The overall goal of this procedure is to demonstrate an optimized protocol for successful visualization of sound perception and processing in auditory and song control regions in the songbird brain Using functional MRI, this is accomplished by first preparing good quality auditory stimuli, which can be incorporated into an on-off block paradigm to present to the bird during functional MRI or FMRI scanning. The second step is to prepare the bird for scanning by first anesthetizing it, and then by positioning it on an adapted animal bed, which is subsequently placed in the bore of the MRI scanner. Next FMRI images can be acquired during auditory stimulation.
The final step is to pre-process the acquired MRI data, including normalization of the images to the zebra finch atlas. Ultimately, statistical voxel based analysis is used to allow visualization of the obtained results and localize the brain activations induced by the auditory stimuli. The main advantage of this technique over other more commonly used methods in sunbird research like electrophysiology, histology, and molecular mapping techniques, is that FMRI is a non-invasive in vivo method, which allows one to study the same subject over time and to address the whole brain at once without prior hypothesis on the localization of the neural substrates involved in the investigated process.
Individuals new to this method may struggle because implementing functional MRI in birds is very different and far more challenging than using the same method in humans or rodents. This is not only due to the small size of their brains of these subjects, but also because their skull comprised of numerous air cavities, which induces susceptibility artifacts, and by using a spin echo sequence sequence less susceptible to these artifacts compared to a gradient sequence, we can do whole brain functional imaging of the songbird with very good spatial specificity. The scanner bore is a confined space that can distort the auditory stimuli.
Therefore, to begin preparation for this protocol, record the sound stimuli while being played inside the bore of the seven T MR system using a fiber optic microphone and verify the distortions for the used MRI system. Here we can see the frequencies enhanced and suppressed as shown by recordings of white noise made outside the scanner and at the location of the bird's head within the magnet bore. To compensate for this artificial enhancement, apply an equalizer function to each stimulus using wave lab software for the setup seen.
In this protocol, the function consists of a Gaussian kernel with the parameters seen on screen. Here, the song stimuli should be composed of several individual song motifs of a specific bird interleaved with periods of silence adjust the duration of these silent periods to keep the total amount of sound and silence identical over all stimuli. The total length of each stimulus is 16 seconds, which corresponds to the acquisition time of two FMRI Images.
Using prep software normalize the intensity of each song in terms of matched root means square and high pass filter at 400 hertz before integrating into the complete stimulus. The experiment can be designed using presentation software using an on off block design, alternating auditory stimulation periods with resting periods. Each block should last 16 seconds corresponding to the acquisition time of two images present each stimulus type 25 times resulting in the acquisition of 50 images per stimulus and per subject.
The presentation order of the conditions should be randomized within and between subjects. Also set the experimental paradigm up so that it will receive triggers from the scanner. This assures that while scanning at every repetition within the FMRI sequence, the scanner software will send a trigger to the auditory presentation software, which in turn registers the scan number and executes the corresponding command.
Here we present a protocol specifically adapted to the use of adult zebra finches. Zebra finches should be housed in aviaries under a 12 hour light, 12 hour dark photo period and have access to food and water at libido throughout the study. The minimal number of individuals per experiment is 15.
This number takes into account the sensitivity of spin echo FMRI and the natural inter-individual variability of biological phenomena measured in the experiment. To prepare for the experiment, first, connect the beak mask to the gas controller device with plastic tubes and install it on the MRI bed of a seven T MR system. Open both oxygen and nitrogen gas bottles and switch on the gas controller device.
Also switch on the feedback controlled system and warm airflow device. Next, anesthetize the zebra finch with 3%isof fluorine in a mixture of oxygen and nitrogen by introducing its beak into the mask and holding the head until the bird is fully anesthetized. This can be verified by pulling the paw softly.
When the bird is fully sedated, the paw will not be retracted. In addition, the eyes of the bird will be closed. Introduce a cloacal temperature probe to screen body temperature throughout the experiment and place a pneumatic sensor underneath the belly to monitor breathing rate.
Then place a jacket to restrain the body of the bird throughout the experiment. Maintain the breathing rate within 40 to 100 breaths per minute and keep body temperature constant at 40 degrees. If the breathing range is too low or high, adjust the level of anesthesia accordingly.
If the problem persists, the experiment should be stopped and the animal removed from the setup. In order to recover, position non-magnetic dynamic speakers on either side of the zebra finch head and connect them to the amplifier. Make sure that the wires of the speakers are led away from the temperature probe because this can influence the temperature reading.
When too close, place the surface RF coil on top of the zebra finch head. Then position the zebra finch in the center of the magnet. Finally, reduce the anesthesia level to 1.5%ISO fluorine prior to scanning.
To begin scanning, first, determine the position of the brain in the magnet. Acquire one set of single slice gradient, echo scout images in sagittal, horizontal, and coronal planes, as well as multis slice scout images used afterwards for the positioning of the functional scans. Then decrease the noise of the gradients by increasing their ramp times to 1000 microseconds.
Next, prepare a rapid acquisition relaxation enhanced, otherwise known as rare T two weighted FMRI. Sequence then prescribe 15 sagittal slices covering nearly the whole brain. Now select the auditory protocol in the presentation software to ensure that the auditory presentation software does not miss any trigger from the scanner.
Initiate the auditory protocol first. Then once the protocol is fully loaded, start the FMRI sequence. Each FMRI experiment should be proceeded by the acquisition of 10 dummy images to allow the signal to reach a steady state before starting auditory stimulation.
After acquisition is complete zero. Fill the data to 64 by 64 To determine the quality of the data, take a preliminary look at the results. Using the functional tool.
In ParaVision software, calculate the differential bold response between all on blocks and baseline. If no activation is seen in the primary auditory areas, the bird probably did not hear or process the auditory stimuli to technical problems or anesthesia level. The setup should be verified and the measurement repeated.
After functional acquisition, run an anatomical 3D rare T two weighted sequence in the same orientation as the previous FMRI scans. Then zero. Fill this data to 2 56 by 2 56 by 2 56.
Finally remove the zebra finch from the MRI bed and allow it to recover from anesthesia in a cage under a heat lamp. Normally the recovery of a zebra finch from ISO fluorine anesthesia takes under five minutes. After only a few minutes, the bird will try to stand up, and once the bird is fully recovered, it will perch on a branch instead of sitting on the bottom of the cage to begin data processing.
First, convert the MR data into analyze or nifty format. Then before starting data processing in SPM, it is necessary to adapt the FMRI songbird data by multiplying the voxel size by 10. Since the FMRI analysis software has been developed to process FMRI data acquired in humans, the simplest way to proceed is to artificially increase the voxel size of bird FMRI data using MRI Crow software.
Note that this adjustment does not influence the data itself. No re-sampling or any other modifications to the data is applied. Next, the acquired data are pre-processed.
Using SPM eight first realign the FMRI data and coregister the anatomical 3D dataset to the FMRI time series. Then normalize the 3D data and the COREGISTERED FMRI time series to the zebra finch brain. MRI Atlas.
Apply the transformation matrix to the FMRI dataset and smooth the data with a 0.5 millimeter with Gaussian kernel. Then carry out statistical voxel based analysis in SPM eight, first on the level of the individual birds, followed by analysis at the group level. Finally, to visualize the results and localize the functional activations project, the statistical parametric map onto the zebra finch atlas, the high resolution 3D zebra finch brain atlas, and stereotactic coordinates used in the described analysis of the zebra Finch FMRI results was designed for this experiment at the Bioimaging lab, the MRI Atlas dataset.
The brain delineations and the nuclei delineations are freely available through the website seen here. Here a lateral view is shown of a 3D representation of the left hemisphere with delineated structures from the atlas projected on its midsagittal slice. Here we can see an example of an FMRI bold response in the primary auditory region field L, and adjacent secondary auditory regions evoked by different auditory stimuli compared to the rest condition.
T values are color coded according to the scale displayed in the figure, and only of voxels in which the T-test was found to be significant are displayed. This is another example of an FMRI experiment in zebra finches. The images show selective responses to the bird's own song in the brain region's area XMLD and HVC.
The numbers under the images indicate coordinates in millimeters from the midline. The positive sign indicates that slices and statistical results are from the right hemisphere. The protocol described in this video can be easily adapted for applications in other songbirds such as starlings.
This is an example of the results of such an FMRI experiment. In a Starling, the figure shows activations induced by auditory stimuli versus rest in the same individual across seasons. The implementation and optimization of functional MRI in songbirds open new research avenues in the field of songbird research.
It has already been shown to be a valuable technique for the experimental analysis of complex sensory motor and cognitive processes underlying vocal communication in these animal models. In addition, thanks to its non invasiveness and whole brain approach, all FMRI can be used as a complementary method to inform more commonly performed local and invasive methods. The growing number of FMRI experiments in SUNBIRDS demonstrates its applicability to cognitive research questions, a future effort which should be made in order to further extend the MRI experiments Beyond auditory processing, which is a process that remains active under anesthesia, is the use of awake subjects habituated to the imaging protocol.
This might offer new research possibilities where birds can actively participate in EMI experiments and where the attention of the birds can be experimentally controlled.
