These methods can be used to characterize neuroinflammatory and hemodynamic responses to brain injury as part of a multivariate system analysis using partial least squares regression. These techniques are used to give more of a holistic view of the brain's response to injury. To induce an injury, confirm a lack of response to toe pinch in an anesthetized mouse and place mouse in the prone position on the center of a thin membrane.
Using both hands to hold the tissue taught, secure the mouse's tail under a thumb and position to mouse head under the guide tube. When the mouse is in position, aiming for the impact between the back of the eyes and the front of the ears, drop the bolt from the top of the guide tube onto the dorsal aspect of the mouse's head. Upon impact, the mouse will break through the tissue, allowing rapid acceleration of the head about the neck.
After the injury, allow the mouse to recover in the supine position on a 37-degree Celsius warming pad. After a two-minute period of stabilization, gently rest the DCS sensor over the right hemisphere of the mouse's skull, such that the top edge of the optical sensor lines up with the back of the eye and the side of the sensor lines up along the midline. Cup a hand over the sensor to shield it from room light and acquire five seconds of data.
Then, reposition the sensor over the left hemisphere and acquire five seconds of left hemisphere data. To assess multiplexed cytokine and phospho-protein production, add 150 microliters of mixed lysis buffer per approximately three milligrams of harvested animal brain tissue. And use a 1, 000-microliter pipette tip to mechanically triturate the tissue 15 to 20 times.
Place the homogenized samples on a rotator for 30 minutes at four degrees Celsius before collecting the tissue debris by centrifugation. Transfer the supernatants into new tubes and centrifuge the samples to remove any remaining precipitate. Transfer the supernatants to new tubes and determine the protein concentration by BCA assay according to standard protocols.
Prepare 25 microliters of each sample at the optimal protein concentration as determined by the linear range analysis in new tubes. Using assay buffer to normalize the total volume of each sample to 25 microliters. To prepare an assay plate, add 200 microliters of wash buffer to each well of a 96-well plate and shake the plate on a shaker for 10 minutes at 750 revolutions per minute.
At the end of the incubation, decant the wash buffer and tap the plate onto a paper towel to remove any residue. Next, add 25 microliters of assay buffer to each well, followed by 25 additional microliters of buffer to the background wells. Add 25 microliters of the diluted samples to the appropriate sample wells and vortex the via of multiplexed magnetic beads for one minute, before adding 25 microliters of the thoroughly re-suspended beads to each well.
When all of the beads have been added, seal the plate with a plate sealer and cover the plate with foil for an overnight incubation with shaking at two to eight degrees Celsius. The next morning, place the plate on a magnetic separator, making sure that the wells are aligned with the magnets. After two minutes, decant the well contents with the plate still attached to the magnetic separator and add 200 microliters of wash buffer to each well.
After keeping it shaking for two minutes, place the plate onto the magnetic separator for two minutes. Then, decant the well contents with the plate still attached to the magnetic separator and wash the plate with a fresh 200 microliters of wash buffer per well, as just demonstrated. After the second wash, add 25 microliters of detection antibody per well and re-cover with foil for a one-hour incubation at 750 revolutions per minute at room temperature.
At the end of the incubation, add 25 microliters streptavidin phycoerythrin to each well and return the plate to the shaker for an additional 30 minutes at room temperature. At the end of the incubation, place the plate onto the magnetic separator for two minutes before decanting the well contents. Wash the plate two times with 200 microliters of fresh washing buffer per wash and add 75 microliters of the appropriate drive fluid to each well.
Then, re-suspend the beads on the plate shaker for five minutes at room temperature and read the plates on the analyzer according to the manufacturer's instructions. In this analysis, the cerebral blood flow was measured with diffuse correlation spectroscopy four hours after the last injury. The cerebral blood flow index and mean cerebral flow index for each hemisphere can then be determined.
A linear range analysis can be conducted to determine an appropriate protein loading mass prior to collecting data from all samples. Cytokine data can be prepared by subtracting the background measurements from the sample data, then computing Z-score data for each anolyte. Partial least squares regression can be conducted by using a phagocyte microglial activation marker or other variable of interest as the response variable, and the cytokine measurements as the predictor variables.
Varimax rotation can be performed to maximize the covariance of the data on latent variable one, with the activation marker measurements. High loading weights in latent variable one correspond with the cytokine expressions most associated with high expression of the activation marker. Linear regressions between the activation marker and cytokines illustrate that those cytokines with the greatest loading weights in latent variable one were also statistically significant for this analysis.
Although we focused this protocol on traumatic brain injury, these methods are widely generalizable to the study of a plethora of pathological conditions that affect the brain. We use these techniques to help us identify tractable targets to modulate in future experiments to establish causal mechanistic relationships and ultimately with the goal of developing novel therapeutic strategies for mild traumatic brain injury.
