Intracranial microstimulation (ICMS) is used to identify appropriate sites in the MI whisker region for injecting anterograde tracers. A pipette puller is used to produce a glass pipette that has a tip diameter of 1 μm and, after it is filled with hypertonic (3 M) saline, an impedance of 1 MΩ. The electrode’s impedance is always measured before placing it in the brain to insure its suitability for ICMS. At least three stimulating electrodes are made just prior to each experiment, and they are temporarily stored in a saline-filled electrode-holding jar to prevent the formation of salt crystals at the electrode tip.

For ICMS, rats are anesthetized with intramuscular injections of ketamine HCl (20 mg/kg) and xylazine (6 mg/kg). After immobilizing the rat’s head in a stereotaxic instrument, a craniotomy is made over the MI cortical region and a small ground screw is placed in the cranium overlying the cerebellum (see Note 1). As shown by Fig. 2, the constant current source must be connected with both the stimulating electrode and the ground screw to form a complete circuit for ICMS.

Muscle twitches can be evoked from MI cortex only if the animal is in a shallow plane of anesthesia. If the animal is deeply anesthetized, ICMS will not evoke muscle twitches even if high stimulation currents are administered. Furthermore, repeated application of a large stimulation current may destroy the neuropil at the electrode tip. Within 60–90 min of the initial anesthetic injection, the rat usually enters a shallow anesthetic plane that is characterized by small, spontaneous whisker movements in which the peak-to-peak whisker motion is no more than 1 mm. This indicates the moment when ICMS is most likely to be successful in evoking a peripheral movement (see Note 2).

Microstimulation mapping begins by lowering the stimulating electrode into the deep layers of MI cortex, which are located 1.2–1.7 mm below the pial surface. Initially, microstimulation consists of short trains (80 ms in length) of 0.7 ms current pulses that are separated by intervals of 3.3 ms. Hence, the experimenter activates the programmable timer so that each button press on the timer delivers a train of 20 pulses at a rate of 250 Hz (see Note 3). While observing the whiskers, the experimenter delivers a series of trains, usually at a rate of 1 Hz or less. Microstimulation should begin with suprathreshold positive currents of 100–150 μA. If the rat is lightly anesthetized and the tip of the stimulating electrode is in the deep layers of the MI whisker region, such currents will evoke robust movements of multiple whiskers, possibly on both sides of the face. The current level is gradually reduced to the lowest level that barely evokes twitches of a small number of whiskers on the contralateral face. The lowest level for evoking a single whisker twitch is usually about 10–20 μA. If ICMS does not evoke whisker twitches, the circuit connections and other factors must be carefully examined (see Note 4).

Rat MI cortex contains two whisker regions that are distinguished by their responses to short and long pulse trains. Brief whisker retractions are evoked when short train pulses are administered to the medial agranular cortex, which is known as the retraction area in the MI whisker region (MI–Re). The MI–Re region is located about 1.7–2.5 mm rostral and 1.5–2.2 mm lateral to bregma in the adult rat. A second, more caudal whisker region evokes whisker protractions when it is stimulated by short (80 ms) pulse trains but evokes rhythmic (5–15 Hz) whisking movements when stimulated by long (1,400 ms) trains administered at 60 or 100 Hz (i.e., 0.7 ms pulses separated by 16 or 9.7 ms). This region, which is called the rhythmic whisking area (MI-RW), is located about 0.7–1.2 mm lateral and 0.6–2.4 mm rostral to bregma (6). To verify that a stimulation site is in one of these regions, we sequentially administer a series of short pulse trains and then a series of long pulse trains at each electrode penetration. The digital timer allows multiple programs to be stored in memory, and switching between short and long pulse trains is done quickly if both sequences of pulse trains have been programmed and stored in memory prior to the start of the experiment.

Locating a suitable site for a tracer injection in MI cortex needs to be done rapidly because the animal’s plane of anesthesia is relatively shallow and reflexive movements could appear if the ICMS is not conducted efficiently. Furthermore, in cases where different anterograde tracers are injected into homotopical regions bilaterally, it is necessary to locate the same functional region in both hemispheres as quickly as possible. Another constraint is posed by the fact that even small tracer injections produce some diffusion, but the labeling data will not be interpretable if the tracer diffuses across the borders between the whisker and forelimb regions in MI.

For these reasons, we use threshold ICMS currents to locate the B, C, or D whisker rows in MI because these representations are in the center of the MI whisker region. Small injections at these sites should reduce the likelihood that tracer will diffuse beyond the whisker representation. We usually stimulate 4–8 sites in each hemisphere. After locating the center part of the MI whisker region, ICMS is administered more laterally in 500 μm steps until we identify a site that elicits forelimb movements (see Note 5). After locating the forelimb-whisker border, the tracer is injected into a site located at least 500 μm away, thereby insuring that the tracer does not diffuse across this functional boundary.

In rare cases, we have used ICMS to identify the whisker representations in subcortical brain regions that were injected with a retrograde tracer. For example, to determine whether the MI whisker region projects directly to the facial nucleus, which contains the cell bodies of the peripheral nerve fibers that innervate the whisker pad, we have injected a retrograde tracer into the lateral facial nucleus (see Fig. 3). In these cases, ICMS is used to locate sites in the facial nucleus that evoke whisker movements.

For most forebrain targets of MI, ICMS is not effective for evoking whisker twitches. Tracer injections in these brain regions must rely on stereotaxic coordinates alone. Thus, after identifying regions in the thalamus, dorsolateral striatum, and claustrum that received projections from the MI whisker region, stereotaxic coordinates indicate where to place retrograde tracers in these regions (see Figs. 3 and 4).