Data download and processing

We used the GEO database for transplantation retrieval. Neutrophils in peripheral blood interact with damage-associated molecular patterns (DAMPs) through pattern recognition receptors, thereby exacerbating graft damage (Braza et al. 2016; McDonald et al. 2010). Based on the role of neutrophils in transplant immunity, we selected the GSE14346 dataset from the GEO database for our study. GSE14346 (Li et al. 2012) consists of peripheral blood samples from kidney transplantation, which include samples exhibiting acute rejection as well as those without rejection (Named STA in the database). There were 38 samples in the AR group and 37 samples in the STA group. Differentially expressed genes (DEGs) in dataset microarrays were screened using the limma package in R software (Ritchie et al. 2015). The cutoff conditions were set to an adjusted P value < 0.05 and an absolute value of log-fold change| log2FC| > 0.8.

GEO2R is an interactive online software application that allows users to compare two or more groups of samples in a certain platform in order to identify genes that are differentially expressed across different subtypes or different diseases. The datasets (GSE131179 and GSE142667) were used to identify that FPR1 and CSF3R were high expressed in acute rejection group by GEO2R (Verma et al. 2020).

We used the public dataset of scRNA-seq from the NCBI BioProject (accession number PRJNA974568) (Shi et al. 2023). Sequencing data were downloaded using Aspera, aligned to the human reference genome (GRCh38), and processed using the Cell Ranger 8.0.0 count pipeline (10x Genomics) to obtain the gene expression matrix. The raw gene expression matrix was processed to remove doublets using the DoubletFinder R package (McGinnis et al. 2019). Cells were filtered using the Seurat R package (Hao et al. 2024) based on the following criteria: cells with > 200 genes and < 20% mitochondrial gene expression in UMI counts. The SCTransform (Lause et al. 2021) normalization method was applied with the top 3000 highly variable genes. Batch effect was removed by harmony R package. Thirty principal components were selected for cell clustering and UMAP visualization. Cell major populations were defined according to known cell-specific markers. For macrophages, the corresponding cell subsets were extracted for reanalysis, including data normalization and cell clustering.

Functional enrichment analysis of DEGs

The ggplot2 package in R was used to generate a volcano map for differential gene analysis. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed using the clusterProfiler package in R. The org.Hs.eg. db package in R was used for ID conversion. The ClusterProfiler package in R software was used for Gene Set Enrichment Analysis (GSEA) (Subramanian et al. 2005), and the gene set database was MSigDB Collections. Visualization of GSEA was performed using the ggplot2 package in R software.

Evaluation of the relationship between immune cells and samples

CIBERSORT is a method for calculating immune infiltration (Newman et al. 2019). CIBERSORT uses linear support vector regression to mix data deconvolution (Newman et al. 2015). CIBERSORT was used to assess the status of immune cells in the samples. The sample dataset was used as the input of gene expression, and LM22 (22 immune cell types) was set as the signature gene file. A follow-up analysis of the immune cells was then performed.

Screening of neutrophil-related genes

Weighted Gene Co-Expression Network Analysis (WGCNA) is a systems biology method used to describe gene association patterns between different samples. It uses gene information to identify gene sets of interest and conduct significant association analysis with phenotypes (Langfelder and Horvath 2008). We used the presentation data retrieved from GEO for WGCNA. The R package “WGCNA” was applied to search for modules and central genes associated with immune cells. The adjacency matrix was transformed into a topological overlap matrix (TOM). Genes were divided into different gene modules based on TOM-based dissimilarity measures. We clustered the samples, selected cutHeight = 200 to filter the cluster tree, and set the minimum number of samples for each cluster to 10. Here, we set the soft threshold power to 14 to identify critical modules. The module with the highest correlation with immune cells was selected for hub gene screening. Module membership indicates the correlation between genes and modules, while Gene Significance reflects the correlation between genes and phenotypic traits. These two parameters are utilized for the identification of key genes. Hub genes were defined as having gene significance > 0.3 and module membership > 0.8.

Protein–protein interaction (PPI) network analysis was performed on the screened module genes using the STRING website (Szklarczyk et al. 2015). CytoHubba in Cytoscape was used to screen hub nodes (Shannon et al. 2003), and the top three genes were selected as hub genes for further analysis.

Assessing the diagnostic efficacy of hub genes

ROC analysis by R was performed to predict the diagnostic validity of biomarkers. The area under the ROC curve (AUC) value was used to determine the diagnostic validity of distinguishing rejected and nonrejected samples after kidney transplantation in the GSE14346 dataset.

Mouse kidney transplantation

Intraperitoneal heterotopic kidney transplantation was conducted, as mentioned previously (Wang et al. 2017). Briefly, the donor’s kidneys were taken from BALB/c or C57BL/6 mice and transplanted into C57BL/6 recipients’ abdominal cavities by microsurgical techniques. Kidney transplantation from C57BL/6 to C57BL/6 mice was considered a syngeneic group. Kidney transplantation from BALB/c to C57BL/6 mice was considered the allogeneic group. fMLF is an endogenous chemotactic peptide and serves as an agonist for FPR122. HCH6-1 is a potent, competitive dipeptide antagonist of FPR1. HCH6-1 inhibits the specific activation of neutrophils by fMLF (Yang et al. 2017). Therefore, the allogeneic group was treated with HCH6-1 and was considered the allogeneic + HCH6-1 group. HCH6-1 (50 mg/kg) was intraperitoneally injected into mice one day before surgery and one hour after surgery. On the seventh day after the operation, the mice were euthanized for follow-up experiments.

Extraction and grouping of neutrophils from mouse bone marrow

Neutrophils were extracted from mouse bone marrow according to the previous scheme (Boxio et al. 2004). Following mice euthanization, the femurs and tibia were extracted. Cells in bone marrow were filtered with a 70 µM cell filter and centrifuged (600 x g, 4 °C, 5 min), then re-suspended in 3mL neutrophil separation buffer. The cell solution was positioned on an intermittent Percoll gradient, and the 1500xg was rotated for 30 min at 4 °C. Neutrophils were absorbed between 78% and 69% of the percoll density. Neutrophils were collected with a pipette and washed with a neutrophil washing solution. Neutrophils were blocked for Fc receptors (anti-CD16/32 antibody, BioLegend, USA) and then stained with (Zombie dye, APC-Cy7, BioLegend, USA) and anti-Ly6G antibody (FITC, BioLegend, USA). The purity of neutrophils was above 85% as determined by flow cytometry. To simulate the inflammatory environment in the body, neutrophils were divided into two groups: (1) Control group (Con): neutrophils without treatment. (2) Stimulation group (Sti): neutrophils pretreated with LPS (1 µg/mL) for 1 h.

HCH6-1(MCE, China) is an effective competitive dipeptide antagonist of FPR1. To verify the role of HCH6-1 in the inflammatory environment, neutrophils were divided into three groups: (1) LPS group (LPS): Neutrophils from bone marrow were supplemented with LPS (1 µg/mL) for 1 h. (2) LPS + fMLF group (LPS + fMLF): LPS-activated neutrophils from bone marrow (1 µg/mL) for 1 h, and then treated with fMLF (30nM) for 12 h. (3) LPS + fMLF + HCH6-1 group (LPS + fMLF + HCH6-1): Neutrophils were pretreated with HCH6-1 (3µM) for 15 min, and then the operation was the same as in LPS + fMLF group.

Real-time polymerase chain reaction (RT-PCR)

RT‒PCR was utilized to identify gene expressions mentioned in this article. The specimens and cell lines were fully lysed utilizing TRIzol (Invitrogen, CA, USA), and total RNA was extracted. Utilizing a cDNA reverse transcription kit (Vazyme, Nanjing, China), isolated RNA was reverse-transcribed into cDNA. Genes associated with expression in this study were analyzed using the Real Universal Color PreMix (SYBR Green) RT-PCR kit (Vazyme, Nanjing, China). The mRNA primers were designed by Keyybio (Jinan, China).

The relative expression of genes was analyzed using the 2-ΔΔCt method and was replicated three times for each group and analyzed using T-test or ANOVA. The primer sequences are as follows:

GAPDH (Mus)

Forward: GGTGAAGGTCGGTGTGAACG

Reverse: CTCGCTCCTGGAAGATGGTG

FPR1 (Mus).

Forward: CCCTGGACCAAAGATCCTGT

Reverse: ATGGACATGGGAGTGCTGAA

CSF3R (Mus)

Forward: GTTACCCAGCTTCTGGGACT

Reverse: TTGGAGCACATAGGCCTGAA

Ftl (Mus).

Forward: CTTCCAGGATGTGCAGAAGC

Reverse: CAGATCCAAGAGGGCCTGAT

Fth (Mus)

Forward: TTGACCGAGATGATGTGGCT

Reverse: CAGTGCACACTCCATTGCAT

Tfrc (Mus).

Forward: ACCTGGGCTATTGTAAGCGT

Reverse: CTCAGCTGCTTGATGGTGTC

Ferroportin (Mus)

Forward: CCACTCTCTCTCCACTTGGG

Reverse: AGCTGTCAAGAGAAGGCTGT

H&E staining

Kidney samples were preserved with 10% formalin for 48 h. Fixated tissues were encapsulated in paraffin wax and prepared into 5 μm sections for H&E staining. The Paller score, which was used to evaluate the renal tubules, was then used to blindly measure the degree of kidney injury (Zhang et al. 2022). The specific scoring criteria are summarized as follows: A total of 100 cortical tubules from at least 10 different regions of each kidney are evaluated, with higher scores indicating more severe damage. The assessment includes the presence and extent of tubular epithelial cell flattening, loss of brush border, formation of membrane vacuoles, interstitial edema, cytoplasmic vacuolization, cell necrosis, and tubular lumen obstruction.

Enzyme-linked immunosorbent assay (ELISA)

To fully assess the inflammatory environment in each group, serum was collected and measured using ELISA kits (DAKEWE, Shenzhen, China) for TGF-β1, TNF-α, and IL-6 levels. All the above indicators are tested per the manufacturer’s guidelines.

Detection of nets formation in vitro

To assess neutrophil Nets formation, cell supernatant and serum were collected, and NE and MPO levels were assessed using ELISA kits (CLOUD-CLONE, Wuhan, China). Besides, we employed Zombie NIR (an amine-reactive dye, APC-Cy7, Biolegend, USA) to exclude necrotic cells and then a DNA dye (7-AAD, Percp-cy5.5, Yeasen Biotechnology, Shanghai, China) to detect extracellular DNA as a symbol of Nets formation by using flow cytometry [20, 21].

Immunohistochemistry staining

Following dewaxing, rehydration, antigen repair, and endogenous peroxidase inactivation, the prepared kidney tissue sections were incubated overnight at 4℃ with primary antibodies: anti-Ly6 G (Servicebio, Wuhan, China), FPR1 (Bioss, Beijing, China), CSF3R (Bioss, Beijing, China), NE (Servicebio, Wuhan, China), MPO (Servicebio, Wuhan, China) monoclonal antibody. The second day was then washed, and the secondary antibody (Santa Cruz Biotechnology, USA) was applied. Following sections were rewashed in phosphate buffer; they were incubated with a streptavidin-biotin peroxidase complex at 37℃ for 30 min. Slices were reacted with diaminobenzidine (Sigma, USA) as a chromogenic agent and then stained with hematoxylin. Specific brown staining observed in immunohistochemical staining experiments under a light microscope was considered as positive cells. Under a light microscope, the numbers of FPR1-, CSF3R-, MPO-, NE- and LY6G- positive cells in 5 diferent felds in each section were counted by two different pathologists, and finally the individual samples were analyzed with the mean of the positive cell counts.

Immunofluorescence staining

To clarify ROS levels in the kidneys of each group, Kidney tissue was collected and made into frozen slices. The histochemical pen drew a circle around the tissue. The autofluorescence quenching agent was added for 5 min. The tissue was then rinsed with running water. The ROS probes (Dihydroethidium, BestBio, Nanjing, China) were dripped into the circle. After that, it was incubated in an incubator for 30 min. The glass slides were rinsed with PBS three times, each time 5 min. DAPI dye solution was applied and incubated at room temperature (RT) without light for 10 min. They were observed and photographed under a fluorescence microscope. The fluorescence intensity of ROS in each group of grafts was calculated by using image J and used for subsequent analysis.

Western blotting

Pretreated neutrophils were fully lysed in RIPA lysis buffers containing protease inhibitors(phenyl methane sulfonyl fluoride, Servicebio, Wuhan, China). Before heating denaturation, protein solution sticentration was determined utilizing a BCA kit (Yeasen Biotechnology, Shanghai, China) and adjusted to the same concentration. Each group’s total protein was separated utilizing SDS-PAGE and then translocated to a PVDF membrane. PVDF membranes were blocked with 5% skimmed milk before being incubated with the following primary antibodies: anti-ERK1/2 (1:1000, Affinity Biosciences); Anti-Phospho-ERK1/2 (1:1000, Affinity Biosciences); anti-Ferritin Light Chain (1:1000, Abclonal); anti-Ferroportin 1 (1:1000, Affinity Biosciences); anti-Anti Ferritin Heavy Chain (1:1000, Abclonal) and anti-Transferrin receptor protein 1 (1:1000, Affinity Biosciences) at 4℃ overnight. On the second day, a secondary antibody (anti-rabbit, 1:1000, Affinity Biosciences) was treated on PVDF membranes for 30 min at RT. The strips were soaked with ECL solution and exposed to preserve the photos.

Protein expression levels of FPR1(anti-FPR1 antibody, 1:1000, Affinity Biosciences), Ly6G(anti- Ly6G antibody,1:1000, Affinity Biosciences) and CSF3R(anti-CSF3R antibody,1:1000, Affinity Biosciences) were examined in kidney grafts according to the above procedure.

Image J software was used to calculate and analyze the gray value of the acquired images.

Flow cytometry analysis

Splenic T cells are an important reference index for observing immune rejection in kidney transplantation (Zimmerer et al. 2022; Rosales et al. 2022; Zhang et al. 2021). Spleen samples were collected from each group of mice and cell suspensions were prepared for flow staining. Total T cells were labeled with fluorescent antibodies anti-CD3 (FITC), anti-CD4 (APC), and anti-CD8 (PerCP-Cyanine 5.5). All cell staining procedures were as per the manufacturer’s guidelines. The fluorescent antibody was purchased from BioLegend (USA). FlowJo software was used to analyze the results.

Determination of ROS generation

ROS produced in neutrophils was detected by ROS detection kit (Beyotime, Shanghai, China) according to the manufacturer’s protocol. Each group’s fluorescence intensity was measured by a fluorescence enzyme labeling instrument.

Chemotaxis assay

A 24-well microchemotactic chamber was used to measure cell migration (Corning, USA). After LPS stimulation, neutrophils were suspended at an overall concentration of 5 × 106 cells /ml. The LPS group was supplemented with LPS only. The LPS + fMLF group was treated with LPS, and fMLF (30nM) was added to the lower chamber. The LPS + fMLF + HCH6-1 group was pretreated with HCH6-1 for 15 min in advance, and the treatment was the same as LPS + fMLF group. After incubation at 37 ℃ for 8 h, five random fields were selected, and migrated neutrophils number was counted by microscope (Olympus, Japan).

Detection of ferrous ions in neutrophils

Iron, as an essential transition metal element in biological systems, participates in various physiological activities. Free iron typically exists in its stable redox states, namely ferrous ions and ferric ions. FerroOrange (Dojindo, Japan), as a fluorescent probe for the detection of ferrous ions, enables the assessment of ferrous ion levels in living cells.The methodology is summarized as follows:

Add DMSO to the tube containing FerroOrange, and vortex to mix thoroughly, resulting in a 1 mmol/L solution of FerroOrange. Dilute the 1 mmol/L FerroOrange solution with serum-free culture medium to prepare the desired concentration. Neutrophils were extracted from mice and subjected to corresponding treatments for each group. A solution of FerroOrange at a concentration of 1 micromolar/L was added, and the samples were incubated in a cell culture incubator. The fluorescence intensity of neutrophils in each group was measured using a fluorescence microplate reader (emission wavelength: 543 nm; excitation wavelength: 580 nm).

Experimental animals and ethics statement

In this study, 20–25 g healthy adult male BALB/C mice and C57BL/6 J were used. Mice had unrestricted access to food and water and were acclimatized for one week. The Tianjin Medical University General Hospital Animal Ethical and Welfare Committee accepted the above animal usage protocol (IRB2023-DW-08).

Statistical analysis

The data are represented as the mean ± standard deviation. The statistical analyses were done with R (Version 3.6.3). Group comparisons were made utilizing a one-way analysis of variance. A p-value < 0.05 is regarded as statistically significant.