeling kit that covalently attaches a biotin molecule every 20-60np of a nucleic acid. I am attempting to label a couple of DNA oligos to start that range from 50ish-60mer length. I am testing the efficiency of labeling after purifying the biotinylated oligo by Zymo‘s clean and concentrator kit (G-50 style spin column).

Here’s my general protocol for the dot blot to test labeling efficiency.

Dilute biotinylated oligo stock to 500pg/uL. Perform 2-fold serial dilutions in DEPC water.

Spot 1uL of each dilution, starting with ~500pg, on a 0.2um positively charged Hybond nylon membrane (not treated). Air-dry for 10min then stick membranes in an 80C incubator for 1hr to ”fix” the dots (probably not necessary given the type of membrane?)

Blocking the membrane:

Tried different formulations, mostly TBS-T (0.05% Tween-20) based since AP activity may be inhibited by PBS. I cannot use milk or casein since they naturally contain biotin.

Types of protein blockers used HRP/AP and antibody dilutions :

1) 3% or 5% BSA - a lot of background with either 1hr RT or O/N 4C incubation

2) 1% PVP (non-protein based blocker)

3) TBS-T (0.05%) only

in all cases, for HRP detection, the background is way too strong, though PVP provides the least amount of background. The whole membrane lights up and even a 1s exposure to x-ray film yields an all-black image. I can let the ECL reagent dwindle and can capture some signal but there is way too much background even after 10min.

Now, I get a lot less background when using AP detection, but this can only reveal blots at ~300pg with our in-house labeling. I can achieve detection down to fg level when using a commercially labeled oligo.

Basically, I’m wondering if there is a better way to detect biotinylated oligos : protein interactions with a non-radioactive approach . Again, this is just optimizing the detection. I will have to optimize the DNA:protein binding reaction, gel electrophoresis and electro transfer steps as well, all which can take a 1+ day to get through before detection. of course, I think the labeling kit isn’t as efficient as other random-priming methods, but this is what I’m working with. I’ve scoured through countless papers but maybe I’m missing something. If anyone has a better approach to this, I’d love to discuss!

- Mike