Intranasal Delivery of Mesenchymal Stem Cell Exosomes Modulates the Immune Response of Allergic Rhinitis via the MAPK Pathway in a Mouse Model

Background Allergic rhinitis (AR) is a non-infectious chronic inammatory disease of the nasal mucosa, which is mainly mediated by IgE after the body is exposed to allergens. It has been shown that transplantation of human mesenchymal stem cells (MSCs) into AR animal models can improve the AR behavioral phenotype. Furthermore, there are recent studies that states exosomes are the main mediators of MSC therapy. However, the effect of exosomes on AR has not been investigated. different treatments. Four independent QPCR experiments were performed;


Abstract Background
Allergic rhinitis (AR) is a non-infectious chronic in ammatory disease of the nasal mucosa, which is mainly mediated by IgE after the body is exposed to allergens. It has been shown that transplantation of human mesenchymal stem cells (MSCs) into AR animal models can improve the AR behavioral phenotype. Furthermore, there are recent studies that states exosomes are the main mediators of MSC therapy. However, the effect of exosomes on AR has not been investigated.

Methods
In the established AR mouse model, different concentrations of MSC-derived exosomes (MSCs-Exo) were applied via intranasal delivery. The AR symptom scores, the eosinophils in the nasal mucosal section, the in ammation in ltration in the spleen section, and IgE, ovalbumin-speci c IgE (OVA-sIgE), histamine, IgG1, IgG2a and other Th1/Th2 related in ammatory factors were evaluated by Hematoxylin-eosin staining, Elisa, real-time PCR.

Results
We found that intranasal administration of MSCs-Exo could not only reduce the behavior of nasal scratching and sneezing in mice, but also cause a decline in related immune indicators via the MAPK pathway, including the spleen index, tissue staining, and the expression of in ammation-related cytokines. Moreover, we found that the optimal concentration of MSCs-Exo was 4×10 8 /mL.

Conclusion
The signi cant bene cial effects of exosomes may be exploited to develop a new, non-invasive treatment strategy for AR.

Background
Allergic rhinitis (AR) is a non-infectious in ammatory disease of the nasal mucosa which is mainly mediated by IgE and driven by Th2 cells after exposure to allergens. The main clinical symptoms are sneezing, a runny nose, nasal itching, and nasal congestion 1 . It has been reported that 10-40% of adults worldwide suffer from AR, and in developing countries, the situation is even worse, as the number of cases of AR rapidly increased in recent decades, especially in China 2 . The main treatment methods for AR are allergen avoidance, drug therapy, and speci c immunotherapy 3 . However, drug therapy can only manage symptoms and cannot reverse or alleviate the immune imbalance. Although speci c immunotherapy can desensitize patients to allergens and prevent the development of lesions, it has a long treatment cycle, which tends to result in poor patient compliance and unproductive treatments 4 . Therefore, it is urgent to nd new directions for effective treatment.
Mesenchymal stem cells (MSCs) have the potential for self-renewal and multi-directional differentiation 5 and have immunomodulatory functions 6 . Previous studies have shown that transplantation of MSCs via the tail vein into an AR mouse model can improve symptoms 7,8 . Xenogeneic MSCs have also been proven to exert similar therapeutic effects in AR mice 9 . In addition, MSCs can also reduce the immunological indicators and adjust the Th1/Th2 reaction balance in AR patients 10 . The application of MSCs has resulted in de nite curative effects. However, MSCs have several disadvantages, including vascular obstruction caused by the large cells 11 , unpredictable differentiation in the host 12 , and reported concerns regarding the cancer progression 13,14 .
It has also been proven that exosomes secreted by MSCs (MSCs-Exo) have the same e cacy as MSCs; more importantly, transplantation with its associated risks can be avoided 15 . Exosomes are homogenous vesicles with a diameter of 30-150 nm secreted by a variety of cells 16 . The vesicles contain different types of proteins, nucleotides, and other substances. Cells can swallow the vesicles through endocytosis so that the active substances in the vesicles can function inside the cells 17 . MSCs-Exo can mimic the functions of MSCs and exert similar immunomodulatory effects 18 . Moreover, exosomes do not have a nuclear structure and cannot be ampli ed in the host, and thus they are safer in clinical application 19 .
However, the clinical use of MSCs-Exo to treat AR has never been investigated. In this paper, we show that AR mice treated with MSCs-Exo by intranasal administration showed a signi cant improvement in behavior and histology via MAPK signaling.

Animals
Female BLAB/c mice were purchased from Nanjing Medical University and were bred in Nanjing Drum Tower Hospital. All mice were nurtured in ventilated cages and had free access to water and food. The protocols were approved by the Animal Care and Use Committee of Nanjing Medical University and animals were treated following the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All operations were carried out in accordance with the approved procedures.
AR model and MSCs-Exo treatment BALB/c mice were sensitized with ovalbumin (OVA) (Sigma-Aldrich, A5503) and aluminum hydroxide gel (Thermo Scienti c, 77161) to establish the AR model. At 4 weeks of age, the mice were randomly divided into the control group, the AR group, and the MSCs-Exo treatment group. AR mice were given an intraperitoneal injection of 100 μg OVA in 100 μl PBS combined with 100 μl aluminum hydroxide gel on days 1, 5, and 10 and were treated intranasally with 20 μl 10% OVA in PBS per nostril every day from day 15 to 21. Mice in the control group were given an intraperitoneal injection of 100 μl PBS combined with 100 μl aluminum hydroxide gel, and they were treated intranasally with 20 μl PBS per nostril every day.
Mice in the MSCs-Exo treatment group were given an intraperitoneal injection of 20 μg of OVA in 100 μl PBS combined with 100 μl aluminum hydroxide gel on days 1, 5, and 10; were treated intranasally with 20 μl 10% OVA in PBS per nostril every day from day 15 to 21; and were treated intranasally with 20 μl MSCs-Exo at different concentrations per nostril every day. Administration was performed using a 10-μL pipette with no anesthetics.

Behavioral tests
After the last intranasal administration of 10% OVA on day 21, mice were separately placed in a 10×10×15 cm 3 cage. After waiting for 10 min, their symptoms were observed for 10 min. The behaviors that were observed and quanti ed included nose scratching and sneezing.

Determination of spleen index
The spleen index was calculated according to the following formula: spleen index (mg/g) = spleen weight (mg) ÷ animal body weight (g).

Nasal lavage uids and serum collection
At 24 h after the last intranasal administration, mice were sacri ced by cervical dislocation. PBS (1 ml) was injected into the posterior nostril and a 1.5-mL EP tube was used to collect the nasal lavage uid at the anterior nostril. The nasal lavage uid was centrifuged at 2000 rpm for 7 min at 4°C and stored at −20°C until analysis. Blood samples were collected from the orbital venous plexus by capillaries. The blood samples were rested for 60 min, centrifuged at 4000 rpm for 10 min at 4°C, and stored at −20°C until analysis.

Isolation of exosomes
The human umbilical cord was taken from a full-term infant, cut into small pieces, and cultured. Cells were split when the cells reached 80-90% con uence. In the following procedures, only P5 generation cells were used. After the cells grew to 70% con uence, the culture medium was discarded, the cells were washed thoroughly with PBS, and the medium was replaced with serum-free medium for 48 h. Next, the conditioned medium was collected, centrifuged at 1000 ×g for 20 min to remove cell debris, and centrifuged at 2000 ×g for 20 min, and the supernatant was ltered using a 220-nm lter. Next, the ltered supernatant was centrifuged at 10,000 ×g for 60 min at 4°C, and the supernatant was ltered using a 220-nm lter again. The ltered supernatant was centrifuged at 100,000 ×g for 3 h at 4°C, the supernatant was discarded, the pelleted exosomes were diluted in 1 ml of pulse medium, and the effective diameter of exosomes was assessed by dynamic light scattering (DLS, Brookhaven Instruments).

Transmission electron microscopy
For transmission electron microscopy (TEM), exosomes were isolated as described and directly xed in 200 μl of 2% paraformaldehyde. Exosome preparations (20 μl) were allowed to adsorb in a 75 mesh Formvar/carbon coated grid for 30 min at room temperature. Grids were then washed with PBS (membrane side facing down) and dried using a lter paper. TEM was performed using a Hitachi H-9000 transmission electron microscope at 300 kV, and images were captured using a slow-scan CCD camera.

RNA extraction and real-time PCR
The spleen was taken from the sacri ced mice for RNA extraction using TRIzol™ Reagent (ThermoFisher, 15596026). RNA was reverse transcribed into cDNA using the RevertAid First Strand cDNA Synthesis Kit ELISA Standard wells and blank wells were set in accordance with the protocol of the Bioswamp ELISA kit. Serum (40 μl) was added to each sample well, and 10 μl biotinylated antibody and 50 μl HRP-conjugated reagent were added to each well except the blank well. After incubating for 30 min at 37°C, washing buffer was added to every well and samples were rested for 30 s ve times. Then, chromogen solution was added to each well and samples were incubated for 10 min at 37°C. Lastly, stop solution was added to terminate the reaction and the optical density (OD) at 450 nm was measured within 15 min.

Hematoxylin-eosin staining
After mice were sacri ced, their entire heads and spleens were soaked in 10% neutral formalin xative for 24 h, because the nasal cavity of BALB/c mice is small. Afterwards, the heads were soaked in 0.5 M ethylene diamine tetraacetic acid for decalci cation for 4 weeks. After the mouse skull was softened, the tissues were conventionally dehydrated, embedded in para n, and sectioned in the coronal position (about 3 μm thick slices). The histopathological changes of the nasal mucosa were visualized with hematoxylin-eosin staining. The same steps were applied to the spleens.

Statistical analysis
At least three independent experiments were performed for each experimental condition. Data were analyzed by GraphPad Prism 5 software and are shown as mean ± SEM. In addition, two-tailed, unpaired Student t-tests were used. Animals were randomly assigned during collection, but the data analysis was single-masked.

Characterization of MSCs-Exo
MSCs-Exo have been recognized as a promising substitution for MSCs. Our size distribution analysis revealed that the MSCs-Exo were around 100 nm in diameter ( Figure 1B). TEM analysis of the morphology further con rmed the exosomes' identity ( Figure 1A).

Intranasal delivery of MSCs-Exo relieves AR symptoms
We used an AR mouse model to investigate the role of MSCs-Exo in AR (Figure 2A). We measured the number of sneezing and nose rubbing motions during 10 min after the last intranasal administration of OVA. The results showed that the mice in the AR group sneezed and rubbed more frequently than those in the control group. Moreover, the numbers of sneezing and rubbing events were signi cantly lower in the MSCs-Exo treatment group compared to the AR group ( Figure 2B). In conclusion, intranasal treatment with MSCs-Exo could notably reduce sneezing and rubbing frequencies, thus indicating that MSCs-Exo could relieve the symptoms in AR mice.

Intranasal delivery of MSCs-Exo reduces AR in ammation
To further understand the effects of MSCs-Exo on AR, the levels of different in ammation-related indicators were measured. As shown in Figure 2C, sections of the control group revealed normal nasal cavity mucosa. In the AR group, we observed more in ltrating eosinophils and lymphocytes, glandular secretion, and mucosal swelling ( Figure 2D). After MSCs-Exo treatment, there were signi cantly fewer in ltrating cells and less glandular secretion and mucosal swelling in the nasal mucosa ( Figure 2E). Additionally, the quanti cation of eosinophils ( Figure 2F) suggested that MSCs-Exo have a therapeutic effect on AR mice. The spleen index, which also re ects the in ammatory status, demonstrated similar differences between the three groups ( Figure 3A). Spleen sections from different groups are shown in Figure 3B-D. We observed apparent splenic sinus edema, splenic capsule incrassation, and vasodilation in the AR group ( Figure 3C), and MSCs-Exo administration reduced in ammation.

Optimal concentration of MSCs-Exo
After clarifying the therapeutic effect of MSCs-Exo, we used different concentrations of MSCs-Exo to nd the optimal concentration for AR treatment. We used 4×10 6 /mL, 4×10 7 /mL, 4×10 8 /mL, 4×10 9 /mL, and 4×10 10 /mL exosomes for intranasal administration in AR mice ( Figure 5A). We found that the effect was dose-dependent ( Figure 5B-E). At a concentration of 4×10 6 /mL, we observed basically no effect on the sneezing and rubbing events and in ammation-related indicators. At concentrations higher than 4×10 8 /mL, no signi cant further improvement in the in ammatory symptoms of the mice was observed.
This result suggests that the optimal concentration of MSCs-Exo to reduce allergic in ammation in mice is around 4×10 8 /mL.

MSCs-EXO suppress in ammation via the MAPK pathway
As shown in Figure 6, the mRNA levels of JUN, FOS, ERK, JNK, IKB, and IKK were signi cantly higher in the AR group than in the control group, while the mRNA levels of JUN, FOS, ERK, and JNK were signi cantly lower in the MSCs-Exo group than in the AR group but higher than in the control group. Furthermore, there were no differences in the mRNA levels of IKK and IKB between the AR group and the MSCs-Exo group. These results imply that AR might stimulate in ammation by activating the MAPK and NF-κB pathways, while MSCs-Exo alleviates in ammation by inhibiting the MAPK pathway.

Discussion
We have veri ed that the AR mouse model was successfully established. We rst showed that intranasal delivery of MSCs-Exo could not only relieve the main AR symptoms, sneezing and runny nose, but also decrease various in ammation indicators, including the spleen index, tissue staining, and the expression of in ammation-related cytokines. Moreover, we found that the optimal concentration of MSCs-Exo was 4×10 8 /mL. Besides, we demonstrated that the MAPK and NF-κB pathways play an important role in in ammation, and that MSCs-Exo ameliorated AR by inhibiting the expression of MAPK pathway proteins, which modulated the immune response in the AR mouse model. MSCs exert immunoregulatory effects by inhibiting T cell proliferation, inactivating allogeneic T cells, in uencing T cell apoptosis 22 , and inhibiting B cell proliferation to decrease antibody secretion 23 . Furthermore, MSCs inhibit the differentiation and maturation of dendritic cells, resulting in a decline in the ability to activate T cells 24 . MSCs exert the above immunomodulatory effects mainly through the paracrine secretion of exosomes 25 . The presence of microvilli and columnar cells in the nasal cavity can intensify drug absorption 26 , and intranasal administration is not only safe but also convenient to operate.
These advantages increase its feasibility for clinical application. Therefore, we chose to study the therapeutic effect of MSCs-Exo on AR by intranasal delivery; our results verify that intranasal administration of MSCs-Exo works not only locally but also systemically.
After exposure to the speci c allergen, IgE mediates mast cell degranulation, causing itching and sneezing and further promoting hemangiectasis and increasing secretion of submucosal glands, thus resulting in obstruction and rhinorrhea. The immunohistological characteristics of AR include the in ltration of eosinophils and the predominant expression of cytokines secreted by Th2 cells, with a reduced activity level of Th1 cells, which contributes to the late-phase response such as nasal obstruction and hyperreactivity 27,28 . Our results correspond with the mechanism underlying AR described above. In the AR model, IgE, OVA-IgE, IgG1, and histamine levels in BALF and serum and Th2 cytokine levels increased, corresponding to severe local and systematic symptoms. After intranasal administration of MSCs-Exo, the decrease in IgE, OVA-IgE, and histamine levels relieved the classic symptoms; the decrease in CCL-11 levels is linked to lower eosinophil in ltration; the change in ICAM-1 and Th2 cytokine levels is linked to reduced in ammation; and increased IFN-γ levels can stimulate the expression of IgG2a and inhibit the production of IgG3, IgG1, IgG2b, and IgE 29 , thus further exerting a therapeutic effect on AR. Many studies have also demonstrated that after intravenous treatment with different kinds of exosomes in an asthma mouse model, in ammation of lung tissues decreased, and Th2 cytokine and in ammatory cytokine levels were lower in serum and BALF [30][31][32] . All these results imply that intranasal delivery of MSCs-Exo has an effect similar to that of intravenous delivery to suppress allergic reactions.
In order to pave the way for clinical application in the near future, we examined the effects of different doses, and found that the optimal concentration is 4×10 8 /mL. When the concentration of MSCs-Exo is lower than 4×10 8 /mL, the e cacy decreases signi cantly. Once the concentration is higher than 4×10 8 /mL, the curative effects do not further increase signi cantly. The data show that MSCs-Exo works in a dose-dependent manner, which supports future clinical research.
The activation of NF-κB plays an important role in allergic diseases, inducing the accumulation of in ammatory cells 33 . When activated by in ammatory signals, the released NF-κB induces the transcription of many in ammatory cytokines, which are related to the pathogenesis of asthma 34 . MAPKs are involved in many different cellular events, including allergic diseases 35 . Mammalian MAPKs are mainly divided into three categories: extracellular signal-regulated kinase (ERK), c-Jun N-terminal kinase (JNK), and p38 36 . The activation of MAPK is critical for the production of in ammatory cytokines 37 , and functional differentiation into Th1 or Th2 subsets 38 . Notably, inhibition of ERK promotes the transition of Th2 lymphocytes to Th1 39 . Our results showed that the expression of genes related to the NF-κB and MAPK pathways increased signi cantly in AR mice, but after MSCs-Exo treatment, the expression of genes related to the NF-κB pathway did not decrease signi cantly, while MSCs-Exo treatment signi cantly inhibits ERK, JNK, JUN, and FOS expression. The results con rm that the NF-κB and MAPK pathways play important roles in AR and indicate that MSCs-Exo do not reduce the activation of NF-κB to alleviate the symptoms of AR. MSCs-Exo inhibit in ammation by restricting OVA from affecting the MAPK pathway in AR mice.
While our results have demonstrated that intranasal delivery of MSCs-Exo is an effective therapy for AR, the speci c mechanism has not been adequately revealed. It has not been researched how the MAPK pathway regulates the expression of cytokines to play a therapeutic role in AR. The speci c functional components of MSCs-Exo have not been identi ed, and it is unknown how long they remain active. Previously, the effects of intranasal and intravenous delivery of MSCs-Exo on the brain were compared, and intranasal delivery was more effective 40 . For AR, no direct comparisons between intravenous and intranasal delivery have been made. Note also that we only studied intranasal administration of MSCs-Exo isolated from human umbilical cord; and while this extends the current treatment methods, it will be interesting to test exosomes secreted by other cells to examine the different effects.
Intranasal administration of MSCs-Exo has been researched in various diseases, including complete spinal cord injury 41 , microglia-mediated neuroin ammation 42 , and autism spectrum disorders 43 . It is noteworthy that intranasal delivery of MSC-Exo could substantially expand pulmonary IL-10-producing interstitial macrophages to protect against allergic asthma in mice 44 . However, research on intranasal delivery of MSC-Exo in AR is still scarce. Our results offer a theoretical and experimental basis for the future clinical local application of MSCs-Exo in the nasal cavity for the treatment of AR, which can effectively alleviate pain in AR patients and has signi cant clinical and social value.

Conclusions
In conclusion, the effect of exosomes on AR has rstly been investigated in this study, we found that intranasal administration of MSCs-Exo could not only reduce the behavior of nasal scratching and sneezing in mice, but also cause a decline in related immune indicators via the MAPK pathway. The signi cant bene cial effects of exosomes may provide a new, non-invasive treatment strategy for AR.

Availability of data and materials
All data generated or analyzed during this study are available from the corresponding author upon reasonable request.

Ethics approval and consent to participate
The animal procedures were approved by the Animal Care and Use Committee of Nanjing Medical University and animals were treated following the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All operations were carried out in accordance with the approved procedures. All umbilical cord samples were taken after informed and written consent, and the study was approved by the Research Ethics Board of Nanjing Drum Tower Hospital (permit number 2017-161-01).