Identification and Diagnostic Significance of MPO, PRTN3, and CTNND1 as Biomarkers in Acute Hematogenous Osteomyelitis in Children: A Comprehensive Analysis Using Machine Learning Algorithms

doi:10.21203/rs.3.rs-4943707/v1

Download PDF

Research Article

Identification and Diagnostic Significance of MPO, PRTN3, and CTNND1 as Biomarkers in Acute Hematogenous Osteomyelitis in Children: A Comprehensive Analysis Using Machine Learning Algorithms

https://doi.org/10.21203/rs.3.rs-4943707/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Acute hematogenous osteomyelitis (AHO) is a severe bacterial bone infection predominantly affecting children. Early diagnosis is crucial to prevent the progression to chronic osteomyelitis. However, current diagnostic methods are limited in sensitivity and specificity, underscoring the need for reliable biomarkers.

Methods

This study utilized gene expression data from the Gene Expression Omnibus (GEO) to identify differentially expressed genes (DEGs) associated with AHO. We employed three machine learning algorithms—The Least Absolute Shrinkage and Selection Operator (LASSO) regression, Support Vector Machine-Recursive Feature Elimination (SVM-RFE), and Random Forest (RF)—to screen for potential diagnostic markers. The expression levels of key genes were validated using clinical samples from pediatric AHO patients. Receiver operating characteristic (ROC) curve analysis was performed to assess the diagnostic accuracy of these biomarkers.

Results

Our analysis identified five candidate genes, among which Myeloperoxidase (MPO), Serum proteinase 3 (PRTN3), Catenin delta 1 (CTNND1) were significantly associated with AHO, MPO and PRTN3 were upregulated, while CTNND1 was downregulated in AHO samples compared to healthy controls. ROC curve analysis demonstrated that CTNND1 (AUC = 0.8832), MPO (AUC = 0.9803) and PRTN3 (AUC = 0.9767) exhibited strong diagnostic potential. Importantly, the expression levels of MPO and PRTN3 positively correlated with disease severity as classified by the Cierny-Mader staging system, whereas CTNND1 expression showed a negative correlation.

Conclusion

MPO, PRTN3, and CTNND1 are promising biomarkers for the diagnosis and monitoring of AHO in children. Their expression levels correlate with disease severity, making them valuable tools for assessing the progression and treatment efficacy in pediatric AHO. Further research is warranted to explore their potential in clinical applications.

Acute Hematogenous Osteomyelitis

Biomarkers

CTNND1

PRTN1

MPO

Acute hematogenous osteomyelitis (AHO) is a bacterial bone infection that arises from hematogenous spread, primarily affecting young children [1], In high-income countries, the incidence is approximately 8 per 100,000 children annually [1–2]. AHO is the most common form of osteomyelitis in pediatric patients and often necessitates hospitalization, invasive diagnostic procedures, surgical intervention, and prolonged antimicrobial therapy [3]. Staphylococcus aureus (S. aureus) is the predominant pathogen, with both methicillin-sensitive Staphylococcus aureus and methicillin-resistant Staphylococcus aureus being frequently implicated [4]. Current treatment strategies emphasize antimicrobial therapy and meticulous debridement to manage these infections. However, as the disease progresses, complications such as antibiotic resistance and abscess formation can significantly compromise treatment efficacy [5]. Thus, early diagnosis and timely intervention to prevent the transition from acute to chronic osteomyelitis are critical.

The clinical diagnosis of acute osteomyelitis typically involves a multifaceted approach, including the evaluation of clinical signs, radiological imaging, and laboratory test results. Bone biopsy and microbiological culture are often employed to confirm the diagnosis [6]. Although X-rays are commonly the initial imaging modality, their sensitivity in detecting early osteomyelitis is limited, diminishing their diagnostic utility. In pediatric osteomyelitis, radiographic changes typically become apparent around two weeks after disease onset, whereas in adults, these changes may take longer to manifest. Magnetic resonance imaging is considered the gold standard for osteomyelitis imaging due to its high sensitivity in detecting early-stage lesions. However, its high cost and limited impact on treatment outcomes present significant challenges [7]. The lack of rapid, direct, and specific diagnostic tools creates substantial obstacles in the timely and accurate diagnosis of osteomyelitis, highlighting the urgent need for the identification of suitable non-invasive biomarkers in AHO patients.

With the advancement of high-throughput sequencing technologies, high-throughput genetic microarray analysis of various sample types has enabled the investigation of diseases at multiple levels, including DNA, RNA, and protein. Recent studies have identified numerous specific genes involved in the progression of osteomyelitis. In a recent study involving a Chinese population, a comparison of gene polymorphisms between cases of post-traumatic osteomyelitis and healthy controls revealed that polymorphisms in the NLR family pyrin domain containing 3, elongator complex protein 2, signal transducer and activator of transcription 3, caspase 1, nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, alpha, nuclear factor of kappa light polypeptide gene enhancer in B-cells 1, caspase recruitment domain family member 8, and cluster of differentiation 14 genes are associated with an increased risk of post-traumatic osteomyelitis. Specifically, variants rs10754558 and rs7525979 of the NLRP3 gene were found to significantly increase the risk of developing post-traumatic osteomyelitis.[8]. Further cellular studies demonstrated that S. aureus induces the activation of the NLRP3 inflammasome, and inhibition of NLRP3 inflammasome activity by MCC950 reduces S. aureus-induced bone resorption and the expression of osteoclast-specific genes, including tartrate-resistant acid phosphatase, matrix metallopeptidase 9 (MMP9), cathepsin K, calcitonin receptor, and vacuolar ATPase V0 subunit d2 [9]. Additionally, Campbell et al. revealed that S. aureus significantly upregulates the expression of the receptor activator of nuclear factor kappa-Β ligand (RANKL) gene, and treatment with denosumab completely prevents severe cortical bone destruction caused by the infection, indicating that RANKL-mediated osteoclastogenesis is essential for bone loss induced by S. aureus infection [10]. These findings suggest the critical role of certain functional genes in the progression of osteomyelitis. However, the diagnostic value of many genes in osteomyelitis remains to be explored.

In this study, we utilized data from the Gene Expression Omnibus (GEO) to identify differentially expressed genes (DEGs) associated with AHO. We constructed a nomogram model to evaluate the diagnostic potential of key genes in osteomyelitis and assessed the model's diagnostic accuracy through receiver operating characteristic (ROC) curve analysis. The reliability of the model was further validated using an independent dataset.

Data preprocessing

The gene expression data for a pediatric MRSA infection dataset (GSE30119) and its corresponding platform files were obtained from the GEO database (https://www.ncbi.nlm.nih.gov/geo/). The GSE30119 dataset includes 121 MRSA infection samples and 22 healthy control samples. Among these, 12 cases were identified as MRSA infections confined to bone and joint, with no concurrent infections in other body sites. The data were processed following platform-specific protocols, ensuring the normalization and quality control of the gene expression profiles. This preprocessing step is critical for accurately reflecting the transcriptional differences between MRSA-infected and healthy samples, particularly in cases with localized bone and joint infections.

Identification of DEGs

Differential expression analysis was conducted on 48,003 AHO-related genes using the "limma" package to identify differentially expressed genes between osteomyelitis samples and healthy controls. Visualizations were generated using the "Matplotlib" and "pheatmap" packages, resulting in a volcano plot and a heatmap, respectively. The differentially expressed genes were then further filtered based on an adjusted p-value of less than 0.05 and a log-fold change (logFC) greater than |1|.

Functional Enrichment Analysis

The potential biological functions of the differentially expressed AHO genes were investigated through enrichment analysis. This analysis included Gene Ontology (GO) categories—biological process (BP), cellular component (CC), and molecular function (MF)—as well as Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. The enrichment analysis was performed using the online platform Metascape (https://metascape.org). The top five significantly enriched functions and pathways were identified and visualized in the form of enrichment network diagrams, with a P-value of less than 0.05 considered indicative of significant enrichment.

Candidate Diagnosis Marker Selection

Three machine learning algorithms were applied to identify potential diagnostic factors and predict AHO status. The Least Absolute Shrinkage and Selection Operator (LASSO), a regression analysis algorithm, was employed to improve prediction accuracy through regularization. LASSO regression analysis was conducted using the "glmnet" library in Python to identify genes significantly associated with the discriminative power between AHO and healthy samples. The Support Vector Machine (SVM), a supervised machine learning technique widely used for classification and regression analysis, was also utilized. To avoid overfitting, the Recursive Feature Elimination (RFE) algorithm was applied to select the optimal genes from the metadata cohort, with SVM-RFE being used specifically to identify features with the greatest discriminatory ability. Additionally, the Random Forest (RF) algorithm, an ensemble learning method based on decision trees, was implemented using the "randomForest" library in Python to enhance classification and prediction accuracy by constructing multiple decision trees. This algorithm was further employed to filter key genes associated with AHO status.

Patient and Ethics statement

The participants in this study were AHO patients treated at the Affiliated Hospital of Zunyi Medical University. Between July 2022 and July 2024, a total of 61 patients were treated in the Department of Pediatric Surgery at the Affiliated Hospital of Zunyi Medical University. The healthy control group consisted of children who underwent physical examinations at our hospital, excluding those with infectious diseases. The research protocol was approved by the Ethics Committee of the Affiliated Hospital of Zunyi Medical University (authorization number: 2022 Review No. [521]). All procedures performed in this study adhered to the institution’s ethical standards, the Declaration of Helsinki, and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study.

Biochemical assessment

Peripheral blood was collected from each patient before any AHO treatment and on the third day post-operation. Serum was obtained by centrifugation at 3000 rpm for 5 minutes and stored at -80°C until testing. All serum samples were stored in duplicate. White blood cell (WBC) counts were measured using an automated blood analyzer. C-reactive protein (CRP) levels were determined by immunoturbidimetry. The erythrocyte sedimentation rate (ESR) was measured using the Westergren method, and procalcitonin (PCT) levels were assessed by electrochemiluminescence immunoassay. MPO levels were determined using a time-resolved fluorescence lateral flow immunoassay based on the principle of immunochromatography and a double antibody sandwich method (Eachy Biopharma, China). Serum PRTN3 concentrations were measured using a human PRTN3 enzyme-linked immunosorbent assay (ELISA) kit (ab226902; Abcam, Tokyo, Japan) according to the manufacturer’s instructions. Serum CTNND1 concentrations were measured using an ELISA kit (SEE901Mu; Cloud-Clone Corp, Wuhan, China) according to the manufacturer’s instructions

Statistical analysis

All experimental data were analyzed using SPSS software (version 22.0; IBM, Armonk, NY, USA) and GraphPad Prism 5.0 (GraphPad Software, La Jolla, CA, USA). To evaluate the classification performance of key genes in AHO and healthy samples, ROC curves were generated to assess their diagnostic value. A P-value of less than 0.05 was considered statistically significant.

Screening for diferentially expressed AHO‑associated genes

In the GSE30119 dataset, after screening, 22 healthy controls and 12 cases of osteomyelitis or pyogenic arthritis without other comorbidities were identified. Differential expression analysis between these two groups was conducted to identify potential biomarkers for AHO. A total of 279 genes were found to be significantly differentially expressed between the normal and AHO groups (|LogFC| > 1), with 125 genes upregulated and 154 genes downregulated in AHO. The corresponding volcano plot and heatmap are shown in Fig. 1A and 1B, respectively.

Functional Enrichment Analyses

To elucidate the biological functions of these regulators in AHO, we conducted GO enrichment analysis and KEGG pathway analysis using the Bioconductor package clusterProfiler in R. To better distinguish the biological functions of upregulated and downregulated genes, GO and KEGG analyses were performed separately for the upregulated and downregulated gene sets.

First, we analyzed the upregulated genes. The GO enrichment analysis revealed that, in terms of BP, AHO was predominantly associated with positive regulation of locomotion, defense response to bacterium, and wound healing. For CC, these genes were mainly involved in the secretory granule lumen and specific granule. Regarding MF, the upregulated genes were primarily linked to glycosaminoglycan binding and endopeptidase activity (Fig. 2A). Pathway analysis (Fig. 2C) showed significant enrichment in the Focal adhesion, Transcriptional misregulation in cancer, and Proteoglycans in cancer pathways.

Next, we analyzed the downregulated genes. The GO enrichment analysis indicated that, in terms of BP, AHO was associated with neuron projection morphogenesis, behavior, and circulatory system processes. For CC, these genes were involved in the cell-cell junction, perinuclear region of the cytoplasm, and Schaffer collateral-CA1 synapse. Regarding MF, the downregulated genes were primarily linked to signaling receptor activator activity, purinergic nucleotide receptor activity, and morphogen activity (Fig. 2B). Pathway analysis (Fig. 2D) revealed significant enrichment in the PI3K-Akt signaling pathway, Calcium signaling pathway, and AGE-RAGE signaling pathway in diabetic complications.

Screening target genes by LASSO regression, SVM‑RFE and RF analysis

The DEGs were identified using the LASSO regression algorithm, which resulted in the identification of 15 variables as diagnostic markers for AHO (Fig. 3A, 3B). Additionally, we identified 15 key genes using the RF algorithm (Fig. 3C). A subset of 39 features among the DEGs was identified using the SVM-RFE algorithm (Fig. 3D). The three overlapping features (MPO, PRTN3, CTNND1, VARS, LOC401525) among these three algorithms were ultimately selected (Fig. 3E). These five genes may be critical in the progression of AHO.

The Expression and Diagnosis Significance of candidate genes in AHO

Our team found that there was no statistically significant difference in the expression level of LOC401525 between the control and experimental groups (P > 0.05). The expression levels of Variance Adjusted Response Selection (VARS) and CTNND1 were significantly downregulated in AHO samples compared to healthy samples (P < 0.05) (Fig. 4A-C), while the expression levels of MPO and PRTN3 were significantly upregulated in AHO samples (P < 0.05) (Fig. 4D-E). To further explore the diagnostic value of these candidate genes, we performed ROC curve analyses. The results showed that VARS (Fig. 4B, AUC = 0.761), CTNND1 (Fig. 4C, AUC = 0.811), MPO (Fig. 4D, AUC = 0.917), and PRTN3 (Fig. 4E, AUC = 0.894) exhibited varying degrees of diagnostic ability in distinguishing AHO samples from normal samples.

General characteristics of the study population

A total of 61 patients with AHO and 25 healthy patients were recruited into the study. The demographic, clinical, and biochemical characteristics are provided in Table 1. There were no significant differences between the two groups regarding sex, age, height, or weight (p > 0.05 for all). However, in the AHO group, the levels of WBC, CRP, ESR, and PCT were significantly higher compared to the healthy control group (p < 0.05). According to the Cierny-Mader staging system, AHO is classified into four stages. However, due to the rarity of Stage IV cases, they were not included in the statistical analysis of this study. As shown in Table 2, there were no significant statistical differences in age, gender, height, or weight (p > 0.05). However, WBC, CRP, ESR, and PCT levels showed statistically significant differences, indicating that these indicators are related to the severity of the disease.

Table 1

Baseline characteristics and factors of all the patients included in the study
Demographic data	Healthy	Osteomyelitis	p-value
Age (years)	4.8 ± 2.1	3.8 ± 2.2	0.067
Sex (males, n)	14(56%)	39(64%)	0.500
Height (cm)	106.2 ± 13.9	100.4 ± 18.3	0.154
Weight (Kg)	18.4 ± 6.1	15.9 ± 5.6	0.071
WBC (/ul)	8416.0 ± 1820.0	16027 ± 6435.0	<0.001
CRP (mg/l)	4.0 ± 1.6	69.0 ± 43.2	<0.001
ESR (mm/h)	6.8 ± 3.2	60.6 ± 42.0	<0.001
PCT (ng/mL)	0.03 ± 0.01	1.65 ± 2.12	<0.001

Table 2

Characteristics of AHO Patients Based on the Cierny-Mader Staging System
Demographic data	Stage I	Stage II	Stage III	p-value
Age (years)	3.8 ± 2.2	4.0 ± 2.2	3.6 ± 2.3	0.879
Sex (males, n)	19(65%)	17(59%)	17(67%)	0.878
Height (cm)	99.8 ± 19.0	101.9 ± 18.0	99.8 ± 18.4	0.920
Weight (Kg)	15.1 ± 5.4	16.6 ± 5.3	16.5 ± 6.5	0.598
WBC (/ul)	10737.9 ± 2061.1	17500.0 ± 4548.4	24586.7 ± 2499.1	<0.001
CRP (mg/l)	29.8 ± 13.2	62.3 ± 15.3	113.3 ± 23.3	<0.001
ESR (mm/h)	23.9 ± 14.7	84.6 ± 21.7	112.3 ± 18.9	<0.001
PCT (ng/mL)	0.21 ± 0.26	2.07 ± 1.85	3.94 ± 2.24	<0.001

MPO, PRTN3 and CTNND1 expression level in peripheral blood cells is an independent predictive biomarker for AHO diagnosis

We performed clinical tests on peripheral blood from AHO patients and healthy individuals and found that MPO, PRTN3, and CTNND1 exhibited significant statistical differences, with p-values less than 0.05 (Fig. 5A-C). These indicators were significantly elevated in AHO patients. To further validate the diagnostic efficiency of MPO, PRTN3, and CTNND1 for AHO patients, we plotted ROC curves, where the AUC for MPO, PRTN3, and CTNND1 was 0.9803, 0.9767, and 0.8832, respectively (Fig. 5D-F). According to the Cierny-Mader staging system, we classified osteomyelitis into three stages. MPO and PRTN3 levels in serum increased with the progression of the stages, showing statistically significant differences between each stage. In contrast, CTNND1 levels in serum decreased as the stage advanced, with statistically significant differences observed between Stage I, Stage II, and Stage III (Fig. 5G-I).

Next, we evaluated the expression levels of MPO, PRTN3, and CTNND1 in peripheral blood cells obtained from AHO patients before and after surgery. Interestingly, the expression levels of MPO and PRTN3 in peripheral blood were significantly lower in the postoperative state compared to the preoperative state. In contrast, the expression level of CTNND1 was significantly higher after surgery (P < 0.05; Fig. 6A-C).

AHO is the most common cause of osteomyelitis in children [11]. If not treated promptly, it can progress to chronic osteomyelitis, severely impacting the quality of life of affected children [12]. Therefore, timely diagnosis of AHO is crucial for managing the disease's progression. However, an early diagnostic method is still lacking in clinical practice. In this study, we analyzed datasets from the GEO database and identified 125 genes that were upregulated in AHO. GO analysis indicated that these genes are primarily involved in positive regulation of locomotion, defense response to bacteria, wound healing, secretory granule lumen, extracellular matrix, glycosaminoglycan binding, and endopeptidase activity. KEGG pathway analysis revealed significant enrichment in pathways related to focal adhesion and proteoglycans in cancer. These findings suggest that these genes actively participate in the inflammatory process and may be critical to the development of AHO. Additionally, 154 genes were identified as downregulated in AHO. GO analysis indicated that these genes are mainly involved in neuron projection morphogenesis, behavior, cell-cell junctions, the perinuclear region of the cytoplasm, signaling receptor activator activity, and purinergic nucleotide receptor activity. KEGG pathway analysis revealed significant enrichment in the PI3K-Akt signaling pathway and Calcium signaling pathway. These findings suggest that these genes are inhibited during inflammation and may play a crucial role in mitigating the progression of AHO.

To identify potential diagnostic biomarkers for AHO, we analyzed the aforementioned 279 DEGs using three machine learning algorithms and ultimately identified five key genes: LOC401525, VARS, CTNND1, MPO, and PRTN3. Among these, LOC401525, VARS, and CTNND1 were downregulated, while MPO and PRTN3 were upregulated. Currently, there are no literature reports on the function of LOC401525. VARS is a valyl-tRNA synthetase, primarily responsible for attaching valine to tRNA during protein synthesis [13]. Current research suggests that VARS is mainly associated with hereditary neurological diseases [14]. The CTNND1 gene encodes a protein called p120-catenin, which is a key component of adherens junctions [15]. These structures play a crucial role in maintaining cell-cell adhesion and tissue structural integrity [16]. In this study, we hypothesize that Staphylococcus aureus infections typically invade and spread by disrupting the adhesive structures of host cells. In the context of osteomyelitis, CTNND1 may play a role in maintaining cellular structure and resisting the spread of infection. If CTNND1 function is impaired or its expression is reduced, it may lead to weakened cell-cell adhesion, thereby facilitating the propagation and spread of Staphylococcus aureus infection.

Using machine learning algorithms, we identified LOC401525, VARS, and CTNND1 as downregulated genes in AHO samples. However, cohort studies using sequencing data revealed that LOC401525 had a p-value greater than 0.05, indicating no statistically significant difference. ROC analysis of the data for VARS and CTNND1 showed AUC values of 0.7614 and 0.8106, respectively. Given the current literature on VARS and the observation that the AUC curve for VARS was significantly lower than that of the other three genes during ROC analysis, further testing of VARS was not pursued in subsequent human experiments. Based on related research and our experimental studies, we believe that CTNND1 is a potential diagnostic biomarker for AHO.

MPO is a heme-containing enzyme primarily produced by neutrophils and monocytes, playing a critical role in immune responses and inflammation [17]. It is predominantly located in the lysosomes of these cells. During an inflammatory response, neutrophils generate hydrogen peroxide through a respiratory burst, and MPO utilizes this hydrogen peroxide to react with chloride ions, producing hypochlorous acid. Hypochlorous acid is a potent oxidant capable of killing a wide range of pathogens, including bacteria, fungi, and viruses [17–18]. However, the hypochlorous acid and other oxidative products generated by MPO not only exert bactericidal effects but also have the potential to damage host cells and tissues [19]. Current research indicates that MPO contributes to the pathogenesis of osteomyelitis not only through its antimicrobial activity but also by exacerbating local inflammatory responses via the production of oxidative products. These oxidative products can induce oxidative stress, leading to tissue damage, further intensifying inflammation, and consequently delaying the healing of bone tissue [20–21]. In chronic osteomyelitis, sustained MPO activity may result in a prolonged inflammatory state, hindering bone tissue repair, and ultimately leading to persistent tissue destruction and functional loss. Additionally, elevated MPO levels in osteomyelitis patients have been found to correlate with disease severity [22]. PRTN3 is a serine protease primarily located in the granules of neutrophils. In the antibacterial defense mechanism of neutrophils, PRTN3 plays a crucial role by cleaving microbial proteins and directly killing pathogens [23]. Moreover, PRTN3 can activate other antimicrobial molecules, further enhancing the bactericidal capacity of neutrophils [24]. Additionally, PRTN3 is capable of cleaving and modulating various cytokines and chemokines, thus influencing the intensity and duration of the inflammatory response [24]. However, overexpression of PRTN3 may disrupt the bone matrix and soft tissues, impeding bone tissue repair and reconstruction, thereby delaying the recovery process [25]. In this study, we observed significant differences in the expression levels of PRTN3 and MPO between AHO samples and normal samples. ROC analysis further validated their robust ability to distinguish AHO samples from normal samples. Importantly, in our cohort, we confirmed that the expression levels of PRTN3 and MPO were significantly upregulated in AHO samples. Our study also reveals that MPO and PRTN3 are positively correlated with the severity of AHO, and their levels show a decreasing trend after treatment. This suggests that MPO and PRTN3 can serve as indicators for evaluating the effectiveness of AHO treatment. Additionally, current literature suggests that a sustained inflammatory response involving MPO and PRTN3 may contribute to the transition from acute to chronic osteomyelitis. Further experiments could be conducted to explore the mechanisms of chronic osteomyelitis formation by regulating MPO and PRTN3.

In conclusion, this study found that the expression levels of MPO, PRTN3, and CTNND1 play important roles in the serum of children with AHO. According to the Cierny-Mader staging system for osteomyelitis, PRTN3 and CTNND1 were positively correlated with the severity of the disease, while CTNND1 was negatively correlated. Therefore, MPO, PRTN3, and CTNND1 may serve as potential biomarkers for predicting S. aureus-induced acute osteomyelitis in children.

AHO Acute hematogenous osteomyelitis

GEO Gene Expression Omnibus

DEGs Differentially expressed genes

LASSO The Least Absolute Shrinkage and Selection Operator

SVM-RFE Support Vector Machine-Recursive Feature Elimination

RF Random Forest

ROC Receiver operating characteristic

MPO Myeloperoxidase

PRTN3 Proteinase 3

CTNND1 Catenin delta 1

RANKL Receptor activator of nuclear factor kappa-Β ligand

GO Gene Ontology

BP Biological process

CC Cellular component

MF Molecular function

KEGG Kyoto Encyclopedia of Genes and Genomes

WBC White blood cell

CRP C-reactive protein

ESR Erythrocyte sedimentation rate

PCT Procalcitonin

ELISA Enzyme-linked immunosorbent assay

Acknowledgements

We extend thanks to our patients.

Authors’ contributions

All authors have contributed to the study conception and design, material preparation, data analysis, and writing, reviewing, and editing of the fnal draft. The authors have read and agreed to publish final version of the manuscript. This work was performed at the Affiliated Hospital of Zunyi Medical University, Zunyi Medical University, Zunyi, China. Kezhi Chen: performed measurements, manuscript preparation. Jihang Luo: collected the data. Jiafei Yang: performed the statistical analysis, and wrote the draft. Tianjiu Zhang, Song Yu: reviewed and approved the article.

Funding

Guizhou Provincial Science and Technology Department, Qian Ke He support (2021) General 076. 2. Guizhou Provincial Science and Technology Department, Guizhou Science and Technology Cooperation Platform Talents (2020) 6015-2.3. 3. Health Commission of Guizhou Province (Grant No.gzwkj2024-036).

Availability of data and materials

Author Jiafei Yang can provide original data upon reasonable request. His email is [email protected].

Ethics approval and informed consent section

This study protocol was reviewed and approved by the Ethics Committee of the Affiliated Hospital of Zunyi Medical University (Approval No: 2022 Review No. [521]). Informed consent was obtained from all subjects involved in the study.

Consent for participate

Not applicable.

Competing interests

The authors declare no competing interests.

Thakolkaran N, Shetty A K. Acute hematogenous osteomyelitis in children[J]. Ochsner Journal, 2019, 19(2): 116-122.
Woods C R, Bradley J S, Chatterjee A, et al. Clinical practice guideline by the Pediatric Infectious Diseases Society and the Infectious Diseases Society of America: 2021 guideline on diagnosis and management of acute hematogenous osteomyelitis in pediatrics[J]. Journal of the Pediatric Infectious Diseases Society, 2021, 10(8): 801-844.
McNeil J C. Acute hematogenous osteomyelitis in children: clinical presentation and management[J]. Infection and drug resistance, 2020: 4459-4473.
Arnold S R, Elias D, Buckingham S C, et al. Changing Patterns of Acute Hematogenous Osteomyelitis and Septic Arthritis: Emergence of Community-associated Methicillin-resistant: Staphylococcus aureus[J]. Journal of Pediatric Orthopaedics, 2006, 26(6): 703-708.
Urish K L, Cassat J E. Staphylococcus aureus osteomyelitis: bone, bugs, and surgery[J]. Infection and immunity, 2020, 88(7): 10.1128/iai. 00932-19.
Gornitzky A L, Kim A E, O’Donnell J M, et al. Diagnosis and management of osteomyelitis in children: a critical analysis review[J]. JBJS reviews, 2020, 8(6): e19.
Jha Y, Chaudhary K. Diagnosis and treatment modalities for osteomyelitis[J]. Cureus, 2022, 14(10).
Qu Y, Li J, Zhang W, et al. Posttraumatic osteomyelitis Risks Associated with NLRP3 gene polymorphisms in the Chinese Population[J]. Journal of Personalized Medicine, 2023, 13(2): 253.
Yao L, Huang C, Dai J. Staphylococcus aureus enhances osteoclast differentiation and bone resorption by stimulating the NLRP3 inflammasome pathway[J]. Molecular Biology Reports, 2023, 50(11): 9395-9403.
Campbell M J, Bustamante-Gomez C, Fu Q, et al. RANKL-mediated osteoclast formation is required for bone loss in a murine model of Staphylococcus aureus osteomyelitis[J]. Bone, 2024, 187: 117181.
Saavedra-Lozano J, Falup-Pecurariu O, Faust S N, et al. Bone and joint infections[J]. The Pediatric infectious disease journal, 2017, 36(8): 788-799.
Peltola H, Pääkkönen M. Acute osteomyelitis in children[J]. New England Journal of Medicine, 2014, 370(4): 352-360.
Wu Q, Tavazoie S F. Translational control by VARS in melanoma[J]. Nature Cell Biology, 2024: 1-2.
Stephen J, Nampoothiri S, Banerjee A, et al. Loss of function mutations in VARS encoding cytoplasmic valyl-tRNA synthetase cause microcephaly, seizures, and progressive cerebral atrophy[J]. Human genetics, 2018, 137: 293-303.
Kourtidis A, Ngok S P, Anastasiadis P Z. p120 catenin: an essential regulator of cadherin stability, adhesion-induced signaling, and cancer progression[J]. Progress in molecular biology and translational science, 2013, 116: 409-432.
Duñach M, Del Valle-Pérez B, Garcia de Herreros A. p120-catenin in canonical Wnt signaling[J]. Critical Reviews in Biochemistry and Molecular Biology, 2017, 52(3): 327-339.
Lin W, Chen H, Chen X, et al. The roles of neutrophil-derived myeloperoxidase (MPO) in diseases: The new progress[J]. Antioxidants, 2024, 13(1): 132.
Pravalika K, Sarmah D, Kaur H, et al. Myeloperoxidase and neurological disorder: a crosstalk[J]. ACS chemical neuroscience, 2018, 9(3): 421-430.
Lin W, Chen H, Chen X, et al. The roles of neutrophil-derived myeloperoxidase (MPO) in diseases: The new progress[J]. Antioxidants, 2024, 13(1): 132.
Frangie C, Daher J. Role of myeloperoxidase in inflammation and atherosclerosis[J]. Biomedical reports, 2022, 16(6): 1-11.
Massaccesi L, Galliera E, Pellegrini A, et al. Osteomyelitis, oxidative stress and related biomarkers[J]. Antioxidants, 2022, 11(6): 1061.
Severe chronic non-bacterial osteomyelitis in combination with total MPO deficiency and responsiveness to TNFalpha inhibition.
Guo D, Zhang B, Wu D, et al. Identification of PRTN3 as a novel biomarker for the diagnosis of early gastric cance[J]. Journal of Proteomics, 2023, 277: 104852.
Korkmaz B, Horwitz M S, Jenne D E, et al. Neutrophil elastase, proteinase 3, and cathepsin G as therapeutic targets in human diseases[J]. Pharmacological reviews, 2010, 62(4): 726-759.
Pham C T N. Neutrophil serine proteases: specific regulators of inflammation[J]. Nature Reviews Immunology, 2006, 6(7): 541-550.

No competing interests reported.

Download PDF

Editor assigned by journal
21 Aug, 2024
Submission checks completed at journal
21 Aug, 2024
First submitted to journal
20 Aug, 2024

You are reading this latest preprint version

Identification and Diagnostic Significance of MPO, PRTN3, and CTNND1 as Biomarkers in Acute Hematogenous Osteomyelitis in Children: A Comprehensive Analysis Using Machine Learning Algorithms

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Introduction

Materials and Methods

Data preprocessing

Identification of DEGs

Functional Enrichment Analysis

Candidate Diagnosis Marker Selection

Patient and Ethics statement

Biochemical assessment

Statistical analysis

Result

Screening for diferentially expressed AHO‑associated genes

Functional Enrichment Analyses

Screening target genes by LASSO regression, SVM‑RFE and RF analysis

The Expression and Diagnosis Significance of candidate genes in AHO

General characteristics of the study population

Discussion

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1