Abstract

Objectives

   This study aimed to compare systematically strain-specific immune
   differences between BALB/c and C57BL/6 mice in a local tolerance model
   and explore the underlying mechanisms.

Methods

   BALB/c and C57BL/6 mice received daily intranasal ovalbumin (OVA; 25
   mg/ml, 10μl/nostril) or PBS for 15 weeks. Systemic responses (serum
   OVA-specific IgE, IgG1, IgG2a; splenocyte cytokine secretion: IL-4,
   IL-10, IFN-γ) and local nasal responses (symptoms, histopathology:
   polymorphonuclear/goblet cell infiltration; immunohistochemistry:
   TGF-β, IL-10, eotaxin; RNA-seq transcriptomics of nasal mucosa) were
   assessed at the 8th and 15th weeks.

Results

   BALB/c mice initially exhibited worsening nasal symptoms, which was
   followed by significant alleviation. In contrast, C57BL/6 mice showed a
   significant worsening of symptoms. Serum levels of IgE, IgG1, and IgG2a
   increased significantly over time in BALB/c mice. In C57BL/6 mice,
   serum IgE and IgG1 levels also increased significantly, while IgG2a
   levels showed no significant changes. In splenocyte supernatants,
   BALB/c mice showed IL-4 levels that initially increased significantly
   but later decreased significantly, whereas IL-10 levels were
   significantly elevated and sustained. Conversely, C57BL/6 mice
   exhibited no significant changes in these splenocyte cytokines. Within
   nasal mucosa, BALB/c mice displayed polymorphonuclear cell infiltration
   and significantly elevated eotaxin levels, which subsequently
   stabilized, alongside significant upregulation of TGF-β and IL-10
   expression. At 15th week, C57BL/6 mice demonstrated significantly
   higher nasal PMN infiltration and eotaxin levels compared to BALB/c
   mice, but showed no significant increase in TGF-β or IL-10 compared to
   controls. RNA-seq analysis of nasal mucosa revealed that BALB/c mice at
   15th week exhibited significant upregulation of genes involved in
   biological processes, tolerance-related signaling pathways, and
   negative regulatory pathways. Conversely, C57BL/6 mice showed
   significant upregulation of genes associated with cell synthesis- and
   secretion-related pathways.

Conclusion

   Based on the criteria defining “local tolerance” in this
   model—significant symptom attenuation despite allergen escalation
   coupled with downregulation of nasal inflammatory markers (eotaxin,
   polymorphonuclear cell infiltration)—local tolerance was successfully
   induced in BALB/c mice by long-term OVA stimulation, but not in C57BL/6
   mice. The normal function of T regulatory cells is key to establishing
   local tolerance.

   Keywords: Allergic rhinitis, Animal model, Local tolerance, IL-10,
   TGF-β, RNA-seq, Eotaxin, BALB/c, C57BL/6, IgE

Introduction

   Allergic rhinitis (AR) is a common respiratory disease. Epidemiological
   studies have shown that the prevalence of AR is increasing every year
   and poses a burden to society.[38]^1 BALB/c and C57BL/6 mouse strains
   are 2 commonly used animal models for the study of AR. Our previous
   study established stable animal models of local sensitizing AR in
   BALB/c and C57BL/6 mice by nasal allergen drip and found significant
   differences in immune responses between the 2 mice; that is, BALB/c
   mice showed relatively weak local allergic inflammatory responses and
   strong systemic immune responses, whereas C57BL/6 mice showed
   relatively strong local allergic inflammation and relatively mild
   systemic immune responses.[39]^2

   Extensive clinical studies have established that subcutaneous
   immunotherapy (SCIT) with grass pollen allergens induces immune
   tolerance in humans, manifesting as significant improvement in
   nasal/ocular symptoms, enhanced quality of life, prolonged symptom
   remission after treatment cessation, and elevated tolerance thresholds
   in nasal provocation tests.[40]^3 Mechanistically building upon these
   clinical observations, we defined “local tolerance” in our murine model
   as: (1) significant symptom attenuation (eg, reduced sneezing
   frequency) despite progressive allergen dose escalation, coupled with
   (2) downregulation of inflammatory markers (eotaxin and
   polymorphonuclear cells infiltration)—criteria systematically validated
   through histopathological and transcriptomic (RNA-seq) analyses. Our
   experimental results demonstrated that BALB/c mice progressively
   developed this local tolerance state, showing gradual symptom
   resolution approaching baseline levels following prolonged nasal
   allergen exposure, along with upregulated systemic immune tolerance
   responses. In striking contrast, C57BL/6 mice subjected to identical
   protocols failed to establish either local or systemic tolerance,
   maintaining persistent allergic inflammation throughout the observation
   period. Alan et al used enterohemorrhagic Escherichia coli to infect
   BALB/c and C57BL/6 mice. They found that C57BL/6 mice had a worse
   prognosis than BALB/c mice, with severe intestinal damage, epithelial
   barrier dysfunction, and impaired renal function, leading to an
   increased percentage of their deaths, demonstrating that the high
   survival rate of BALB/c mice is associated with the immune tolerance
   mechanism.[41]^4

   Studies in humans and mice suggest that regulatory T cells (Tregs),
   regulatory antigen-presenting cells (APCs), other suppressor cells, and
   cytokines make important contributions to the induction and maintenance
   of immune tolerance.[42]5, [43]6, [44]7 For example, dendritic cells
   (DCs), which are specialized APCs, are the initial cells in asthma and
   other types of allergic inflammation. DCs can induce both allergic
   responses and immune tolerance.[45]^8^,[46]^9 Treg cells play a crucial
   role in immune tolerance by producing cytokines such as IL-10 and
   TGF-β, and expressing suppressive surface molecules, including the
   inhibitory immune checkpoint molecules cytotoxic
   T-lymphocyte-associated protein 4 (CTLA-4) and programmed cell death
   protein 1 (PD-1). CTLA-4, a competitive homolog of CD28, binds CD28
   ligands CD80/CD86 to inhibit T cell activation and mediates their
   endocytosis. PD-1 binds its ligand PD-L1 to negatively regulate T cell
   activation. Both CTLA-4 and PD-1 expression depend on TCR activation,
   and their intracellular signaling via the phosphatase Src homology
   region 2-containing protein tyrosine phosphatase (SHP-2) inhibits
   downstream PI3K signaling. Through these distinct mechanisms, CTLA-4
   and PD-1 suppress T cell activation and promote exhaustion. These
   molecules are essential for negatively regulating immune responses to
   prevent excessive activation and autoimmunity, while also serving as
   critical targets for immune checkpoint therapies.[47]10, [48]11,
   [49]12, [50]13, [51]14, [52]15, [53]16 However, the mechanisms
   underlying immune tolerance remain relatively unknown.

   In this study, we administered a specific dose of ovalbumin (OVA) to
   induce long-term continuous nasal drip in both mouse strains to
   systematically compare the immunological differences, particularly
   those related to immune tolerance, and to provide further evidence on
   the mechanisms of local and systemic immune tolerance.

Materials and methods

Experimental animals

   Forty-eight female mice (24 BALB/c and 24C57BL/6), aged 5 weeks, SPF
   grade, were purchased from Beijing Vital River Laboratory Animal
   Technology Co., Ltd. This study protocol was reviewed and approved by
   the Committee on the Ethics of Animal Experiments of Huazhong
   University of Science and Technology, Wuhan, China, approval number
   2844.

Model construction

   BALB/c and C57BL/6 mice were randomly divided into 4 groups: BALB/c PBS
   group, BALB/c OVA group, C57BL/6 PBS group, and C57BL/6 OVA group. Each
   group contained 12 animals. The mice were dripped with phosphate
   buffered solution (PBS) or 25 mg/ml OVA daily, 10ul on each side of the
   nasal cavity. Half of the mice in each group were sacrificed on 8th
   week, and the remaining mice were sacrificed on 15th week, ([54]Fig.
   1). Finally, 8 subgroups of mice were formed, ie, 8th BALB/c PBS group,
   8th BALB/c OVA group, 8th C57BL/6 PBS, 8th C57BL/6 OVA, 15th BALB/c
   PBS, 15th BALB/c OVA, 15th C57BL/6 PBS, and 15th C57BL/6 OVA groups.

Fig. 1.

   [55]Fig. 1
   [56]Open in a new tab

   General flow diagram of experiment.

   According to a previous study, the symptoms of BALB/c mice peak at the
   8th week, and the symptoms of C57BL/6 mice also peak at 8th∼9th week.
   Serum-specific antibodies and localized nasal mucosal polymorphonuclear
   cells in both mice also increased significantly.[57]^2 At the 15th
   week, the symptoms of the BALB/c mice were greatly relieved and
   approached normal. However, the symptoms in C57BL/6 mice were still
   severe, and the risk of death in C57BL/6 mice increases with time.
   Therefore, we chose the time points at which the mice were sacrificed
   at the 8th week and 15th week.

Allergic symptoms

   Sneezing was the most prominent symptom; therefore, we counted the
   number of sneezes to reflect the severity of nasal symptoms. We
   recorded the number of sneezes within 10 min after nasal drip on the
   first day at 0th, 4th, 8th, 12th, and 15th week.

Detection of serum OVA-specific antibody levels

   Blood was collected from the tail vein of the mice every 3 weeks and
   within 24h after the last nasal drip at the 15th week. The blood was
   centrifuged at 2500 rpm for 5 min by ultracentrifugation, and the upper
   layer of yellowish serum was stored at −20 °C for examination. The OD
   values of OVA-specific IgE, IgG1, and IgG2a were determined using ELISA
   (eBioscience, San Diego, CA, USA).

Splenocyte culture and cytokine level assay

   After deep anesthesia with isoflurane, the mice were treated with
   cervical dislocation and disinfected with 75% alcohol. The spleens were
   removed from mice on an ultra-clean bench (Aintech, Suzhou, China) and
   ground into centrifuge tubes using RPMI-1640 (Hyclone, Beijing, China)
   with a 200-mesh filter. After centrifugation for 10 min, the
   supernatant was discarded, the cell precipitate was resuspended in
   RPMI-1640, and lymphocytes were isolated using a Mouse Spleen
   Lymphocyte Isolation Kit (TBDscience, Tianjin, China). Lymphocytes
   (5∗10^6/ml) were cultured in culture medium containing 20% serum, 1%
   double antibody, and 30ug/ml OVA. Splenocyte supernatants were
   collected after 72h in an incubator and stored at −20 °C. IL-4, IFN-γ,
   and IL-10 levels in splenocyte supernatants were measured using ELISA
   (R&D Systems, Minneapolis, MN, USA). The OD values at 450 nm were
   measured using an automated enzyme marker (BioTek, Beijing, China) and
   the concentrations were analyzed.

Histology of the nasal mucosa

   The nasal mucosa was extracted according to the Dunston method.[58]^17
   The skin was removed. The muscles and jaws around the nose were cut off
   with scissors, and the nasal bone on one side was cut slowly until the
   nasal mucosa was completely exposed. The nasal mucosa was quickly
   placed in a sterile centrifuge tube and stored at −80 °C for RNA
   sequencing. The unexposed nasal mucosa on the other side was fixed in
   paraformaldehyde overnight. The nasal mucosa was removed the following
   day for decalcification in 10% EDTA for 1 week. The nasal mucosa was
   embedded in paraffin wax. HE and PAS staining were performed after
   parallel sectioning and roasting. Under light microscopy (1000 × ), the
   morphological structure of nasal mucosal tissue and the infiltration of
   polymorphonuclear cells and goblet cells were observed. Assessments
   were performed using a blind method. Five fields of view were counted,
   and the average value of these 5 fields was calculated. Goblet cells
   were graded according to Judith A:[59]^18 grade 0: Absence of
   PAS-positive staining; grade 1: Focal PAS positivity (≤25% epithelial
   area); grade 2: Moderate PAS positivity (26–50% epithelial area); grade
   3: Extensive PAS positivity (51–75% epithelial area); grade 4: Diffuse
   PAS positivity (>75% epithelial area).

Immunohistochemical staining

   Paraffin sections of the nasal mucosa were sequentially dewaxed,
   rehydrated, and antigen-repaired. The sections were then treated with a
   blocking solution containing 3% bovine serum albumin for 45 min.
   Sections were incubated overnight at 4 °C with TGF-β antibody (Abcam,
   Shanghai, China), IL-10 antibody (R&D Systems, Minneapolis, MN, USA),
   and eotaxin antibody (Proteintech, Wuhan, China) and then incubated
   with anti-rabbit IgG-HRP (Boster, Wuhan, China), followed by DAB color
   development. Subsequently, hematoxylin was re-stained, dehydrated, made
   transparent, sealed, and placed under a light microscope for
   examination. Expression levels of eotaxin, IL-10, and TGF-β were
   quantified using IOD/area (integrated optical density per unit area)
   analysis.

Gene sequencing

   Lyophilized nasal mucosa was subjected to RNA sequencing analysis.
   Total RNA was extracted using Trizol reagent and subjected to DNA
   digestion. RNA quality was determined using a NanodropTM OneC
   spectrophotometer (Thermo Fisher Science Inc., USA), and RNA integrity
   was confirmed using 1.5% agarose gel electrophoresis. Qualified RNA was
   quantified using the Qubit 3.0. RNA sequencing libraries were prepared
   using a Ribooff rRNA Depletion Kit (human/mouse/rat). Sequencing was
   performed using a DNBSEQ-T7 sequencer (MGI Tech Co., Ltd., China).

   After the raw data were filtered, genes differentially expressed
   between groups were identified using the edgeR package (3.12.1), and P
   < 0.05 and logFC cutoff values were set to 2 to determine the
   statistical significance of gene expression differences. This stringent
   cutoff was selected to prioritize biologically significant changes, as
   a logFC of 2 represents a 4-fold change in expression, reducing false
   positives while capturing key regulatory genes in tolerance pathways.
   Gene ontology (GO) analysis, Kyoto Encyclopedia of Genes and Genomes
   (KEGG) analysis, and Gene Set Enrichment Analysis (GSEA) of
   differentially expressed genes were performed using KOBAS software
   (version 2.1.1). Differentially expressed genes were entered into the
   search tool for the retrieval of interacting genes (STRING) database to
   construct a protein-protein interaction (PPI) network of differentially
   expressed genes. The CytoHubba plug-in in the Cytoscape software was
   used to calculate the connectivity of protein nodes and to screen for
   significantly different proteomic modules. The top 10 genes were
   identified as core genes that were considered statistically
   significant.

Statistical analysis

   GraphPad Prism 9.0 and SPSS 22 (IBM, Armonk, NY) were used for
   statistical analysis. Allergic symptoms, serum-specific antibodies,
   cytokines in the splenic culture supernatant, and the number of
   polymorphonuclear cells were expressed as mean ± standard deviation
   (SD). Tukey's multiple comparison method in one-way ANOVA was used to
   assess the statistical significance of symptoms, serum antibodies,
   cytokine concentrations, polymorphonuclear cell and goblet cell numbers
   and immunohistochemical indices. The sample sizes required for each
   experiment are shown in [60]Supplemental Table 1. P < 0.05 was
   considered statistically significant.

Results

Nasal symptoms

   The BALB/c PBS and C57BL/6 PBS groups maintained low symptom levels
   throughout the study. In contrast, both OVA groups (BALB/c and C57BL/6)
   developed symptoms starting in 4th week. Within the BALB/c OVA group,
   nasal symptoms peaked during 7-8th weeks (0th week vs. 8th week, p <
   0.01) and then decreased significantly, remaining lower from 10th week
   to 15th week (8th week vs 15th week, p < 0.05). Conversely, the C57BL/6
   OVA group reached peak symptoms during 8-9th week (0th week vs. 8th
   week, p < 0.01) and exhibited a significant increase thereafter (8th
   week vs 15th week, p < 0.01), with symptoms persisting near peak
   levels. At 15th week, sneezing was significantly higher in C57BL/6 OVA
   mice compared to BALB/c OVA mice (p < 0.01) ([61]Fig. 2A).
   [62]Supplementary Table 2 contains detailed within- and between-group
   comparison results. [63]Fig. 2B specifically illustrates the following
   OVA group comparisons: 0th week versus 8th week and 8th week versus
   15th week (within each strain), along with the 15th week comparison
   between BALB/c OVA and C57BL/6 OVA groups.

Fig. 2.

   [64]Fig. 2
   [65]Open in a new tab

   (A) Trends in the number of sneezes at 0th week, 4th week to 15th week
   for each group. (B) Comparisons of OVA groups: 0th week vs. 8th week
   and 8th week vs. 15th week (within each strain), and the 15th week
   comparison between BALB/c OVA and C57BL/6 OVA groups. n = 12 for each
   group in the first 8 weeks and n = 6 for each group from 9th week to
   15th week. Data represent mean ± SD. Within-group comparisons: ∗, P <
   0.05. ∗∗, P < 0.01

Serum OVA-specific antibodies

   From the 0th week to the 15th week, OVA-specific IgE, IgG1, and IgG2a
   in the BALB/c OVA group showed an increasing trend over time. In the
   C57BL/6 OVA group, OVA-specific IgE tended to increase with longer
   nasal drip time, sIgG1 peaked at 12th week and remained stable at 15th
   week, while IgG2a did not show significant changes throughout the
   course. Serum antibody concentrations were higher in the BALB/c OVA
   group than in the C57BL/6 OVA group at all time points ([66]Fig. 3).

Fig. 3.

   [67]Fig. 3
   [68]Open in a new tab

   Dynamics of OVA-specific antibody responses in BALB/c and C57BL/6 mice
   following intranasal OVA challenge. Serum levels of IgE, IgG1, and
   IgG2a antibodies reflect the immunological response: IgE and IgG1
   represent Th2-type immune responses, while IgG2a indicates a Th1-type
   immune response. Data are presented as mean ± SD. Sample size: n = 12
   per group during the initial 8 weeks, subsequently reduced to n = 6 per
   group from the 9th to 15th week

Cytokines in splenic culture supernatants

   In the BALB/c OVA group, the concentrations of IL-4 and IL-10 were
   significantly higher in splenic culture supernatants at 8th week
   compared than in the BALB/c PBS group. The IL-4 concentration was lower
   and the IL-10 concentration was higher in the 15th BALB/c OVA group
   than in the 8th BALB/c OVA group. The concentrations of IL-4, IL-10,
   and IFN-γ were not changed in the C57BL/6 OVA group at 8th and 15th
   week compared to in the C57BL/6 PBS group ([69]Fig. 4).

Fig. 4.

   [70]Fig. 4
   [71]Open in a new tab

   Cytokines (IL-4, IFN-γ, IL-10) in splenocyte supernatants. Subgroups:
   8th BALB/c PBS group, 8th BALB/c OVA group, 8th C57BL/6 PBS, 8th
   C57BL/6 OVA, 15th BALB/c PBS, 15th BALB/c OVA, 15th C57BL/6 PBS, and
   15th C57BL/6 OVA groups. ∗P < 0.05, ∗∗P < 0.01: Data: mean ± SD

HE and PAS staining of the nasal mucosa

   Nasal polymorphonuclear cell infiltration was observed in both 8th
   BALB/c OVA group and 8th C57BL/6 OVA groups. There was no significant
   difference in polymorphonuclear cell numbers between 8th BALB/c OVA
   group and 15th BALB/c OVA groups. A significant increase in
   polymorphonuclear cell number was observed in the 15th C57BL/6 OVA
   group compared to that in the 8th C57BL/6 OVA group.

   PAS results showed a significant increase in the number of nasal goblet
   cells in both the 8th BALB/c OVA group and the 8th C57BL/6 OVA group.
   There was a significant decrease in the goblet cell number in the 15th
   BALB/c OVA group compared with the 8th BALB/c OVA group, and a
   continued increase in goblet cells in the 15th C57BL/6 OVA group
   compared to the 8th C57BL/6 OVA group. At both time points of 8th week
   and 15th week, the number of goblet cells in the BALB/c OVA group was
   lower than that in the C57BL/6 OVA group ([72]Fig. 5).

Fig. 5.

   [73]Fig. 5
   [74]Open in a new tab

   Histopathological staining of mouse nasal mucosa. Subgroups: (1) 8th
   BALB/c PBS, (2) 8th BALB/c OVA, (3) 8th C57BL/6 PBS, (4) 8th C57BL/6
   OVA, (5) 15th BALB/c PBS, (6) 15th BALB/c OVA, (7) 15th C57BL/6 PBS,
   (8) 15th C57BL/6 OVA. (A) H&E staining: Representative images (1–8)
   showing polymorphonuclear cell infiltration (arrows). (B) PAS staining:
   Representative images (1–8) demonstrating goblet cell distribution
   (arrows). (C) Polymorphonuclear cell counts: Mean values per 5
   high-power fields (HPF). (D) Goblet cell grading: Mean scores per 5
   HPF. Data presented as mean ± SD; ∗P < 0.05, ∗∗P < 0.01. n = 6
   mice/group; 5 fields/mouse; blinded analysis

Immunohistochemical staining of the nasal mucosa

   The expression of eotaxin was higher in the C57BL/6 OVA group than in
   the control group at both the 8th week and 15th week and was even
   higher at the 15th week than at the 8th week. There was an increasing
   trend in eotaxin expression in the 8th BALB/c OVA group, but the
   difference was not statistically significant. Eotaxin staining
   intensity was higher in C57BL/6 nasal mucosa than that in BALB/c mice
   at all time points ([75]Fig. 6A and D).

Fig. 6.

   [76]Fig. 6
   [77]Open in a new tab

   Nasal mucosa immunohistochemistry. Subgroups: (1) 8th BALB/c PBS, (2)
   8th BALB/c OVA, (3) 8th C57BL/6 PBS, (4) 8th C57BL/6 OVA, (5) 15th
   BALB/c PBS, (6) 15th BALB/c OVA, (7) 15th C57BL/6 PBS, (8) 15th C57BL/6
   OVA. (A) Eotaxin: Representative images (1–8) per subgroup. (B) TGF-β:
   Representative images (1–8) per subgroup. (C) IL-10: Representative
   images (1–8) per subgroup. (D–F) Eotaxin, TGF-β, IL-10 expression in
   each group. Quantified expression: IOD/area analysis (positive
   pixels/field). Data presented as mean ± SD; ∗P < 0.05, ∗∗P < 0.01. n =
   3 mice/group; blinded analysis

   No significant changes in TGF-β1 and IL-10 were observed in the nasal
   mucosa of the BALB/c or C57BL/6 OVA groups at 8th week. TGF-β1 and
   IL-10 expression was significantly increased in the 15th BALB/c OVA
   group. In contrast, there was no significant increase in TGF-β1 and
   IL-10 expression in the C57BL/6 OVA group at any time point. ([78]Fig.
   6B,C,6E,6F).

Nasal mucosal gene sequencing

   We have previously studied the genetic alterations in the nasal mucosa
   of BALB/c and C57BL/6 mice in the 8th week.[79]^2 In this study, we
   focused on nasal mucosal gene alterations at 15th week. We found 3903
   differentially expressed genes in the 15th BALB/c OVA group compared
   with the control, including 2063 upregulated and 1840 downregulated
   genes. GO-KEGG analysis showed significant enrichment in intrinsic
   immune response, immune system response, and inflammatory response.
   There were 225 differential genes in the 15th C57BL/6 OVA group
   compared to the control group, including 83 upregulated genes and 142
   downregulated genes. GO-KEGG analysis showed significant enrichment in
   the acute inflammatory response and B-cell proliferation.

   When comparing the 15th BALB/c OVA group and the 15th C57BL/6 OVA
   group, there were 4299 differential genes, including 2354 upregulated
   genes and 1945 downregulated genes. GO-KEGG analysis showed that
   pathways of BALB/c mice were significantly enriched in the T-cell
   receptor signaling pathway, B-cell receptor signaling pathway, cytokine
   receptor interaction, TGF-β signaling pathway, Fc epsilon RI signaling
   pathway, TH1 and TH2 differentiation, and antigen presentation
   pathways. Pathways of C57BL/6 mice were significantly enriched in the
   HIF-1 signaling pathway, PI3K-Akt signaling pathway, calcium signaling
   pathway, ABC transporter protein, cAMP signaling pathway, protein
   processing in the endoplasmic reticulum, amino acid biosynthesis,
   protein export, and RNA transport response pathway. GSEA results showed
   that BALB/c mice were significantly enriched in the TGF-β pathway and
   negatively regulated gene expression.

   We constructed a PPI network of differentially expressed genes between
   the 15th BALB/c OVA group and the 15th C57BL/6 OVA group, which
   generated 8848 nodes and 99525 edges, and obtained the top 10 HUB
   nodes: Rps27a, Uba52, Ubc, Kng2, Gnal, C3, Gng7, Gnb3, Reep5, and Reep6
   ([80]Fig. 7).

Fig. 7.

   [81]Fig. 7
   [82]Open in a new tab

   Nasal mucosa transcriptomics. (A) Volcano plots of DEGs: (a) 15th
   BALB/c OVA vs. 15th BALB/c PBS; (b) 15th C57BL/6 OVA vs. 15th C57BL/6
   PBS; (c) 15th BALB/c OVA vs. 15th C57BL/6 OVA. Upregulated genes are
   shown in red, downregulated in blue; significance thresholds: |logFC| >
   2, adjusted P < 0.05. (B) GO-KEGG pathway enrichment: (a) Top 20
   immune-related upregulated pathways in BALB/c and C57BL/6 OVA groups
   compared to their respective PBS controls, ranked by P-value; (b)
   Comparison between BALB/c and C57BL/6 OVA groups, highlighting their
   respective upregulated pathways (|logFC| > 2, adj. P < 0.05). (C) GSEA
   pathway enrichment analysis comparing BALB/c OVA and C57BL/6 OVA
   groups, focusing on pathways upregulated in BALB/c OVA (|logFC| > 2,
   adj. P < 0.05). (D) The top 10 core genes differentiating BALB/c and
   C57BL/6 OVA groups, ranked by Degree value; colors transition from dark
   to light to indicate ranking from top to lower

Discussion

   In this study, we compared local immune tolerance in BALB/c and C57BL/6
   mice, as well as systemic immune differences, through the use of
   long-term nasal OVA drop stimulation. BALB/c mice were found to develop
   local tolerance: nasal symptoms initially worsened and later improved;
   the systemic TH2 response was enhanced at first and weakened with time,
   and the Treg response was continuously enhanced. In contrast, C57BL/6
   mice did not develop a local tolerance response even with the
   prolongation of nasal drip time, and the local symptoms and
   inflammatory response continued to worsen and remain at a high level;
   the systemic immune response only maintained a certain degree of TH2
   response and the Treg response was not enhanced.

   BALB/c and C57BL/6 mice are 2 inbred strains commonly used in
   immunological studies. Based on the cytokine production tendency,
   BALB/c mice are thought to favor TH2-type responses (e.g., IL-4, IL-5,
   IL-10, and IL-13), whereas C57BL/6 mice favor TH1-type responses (TNF-α
   and IFN-γ).[83]^19 BALB/c mice are susceptible to viral infections and
   exhibit lower cytotoxic responses. When stimulated by allergens, the
   systemic humoral response is stronger, consistent with the cytokine
   profile. In contrast, C57BL/6 mice are more reactive to viral
   infection, exhibiting a higher cytotoxic response and a relatively
   weaker systemic allergic response.[84]^20^,[85]^21 In our previous
   study, we established a local nasal drip allergic rhinitis (AR) model
   in BALB/c and C57BL/6 mice. We observed notable differences in their
   immune responses: BALB/c mice exhibited relatively mild local allergic
   inflammatory responses but strong systemic allergic reactions.
   Conversely, C57BL/6 mice demonstrated more pronounced local allergic
   inflammation with milder systemic responses.[86]^2

   In this study, we observed that the concentrations of OVA-specific IgE,
   IgG1, and IgG2a in BALB/c mice increased with the duration of nasal
   drip exposure. Conversely, in C57BL/6 mice, OVA-specific IgE levels
   rose over time, sIgG1 initially increased and then plateaued at the
   12th week, while IgG2a levels remained unchanged throughout the entire
   course. Meanwhile, in BALB/c mice, the splenic culture supernatant
   results showed that IL-4 concentration increased at 8th week then
   decreased at 15th week, and IL-10 concentration continued to increase
   with longer nasal drip. In contrast, C57BL/6 mice showed no change in
   IL-4, IFN-γ, and IL-10 levels during the entire course of nasal OVA
   drip. These results suggest that when BALB/c mice were stimulated with
   a certain dose of allergen for a long time, their allergic reaction
   (TH2 function) was aggravated, but their TH1 and Treg functions also
   tended to increase with time and eventually suppressed TH2 function. In
   contrast, although the increase in serum sIgE (TH2 function) in C57BL/6
   mice was lower than that in BALB/c mice, Treg function was not
   significantly enhanced with time, so the overall response was allergic.
   Some reports have focused on the immunity and tolerance between BALB/c
   and C57BL/6 mice. For example, Alan et al. found that BALB/c mice
   produced more specific antibodies than C57BL/6 mice and showed more
   significant tolerance to enterohemorrhagic Escherichia coli.[87]^4 Gera
   et al. demonstrated that BALB/c mice were more tolerant to colitis than
   C57BL/6 mice, possibly because of enhanced BALB/c retinoic acid
   signaling and the consequent increased ability to fight
   mucosa-associated lesions.[88]22, [89]23, [90]24, [91]25, [92]26,
   [93]27 Chen et al. found that BALB/c thymus and peripheral lymphoid
   organs had more Treg cells and showed a more potent suppressive effect
   compared to C57BL/6 mice.[94]^28 In a study by Ki-Il et al., it was
   also found that IL-4, IFN-γ, and IL-10 were not significantly increased
   in the spleen cell culture supernatant of C57BL/6 mice sensitized by
   dust mites, and only IL-6 was mildly increased. In contrast, all of the
   above cytokines were significantly increased in BALB/c mice, showing
   active TH1 responses, TH2 responses, and Treg responses.[95]^29

   In this study, we found that polymorphonuclear cells (neutrophils and
   eosinophils are difficult to distinguish using HE staining alone; their
   differentiation mainly relies on morphological characteristics and
   additional markers. Based on other research findings, we speculate that
   these cells are likely eosinophils) and goblet cells increased in the
   nasal mucosa of BALB/c mice at 8th week, while the number of goblet
   cells decreased and the number of polymorphonuclear cells stopped
   increasing at 15th week. The expression of local IL-10 and TGF-β also
   increased over time, showing the characteristics of local tolerance. In
   C57BL/6 mice, local polymorphonuclear cells and goblet cells increased
   continuously from 8th week to 15th week, and eotaxin expression
   increased significantly. However, the local IL-10 and TGF-β levels did
   not change throughout the course. As IL-10 and TGF-β are important
   cytokines for immune tolerance, the lack of IL-10 and TGF-β may
   influence the formation of local tolerance.[96]^30

   Our previous research focused on comparing the gene expression profiles
   of the nasal mucosa in BALB/c and C57BL/6 mice at week 8, during the
   peak of inflammation, following OVA stimulation. The results showed
   that, compared to BALB/c mice, C57BL/6 mice exhibited a significantly
   enhanced activation of pro-inflammatory pathways, particularly the
   JNK/MAPKK signaling pathway.[97]^31 In this tolerance study, we
   examined gene expression changes at week 15, during the tolerance
   phase. The findings indicated that BALB/c mice could develop local
   immune tolerance after prolonged OVA exposure, with a more active local
   immune response evidenced by the presence of 3903 activated genes. In
   contrast, the C57BL/6 mice had only 225 activated genes, much fewer
   than the BALB/c mice. This difference suggests that local tolerance in
   BALB/c mice may be an actively regulated immune process. C57BL/6 mice
   displayed weaker Th2 responses at week 8, the lack of effective
   regulatory and inhibitory mechanisms led to persistent inflammation
   symptoms. In the gene analysis at week 15, we observed significant
   differences between BALB/c and C57BL/6 mice in the classical Wnt
   signaling pathway. Wnt signaling is known to negatively regulate
   Tregs.[98]^32 And additionally, the negative regulation of c-Jun
   N-terminal kinase (JUN kinase) activity was more pronounced in BALB/c
   mice. Since the JUN kinase pathway is positively involved in AR, the
   combined negative regulation of these 2 pathways may be a key mechanism
   underlying immune tolerance in BALB/c mice.

   To further explore these differences, we constructed a PPI network
   based on the differentially expressed genes (DEGs) between the BALB/c
   and C57BL/6 OVA groups at week 15. The top 10 hub nodes identified
   include: ribosomal protein S27a (Rps27a), ubiquitin A-52 amino acid
   fusion protein (Uba52), ubiquitin C (Ubc), kininogen 2 (Kng2), cyclic
   nucleotide-gated channel subunit alpha (Gnal), complement component 3
   (C3), cyclic nucleotide-gated channel subunit gamma 7 (Gng7), GNB3,
   receptor expression-enhancing protein 5 (Reep5), and Reep6.Recent
   reports have found that Rps27a is involved in protein synthesis,
   post-translational modification of proteins, and transcription of
   genes, and is highly expressed in tumor tissues such as breast
   fibroids, colorectal cancer and kidney.[99]^33^,[100]^34 Ubiquitin
   ribosomal fusion protein (UBA52) is sheared into ubiquitin molecules
   and ribosomal protein L40 by the action of cell-intrinsic enzymes,
   where ubiquitin molecules are mainly involved in ubiquitination
   processes in vivo, such as the degradation of proteins, regulation of
   the TGF-β pathway and immune response.[101]^35^,[102]^36 UBC genes play
   key roles in maintaining ubiquitin (Ub) homeostasis.[103]^37 CAMP
   signaling induces kininogen 2 (Kng2) production in brown adipose
   tissue, which inhibits brown adipose tissue thermogenesis and may be
   related to local metabolic function.[104]^38 Gnal encodes the
   stimulatory G protein Gαolf, which is essential for activating the cAMP
   pathway in striatal projection neurons and is associated with dystonia
   and hyperactivity disorders.[105]^39^,[106]^40 G protein subunit γ7
   (Gng7), a member of the G protein family, was moderately to highly
   correlated with the degree of infiltration of CD4^+ T cells in the
   peripheral blood of patients with colorectal cancer. Additionally, GNG7
   expression positively correlated with the degree of infiltration of B
   cells, macrophages, neutrophils and DCs.[107]^41 Gnb3 encodes the G
   protein β3 subunit and is associated with obesity, diabetes, depression
   and tumors.[108]42, [109]43, [110]44 REEP5 and REEP6 play important
   roles in IL-8-stimulated activation of CXCR1.[111]^45 CXCR1 can be used
   as a marker of AR.[112]^46 Liliane et al. found that C3 induces mice to
   form a peptide-mediated skin tolerance model by regulating DC function,
   and that C3-deficient mice fail to produce Treg cells required for the
   induction of tolerance.[113]^47 While current literature implicates C3
   and UBA52 as potential mediators of local immune tolerance, definitive
   confirmation of their functional significance awaits experimental
   validation via CRISPR-based gene editing or antibody neutralization
   studies.

   While this study elucidates tolerance mechanisms in specific mouse
   strains, it is crucial to acknowledge the inherent limitations of the
   mouse model. These limitations include differences in the immune
   systems between species, variations in nasal cavity anatomy, and the
   complexity of the immune environment—all of which are significantly
   less intricate than in humans. Consequently, these factors restrict the
   direct extrapolation of our findings to human conditions. Moreover, the
   long-term nasal drip model cannot fully replicate the multifactorial
   immune regulation and chronic disease states influenced by multiple
   factors in humans. This is partly because clinical allergen
   immunotherapy (AIT) employs standardized doses of allergens combined
   with adjuvants, whereas our model uses only a single antigen OVA
   without adjuvants. Additionally, human AIT is typically administered
   via subcutaneous or sublingual routes to induce systemic immune
   modulation, while the localized nasal delivery in mice mainly aims to
   mimic mucosal responses. These differences suggest that the immune
   tolerance mechanisms observed in specific mouse strains should not be
   directly generalized to human immunotherapy. Our conclusions should
   consider the study design. Sacrificing mice in the 8th and 15th weeks
   for tissue analysis reduced group sizes from 12 to 6 after 8th week.
   While enabling critical time-point analyses, the smaller sample in
   later phases may reduce sensitivity to detect longitudinal changes,
   particularly in symptom-variable C57BL/6 mice. However, key strain
   differences in tolerance (e.g., IL-10/TGF-β dynamics) showed strong
   effects (p < 0.01) and were confirmed by histology and RNA-seq. Larger
   terminal cohorts in future studies would strengthen these findings. Our
   study mainly monitored immune changes and clinical manifestations over
   15 weeks; thus, we cannot exclude the possibility of late-phase
   symptoms. Future studies will extend observation periods to achieve a
   more comprehensive understanding of the dynamic immune responses.

Conclusion

   In summary, prolonged OVA exposure induces local tolerance in BALB/c
   mice but not in C57BL/6 mice, mainly due to strain-specific differences
   in Treg functionality. The tolerance observed in BALB/c mice is
   associated with dynamic upregulation of IL-10 and TGF-β, as well as
   negative regulation of pro-inflammatory pathways such as Wnt and JUN.
   In contrast, C57BL/6 mice fail to exhibit similar regulatory responses,
   leading to immune dysregulation. Our study enhances understanding of
   the systemic and local immune differences between these 2 strains and
   emphasizes the advantage of BALB/c mice in inducing Tregs, which aligns
   with the response characteristics seen in human AIT, confirming their
   utility in studying tolerance mechanisms. Conversely, C57BL/6 mice
   exhibit persistent inflammation, resembling severe, refractory human
   AR, which can be valuable for drug screening targeting
   difficult-to-treat conditions.

Data availability statement

   All relevant data are within the paper.

Authorship consent

   All authors have read and approved the manuscript for submission and
   publication in the World Allergy Organization Journal.

   Jianjun Chen PhD, designed and follow-up the whole study. Qidi Zhang,
   MD, completed most animal experiments. Wanting Zhu, MD, Zhixing Zou,
   MD. Ziyi Long, MD, assisted in some experiments and data collection.
   Wenting Yu, PhD, Pei Gao, PhD, Ying Wang, PhD helped with data analysis
   and grammar revising. All co-authors have approved the current work.

Ethics statement

   The experimental protocols and operations were conducted in accordance
   with the guidelines issued by the Experimental Animal Ethics Committee.

Declaration of Generative AI and AI-assisted technologies in the writing
process

   This manuscript used the generative AI tool deepseek for language
   polishing, solely to enhance text readability and expression. The tool
   was not involved in study design, data analysis, conclusion derivation,
   image processing, or generating any scientific/medical insights.

   Authors have rigorously reviewed and edited AI-generated content, and
   bear full responsibility for the manuscript's scientific validity and
   completeness.

   (Per journal policy, AI tools are not listed as authors.)

   Confirmation of unpublished work. But this research article was
   published on the preprint server (DOI:
   [114]10.21203/rs.3.rs-4279311/v1).

Funding

   This study is supported by National Natural Science Foundation of China
   (No. 82371124) and National Key R&D Program of China (2022YFC2504100).

Declaration of competing interest

   All authors have no conflict of interest in this paper.

Acknowledgment