Introduction

Osteonecrosis of the femoral head (ONFH), also referred to as ischemic necrosis of the femoral head, is a frequently seen and challenging condition in orthopedics. It is characterized by the interruption or impairment of blood supply to the femoral head, leading to the death and insufficient repair of bone cells and bone marrow cells. Subsequently, this causes structural changes in the femoral head and its eventual collapse, resulting in joint pain and dysfunction [1]. According to a large-scale epidemiological survey, it was found that the estimated cumulative number of non-traumatic ONFH patients in China has reached 8.12 million, with males showing a significantly higher prevalence compared to females, and urban areas having a higher prevalence than rural areas [2]. Nowadays, various methods have been used in the clinical treatment of ONFH, such as osteotomies [3], total hip arthroplasty (THA) [4], and bone marrow-derived cell therapies (BMCTs) combined with core decompression (CD) [5], but their effectiveness and safety are limited. A comprehensive elucidation of the pathogenesis of ONFH can provide important directions for its precise treatment. Previous studies have revealed that multiple factors, including single nucleotide polymorphisms (SNPs), may be closely associated with its occurrence [6,7,8]. Hence, the investigation of the influence of SNPs on the risk of ONFH holds great significance in terms of preventing, diagnosing, and effectively treating this condition.

Estrogen receptor (ER) is a group of nuclear receptor superfamily members that function as transcription factors in the nucleus, consisting of two types: ERα and ERβ [9]. The Estrogen Receptor 1 (ESR1) gene encodes the ERα, which is localized within the bone and plays a crucial role in regulating bone metabolism [10]. Relevant studies have shown that estrogen can promote osteoblast proliferation and inhibit osteoclast activity by binding with ESR1, thus playing a physiological role, while estrogen deficiency may cause changes in bone microstructure, resulting in increased cortical bone vascular aperture, increased bone trabecular separation and decreased bone trabecular number, thus leading to the occurrence of ONFH [11,12,13]. Currently, there has been research conducted to investigate the impact of ESR1 polymorphisms on the quality of the femoral head in patients with Turner syndrome, revealing that ESR1 rs2234693 is potentially linked to decreased bone mineral density (BMD) in the femoral neck and total hip regions [14]. However, the available research on the association between ESR1 gene polymorphisms and the risk of ONFH is scarce.

Apolipoprotein E (APOE), located on chromosome 19, plays a crucial role in plasma lipid metabolism by facilitating the hepatocyte-mediated uptake and removal of chylomicrons, very low-density lipoproteins (VLDL), and high-density lipoproteins (HDL) lipoproteins [15]. It is reported that APOE-deficient individuals may exhibit severe hyperlipidemia, which can potentially influence the microcirculation of the femoral head and contribute to the development of ONFH [16, 17]. At present, there is a dearth of research investigating the potential link between APOE gene polymorphisms and the risk of developing steroid-induced ONFH. Some studies have suggested a possible association between the rs7412 C/T and rs429358 T/C loci and an elevated risk of SONFH [18]. Nevertheless, as of now, there is still no reported evidence regarding the specific relationship between APOE gene polymorphisms and ONFH.

Given that ESR1 and APOE gene polymorphisms may be linked to the ONFH, this study seeks to assess the influence of ESR1 gene (rs2982573 C < T, rs10872678 C < T and rs9322332 A < C) and APOE gene (rs7259620 A < G and rs769446 C < T) polymorphisms on the susceptibility to ONFH. The findings of this study are anticipated to shed light on potential biomarkers for the diagnosis and treatment of ONFH.

Materials and methods

Study participants

This study included a sample of 505 individuals diagnosed with ONFH and 512 healthy controls, all of whom were obtained from the Affiliated Hospital of Weifang Medical University and the Second Affiliated Hospital of Inner Mongolia Medical University. The inclusion criteria for ONFH cases were as follows: (1) Patients experiencing pain in the hip joint, buttock, or groin area, accompanied by pain in the knee joint and restricted internal and external rotation of the hip joint; (2) Diagnosis of ONFH confirmed by X-ray, computed tomography (CT), and magnetic resonance imaging (MRI); (3) No history of direct trauma, osteoarthritis, ankylosing spondylitis, hip joint-related diseases (e.g., hip joint synovitis), cardiovascular and cerebrovascular diseases, metabolic disorders, or bone metastasis. The group of healthy controls consisted of individuals who met the following inclusion criteria: (1) No pain in the hip joint, buttock, or groin area; (2) No evidence of lesions on imaging examinations; (3) No chronic alcohol use or steroid use; (4) No history of direct trauma, osteoarthritis, ankylosing spondylitis, cardiovascular and cerebrovascular diseases, metabolic disorders, or bone metastasis. Demographic characteristics and clinical data of the study participants were collected through questionnaires and a review of patient medical records. Additionally, this study was approved by the ethics committee of the Affiliated Hospital of Weifang Medical University and the ethics committee of the Second Affiliated Hospital of Inner Mongolia Medical University, and conducted in compliance with the principles outlined in the Declaration of Helsinki. Informed consent was obtained from all participants before the commencement of the experiment.

DNA extraction and genotyping

Three SNPs (rs2982573, rs10872678, and rs9322332) in the ESR1 gene, as well as two SNPs (rs7259620 and rs769446) in the APOE gene, were selected for genotyping. These SNPs were selected based on the following process: (1) Data on ESR1 and APOE gene polymorphisms were obtained from the 1000 Genomes Project database; (2) Screening criteria were implemented, necessitating a minor allele frequency (MAF) above 5% and Hardy–Weinberg equilibrium (HWE) exceeding 0.01; (3) The SNPs to be studied were identified through a combination of primer design and an extensive literature search. After a 12-h fasting period, 5 mL of peripheral venous blood was collected from the study participants using a vacuum blood collection tube containing EDTA-K2 anticoagulant, and mixed upside down and used in subsequent experiments. Genomic DNA extraction was performed using a genomic DNA isolation kit (GoldMag Biotechnology) following the manufacturer’s instructions. The concentration of the DNA was measured using a Nanodrop 2000 spectrophotometer (Thermo, USA). SNP genotyping was carried out utilizing the Agena MassARRAY platform (Agena Bioscience, USA), and the data analysis was performed using the Agena Typer 4.0 software. Additional file 1: Table S1 contains the listed sequences of the primers.

Statistical analysis

The characteristics of the study participants were analyzed using the t-test for continuous variables and χ2 test for categorical variables. HWE in controls was calculated using the χ2 test to further explain the good representativeness of the study population. Logistic regression analysis with odds ratios (ORs) and corresponding 95% confidence intervals (CIs) was used to assess the association between ESR1 and APOE gene polymorphisms and the risk of ONFH. Multifactor dimensionality reduction (MDR) analysis was employed to explore SNP-SNP interactions. The reliability of significant findings was assessed using false positive report probability (FPRP) analysis. Statistical significance was set at a p value of less than 0.05. The flowchart of this study is shown in Fig. 1.

Fig. 1
figure 1

Flowchart illustrating the analysis of the association between ESR1 and APOE gene polymorphisms and ONFH

Full size image

Results

Demographic characteristics of the study participants

This case–control study encompassed a total of 1,017 participants, with 505 cases of ONFH (284 males and 221 females) and 512 healthy controls (305 males and 207 females). The average age of ONFH patients was 51.65 ± 14.43, whereas the average age of healthy controls was 50.59 ± 14.39. It is noteworthy that a statistically significant difference in smoking (p = 0.004) factor was observed between the cases and controls, whereas there were no significant differences in terms of age (p = 0.242) and gender (p = 0.282) distribution between the two groups. The detailed characteristics of the study participants are shown in Table 1.

Table 1 Basic characteristics of study participants
Full size table

Association of ESR1 and APOE allele frequencies with ONFH risk

The basic information and MAF of SNPs for ESR1 and APOE are provided in Table 2. It is worth noting that all five SNPs in the control group followed HWE. Through allele model analysis, a χ2 test identified one SNP that exhibited a close association with ONFH. Specifically, the A allele of rs9322332 was found to be significantly associated with a reduced risk of ONFH, showing a 0.81-fold decrease (OR = 0.81, 95% CI [0.68–0.97], p = 0.020).

Table 2 Basic information and allele frequencies of rs2982573, rs10872678, rs9322332, rs7259620, and rs769446
Full size table

Overall analysis of the association of ESR1 and APOE gene polymorphisms with ONFH risk

This study investigated the correlation between ESR1 and APOE gene polymorphisms and the risk of ONFH using different genetic models, including co-dominant, dominant, recessive, and additive models. The findings of the overall analysis examining the association of ESR1 and APOE gene polymorphisms with ONFH risk are summarized in Fig. 2 and Table 3. The findings indicated a significant association between ESR1-rs9322332 and a decreased risk of ONFH in the overall analysis, especially in homozygous (OR = 0.69, 95% CI [0.53–0.90], p = 0.006), as well as in dominant (OR = 0.70, 95% CI [0.54–0.90], p = 0.006) and additive (OR = 0.79, 95% CI [0.66–0.95], p = 0.013) models. However, this study found no association between other SNPs of ESR1 and APOE genes and ONFH risk.

Fig. 2
figure 2

Forest maps of associations between ESR1 and APOE gene polymorphisms and ONFH in different genetic models

Full size image
Table 3 Association between ESR1 and APOE gene polymorphisms and ONFH risk
Full size table

Stratified analysis of the association of ESR1 and APOE gene polymorphisms with ONFH risk

To further investigate the correlation of ESR1 and APOE gene polymorphisms with ONFH risk, we performed stratified analyses based on age (Table 4), gender (Table 5), smoking status (Table 6), and clinical staging (Additional file 1: Table S2). Under different genetic models, the association between rs9322332 and a reduced risk of ONFH was observed in specific subgroups, including individuals older than 51 years (CA vs CC: OR = 0.47, 95% CI [0.28–0.80], p = 0.005; AA vs CC: OR = 0.52, 95% CI [0.35–0.77], p = 0.001; CA + AA vs CC: OR = 0.51, 95% CI [0.35–0.74], p < 0.001; Additive: OR = 0.65, 95% CI [0.50–0.84], p = 0.001), females (AA vs CC: OR = 0.51, 95% CI [0.34–0.78], p = 0.002; CA + AA vs CC: OR = 0.55, 95% CI [0.37–0.82], p = 0.004), and non-smokers (CA + AA vs CC: OR = 0.68, 95% CI [0.47–0.99], p = 0.042; Additive: OR = 0.77, 95% CI [0.59–0.99], p = 0.042). However, this study did not observe any associations between several other SNPs (rs2982573, rs10872678, rs7259620, and rs769446) and ONFH when analyzed using different genetic models.

Table 4 Association between ESR1 and APOE gene polymorphisms and ONFH risk stratified by age
Full size table
Table 5 Association between ESR1 and APOE gene polymorphisms and ONFH risk stratified by gender
Full size table
Table 6 Association between ESR1 and APOE gene polymorphisms and ONFH risk stratified by smoking status
Full size table

SNP-SNP interaction analysis based on MDR analysis

MDR software was used to analyze SNP-SNP interactions among ESR1 and APOE gene polymorphisms (Table 7). Consequently, the most effective single-locus prediction model identified was rs9322332, achieving a cross-validation consistency (CVC) of 10/10 and a testing balanced accuracy of 0.539 (p = 0.009). Furthermore, the optimal multi-locus prediction model was a combination of five loci (rs2982573, rs10872678, rs9322332, rs7259620, and rs769446), demonstrating a CVC of 10/10 and a testing balanced accuracy of 0.551 (p < 0.001). Additionally, the interaction between each locus was demonstrated in the dendrogram (Fig. 3A) and the circle graph (Fig. 3B). In Fig. 3B, the most significant interaction was observed between rs2982573 and rs7259620, with an information gain (IG) value of 0.30%.

Table 7 SNP-SNP interaction models of candidate SNPs analyzed by the MDR method
Full size table
Fig. 3
figure 3

Dendrogram A and circle graph B of SNP-SNP interaction among ESR1 and APOE gene polymorphisms by MDR method

Full size image

FPRP analysis

FPRP analysis was employed to validate the reliability of the observed associations between ESR1 and APOE SNPs and the risk of ONFH (Table 8). The associations, reflected by FPRP values below 0.2, are notable findings of significance. Significantly, in both the overall analysis and subgroup analyses based on age (> 51 years), females, and non-smoking status, rs9322332 demonstrated a noteworthy association with a reduced risk of ONFH, particularly at a prior probability of 0.25. With a prior probability of 0.1, the association between rs9322332 and a lowered risk of ONFH remained significant in the overall analysis (A vs C: FPRP = 0.167, power = 0.983; AA vs CC: FPRP = 0.085, power = 0.600; CA + AA vs CC: FPRP = 0.070, power = 0.648; Additive: FPRP = 0.103, power = 0.964), as well as in stratified analyses based on age (> 51) (AA vs CC: FPRP = 0.084, power = 0.107; CA + AA vs CC: FPRP = 0.043, power = 0.079; Additive: FPRP = 0.021, power = 0.423) and females (AA vs CC: FPRP = 0.136, power = 0.108; CA + AA vs CC: FPRP = 0.149, power = 0.173). Even with a prior probability of 0.01, rs9322332 continued to exhibit an association with a reduced risk of ONFH among individuals older than 51 years (Additive: FPRP = 0.188, power = 0.423) under the additive model. Based on the results of the FPRP analysis, this study provided further evidence for a strong association between ESR1-rs9322332 and a reduced risk of ONFH.

Table 8 Results of FPRP analysis for significant findings
Full size table

Discussion

In the present study, we investigated the potential link between ESR1 and APOE gene polymorphisms and the risk of ONFH. Our results suggested a significant association between ESR1-rs9322332 and a significantly decreased risk of ONFH under the homozygous (AA vs.CC: OR = 0.69, 95% CI [0.53–0.90], p = 0.006), dominant (CA + AA vs. CC: OR = 0.70, 95% CI [0.54–0.90], p = 0.006), and additive (OR = 0.79, 95% CI [0.66–0.95], p = 0.013) models. The stratified analysis revealed that this polymorphism has a protective effect against ONFH in non-smoker and aged over 51 years old. These results emphasized the significance of ESR1-rs9322332 in the pathogenesis and advancement of ONFH and suggest its potential as a novel biomarker for ONFH treatment.

Regarding the ESR1 rs2982573, rs10872678, and rs9322332, currently, there is limited research available. Liu et al. conducted a study showing that individuals in the Taiwanese population carrying the TC + CC genotypes of ESR1 rs2982573 had a lower likelihood of developing osteoporosis when consuming at least three cups of coffee per week [19]. However, our study did not observe any relationship between ESR1 rs2982573 and the risk of ONFH. This discrepancy could be attributed to differences in the geographical locations of the study participants, as our study focused on individuals from inland China. For rs9322332, Andrew May et al. discovered that carrying the C allele of ESR1-rs9322332 was associated with a decrease in bone mineral content among black South African children [20]. Interestingly, our study revealed a correlation between carrying the A allele of ESR1-rs9322332 and a lowered risk of ONFH, indicating a potential association between the rs9322332 mutation’s impact on bone mineral content and the reduced risk of ONFH. However, there are no studies on the association between rs10872678 and ONFH or other diseases.

Regarding the APOE rs7259620 and rs769446, there has been a significant amount of research conducted thus far. Cai et al. investigated the correlation between APOE gene polymorphisms, diet, and dyslipidemia in the Yao minority area, and found no significant association between rs7259620 and dyslipidemia [21]. Park et al. conducted a genetic variation selection study associated with the risk of hyper-LDL-cholesterolemia and found that rs7259620 is associated with a reduced risk of hyper-LDL-cholesterolemia [22]. Furthermore, it has been reported that rs7259620 is associated with both Alzheimer’s disease and coronary heart disease [23, 24]. In regards to rs769446, Ereqat et al. investigated the impact of APOE gene variations on the risk of dyslipidemia in diabetes, and no statistical differences were observed in rs449647 variants among T2D patients with and without dyslipidemia [25]. Moreover, multiple studies have also explored the association of rs7259620 with Alzheimer’s disease and coronary heart disease, but no significant association was observed [26,27,28,29]. In our results, we explored the association between two SNPs of the APOE gene and the risk of ONFH, but no significant association was found between them. We speculate that such results may be related to the study participants’ region of residence, race, type of disease, and a variety of other factors.

The onset of ONFH is affected by many factors, including age, gender, and smoking status. ONFH is a disabling condition that primarily affects young to middle-aged individuals. In China, the mean age at diagnosis is 50.40 ± 13.71 years, with a predominant number of patients falling within the age range of 41 to 60 years old [30]. Additionally, epidemiological data indicates that the incidence of ONFH is higher in males compared to females [30, 31]. Considering the significance of age and gender as risk factors for ONFH, we conducted a stratified analysis to explore the influence of ESR1 and APOE gene polymorphisms on the risk of developing ONFH. Our findings indicated that the ESR1-rs9322332 polymorphism was linked to a decreased risk of ONFH among individuals aged over 51 years, suggesting an age-dependent effect of ESR1-rs9322332 on ONFH risk. We also found a significant correlation between ESR1-rs9322332 polymorphism and a reduction in ONFH risk among female participants. However, considering that the HWE p value < 0.05 in the female population (Additional file 1: Table S3), further validation is needed to determine whether the correlation between ESR1-rs9322332 and ONFH risk depends on gender. In addition, previous research has also demonstrated a positive association between smoking and an elevated risk of ONFH. As demonstrated by the study conducted by Takahashi et al., current smokers, individuals with a smoking consumption exceeding 20 cigarettes per day, and those with 26 pack-years or more, have ONFH risks that are 3.89 (95% CI 1.46–10.4), 3.89 (95% CI 1.22–12.4), and 4.26 (95% CI 1.32–13.7) times higher, respectively, compared to non-smokers [32]. From this observation, it is evident that the more one smokes, the greater the risk of developing ONFH. These findings emphasize the significance of taking behavioral habits into account when investigating the relationship between genetic factors and the risk of ONFH.

Certain limitations exist in this study that should be acknowledged. Firstly, the participants were exclusively recruited from a single hospital, possibly leading to selection bias. Secondly, functional experiments were not performed in this study. In future studies, we will expand the sample size to further investigate the association between ESR1 rs9322332 and the risk of ONFH, and validate our findings through in vivo animal experiments.

Conclusions

This study has established a strong link between ESR1-rs9322332 and a lower incidence of ONFH, particularly among individuals over the age of 51 and non-smokers. However, further validation with a larger sample size is necessary. In summary, this study provides valuable insights into the role of ESR1 gene polymorphisms in the prevention and diagnosis of ONFH.