Investigation of neuronal activation of olfactory bulbs (MEMRI)

House mice of three taxa were divided into fourteen groups (Table. 2). One day prior to testing, the mice were placed in clean ventilated cages (350 × 250 × 120 mm, ZOONLAB GmbH, Germany), with dust-free sawdust as bedding. To test the neuronal activation of MOB and AOB in response to odors 10-μl aqueous solution of 10 mM MnCl² (Sigma-Aldrich Co, MO, USA) was rapidly injected into one nostril using a 20-μl micropipette. After that, the mouse was put into its empty clean cage. Subjects were exposed to either clean “saline” air, or estrus female urinary volatile odors. For the odor stimulation groups, each individual was exposed to one urine odor. All odor and saline exposures were pulsed at 1:3-min on:off (1 min on, 3 min off) intervals taking into account the habituation of glomerular responses in the olfactory bulb following odor stimulation [80]. Each odor was exposed 4 times (16 min in total).

Summary of treatment conditions, odor exposure, and distribution of subjects

For odor exposure, an olfactometer of the following design was used: the air pump (Barbus SB-348A) was connected to the closed cage of the tested individual by a silicone hose with a 1-ml plastic nozzle. A piece of filter paper (0.5 cm × 2 cm) was placed inside the plastic nozzle inserted into the drinking hole in the cage cover and 20 μl of urine / saline of the odor stimulus was applied. During the exposure of the odor stimulus, air was passed through the nozzle at a rate of 200 ml / min. To test each mouse, a new nozzle and a new piece of filter paper were used. For each odor stimulus, a new silicone hose was used.

We used a 11.7 T BioSpec 117/16 USR (Bruker, Germany) MR-scanner for MEMRI study. The mice were immobilized with a gas mixture (4%) of isoflurane (Isofluran, Baxter Healthcare Corp., USA) and air using an anesthesia machine (The Univentor 400 Anaesthesia Unit, Univentor, Malta) 3 min before the experiment. Anesthetized mouse were placed on a heated surface (temperature 30 °C) set in the MR-scanner. Pneumatic sensor for breathing (SA Instruments, Stony Brook, NY, USA) was put under the lower part of animal body.

The neuronal activity of olfactory epithelium and VNO of the male M. spicilegus, M. m. wagneri and M. m. musculus was assessed based on the level of the MRI signal in the glomerular layer of the MOB and in the AOB. Accumulation of manganese ions (Mn^{2 +}) in neurons of MOB and AOB is very reliably correlated with the level of activity of calcium channels of olfactory epithelial cells and VNO [72, 73]. The accumulation of manganese ions in OB cells was expressed as the ratio of the tissue MRI signal level to the level of the MRI signal in the reference, which was a microtubule with phosphate buffer (0.5 ml) placed along the mouse’s head (Fig. 1). MRI scanning was performed at 2 h after the exposure to the odor stimulus or saline.

Scanning process of MEMRI and mapping of the activity in the mouse MOB. Maps demonstrate the neuronal activity of various areas of the mouse MOB glomeruli layer in response to the olfactory stimulus

The distribution of manganese ions within the OB in control experiments and under the influence of odor stimuli was obtained using T1-weighted images using the RARE (Rapid Acquisition with Relaxation Enhancement) method. Parameters of the pulse sequence of the method (TE = 10 ms, TR = 400 ms), image parameters (field of view – 1.8 × 1.8 cm, matrix – 256 × 256 pixel array, thickness of slice – 0.5 mm, 75 μm × 75 μm × 0.5 mm voxel dimensions, the distance between the slices – 0.5 mm, the number of slices is 5, the orientation of the slices is coronary), the total scan time was 7 min.

Preliminary processing of MRI scans was carried out in ImageJ. This procedure consisted of several stages: aligning the images horizontally, isolating the boundaries of the mouse’s brain, resizing the images. Alignment of the brain geometry made it possible to compare automatically the level of the MRI signal in AOB and in certain regions of the MOB in individuals. To analyze the distribution patterns, the globular layer of OB in each MRI slice was divided into 12 regions. MOB was fitted in 5 scans (Fig. ​(Fig.1).1). Thus, MOB was presented in 60 regions (12 × 5) and the original resolution of the scan MRI was reduced to 250 μm × 250 μm × 0.5 mm. Within these 60 regions, the MRI level of the signal was averaged. It gave us an opportunity to make various intergroup comparisons and to evaluate the changes in neuronal activity in response to the odor stimulus. Next, a two-dimensional map of the OB was used to visualize the obtained results. The number of the region (1–12) was plotted along the abscissa axis, the cutoff number (1–5) was plotted along the ordinate axis. Pseudo coloring reflects the value of the Student’s t-criterion, characterizing the reliability of the differences between the two groups (Fig. ​(Fig.1).1). To analyze the activation patterns of the AOB, the following parameters were used: the total number of regions in which the manganese accumulation significantly differs between the two groups, the average value of the t-test and its variance.

To compare the two patterns of the MOB reaction according to the number of regions of the glomerular layer, where the manganese accumulation significantly differs between the two groups (p < 0.05, based on the values of the t-test), the χ2 criterion was used. For the values whose variational series approached the normal one, dispersion analysis was used. For multiple average comparisons, the LSD test (Least Significant Difference) was used. Data were expressed as Mean ± SE.

To evaluate the relationship between the two activation patterns of the OB, Spearman’s nonparametric correlation coefficient was used. To compare the two correlation coefficients, the approach described by Myers and Sirois [81] was used.