For this case–control study, 134 unrelated individual subjects (68 patients with AR, 66 healthy controls) at the discovery stage were enrolled from the outpatient clinic of the First Affiliated Hospital of Xinjiang Medical University from December 2017 to March 2018. In the second cohort, 120 subjects (69 patients with AR, 51 without AR) were recruited from the otolaryngology ward between May 2019 and December 2019. Patients with AR combined with a deviated nasal septum were considered case patients, and those with a deviated nasal septum without AR were selected as the control group. In addition, nasal mucosa tissue samples from 24 patients from the validation cohort (18 patients for mRNA level analysis and 6 patients for protein level analysis) were collected again to study the expression of mitochondrial mutant genes.

There were similar living environments and lifestyle habits of all Han Chinese subjects in the discovery and validation cohorts over three or more years before enrollment within the Xinjiang Uygur Autonomous Region of China.

Patients in the case group had a history of AR symptoms for more than two years based on the 2015 Chinese Guidelines for the Diagnosis and Treatment of Allergic Rhinitis [17]. Other inclusion criteria included: (1) suffering from two or more AR symptoms, including sneezing, rhinorrhea, nasal itching, and nasal blockage, that continued for at least an hour each day, which could be accompanied by other symptoms, such as itchy eyes, tears, red eyes and other eye symptoms; (2) signs including watery nasal discharge, edema and pale nasal mucosa, conjunctival edema, and congestion; and (3) a positive allergen test comprised of a minimum of one skin prick test (SPT) or a positive serum-specific IgE test result. The inclusion criteria of the subjects in the control group are mentioned below: a clinical history of AR, anterior rhinoscopy and/or nasal endoscopy outcome(s), allergen detection and serum-specific IgE test should all be negative. AR diagnoses were made by specially-trained practitioners. None of the subjects experienced malignancy or a serious allergic reaction, underwent an organ transplant, or had liver or kidney failure. Pregnant women and people with immune deficiencies were not included in this study.

Age, sex, family history of AR, history of drug allergies, and other clinical data were extracted from medical records. SPT, serum-specific IgE, and anterior rhinoscopy or nasal endoscopy outcome(s) were recorded and examined. Allergen SPT positivity was defined as 3 mm of growth in a weal pattern or a minor swelling of the skin within 15–20 min of being pricked with histamine versus a negative control using a 0.9% saline solution. Serum-specific IgE positivity was defined as IgE levels ≥ 0.35 kU/L [18]. Venous blood (3 mL) from each individual was collected into ethylenediaminetetraacetic acid (EDTA) test tubes, then temporarily stored at − 80 °C until the next steps. In consideration of invasiveness and the impact of the COVID-19 pandemic, 24 patients (18 patients for mRNA level analysis and 6 patients for protein level analysis) in the validation cohort with deviated nasal septums who were awaiting surgery were selected for nasal mucosal tissue collection to minimize trauma to the patients. At the same time, fresh tissues that the electric knife had not cauterized were collected as often as possible to ensure the availability of tissue specimens. The nasal mucosa was clamped and cleaned during the operation and quickly dispensed into 5 mL lyophilization tubes to avoid exposing the tissues to air for too long, then put into liquid nitrogen tanks.

Genomic DNA was isolated from EDTA anti-coagulated venous blood samples using the Wizard Genomic DNA Purification Kit (Promega, Beijing, China), following the manufacturer’s instructions. The DNA samples were determined by electrophoresis and quantified using NanoDrop 2000. Shanghai Genesky Biotechnologies captured the mitochondrial genome. Long-range PCR was designed to amplify the mtDNA using specific primers for the human mitochondrial genome. Agencourt AMPure XP-PCR Purification kits (Beckman, Germany) were used to pool and purify the six partially overlapping PCR products, each measuring roughly 5 kb. After that, ultrasonography was used to fragment the PCR results into 100–500 bp fragments (ME220, Woburn, MA). The NEBNext DNA Library Prep Reagent Set from Illumina (New England Biolabs, Ipswich, MA) was used to perform end-repairing, A-tailing, and adaptor ligation after DNA fragmentation. Following PCR amplification, the amplicons were again analyzed and quantified using the BioAnalyzer 2100 before being submitted to 2 × 150 bp paired-end massively parallel sequencing on an Illumina NovaSeq System (Illumina).

The sequencing of human mitochondrial genome is to observe the mutation in individual mitochondrial genomes by double-ended sequencing of the whole mitochondrial genome sequence of each sample. FastQC raw files(version0.11.5) were collected for the sequencing quality evaluation, including the base mass distribution and base composition distribution of each sample. The sequences were aligned against the mitochondrial reference data (revised Cambridge Reference Sequence, rCRS), and sequence alignment files were generated by the Burroughs-Wheeler Aligner (BWA). Variant calling was performed utilizing Mutect2 and Haplotype Caller from the Genome Analysis Toolkit (GATK). Further, variations were sorted based on the threshold (mapping quality 20 and base quality 20). Then, a mitochondrial DNA library was constructed and the length distribution of inserted fragments was counted. The average coverage depth of the genome sequence of samples in this study exceeded 100X, so it is considered that the SNV detected at this site is relatively reliable. Otherwise, variants with less 100X coverage were deleted or re-tested. Variants with frequency levels between 0.1 and 0.9 were regarded as heteroplasmic. The average number of distinct heteroplasmic variations found in the designated site across all samples within the group was used to calculate the number of mtDNA variants.

The candidate single nucleotide polymorphisms (SNP) loci were subjected to the SNaPshot SNP test for genotyping. The sequences of the PCR primer pairs utilized to amplify sequences of genomic DNA are described in Additional file 1: Table S1. An ABI3130XL sequencer and GeneMapper 4.0 software (Applied Biosystems, Co. Ltd., USA) were used to analyze the collected data. To confirm the accuracy of genotyping quality, random samples of 5% of cases and controls were genotyped twice by blinded laboratory personnel. The results of the repeated samples were more than 99% similar.

Total RNA was extracted from frozen nasal mucosal tissue samples stored at − 80 ℃ using Trizol reagent (Invitrogen). One ug of RNA was reverse-transcribed using random primers, and one ul of the resultant single-stranded cDNA was used as the template. The qRT-PCR assay was designed using a SYBR-Green PCR kit and analyzed in triplicate using the Applied Biosystems 7900HT Fast Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). The primers used for the qRT-PCR assay are detailed in Additional file 2: Table S2. The 2ΔCt formula was used to calculate the outcome data.

The Total Protein Extraction (TPE) kit (Sangon Biotech) was used in accordance with the manufacturer’s instructions to extract total protein from nasal mucosal frozen tissue samples stored at − 80 °C. As previously described [19], proteins were separated using sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride (PVDF) membranes (EMD Millipore, Billerica, Massachusetts, USA). PVDF membrane was exposed using an enhanced chemiluminescence detection method after being blocked with non-fat milk and incubated with primary and secondary antibodies. The bands' intermixed density was measured using Image Lab software (Bio-Rad, Hercules, CA, USA). Anti-MT- ATP6 (CPA1767) was purchased from Cohesionbio, and anti-GAPDH antibody (8245) from Abcam.

The study was approved by the medical ethics committee of the First Affiliated Hospital of Xinjiang Medical University (20,170,316–07), and all participants signed informed consent.

Normally distributed data for continuous variables are described as means with standard deviations, and the two groups were compared by t-test. Non-normal distribution data are presented as the median (interquartile range, IQR), and groups were analyzed using the Mann–Whitney U test. The heteroplasmic level between AR patients and controls was determined using the Mann–Whitney U test if at least three samples from each group were carriers. Count data were expressed as percentages, and the two groups were compared using the Chi-squared test. R statistical software (version 3.6.2) was used to evaluate all statistics. MtSNPs with minor allele frequency above 0.05 were used for single variant association analysis. Using the sequence kernel association test provided by the R package SKAT, mitochondrial gene set-based analyses were performed using mtSNV with a frequency range of 0.1–0.9 and call rate above 0.9. All bilateral tests were based on a P < 0.05, which suggested that the difference between groups was statistically significant.