Materials

Insulin–transferrin–sodium selenite media supplement, retinoic acid, hydrocortisone, gentamicin, epidermal growth factor (EGF) from murine submaxillary gland, avertin (tribromoethanol), pilocarpine, potassium chloride, magnesium chloride, sodium chloride, magnesium sulfate, potassium phosphate monobasic, HEPES, glucose, bovine serum albumin, calcium chloride dihydrate, collagenase, hyaluronidase, insulin and Fura-2 AM were purchased from MilliporeSigma (St. Louis, MO). Dulbecco’s Modified Eagle Medium/Nutrient Mixture F-12 (1:1) [DMEM/F12 (1:1)], penicillin–streptomycin solution, ethanol, formalin, fetal bovine serum (FBS), glutamine, rat anti-CD45 antibody (catalog number:14–0451-82) and E-Gel™ 2% agarose gels with SYBR™ Safe DNA gel stain were purchased from Thermo Fisher Scientific (Newington, NH). Mouse anti-E-cadherin antibody (catalog number: 610182) was purchased from BD Biosciences (San Jose, CA). DAPI, β-tubulin loading control (catalog number: MA5-16,308-1MG), rabbit anti-zonula occludens-1 (ZO-1) antibody (catalog number: 61–7300), Alexa Fluor 488-conjugated anti-rabbit IgG and Alexa Fluor 568-conjugated anti-mouse secondary antibodies (catalog numbers: A-11008 and A-11031) were purchased from Invitrogen (Carlsbad, CA). Alexa Fluor 568-conjugated anti-rat IgG (catalog number: ab175476), Alexa Fluor 488-conjugated anti-goat IgG (catalog number: ab150129), rabbit anti-Ki67 (catalog number: ab15580), mouse anti-Ki67 (catalog number: ab279653) and rabbit anti-AQP5 (catalog number: ab78486) antibodies were purchased from Abcam (Cambridge, MA). HRP-linked anti-rabbit IgG (catalog number: 7074S) and HRP-linked anti-mouse IgG (catalog number: 7076S) antibodies were purchased from Cell Signaling Technology (Danvers, MA). Goat anti-Mucin 10 (Muc10) antibody (catalog number: EB10617) was purchased from Everest Biotech (Oxfordshire, UK). MMK-1 was purchased from Apexbio Technology (Houston, TX). Boc-2 was purchased from MP Biomedicals (Santa Ana, CA). Finally, anti-human FPR2/ALX (extracellular) antibody (catalog number: AFR-002) was purchased from Alomone Labs (Jerusalem, Israel) while AT-RvD1 was purchased from Cayman Chemical (Ann Arbor, MI).

Human Subjects

Human SMG were obtained from the University of Missouri, Department of Otolaryngology, Head and Neck Surgery. Note that tissues were drawn from healthy SMG tissue from head and neck cancer patients undergoing neck dissections that would otherwise be discarded. Usage of all human specimens was conducted under the guidelines and with the approval of the University of Missouri Health Sciences Institutional Review Board (IRB), Project Number 2024461, Review Number 445894, Trial Number: Not Applicable, with informed consent obtained for each patient.

Experimental Animals

All experimental procedures were approved by the University of Missouri’s Institutional Animal Care and Use Committee Approval Number: 42063 and were conducted in accordance with the guidelines outlined in the Guide for the Care and Use of Laboratory Animals. To produce hFPR2 knock-in (KI) humanized mice (i.e., C57BL/6J-Fpr2emlOlgab), C57BL/6J embryo donor females (3 weeks of age) and stud males (10 weeks of age) were obtained from Jackson Laboratory while CD-1 surrogate females (8 weeks of age) were procured from Charles River. Control experiments utilized wild-type C57BL/6J mice acquired (6 weeks of age) also from Jackson Laboratory. Animals were kept in the vivarium for a 2-week acclimation period to alleviate stress before initiating experiments and were used at 8 weeks of age. All mice were housed in ventilated cages (Thoren) and maintained on a 12:12 light cycle with food and water provided ad libitum throughout the experiment.

Single Guide RNA (sgRNA) Design and Synthesis

The genomic sequence of Formyl Peptide Receptor 2 (Fpr2) in the C57BL/6J mouse was obtained from Ensembl.org (assembly GRCm39), while the CRISPR RGEN Tools website (Center for Genome Engineering Institute, South Korea) was used both to design sgRNAs flanking the mouse Fpr2 coding regions and to calculate off-target scores [10]. Furthermore, the CCTop website (Centre for Organismal Studies in Heidelberg) was used to predict the CRISPRater efficiency scores for each sgRNA [11]. Following the design process, the sgRNAs were chemically synthesized and modified using Synthego, a CRISPR design tool. Finally, chemical modifications included 2′-O-methyl analogs and 3′ phosphorothioate internucleotide linkages at the first three 5′ and 3′ terminal RNA residues (Fig. 1A).

Fig. 1

Generation and characterization of humanized FPR2 knock-in mice. A DNA repair template was used to replace the mouse Fpr2 coding region with the human FPR2 sequence, thereby integrating it at the endogenous ATG start site (A). Structural predictions of mouse Fpr2 (red) and human FPR2 (green) were generated using ColabFold v1.5.5 and visualized with the PyMOL Molecular Graphics System (version 3.0). Ligand (AT-RvD1, orange molecule) interactions were modeled using the Molecular Operating Environment (MOE) software. The resulting models revealed overall structural similarity between the two receptors but likewise revealed differences in ligand binding site composition (B). Successful replacement of the mouse Fpr2 gene with the human FPR2 coding sequence was confirmed by PCR performed on cDNA synthesized from total RNA extracted from wild-type mice, humanized mice and human SMG tissue with the wild-type transcript (146 bp) amplified using a forward primer (WT primer F1) and a reverse primer (WT primer R), while the humanized transcript (243 bp) was detected using a knock-in-specific forward primer (WT primer F2) and a reverse primer located within the human FPR2 sequence (KI primer R) (C). Expression of the human FPR2 protein was confirmed by Western blot analysis using an antibody that selectively binds human FPR2. The protein was detected in humanized mouse and human SMG but not in wild-type mouse SMG, with β-tubulin used as a loading control in each of these cases (D). Functional expression of hFPR2 was validated by calcium imaging. SMG cells were loaded with Fura-2 AM and stimulated with AT-RvD1 (100 ng/mL). Intracellular calcium levels ([Ca2+]i) increased rapidly upon stimulation, indicating intact receptor function. Data shown are representative of n = 12 (E)

Repair Template Design and Chemical Synthesis

For CRISPR-mediated knock-in, a repair template was designed to replace the mouse Fpr2 coding region with the human FPR2 coding sequence, thereby integrating the human coding sequence at the native ATG start site. The repair template was synthesized by GeneWiz and inserted into the pUC-GW-Amp plasmid backbone and consisted of dissimilar homology arms measuring 1,180 bp and 748 bp, respectively.

Microinjection and Embryo Transfer

A microinjection mixture consisting of 50 ng/μL for each sgRNA, 50 ng/μL of enhanced-specificity Cas9 protein and 10 ng/μL of circular dsDNA repair template was prepared. CRISPR sgRNA/Cas9 ribonucleoprotein (RNP) complexes were then formed by combining the sgRNAs and Cas9 protein, followed by incubation at room temperature for 10 min. Subsequently, the DNA repair template was introduced into the mixture. Finally, zygotes with injected pronuclei were surgically transferred to pseudopregnant surrogate females on the same day as the microinjection, with offspring naturally delivered and genotyped at 3 weeks of age.

CRISPR Off-Target Analysis

An enhanced-specificity Cas9 (eSPCas9) protein was utilized when making this mouse model to reduce the potential for off-target effects. The eSPCas9 variant has been shown to significantly reduce CRISPR off-targeting [12]. The method used for examining CRISPR off-target sites utilized the CRISPR RGEN Tools website maintained by the Center for Genome Engineering Institute (Korea) and the CCTop website maintained by the Centre for Organismal Studies (Heidelberg) to calculate off-target scores. Any predicted off-target site with less than a 2 bp mismatch (including DNA or RNA bulges) or with less than 3 bp mismatches if no mismatches are in the 12 bp seed region of the sgRNA was PCR amplified and sequenced to ensure no erroneous edits were made. The sequence was aligned with the reference C57BL/6J genomic assembly. Only one location in the mouse genome matched our off-target selection criteria. Sanger sequencing showed no erroneous edits were made in this location (Supplementary Fig. 1).

Genotyping

Tail biopsy samples were collected from 3-week-old mice and genomic DNA extracted using a Qiagen DNeasy Blood and Tissue Kit. Genotyping was performed using knock-in-specific primers (forward primer 5′-GCTATGCTACCCCAGAAAGG-3′ binding to the mouse Fpr2 flanking sequence and reverse primer 5′-GCAGAACAGTGTAGCCAGCA-3′ binding to the human FPR2 coding region), resulting in a 243 bp amplicon. PCR amplification was performed under the following conditions: a) 95 °C for 3 min, b) 95 °C for 30 s, c) 61 °C for 30 s, d) 72 °C for 1 min (steps 2–4 repeated 35 times), e) 72 °C for 7 min and f) hold at 4 °C. PCR products were analyzed using a QIAxcel Advanced capillary electrophoresis system with a 15 bp-3 kb alignment marker (denoted by green lines, Supplementary Fig. 2) and a QX DNA size marker. To further validate the genotype and gene expression patterns, SMG were harvested after euthanasia and total RNA was isolated, reverse transcribed into cDNA and PCR was performed on samples from: a) wild-type mice, b) humanized mice, and c) human SMG tissues. Both wild-type-specific primers (forward 5′-GAGAGCCCTGAGTGAGGATTCTGG-3′ and reverse 5′-AGGCAGAAGTGGAATGGGACTGAA-3′) and knock-in-specific primers (forward 5′-GCTATGCTACCCCAGAAAGG-3′ and reverse 5′-GCAGAACAGTGTAGCCAGCA-3′) were used. Finally, PCR products were analyzed using E-Gel™ 2% agarose gels with SYBR™ Safe DNA gel stain (Fig. 1C).

Western Blot Analysis

SMG were harvested from a) wild-type mice, b) humanized hFPR2 mice and c) human SMG tissue samples. Tissues were homogenized in RIPA lysis buffer (MilliporeSigma, St. Louis, MO) containing protease inhibitors. Protein extracts were mixed with loading buffer containing β-mercaptoethanol and then heated at 95 °C for 5 min to denature proteins. Equal amounts of total protein (15 μg per well) were loaded onto a 12% Mini-PROTEAN TGX precast SDS-PAGE gel (Bio-Rad) and separated at 100 V for 70 min using Tris/Glycine/SDS running buffer. Proteins were transferred onto PVDF membranes using the Trans-Blot Turbo Transfer System (Bio-Rad). Membranes were blocked with EveryBlot Blocking Buffer (Bio-Rad) for 5 min at room temperature and then incubated overnight at 4 °C with anti-FPR2/ALX (extracellular) antibody at a dilution of 1:500 in blocking buffer. The next day, membranes were washed three times with TBST (Tris-buffered saline with 0.1% Tween-20) and incubated with HRP-conjugated secondary antibody (Cell Signaling Technology, anti-rabbit HRP, 1:5000) for 1 h at room temperature. After three additional washes with TBST, protein bands were detected using the SuperSignal™ West Femto Maximum Sensitivity Substrate (Thermo Scientific, Waltham, MA) according to the manufacturer’s instructions. Images were captured using a ChemiDoc™ MP Imaging System (Bio-Rad) to visualize the expression of human FPR2 protein (Fig. 1D). To confirm equal protein loading, membranes were subsequently re-probed using an anti-β-tubulin antibody under the same blocking and detection conditions (Fig. 1D).

Agonist-Induced Calcium Signaling

Primary SMG cells were freshly isolated from wild-type mice, humanized hFPR2 mice and human SMG tissues using cell dissociation media composed of 300 U/mL collagenase, 100 U/mL hyaluronidase, 1 mg/mL insulin, 20 mg/mL bovine serum albumin (BSA), 100 U/mL penicillin, and 100 μg/mL streptomycin [13]. Isolated cells were seeded into a Matrigel-coated 8-well Lab-Tek™ II Chambered Coverglass at a density of 5 × 104 cells per well. After three days of incubation, cells were washed twice with Calcium Flux Assay Buffer, prepared by supplementing Acini Buffer (5.4 mM KCl, 1 mM MgCl2, 140 mM NaCl, 1.2 mM MgSO4, 1.2 mM KH2PO4, 15 mM HEPES, pH 7.4) with 0.18% (w/v) glucose, 0.1% (w/v) BSA, and 1 mM CaCl2. Cells were then loaded with Fura-2 AM by incubating them in Calcium Flux Assay Buffer containing 5 μM Fura-2 AM for 45 min at 37 °C. After dye loading, cells were washed twice again with Calcium Flux Assay Buffer (without Fura-2 AM) and incubated for an additional 30 min to allow intracellular cleavage of the Fura-2 AM ester. For stimulation experiments, cells were treated with 100 ng/mL AT-RvD1 (Fig. 1E) and intracellular calcium ([Ca2+]i) responses were monitored over time by measuring changes in Fura-2 fluorescence intensity using a Leica STELLARIS 5 confocal microscope. All experiments were independently repeated four times to confirm reproducibility.

Treatments

Female humanized hFPR2 mice (i.e., C57BL/6J-Fpr2emlOlgab), as well as wild-type C57BL/6J control mice at 8 weeks of age, were randomly distributed into 3 groups based on weight distribution (16–20 g) and treated with 1Χ PBS (healthy controls; 100 µl/20 g) or LPS (acute sialadenitis-like; 5 mg/kg) via intraperitoneal injection into the left peritoneal cavity. Immediately following the LPS injection, mice were treated with PBS (8.9% ethanol) or AT-RvD1 (0.1 mg/kg AT-RvD1, 8.9% ethanol) also via intraperitoneal injection into the right peritoneal cavity. Doses for LPS and AT-RvD1 were chosen based on previous studies [7, 8]. Two days after treatment, mice were either utilized for saliva secretion studies (as described below) or euthanized using CO2 at a flow rate of 4.0 L per minute followed by abdominal exsanguination and SMG were collected. Finally, all animal usage, anesthesia and tissue collection were performed following the AVMA (American Veterinary Medical Association) Guidelines for the Euthanasia of Animals, and our study design complied with ARRIVE (Animal Research: Reporting of In Vivo Experiments) 2.0 guidelines. A cumulative total of 66 mice (7 mice for saliva secretion, 4 mice for histological analyses with three treatment groups for both wild-type and humanized mice) were used, with group sizes determined by power analysis.

Histological Studies

SMG were collected, washed with PBS, and fixed in formalin for one day. After dehydration with ethanol, tissue blocks were created using paraffin wax, and 5-μm-thick tissue sections were obtained. To assess morphological changes, the widely used H&E staining method was performed [14]. Specifically, SMG tissue sections from each group were deparaffinized with xylene and rehydrated with serial ethanol solutions (100%, 70%, and 50%) followed by distilled water. The rehydrated sections were stained with Harris Hematoxylin for 6 min. Then, hematoxylin-stained sections were washed with distilled water for 2 min, treated with 0.5% (w/v) lithium carbonate (Li2CO3) solution for 1 min, and rinsed with distilled water for 1 min. Slides were then washed with 95% ethanol for 1 min, followed by 1 min incubation with Eosin and washed with 95% ethanol for 1 min. Next, hematoxylin and eosin-stained gland sections were washed three times with 100% ethanol, cleared in xylene, and mounted with a xylene-based mounting medium, samples were examined for a blind histopathological analysis using a Leica DMI6000B inverted microscope (Leica Microsystems, Wetzlar, Germany), and tissue analysis was conducted using ImageJ software.

Confocal Immunofluorescence Analysis

Tissue sections were deparaffinized with xylene and rehydrated through serial ethanol dilutions (100%, 95%, 80%, 70%, and 50%, v/v) followed by a rinse with distilled water. For antigen retrieval, the rehydrated SMG sections were treated either with sodium citrate buffer in a pressure cooker for 20 min (for Muc10 and E-Cadherin staining) or with Tris–EDTA buffer [10 mM Tris, 1 mM EDTA, 0.05% (v/v) Tween® 20, pH 9.0] at 95 °C for 30 min (for Ki67/CD45, ZO-1/E-Cadherin, AQP5/Ki67 and AQP5/CD45). After antigen retrieval, samples were permeabilized with 0.1% (v/v) Triton X-100 in PBS at room temperature for 45 min. Blocking was performed with 3% BSA for Muc10 and E-Cadherin, while a 5% (v/v) goat serum in PBS was used for the other antibodies at room temperature for 1 h. The sections were incubated overnight at 4 °C with the following primary antibodies at their respective dilutions: anti-CD45 (1:100), anti-E-Cadherin (1:100), anti-ZO-1 (1:4000), anti-AQP5 (1:200) and anti-Ki67 (1:100). The following day, sections were washed and incubated for 1 h at room temperature with species-specific secondary antibodies, each diluted 1:500 in blocking buffer: Alexa Fluor 488-conjugated anti-goat IgG, Alexa Fluor 488-conjugated anti-rabbit IgG, Alexa Fluor 568-conjugated anti-mouse IgG and Alexa Fluor 568-conjugated anti-rat IgG. Sections were then counterstained with 300 nM DAPI at room temperature for 5 min. Finally, samples were examined using a Leica STELLARIS 5 confocal microscope, visualized using an artificial intelligence-guided image analysis software (AIVIA) and quantified using ImageJ.

Quantification of Acinar Cell Shrinkage

Levels of acinar cell shrinkage were measured using E-Cadherin immunostaining to delineate cell boundaries (Fig. 2D, upper right panel) and Muc10 immunostaining to identify acinar cells (Fig. 2D, bottom left panel). Next, ImageJ was employed to determine acinar cell size by measuring the staining areas as depicted in Fig. 2D, bottom right. Finally, values of acinar cell size were compared between the different treatment groups using one-way analysis of variance (ANOVA, where *p ≤ 0.05 is considered statistically significant) followed by Dunnett’s post-hoc test for multiple comparisons using GraphPad Prism software (version 8.4.2).

Fig. 2

AT-RvD1 restores morphology in SMG from hFPR2 mice treated with LPS. Rehydrated SMG tissue sections from humanized FPR2 mice treated with LPS (A, E), AT-RvD1 + LPS (B, F) and PBS (C, G) were stained with hematoxylin–eosin and images captured using a Leica DMI6000B. Results show that SMG treated with LPS display a loss of epithelial integrity together with areas of acinar shrinkage (white arrows). In contrast, SMG treated with AT-RvD1 + LPS showed restoration of tissue architecture as evidenced by a reduction of acinar shrinkage compared to those with LPS alone. To quantify these changes, tissue sections were stained with goat anti-Muc10 and mouse anti-E-Cadherin and counterstained with DAPI (blue). Next, acinar cell size for each treatment was calculated using ImageJ (D) and expressed as mean ± SD of results from a total of 4 mice per group in quadruples. Statistical significance was assessed by one-way ANOVA (where *p ≤ 0.05 is considered statistically significant) followed by Dunnett's post-hoc test for multiple comparisons (H). Scale bars represent 50 μm

Quantification of Apical Protein Expression Patterns

To analyze apical protein expression patterns (i.e., ZO-1 and AQP5), a line was drawn across the cells and the normalized signal intensity along this line was measured using ImageJ (Fig. 4D) and plotted relative to cell length (Fig. 4E and 5G). Next, the normalized signal intensity in the apical region (Fig. 4E and 5G, red arrows) was further quantified using GraphPad Prism 8 to determine any statistically significant changes (Fig. 4F and 5H). Finally, intensity values in the apical membrane were compared between the different treatment groups using ANOVA (where *p ≤ 0.05 is considered statistically significant) followed by Dunnett’s post-hoc test for multiple comparisons using GraphPad Prism software (version 8.4.2).

Saliva Flow Rate Measurements

Two days after drug treatment, mice were anesthetized intraperitoneally with an avertin solution (250 mg/kg in PBS; left peritoneal cavity) followed by intraperitoneal injection with pilocarpine (2.5 mg/kg in PBS; right peritoneal cavity) to stimulate the secretion of saliva. Saliva was then collected for 20 min and subsequently centrifuged at 12,000 rpm for 10 min. After this step, the supernatant was carefully transferred to a new tube, and the volume of the collected saliva was measured and calculated using the following equation:

$$\text= \frac\left(\mu \text\right)}\left(\text\right) \times \text\left(\text\right)}$$

Statistical significance was assessed using one-way ANOVA (where *p ≤ 0.05 is considered statistically significant) followed by Dunnett’s post-hoc test for multiple comparisons using GraphPad Prism software (version 8.4.2). Raw data showing total saliva amounts as well as mouse weight are shown in Supplementary Figs. 8 and 9.