Eight-week-old male ddY mice were purchased from CLEA Japan, Inc. (Tokyo, Japan). The average weight of the mice was 38 g. Animals were bred under conditions of 23˚C ± 2˚C, humidity 55% ± 5%, light and dark cycle (light from 9 am to 9 pm), and unlimited access to food and water. All experimental procedures conformed to “Regulations for Animal Experiments and Related Activities at Tohoku University” and were reviewed by the Institutional Laboratory Animal Care and Use Committee of Tohoku University. After habituation for 1 week, OBX mice were prepared as described previously [16]. In brief, mice were anesthetized and their head immobilized by a stereotaxic apparatus (David Kopf Instruments, Tujunga, CA, USA). A small burr hole was drilled on both sides of the olfactory bulbs, which were then removed using a suction pump. Holes were filled with hemostatic sponges to prevent hemorrhagia.

Human UC tissues were collected from full-term pregnant women who provided informed consent at Narita Ladies’ Clinic (Saitama, Japan). The sampling collection has been approved by the Human Ethics Committee in Narita Ladies' clinic as well as BioMimetics Sympathies Inc., and the UC tissues are not be collected without patient consent. UC-MSCs were isolated from the sample, and stromal vascular fraction were passaged with sf-DOT (BioMimetics Sympathies, Inc., Tokyo, Japan), a serum-free culture medium, at 37˚C in an atmosphere containing 5% CO₂. About 1.2 × 10⁶ UC-MSCs of passage 5 were seeded in each T175 CellBIND culture flask containing 30 ml of sf-DOT medium. After 5 days of incubation, resulting MSC-CM was collected, centrifuged at 400 g for 5 min to remove the cell debris, filtered through 0.22-μm filter, and then concentrated (10 times) by ultrafiltration using centrifugal filtering units with a cut-off value of 3 kDa (Amicon Ultra-15; Millipore, MA), according to the manufacturer’s instructions. The concentrated UC MSC-CM was aliquoted and stored at −20˚C until use.

We conducted the following experiments as shown in Figure 1. During 14 days from OBX operation, nasal administration of the control medium or the conditioned medium was performed 5 times. After the treatment, we conducted behavior tests and immunohistochemical analysis. Intranasal administration was conducted for OBX mice in 2nd, 4th, 7th, 10th and 12th day after the operation (Control medium: 20 μL; Conditioned Medium: 10 or 20 μL each day).

Three behavioral tests were conducted 14 days later from OBX operation.

A Y-maze task that can measure spatial memory was conducted as previously described [16]. A mouse was put on the end of the arm and allowed to move freely

for 8 minutes. Alternation behavior was defined as entries into all three arms on consecutive choices. Percentage of alternation was calculated as (alternation)/(total arm entries − 2) × 100.

Novel object recognition task is based on the characteristics that mice prefer novel objects to familiar ones. In the trial session, two similar objects were placed on the mouse cage symmetrically, and then the mouse was put on the cage for 10 minutes. 24 hours later, one of the objects was replaced with a novel one and let the mice explore in the cage for 5 minutes (called “test session”). A discrimination index was evaluated by comparing the difference between exploratory contacts of novel or familiar objects and the total number of contacts, adjusting for differences in total exploration contacts.

Passive avoidance task was conducted as previously described [16]. The apparatus consisted of light and dark compartments with stainless steel rods connected with an electronic stimulator (Muromachi Kikai, Tokyo, Japan). After putting a mouse on the light compartment, When the mouse entered the dark compartment, it received an electric shock (0.3 mA for 2 s) from the grid floor and was removed from the apparatus 30 s later on training trials (fear acquisition). 24 hours after this fear acquisition session, mice were placed in the light compartment and step-through latency was recorded over a period of 300 s, an indicator of retention level (test session).

Mice were perfused with ice-cold phosphate-buffered saline (PBS, pH 7.4) and then with 4% paraformaldehyde. After removal and postfixation of the brain, tissue was sliced into 50 μm-thick coronal sections using a vibratome (Dosaka EM, Kyoto, Japan). TSA immunohistochemistry was performed using TSA Fluorescein System kit (PerkinElmer, MA, USA). The sections were incubated with 0.3% H2O2 in PBS for 30 minutes to inactivate endogenous peroxidase. After blocking with TSA buffer (containing 5 mL 2M Tris-HCl, 0.88 g NaCl and 0.5 g blocking reagent in the kit per 100 mL), mouse anti-ChAT (1:500 in TSA buffer, Millipore) diluted in blocking solution was added and incubated in 4˚C for overnight. Washing with PBS several times, biotinylated anti-mouse antibody (1:500 in TSA buffer) was reacted for 2 hours at 4˚C. Then, streptavidin-HRP (1:500 in TSA buffer) was added and incubated for 2 hours at 4˚C after PBS washing. Finally, the sections were incubated with the fluorescein tyramide reagent (1:100 diluent) for 10 min in room temperature in the dark and enclosed by MAS coated microscope slide immersed in VECTASHIELD® Antifade Mounting Media (Vector Laboratories, CA, United States). Images were acquired by confocal laser scanning microscope (TCS SP8, Leica Microsystems, Wetzlar, Germany).

All data were presented as means ± standard error of the mean (S.E.M). Statistical significance was tested by one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. Pairs of means were compared by Student’s t-test. All the statistical analyses were performed using GraphPad Prism 8 (GraphPad Software, CA, USA). Differences with p < 0.05 were considered statistically significant.

First, we investigated whether conditioned medium can recovery impaired memory of OBX mice by three behavior tests. In Y-maze task, OBX mice showed a decrease of alternation compared with sham-operated mice (53.12% ± 2.89%, p < 0.01 vs. sham; Figure 2(b)). Intranasal administration of conditioned medium (20 μL) restored memory loss without alteration of total arm entries (71.19% ± 3.11%, p < 0.05 vs. OBX; Figure 2(a) and Figure 2(b)). The discrimination index was measured by a novel object recognition task. In the trial session, there are no differences in the discrimination index in all groups (Figure 2(c)). In the test session, OBX mice could not discriminate a novel object from familiar one (48.96% ± 1.87%, p = 0.44 vs. familiar; Figure 2(d)). On the other hand, the administration of conditioned medium restored memory deficit observed in OBX mice (59.65% ± 2.15%, p < 0.001 vs. familiar; Figure 2(d)). In the fear acquisition session of passive avoidance task, mice in all groups entered the dark box immediately (Figure 2(e)). In the test session, latency time of OBX mice was significantly shorter than that of sham-operated mice (16.95 ± 4.75 sec, p < 0.0001 vs. sham; Figure 2(f)) and the memory impairment was recovered by conditioned medium treatment (228.12 ± 50.99 sec, p < 0.0001 vs. OBX; Figure 2(f)).

After examining the behavioral test, we conducted immunostaining of medial septum to analyze the therapeutic effect of conditioned medium on the gene expression of acetyltransferase (ChAT) in OBX mice (Figure 3(a)). Consistent with the previous study, the amount of ChAT protein decreased in the septum in OBX mice (65.66% ± 6.24%, p < 0.05 vs. sham; Figure 3(b)). We further explored if the recovery of OBX memory impairment in the behavioral analysis is related to the restoration of ChAT expression. Intriguingly, treatment of conditioned medium (20 μL) potentiated the gene expression to recover reduced ChAT level in OBX mice (104.15% ± 16.03%, p < 0.01 vs. OBX; Figure 3(b)). These data suggest that UC MSCs-derived conditioned medium has a potent therapeutic effect on the recovery of ChAT gene expression to restore memory functions.