Introduction

BOLD fMRI relies on the fact that neural activity causes increases in blood flow to local brain regions¹. Previously our lab reported that unilateral whisker loss after infraorbital nerve transection (ION) produces an increase in BOLD fMRI response to the intact whisker set. This was reported to be mediated by a post-critical period re-opening of thalamocortical (TC) plasticity in the rat somatosensory barrel cortex (S1BC)². We reproduced these effects in mice, which allows for further characterization of the system using additional tools such as optogenetics. Here we report the expected increase of intact S1BC, mediated by a stronger TC synapse. We also observed a similar bilateral response to unilateral stimulation of the intact whisker set in ION mice and rats. Finally, we report synaptic remodeling after whisker denervation occurs at the callosal input to the deprived S1BC, targeting a specific cell type.

Methods

Adult mice underwent a unilateral sham or ION surgery, which removes input from one whisker pad. Two weeks after ION animals were anesthetized with an IP injection of a ketamine/xylazine mixture and imaged in an 11.7T animal MRI system (Fig. 1A) (30 cm horizonal bore, Magnex Scientific, Oxford, England; MRI Electronics, Bruker Biospin, Billerica, MA). Scan parameters were: TR/TE = 1000/15 ms, .22 mm in plane resolution, .4 mm slice thickness, 7 slices, matrix size 64 x 48. Two 30G needles were placed in the intact whisker pad and 2mA currents were delivered to elicit stimulus-evoked BOLD fMRI. A separate group of mice were unilaterally injected with channel rhodopsin (ChR2) into intact S1BC, underwent ION surgery, were euthanized, and brain slices were prepared for whole-cell slice electrophysiology (Fig. 1A). Principal neurons in the intact and deprived S1BC were patched and synaptic strength was measured between brain regions using optogenetics and strontium evoked desynchronized vesicle release, which enabled quantal analysis of post-synaptic strength.

Results

Unilateral whisker stimulation evoked a reliable BOLD fMRI signal in the contralateral S1BC in sham animals, which was significantly larger in ION animals (Fig. 1B). This is mediated by a stronger TC synaptic connection in the ION group (Fig. 1D). Interestingly the unilateral stimulus provoked a bilateral BOLD response in ION animals (Fig. 1B). We determined that the bilateral activation is due to a stronger callosal connection from the intact to deprived S1BC, which specifically targeted layer 5 (L5) principal neurons (Fig. 2C). The callosum did not strengthen connections to layer 2/3 (L2/3) neurons (Fig. 2B) and local connections between L5 cells were similarly unaltered (Fig. 2D).

Discussion

Neural circuits are known to be plastic in response to changes in sensory experience, even in adult organisms. Our experiments describe a change in the somatosensory circuit which we detected with BOLD fMRI, and then characterized at the synaptic level using whole-cell slice electrophysiology. It will be interesting to determine if the specific synaptic changes also lead to changes in global resting state fMRI in a manner similar to the effects we report in stimulus-evoked fMRI. This has implications for the detection and description of brain disorders in vivo which have a synaptic, circuit level change in function.

Conclusion

Here we demonstrate that stimulus-evoked BOLD fMRI can detect changes in neural activity which can then be described at the cellular and synaptic level with whole cell electrophysiology. These results indicate that BOLD imaging can be a useful tool to identify locations of robust neural circuit reorganization.

Acknowledgements

No acknowledgement found.