*2.1. Animals*

Male eight-week-old C57BL/6J mice were used as experimental subjects and were housed under a 12 h light/dark cycle with food and water ad libitum. Normally, four or five mice were placed in one cage, but mice were housed individually after surgery for at least one week for recovery before further experiments. All experimental procedures complied with the animal care standards of the Xuzhou Medical University Institutional Animal Care and Use Committee (SYXK2015-0030).

#### *2.2. Odorants and Preference*/*Avoidance Behavioral Test*

Odorants were applied in three groups: neutral odorants (isoamyl acetate, 2-heptanone), appetitive odorants (peanut butter, food pellets), and aversive odorants (2,4,5-trimethylthiazole, peppermint oil). The odorants were dissolved in mineral oil at 40% v/v dilution; peanut butter was used in their original states and food was dissolved in saline until the saline was saturated. For each animal, all six odors were tested. The interval between two odors was at least two days. The odors were presented in the order of appetitive odorants (Food, Peanut butter), neutral odorants (Isoamyl acetate, 2-heptanone), and aversive odorants (2,4,5-trimethylthiazole, peppermint oil). During testing, the odorant (50 μL), peanut butter (1 g), food (1 g), or mineral oil (50 μL) was placed on a filter paper in a dish (60 mm × 15 mm) and covered with cage bedding. A custom-designed test chamber (45 cm × 35 cm × 25 cm) with two equally sized compartments was used. Before the test, the mouse was placed into the chamber with empty dishes for 10 min to habituate to the environment. Then, preference for/avoidance of odorants was tested by exposing mice to the two compartments for 10 min, with an odorant in one compartment and mineral oil in the other compartment (the locations of the odorant and mineral oil were randomized) (Figure 1A). Odor preference or avoidance was reflected by the animal's movement trace—the time spent in each chamber was calculated automatically by a computerized recording system.

**Figure 1.** Paradigm for the preference/avoidance test and in vivo recordings. (**A**) Schematic of the preference/avoidance test. (**B**) Representative movement traces illustrating three different behaviors in C57BL/6J mice (left, a neutral odorant: 2-heptanone; middle, an appetitive odorant: food; right, an aversive odorant: 2,4,5-trimethylthiazole). (**C**–**E**) Quantification of time spent by mice in the chamber with the odorant (blue) versus mineral oil (black), for each type of odorant. n = 15 for each pair. (**C**) Mice showed neither preference nor avoidance for neutral odorants compared with mineral oil (paired *t*-tests: isoamyl acetate vs. mineral oil, *<sup>t</sup>*(14) = −0.83, *p* = 0.42; 2-heptanone vs. mineral oil, *<sup>t</sup>*(14) = 0.79, *p* = 0.44). (**D**) Mice showed preference for appetitive odorants compared with mineral oil (paired *t*-tests: peanut butter vs. mineral oil, *<sup>t</sup>*(14) = 2.82, *p* = 0.013; Wilcoxon signed-rank test: Food vs. mineral oil, z = −2.78, *p* = 0.0053). (**E**) Mice showed avoidance for aversive odorants compared with mineral oil (paired *t*-tests: 2,4,5-trimethylthiazole vs. mineral oil, *<sup>t</sup>*(14) = −8.58, *p* < 0.001; peppermint oil vs. mineral oil, *<sup>t</sup>*(14) = −5.80, *p* < 0.001). (**F**) Diagram of the in vivo electrophysiological recordings. The recordings repeated at different metabolism states, including fasted for 0 h, 12 h, and 24 h. (**G**) Schematic representation of the methods for in vivo electrophysiological recordings in awake, head-fixed mice. n.s., not significant. \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001.

## *2.3. Microelectrode Implantation*

The microelectrodes (16-channel, Jiangsu Brain Medical Technology Co. Ltd, Nanjing, China) were implanted into a specific region of the brain as previously described [18,19]. Briefly, mice were anesthetized with pentobarbital sodium (90 mg/kg body weight, i.p.) and positioned in a stereotaxic frame. Eye ointment was applied to the eyes. The skull surface (from the midline of the orbits to the midpoint between the ears) was exposed, and a hole was drilled above the right OB for microelectrode implantation (anterior-posterior (AP): +4.0 mm; medial-lateral (ML): +1.0 mm). Then, the microelectrodes were positioned and lowered through the drilled holes until they reached the OB mitral cell layer at an average depth between 1.8 mm and 2.5 mm. A custom-designed head plate was attached to the skull with small screws and dental acrylic to enable head fixation during recordings. The body temperature of mice was maintained at 37 ± 0.5 ◦C throughout the surgery.

#### *2.4. Spike and LFP Recordings*

The recordings were initiated after the mice had recovered from surgery, as in previous studies [18–21]. Briefly, awake mice were head-fixed with two horizontal bars and were able to maneuver on an air-supported free-floating Styrofoam ball (Thinkerbiotech, Nanjing, China). For spike recordings, the signals from the microelectrodes were sent to a headstage, amplified by a 16-channel amplifier (Plexon DigiAmp (Plexon Inc, Dallas, TX, USA); bandpass filtered at 300–5000 Hz, 2000× gain), and sampled at 40 kHz by a Plexon Omniplex recording system. For LFP recordings, LFP signals were amplified (2000× gain, Plexon DigiAmp), filtered at 0.1–300 Hz, and sampled at 1 kHz. Spikes or LFP signals together with odor stimulation event markers were recorded via the same Plexon Omniplex recording system. The fasting started at 21:00 and finished at 21:00 the next day. Recordings were repeated on the same mice under di fferent nutritional states: satiety, fasted for 12 h, and fasted for 24 h.

#### *2.5. Odorant Presentation during Electrophysiological Recordings*

The three sets of odorants described above were also used during electrophysiological recordings. Peanut butter was mixed with mineral oil at a 10% m/v dilution and food was dissolved in saline as described above. The other odorants were dissolved in mineral oil at 1% v/v dilution. During the odor delivery period, an odor delivery system (Thinkerbiotech, Nanjing, China) was used, as previously described [20,21]. There were 15 trials for each odorant. The six odorants were presented in a pseudo-randomized order, with no more than two successive presentations of the same odor. Each odor was delivered to the animal for 2 s with an inter-stimulus interval of 20 s. All six odorants were presented passively; the mice were not required to respond to the odors presented.

#### *2.6. Measurement of Sni*ffi*ng Parameters*

The sni ffing patterns of mice during electrophysiological recordings were recorded continuously by placing a cannula into one nasal cavity and connecting it to an airflow pressure sensor (Model No. 24PCEFA6G(EA), 0–0.5 psi, Honeywell) Surgical implantation of the nasal cannula was performed as described in previous studies [22,23]. The pressure transient signals were amplified (100× gain, Plexon DigiAmp) and sampled at 1 kHz by a Plexon Omniplex recording system. A sni ff was defined as the point of transition from exhalation to inhalation.
