Quantification of Protein Interaction Network Dynamics using Multiplexed Co-Immunoprecipitation

Quantitative multiplexed co-immunoprecipitation, or QMI, enables users to simultaneously measure hundreds of binary interactions in a predefined protein interaction network. By performing multiple co-IPs simultaneously in the same lysate, bioinformatics techniques can then examine how groups of co-associations change in a coordinated manner. The assay is time consuming to develop, but once the QMI panel is in hand network scale data can be collected about signal transduction systems in a relatively simply overnight assay.

QMI requires the accurate pipetting of different samples and reagents into 96 well plates. Careful notes should be taken when setting up and running each assay to prevent errors. For sample preparation and immunoprecipitation, begin by lysing the sample in a suitable detergent with protease and phosphatase inhibitors for a 15 minute incubation on ice.

At the end of the incubation, spin down the sample to remove the membranes and debris and perform a BCA assay on the supernatant to determine the protein concentration according to standard protocols. After normalizing the protein concentrations, prepare a magnetic bead master mix that contains approximately 250 magnetic beads of each class per well. Use magnetic separation to wash the magnetic bead mix two times in Fly P buffer before resuspending the beads in 20 microliters of Fly P buffer per sample.

After a thorough pipetting, aliquot 20 microliters of beads into one ice cold microcentrifuge tube per sample and add equal volumes of lysate with normalized protein concentrations to each tube for immunoprecipitation. Then divide each lysate sample into one tube for each plate being run and immunoprecipitate the samples on a rotator at four degrees Celsius overnight protected from light. The next morning use a magnetic bead rack to remove the lysate from the first set of tubes and wash the beads two times in 500 microliters of ice cold Fly P buffer per wash.

After calculating the resuspension volume, resuspend the immunoprecipitates in the calculated volume of ice cold Fly P buffer. Thoroughly resuspend the magnetic beads by gentle pipetting and distribute 25 microliters of beads per well across a flat bottomed 96 well plate on ice. In a different 96 well plate, dilute by attinilated probe antibodies to a two times working concentration.

Use a multichannel pipette to distribute 25 microliters of each probe antibody dilution into the appropriate wells of the magnetic bead containing assay plate and shake at high speed on a horizontal plate shaker to mix and resuspend the magnetic beads. After mixing, reduce the speed for a one hour incubation at four degrees Celsius protected from light. At the end of the incubation, wash the plate three times with fresh Fly P buffer per wash on a magnetic plate washer at four degrees Celsius.

After the last wash, resuspend the beads in 50 microliters of streptavidin PE diluted one to 200 in Fly P buffer. Shake the mix and resuspend the beads before incubating the plate at four degrees Celsius for 30 minutes protected from light. At the end of the incubation, wash the plate three times with Fly P buffer on a magnetic plate washer at four degrees Celsius and resuspend the beads in 125 microliters of Fly P buffer.

Shake the plate for one minute at 900 rotations to thoroughly resuspend the beads and load the plate into a refridgerated flow cytometer. In the flow cytometer software, select High RP1 Target setting, a stop condition of 1, 000 beads per region, and a sample volume of 80 microliters. Pause the run halfway through and resuspend the beads to prevent settling.

At the end of the run, export the data files in the XML format. To perform ANC analysis, fill in the ANC input file to reflect the details of the experimental design and run the program, which will write a CSV file into the active directory. The file ending in hits.

csv will report the protein co-associations that are significantly different at a false positive level of 0.05 between the control and experimental conditions in at least three out of four experimental replicates. For weighted correlation network analysis, paste the column titles of the data file output by Matlab ending in mfi. csv into the first column of a new spreadsheet and add the columns Experiment for experiment number and Treatment for experimental treatment and any other variables to be analyzed.

Save this file as traits. csv and open R Studio. Set the working directory to a folder containing the mfi.

csv and traits. csv files, highlight sections of code, and hit Command Enter to run the R commands as indicated in the commented Command file. The weighted correlation network analysis modules significantly correlated with each experimental trait will be output as a graphic file and the correlation of each interaction with each module will be output as a CSV file.

For each interaction in the ANC output file, identify if that interaction is also significant by CNA and create a new column that indicates if each ANC hit is also a CNA hit. Convert the values to log two fold change and calculate the average value for all ANC union CNA hits. Finally, make a new spreadsheet with each ANC union CNA hit listed by its immunoprecipitation target in one column, its probe target in the second column, and its fold change value in the third column.

Import this file into Cytoscape as a network to create a node edge diagram. High confidence protein interactions identified by the two independent statistical approaches are visualized as node edge diagrams using the open source software, Cytoscape, or as heatmaps using the R code and analysis instructions included in the supplementary material. Throughout the protocol, users need to take care to pipette accurately, to keep meticulous records, and to keep the samples cold at all times.

We have used QMI to understand how signal transduction pathways differentiate between different input types and integrate information from multiple receptors.