Identification of oxidative stress-related diagnostic markers from gestational diabetes mellitus and immune infiltration by machine learning

doi:10.21203/rs.3.rs-3402650/v1

Download PDF

Research Article

Identification of oxidative stress-related diagnostic markers from gestational diabetes mellitus and immune infiltration by machine learning

https://doi.org/10.21203/rs.3.rs-3402650/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

During pregnancy, the instability of glucose tolerance led to gestational diabetes mellitus (GDM) in the expectant mother, resulting in significant harm to both the pregnant woman and the infant. Oxidative stress-related genes (OSRGs) play a crucial role in diabetes. However, it is still not fully understood whether OSRGs also affect GDM. Therefore, this study sought to find hub genes from OSRGs can influence the diagnosis and treatment of GDM.

Methods

GSE203346 and GSE70493 from Gene Expression Omnibus (GEO) database were used as training and verification sets. OSRGs were downloaded from GeneCard database. The differentially expressed genes (DEGs) were identified with the limma package for GO/KEGG/GSEA analysis in the clusterProfiler package. These OSRGs were further analyzed using LASSO and SVM-RFE to identify diagnostic biomarkers for GDM. A diagnostic model was established using the Glm function and verified with ROC curve. The relationship between hub genes and immune infiltration was analyzed.

Results

The study identified 937 DEGs from GSE203346 dataset and 10 differentially expressed oxidative stress-related genes (DEOSRGs) were screened. The LASSO and SVM-RFE results showed that seven genes (DNMT3B, SLC6A2, DNMT1, CACNA1C, PENK, S100B and H2BC3) were identified as OSRGs diagnostic biomarkers for GDM. The AUC value of GDM diagnosis model was 0.9920 in training set and 0.8524 in verification set. Three hub genes (SLC6A2, PENK and S100B) were identified between DEOSRGs and LASSO and SVM-RFE results. ROC analysis showed that AUC values of SLC6A2, PENK and S100B were 0.7024, 0.7119 and 0.6964, respectively. The results of immune infiltration suggest that these SLC6A2, PENK and S100B might have regulatory role in the immune response of GDM. Finally, 19 drugs targeting S100B and SLC6A2 were retrieved from the DrugBank database, some of which may be useful for the treatment of GDM.

Conclusions

By combining DEOSRGs with two machine learning algorithms, three hub genes (SLC6A2, PENK and S100B) were explored. Three hub genes are strongly associated with several infiltrating immune cells, immunoinhibitors, immunostimulators and chemokines. These hub genes could potentially contribute to a better understanding of the development and drug treatment of GDM.

gestational diabetes mellitus

Oxidative stress-related genes

Diagnostic marker

machine learning algorithms

During pregnancy, gestational diabetes mellitus (GDM), as one of health problems threatening pregnant women, is the occurrence of diabetes or impaired glucose intolerance for the first time [1; 2]. Many studies have shown that the blood glucose levels in patients with GDM typically normalize after giving birth. However, patients with GDM during pregnancy are frequently accompanied by complications, including preeclampsia, cesarean delivery, preterm birth and infant macrosomia and a higher susceptibility to type 2 diabetes mellitus (T2DM) versus to healthy pregnant women. [3; 4]. Moreover, during their developmental stage, infants born from mothers with GDM are susceptible to multiple high-risk scenarios such as macrosomia, hypoglycemia, obesity, congenital heart defects, and metabolic syndrome [5; 6]. Between 6.1–15.2% of pregnant women globally are estimated to have GDM, and this number is on an upward trend, highlighting GDM’s growing importance as a public health concern [7; 8]. Up to 14.8% of women experience the occurrence of GDM in China [9]. Currently, there are several indications suggesting that multiple factors contribute to the development and progression of GDM, including age, body mass index, family history of diabetes, diet, physical activity [10]. The oral glucose tolerance test (OGTT) is utilized as a screening tool to identify women who may be at risk of developing GDM during late pregnancy. However, it should be noted that the screening rate of OGTT for GDM is approximately 60%, with compliance rates ranging from 16–19% [11; 12]. Considering the complexity of the pathogenesis of GDM, it is necessary to search for new diagnostic and screening methods.

When the balance of redox in cells is disturbed, it leads to an oxidative stress that causes damage to critical biological substances like proteins, DNA, lipids, and cell membranes [13]. The disturbance of oxidative stress affected the development and progression of plentiful disease [14]. Oxidative stress participates in the development of Alzheimer’s disease by promoting Aβ deposition, tau hyperphosphorylation, and the subsequent loss of synapses and neurons [15]. The occurrence of cardiovascular diseases is caused by arterial hypertension, which is promoted by vasoconstriction and a decrease in the availability of nitric oxide as a result of increased reactive oxygen species (ROS) [16]. In most cases, the balance between the production of ROS and antioxidative defense is disrupted in tumorigeneses [17]. Additionally, the increased of ROS will disrupts the β-cell function, leading to the occurrence of type 1 diabetes [18]. Insulin secretion is inhibited by chronic exposure of β-cells to oxidative stress [19]. In diabetes, hyperglycemia leads to overproduction of ROS, which further increases oxidative stress caused by the imbalance between ROS and antioxidants [20]. The identification of oxidative stress related genes (OSRGs) could potentially become new targets for the treatment and detection of diabetic nephropathy [21]. However, the function of OSRGs in GDM has not yet been fully investigated.

In this study, total 10 differentially expressed oxidative stress-related genes (DEOSRGs) were screened. The LASSO and SVM-RFE results showed that seven genes (DNMT3B, SLC6A2, DNMT1, CACNA1C, PENK, S100B and H2BC3) were identified as OSRGs diagnostic biomarkers for GDM. Three hub genes (SLC6A2, PENK and S100B) were identified between DEOSRGs and LASSO and SVM-RFE results. The results of immune infiltration suggest that these SLC6A2, PENK and S100B might have regulatory role in the immune response of GDM. Finally, drugs targeting S100B and SLC6A2 were retrieved from the DrugBank database. These hub genes could potentially contribute to a better understanding of the development and drug treatment of GDM.

Data download and processing

Gene expression data were obtained from the NCBI Gene Expression Omnibus public database (GEO). GSE203346 (GPL24676) contained 21 cases of placental tissue from GDM patients and 20 controls. As a validation dataset, GSE70493 (GPL17586) contained 32 cases of placental tissue from GDM patients and 31 controls. According to relevance score of greater than 5, total of 883 oxidative stress-related genes were screened.

Identification of DEGs

The overall DEGs were identified (|Fold change| >1, P < 0.05) by “Limma” package in R software. Heat maps and volcano plots of DEGs were created using the “pheatmap” and “ggplot2” packages.

Enrichment Analysis

The DEGs and intersection genes between DEGs and OSRGs were performed to investigate the biological mechanisms, including biological processes (BP), molecular functions (MF), cellular components (CC) and Kyoto Encyclopedia of Genes Genomes (KEGG), in GDM by “clusterProfiler” package in R software.

Screening hub genes by machine learning

We used Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression [22] and Support Vector Machine-Recursive Feature Elimination (SVM-RFE) [23]method to further narrow the range of hub genes. LASSO is a data mining technique that sets the coefficients of insignificant variables to zero, resulting in the creation of the most effective classification model by filtering out important variables by “glmnet” package in R software. By “e1071” package in R software, SVM-RFE is a technique in supervised machine learning that involves removing feature vectors generated by SVM to identify the most appropriate core genes. We evaluated the diagnostic performance of these biomarkers using ROC curves. Then, we also use GSE70493 dataset to validate these biomarkers.

Gene Set Enrichment Analysis (GSEA)

patients were divided into high group and low group according to the median hub genes. GSEA software was performed to evaluate significantly differential pathways. The selected gene sets were “c2.cp.kegg.v2023.1.Hs.symbols.gmt”. Pathways with P < 0.05 were considered significantly enriched.

Immune Infiltration analysis

The cell type identification by estimating relative subsets of RNA transcripts (CIBERSORT) was applied to assess the infiltration levels of GDM and normal groups [24]. The relative proportions of the 22 immune cells in each sample were quantified using expression data from GSE203346 in this study. xCell is a powerful computational technique that transforms gene expression patterns into 64 enrichment scores for immune and stroma cell types in different samples [25]. In this study, the expression of 64 immune and stroma cell types in GDM and normal tissue were compared. The spearman correlation analysis of three hub genes and 22 immune cell types, immunoinhibitors, immunostimulators, and chemokines were analyzed by “corrplot” package in R software.

Drugs screening

DrugBank database (www.drugbank.ca) contains an extensive range of molecular information regarding drugs and their mechanisms, interactions and targets, all of which is accessible through the internet. The drug targeting to the hub gene were analyzed by DrugBank database.

Statistical Analysis

All data has been analyzed and plotted using R software (4.2.1) or GraphPad Prism 9.0. The diagnostic value of hub genes was performed by ROC curve. All data were expressed as mean ± standard deviation. An unpaired t-test was performed for the assessment of hub gene differential expression. P < 0.05 was considered statistically significance.

Identification of oxidative stress related differentially expressed genes (OSRDEGs)

According to the criterion of |Fold change| >1 and P < 0.05, a total of 937 DEGs were screened from GSE203346, including 104 down-regulated genes and 833 up-regulated genes (GDM versus Normal). These DEGs were presented in a volcano plot (Fig. 2A). A heatmap of the top 20 differentially expressed genes was plotted (Fig. 2B). Additionally, those DEGs mainly involved in the cell proliferation, some signaling pathway, DNA binding (Fig. 2D-2G).

To explore the role of OSRGs in the occurrence of GDM, 883 OSRGs were acquired from GeneCard database. Ten OSRDEGs associated with OSRGs and GDM were obtained by taking the intersections of 937 DEGs from GSE203346 and 883 OSRGs (Fig. 3A). histogram showed that S100B was lower expressed in GDM, and MAPK10, SLC6A2, TRPV1, ALOX12, CETP, PENK and CYP2C19 were highly expressed in GDM group from GSE203346 (Fig. 3B). Through GO and KEGG analysis, those gene may participate in the “cellular response to hypoxia”, “positive regulation of gene expression”, “Osteoclast differentiation” and “Inflammatory mediator regulation of TRP channels” (Fig. 3C).

Identification and evaluation of GDM diagnostic biomarkers

LASSO regression and SVM-RFE were utilized to screen hub genes from OSRGs to identify diagnostic biomarkers for GDM. Seven genes were extracted from OSRGs using the LASSO regression algorithm (Fig. 4A and 4B). 708 features were obtained using SVM-RFE algorithm (Fig. 4C). After taking the intersection of the results obtained by these two methods, seven overlapping hub genes were obtained (Fig. 4D). The expression of DNMT3B, SLC6A2, DNMT1, CACNA1C and PENK were increased in GDM, whereas, the expression of S100B and H2BC3 were reduced in GDM (Fig. 4E). The diagnostic model was evaluated using ROC curves, with an AUC value of 0.9920 (95% CI: 0.9728 to 1.000 and P < 0.0001) in the training set, and an AUC of 0.8524 (95% CI: 0.7555 to 0.9494 and P < 0.0001) in the validation set (Fig. 4F and 4G).

To further refine the range of hub genes, we intersected the 10 hub genes that were screened with OSRDEGs. As depicted in Fig. 5A, three overlapping genes (S100B, PENK and SLC6A2) were chosen, indicating their potential significance in the onset and progression of GDM. The expression of S100B was reduced in GDM, whereas, the expression of PENK and SLC6A2 were increased in GDM (Fig. 5B). To explore the diagnostic efficacy of the 3 hub genes, we implemented a ROC curve analysis the AUC values were 0.6964 for S100B, 0.7119 for PENK, and 0.7024 for SLC6A2 (Fig. 5C).

Pathways enriched for the three Hub genes

To understand the potential biological roles of these genes, enrichment analyses were performed. The PENK low expression group was mostly concentrated in enplot_KEGG_ABC_TRANSPORTERS_39 “ABC Transporters”, “Calcium signaling pathway”, “Dilated cardiomyopathy”, “Glycosaminoglycan biosynthesis chondroitin sulfate” and “Taste transduction” (Fig. 6A). The S100B high expression group was highly enriched for “Nitrogen metabolism” (Fig. 6B). The SLC6A2 high group mainly associated with several metabolism-related pathway (Fig. 6C).

Immune infiltration analysis

The microenvironment has an important impact on the sensitivity of clinical treatment and disease diagnosis. In this study, the proportions of 22 immune cells in GDM samples and normal group samples were estimated using the CIBERSORT algorithm, as shown in the histogram (Fig. 7A). Additionally, the correlation of the 22 immune cell types was also investigated (Fig. 7B). The boxplot displays statistical differences of Plasma cells between GDM group and control group, which may collectively regulate the progression of GDM in conjunction with the aforementioned interaction of immune cells (Fig. 7C). What’s more, the expression of Basophils, CD8 + Tem and pro B-cells were increased in GDM group, however, the expression of CD4 + memory T-cells, DC and Preadipocytes were reduced in GDM group (Supplementary Fig. 1).

Next, we further analyzed the correlation between three hub genes and 22 immune cells. The results showed varying degrees of association between hub genes and immune cells (Fig. 7D). Correlation analysis revealed a positive correlation between S100B expression and Neutrophils (R = 0.49), Dendritic cells resting (R = 0.35), Macrophages M2 (R = 0.32) and a negative correlation between S100B expression and Monocytes (R = -0.36). SLC6A2 expression was negatively correlated with Mast cells activate (R = -0.31). PENK expression was positively correlated with T cells regulatory (T regs) (R = 0.54), T cells CD8 (R = 0.43), T cells follicular helper (R = 0.33) and negatively correlated with T cells CD4 naïve (R = -0.35). The correlation heatmaps showed a significant correlation between the three hub genes and most immune factors, including immunoinhibitors, immunostimulators, and chemokines (Figs. 8A-C). Those results suggested that hub genes may be involve in the immune microenvironment of GDM, which also can further deepen our understanding of the relationship between hub genes, immune cells, and GDM.

Drugs From the DrugBank

Based on drug and target information from the DrugBank database, 19 drugs targeting S100B and SLC6A2 were identified (Table 1). Among these medications, most have already been approved for use, which opens up new possibilities for treating GDM. However, it is crucial to conduct further experimental validation to ensure the effectiveness of these medications. Unfortunately, there are currently no drugs specifically targeting PENK.

Table 1

Drugs targeting S100B and SLC6A2 obtained from the DrugBank database.
	ID	Name	DRUG GROUP
S100B	DB01373	Calcium	nutraceutical
	DB05343	Arundic acid	investigational
	DB00768	Olopatadine	approved
	DB06941	(Z)-2-[2-(4-methylpiperazin-1yl) benzyl] diazenecarbothioamide	experimental
	DB07004	2-[(5-hex-1-yn-1-ylfuran-2yl) carbonyl]-N-methylhydrazinecarbothioamide	experimental
	DB04464	N-Formylmethionine	experimental
	DB11093	Calcium citrate	approved
	DB11348	Calcium Phosphate	approved
	DB14481	Calcium phosphate dihydrate	approved
SLC6A2	DB00289	Atomoxetine	approved
	DB00344	Protriptyline	approved
	DB00458	Imipramine	approved
	DB00226	Guanadrel	approved
	DB00285	Venlafaxine	approved
	DB00321	Amitriptyline	approved
	DB00476	Duloxetine	approved
	DB00540	Nortriptyline	approved
	DB00543	Amoxapine	approved
	DB00726	Trimipramine	approved

Over the past few decades, GDM has become an increasingly serious global public health issue [26]. Research has shown that children born to mothers with GDM are at greater risk of developing severe metabolic diseases, which greatly affects their offspring's lives [27]. Although researchers have found some evidence in recent years suggesting that metformin has an effect in treating T2DM, its effectiveness in treating GDM is limited [28]. Furthermore, the prevalence of GDM continues to rise. In current study, we screened 937 DEGs genes, including 104 down-regulated genes and 833 up-regulated genes, between GDM and healthy control samples. Oxidative stress refers to an imbalance between reactive oxygen species and antioxidant defense, which can potentially lead to tissue damage. Previous studies have suggested the potential involvement of oxidative stress in, although the exact mechanism is still uncertain [29]. In order to explore the mechanism of oxidative stress-related genes in the progression of GDM, the 883 oxidative stress-related genes were identified from GeneCard database. After intersecting the DEGs with oxidative stress-related genes, 10 DEOSRGs were identified. Furthermore, 7 hub genes with strong predictive capabilities were selected using machine learning algorithms (LASSO and SVM_RFE) from 883 oxidative stress-related genes. To further refine the range of hub genes, three overlapping genes (S100B, PENK and SLC6A2) were chosen between DEOSRGs and hub genes from machine learning algorithms. In general, S100B, PENK and SLC6A2 can serve as biological biomarkers for GDM, thereby predicting potential therapeutic targets for GDM.

Research shows that oxidative stress-related genes play crucial role in the development of diabetes [30]. The most important aspect is that extensive research has shown the correlation between the NF-KB/IKK13/iNOS pathway and oxidative stress and lipotoxicity in insulin resistance [31; 32]. It has been demonstrated that disrupting or pharmacologically inhibiting the NF-KB pathway can effectively reverse and prevent insulin resistance, consequently decelerating the progression of diabetes [33]. Hyperglycemia can induce an upregulation of iNOS expression along with other factors that cause insulin resistance [34]. In the basic mechanism of diabetic pathophysiology and the development of advanced complications, high blood sugar leads to an excessive production of ROS and oxidative stress. This is accompanied by a decrease in antioxidant response and damage to DNA repair pathways, whereas, oxidative stress also can further worsen the progression of diabetes [35]. Previous research suggests that the development of GDM is heavily influenced by oxidative stress [36]. During normal pregnancy, oxidative stress increases the normal systemic inflammatory response, which is usually well controlled by the body's balanced mechanism detoxified by antioxidant products. However, this balance is easily disrupted, and excessive reactive oxygen species are harmful and may lead to the occurrence of GDM [37]. The levels of serum MDA in patients with GDM showed an increase, accompanied by a decrease in GSH and SOD levels. Additionally, GDM placental tissue exhibited overexpression of HO-1, Nrf2, and NQO1. The pregnancy outcomes of patients with GDM can be affected by abnormal expression of oxidative stress level [38]. The co-supplementation magnesium-zinc-calcium-vitamin D can effectively reduce the level of inflammation markers, lower the level of oxidative stress, thus effectively avoiding the occurrence and development of GDM [39]. Generally speaking, using oxidative stress level reduction as a treatment for GDM may be a limited measure.

S100B protein, which is part of the mutagenic family, comprises two calcium-binding proteins with alpha helix-loop-helix structures that play a role in cytoskeleton development and cellular growth [40; 41]. The scientist has reported that the presence of elevated levels of S100B in the bloodstream of GDM patients is linked to hyperglycemia. Consequently, heightened glucose transfer from the placenta could modify fetal metabolism, inducing fetal hyperglycemia and hyperinsulinemia. This, in turn, can lead to macrosomia and altered tissue growth [42; 43]. Furthermore, hyperglycemia has the potential to impact the release of S100B in the extracellular area, triggering damage or death of CNS cells through the activation of NO synthase or apoptotic mechanisms [44; 45]. PENK serves as an indicator for proenkephalin as it is obtained from its main source, proenkephalin A [46]. A study showed that the DNA methylation level of PENK was lower in placental tissue of GDM, which may explain the higher expression of PENK in placental tissue of GDM [47]. The SLC6A2 is involved in regulating norepinephrine turnover and metabolism, thereby affecting various central and sympathetic nervous system functions, including learning, memory, stress response, and blood flow [48]. The substitution of the nucleotide G1287A in the silencing of the SLC6A2 gene exon 9 may potentially contribute to an increased susceptibility to hypertension among patients diagnosed with T2DM [49]. Additionally, SLC6A2 gene polymorphism might be associated with the risk of T2DM. Generally speaking, S100B, PENK, SLC6A2 may participate in the occurrence process of diabetes, including T2DM and GDM.

The imbalances in immune system also have an impact on the development of GDM. Changes in T cell subtypes and elevated levels of circulating proinflammatory cytokines have been suggested as potential mechanisms involved in the development of insulin resistance in GDM [50]. The percentage of proinflammatory Th17 cells in women with GDM was significantly higher than that in the control group of women without GDM. Changes in homeostatic parameters associated with the maturation and operation of naive and memory T(regs) can lead to a decrease in the suppressive ability of the entire T(reg) population in individuals diagnosed with GDM [51]. During the process of pregnancy, accompanied by an increase in inflammation levels, the number of monocytes gradually decreases, while the incidence rate of GDM and macrosomia increases progressively [51]. In our study, the expression of Plasma cells, Basophils, CD8 + Tem and pro B-cells were increased in GDM group, however, the expression of CD4 + memory T-cells, DC and Preadipocytes were reduced in GDM group. In addition, S100B, PENK, SLC6A2 genes were closely related to some immune cells, immunoinhibitors, immunostimulators, and chemokines. Taken together, the S100B, PENK, and SLC6A2 genes may affect the development process of GDM by regulating the immune microenvironment of GDM. Therefore, targeted therapeutic strategies focusing on immune and oxidative stress are particularly important and promising.

Through bioinformatics approaches and two machine learning algorithms (LASSO regression and SVM-RFE), we obtained three hub genes (S100B, PENK, and SLC6A2) linking oxidative stress and GDM. Exploration of the pathways in which they are engaged will aid in comprehending the progression of GDM. Further, the different of immune cells between GDM and normal tissue were also explored. Last, three hub genes have been found to be associated with various immune factors, suggesting that they may play significant roles in the immune microenvironment of GDM. However, further experimental verification is still needed to verify the functions of these three genes.

Ethics Statement

Not applicable.

Availability of Data and Materials

All data are provided in this study, and raw data can be requested to the corresponding author.

Author Contributions

Jianhua Li and Lihong Chen were the overall principal investigator who conceived the study, revised the manuscript and obtained financial support. Bei Gan, Xiuyan Wu and Xia Zhang collected the data, performed the bioinformatics analysis, drew the figures and tables and wrote the manuscript. All authors critically reviewed the article and approved the final manuscript.

Funding

This study supported by Fujian Provincial Natural Science Foundation (NO.2022J01699).

Competing Interests

The authors have declared that no competing interest exists.

D.R. Coustan, Gestational diabetes mellitus. Clinical chemistry (2013) 59:1310-21. doi.10.1373/clinchem.2013.203331
K. Benhalima, C. Minschart, P. Van Crombrugge, P. Calewaert, J. Verhaeghe et al., The 2019 Flemish consensus on screening for overt diabetes in early pregnancy and screening for gestational diabetes mellitus. Acta clinica Belgica (2020) 75:340–347. doi.10.1080/17843286.2019.1637389
B. Balaji, A. Ranjit Mohan, P. Rajendra, D. Mohan, U. Ram et al., Gestational Diabetes Mellitus Postpartum Follow-Up Testing: Challenges and Solutions. Canadian journal of diabetes (2019) 43:641–646. doi.10.1016/j.jcjd.2019.04.011
L. Mohsen, D.M. Akmal, E.K.E. Ghonaim, N.M. Riad, Role of mean platelet volume and ischemia modified albumin in evaluation of oxidative stress and its association with postnatal complications in infants of diabetic mothers. The journal of maternal-fetal & neonatal medicine: the official journal of the European Association of Perinatal Medicine, the Federation of Asia and Oceania Perinatal Societies, the International Society of Perinatal Obstet (2018) 31:1819–1823. doi.10.1080/14767058.2017.1330329
N.F. Goodman, R.H. Cobin, W. Futterweit, J.S. Glueck, R.S. Legro, AMERICAN ASSOCIATION OF CLINICAL ENDOCRINOLOGISTS, AMERICAN COLLEGE OF ENDOCRINOLOGY, AND ANDROGEN EXCESS AND PCOS SOCIETY DISEASE STATE CLINICAL REVIEW: GUIDE TO THE BEST PRACTICES IN THE EVALUATION AND TREATMENT OF POLYCYSTIC OVARY SYNDROME - PART 2,. Endocrine practice: official journal of the American College of Endocrinology and the American Association of Clinical Endocrinologists (2015) 21:1415-26. doi.10.4158/ep15748.Dscpt2
P. Damm, A. Houshmand-Oeregaard, L. Kelstrup, J. Lauenborg, E.R. Mathiesen et al., Gestational diabetes mellitus and long-term consequences for mother and offspring: a view from Denmark. Diabetologia (2016) 59:1396–1399. doi.10.1007/s00125-016-3985-5
B.E. Metzger, S.G. Gabbe, B. Persson, T.A. Buchanan, P.A. Catalano et al., International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes care (2010) 33:676 – 82. doi.10.2337/dc09-1848
J.M. Lawrence, Women with diabetes in pregnancy: different perceptions and expectations. Best practice & research. Clinical obstetrics & gynaecology (2011) 25:15–24. doi.10.1016/j.bpobgyn.2010.10.003
C. Gao, X. Sun, L. Lu, F. Liu, J. Yuan, Prevalence of gestational diabetes mellitus in mainland China: A systematic review and meta-analysis. Journal of diabetes investigation (2019) 10:154–162. doi.10.1111/jdi.12854
C. Zhang, S. Rawal, Y.S. Chong, Risk factors for gestational diabetes: is prevention possible? Diabetologia (2016) 59:1385–1390. doi.10.1007/s00125-016-3979-3
J. Huynh, G. Xiong, R. Bentley-Lewis, A systematic review of metabolite profiling in gestational diabetes mellitus. Diabetologia (2014) 57:2453-64. doi.10.1007/s00125-014-3371-0
A. Ferrara, Increasing prevalence of gestational diabetes mellitus: a public health perspective. Diabetes care (2007) 30 Suppl 2:S141-6. doi.10.2337/dc07-s206
J.R. Speakman, C. Selman, The free-radical damage theory: Accumulating evidence against a simple link of oxidative stress to ageing and lifespan. BioEssays: news and reviews in molecular, cellular and developmental biology (2011) 33:255-9. doi.10.1002/bies.201000132
H. Sies, Oxidative stress: a concept in redox biology and medicine. Redox biology (2015) 4:180-3. doi.10.1016/j.redox.2015.01.002
Z. Chen, C. Zhong, Oxidative stress in Alzheimer's disease. Neuroscience bulletin (2014) 30:271 – 81. doi.10.1007/s12264-013-1423-y
T. Senoner, W. Dichtl, Oxidative Stress in Cardiovascular Diseases: Still a Therapeutic Target? Nutrients (2019) 11:doi.10.3390/nu11092090
S. Chikara, L.D. Nagaprashantha, J. Singhal, D. Horne, S. Awasthi et al., Oxidative stress and dietary phytochemicals: Role in cancer chemoprevention and treatment. Cancer letters (2018) 413:122–134. doi.10.1016/j.canlet.2017.11.002
F. Prattichizzo, V. De Nigris, E. Mancuso, R. Spiga, A. Giuliani et al., Short-term sustained hyperglycaemia fosters an archetypal senescence-associated secretory phenotype in endothelial cells and macrophages. Redox biology (2018) 15:170–181. doi.10.1016/j.redox.2017.12.001
P. Maechler, L. Jornot, C.B. Wollheim, Hydrogen peroxide alters mitochondrial activation and insulin secretion in pancreatic beta cells. The Journal of biological chemistry (1999) 274:27905-13. doi.10.1074/jbc.274.39.27905
P. Zhang, T. Li, X. Wu, E.C. Nice, C. Huang et al., Oxidative stress and diabetes: antioxidative strategies. Frontiers of medicine (2020) 14:583–600. doi.10.1007/s11684-019-0729-1
M. Xu, H. Zhou, P. Hu, Y. Pan, S. Wang et al., Identification and validation of immune and oxidative stress-related diagnostic markers for diabetic nephropathy by WGCNA and machine learning. Frontiers in immunology (2023) 14:1084531. doi.10.3389/fimmu.2023.1084531
A.J. McEligot, V. Poynor, R. Sharma, A. Panangadan, Logistic LASSO Regression for Dietary Intakes and Breast Cancer. Nutrients (2020) 12:doi.10.3390/nu12092652
X. Lin, C. Li, Y. Zhang, B. Su, M. Fan et al., Selecting Feature Subsets Based on SVM-RFE and the Overlapping Ratio with Applications in Bioinformatics. Molecules (Basel, Switzerland) (2017) 23:doi.10.3390/molecules23010052
A.M. Newman, C.L. Liu, M.R. Green, A.J. Gentles, W. Feng et al., Robust enumeration of cell subsets from tissue expression profiles. Nature methods (2015) 12:453-7. doi.10.1038/nmeth.3337
D. Aran, Z. Hu, A.J. Butte, xCell: digitally portraying the tissue cellular heterogeneity landscape. Genome biology (2017) 18:220. doi.10.1186/s13059-017-1349-1
V. Falcone, G. Kotzaeridi, M.H. Breil, I. Rosicky, T. Stopp et al., Early Assessment of the Risk for Gestational Diabetes Mellitus: Can Fasting Parameters of Glucose Metabolism Contribute to Risk Prediction? Diabetes & metabolism journal (2019) 43:785–793. doi.10.4093/dmj.2018.0218
A.H. Xiang, X. Wang, M.P. Martinez, D. Getahun, K.A. Page et al., Maternal Gestational Diabetes Mellitus, Type 1 Diabetes, and Type 2 Diabetes During Pregnancy and Risk of ADHD in Offspring. Diabetes care (2018) 41:2502–2508. doi.10.2337/dc18-0733
L.R. Mack, P.G. Tomich, Gestational Diabetes: Diagnosis, Classification, and Clinical Care. Obstetrics and gynecology clinics of North America (2017) 44:207–217. doi.10.1016/j.ogc.2017.02.002
S. Zein, S. Rachidi, I. Hininger-Favier, Is oxidative stress induced by iron status associated with gestational diabetes mellitus? Journal of trace elements in medicine and biology: organ of the Society for Minerals and Trace Elements (GMS) (2014) 28:65 – 9. doi.10.1016/j.jtemb.2013.09.009
M.A. Darenskaya, L.I. Kolesnikova, S.I. Kolesnikov, Oxidative Stress: Pathogenetic Role in Diabetes Mellitus and Its Complications and Therapeutic Approaches to Correction. Bulletin of experimental biology and medicine (2021) 171:179–189. doi.10.1007/s10517-021-05191-7
A. Marette, Mediators of cytokine-induced insulin resistance in obesity and other inflammatory settings. Current opinion in clinical nutrition and metabolic care (2002) 5:377 – 83. doi.10.1097/00075197-200207000-00005
S. Sinha, G. Perdomo, N.F. Brown, R.M. O'Doherty, Fatty acid-induced insulin resistance in L6 myotubes is prevented by inhibition of activation and nuclear localization of nuclear factor kappa B. The Journal of biological chemistry (2004) 279:41294-301. doi.10.1074/jbc.M406514200
M. Yuan, N. Konstantopoulos, J. Lee, L. Hansen, Z.W. Li et al., Reversal of obesity- and diet-induced insulin resistance with salicylates or targeted disruption of Ikkbeta. Science (New York, N.Y.) (2001) 293:1673-7. doi.10.1126/science.1061620
A. Ceriello, L. Quagliaro, M. D'Amico, C. Di Filippo, R. Marfella et al., Acute hyperglycemia induces nitrotyrosine formation and apoptosis in perfused heart from rat. Diabetes (2002) 51:1076-82. doi.10.2337/diabetes.51.4.1076
J. Lima, N.C.S. Moreira, E.T. Sakamoto-Hojo, Mechanisms underlying the pathophysiology of type 2 diabetes: From risk factors to oxidative stress, metabolic dysfunction, and hyperglycemia. Mutation research. Genetic toxicology and environmental mutagenesis (2022) 874–875:503437. doi.10.1016/j.mrgentox.2021.503437
M. Lappas, M. Permezel, G.E. Rice, Release of proinflammatory cytokines and 8-isoprostane from placenta, adipose tissue, and skeletal muscle from normal pregnant women and women with gestational diabetes mellitus. The Journal of clinical endocrinology and metabolism (2004) 89:5627-33. doi.10.1210/jc.2003–032097
E.H. Joo, Y.R. Kim, N. Kim, J.E. Jung, S.H. Han et al., Effect of Endogenic and Exogenic Oxidative Stress Triggers on Adverse Pregnancy Outcomes: Preeclampsia, Fetal Growth Restriction, Gestational Diabetes Mellitus and Preterm Birth. International journal of molecular sciences (2021) 22:doi.10.3390/ijms221810122
C. Zhang, Y. Yang, R. Chen, Y. Wei, Y. Feng et al., Aberrant expression of oxidative stress related proteins affects the pregnancy outcome of gestational diabetes mellitus patients. Am. J. translational Res. 11, 269–279 (2019)
M. Jamilian, N. Mirhosseini, M. Eslahi, F. Bahmani, M. Shokrpour et al., The effects of magnesium-zinc-calcium-vitamin D co-supplementation on biomarkers of inflammation, oxidative stress and pregnancy outcomes in gestational diabetes. BMC pregnancy and childbirth (2019) 19:107. doi.10.1186/s12884-019-2258-y
T. Yardan, A.K. Erenler, A. Baydin, K. Aydin, C. Cokluk, Usefulness of S100B protein in neurological disorders. J. Pak. Med. Assoc. 61, 276–281 (2011)
E.A. Grzybowska, Calcium-Binding Proteins with Disordered Structure and Their Role in Secretion, Storage, and Cellular Signaling. Biomolecules (2018) 8:doi.10.3390/biom8020042
D. Gazzolo, M. Bruschettini, V. Corvino, R. Oliva, R. Sarli et al., S100b protein concentrations in amniotic fluid correlate with gestational age and with cerebral ultrasound scanning results in healthy fetuses. Clin. Chem. 47, 954–956 (2001)
D. Gazzolo, M. Lituania, M. Bruschettini, S. Ciotti, R. Sacchi et al., S100B protein levels in saliva: correlation with gestational age in normal term and preterm newborns. Clinical biochemistry (2005) 38:229 – 33. doi.10.1016/j.clinbiochem.2004.12.006
J.F. Plows, J.L. Stanley, P.N. Baker, C.M. Reynolds, M.H. Vickers, The Pathophysiology of Gestational Diabetes Mellitus. International journal of molecular sciences (2018) 19:doi.10.3390/ijms19113342
J. Hu, A. Ferreira, L.J. Van Eldik, S100beta induces neuronal cell death through nitric oxide release from astrocytes. Journal of neurochemistry (1997) 69:2294 – 301. doi.10.1046/j.1471-4159.1997.69062294.x
R. Beunders, J. Struck, A.H.B. Wu, A. Zarbock, S. Di Somma et al., Proenkephalin (PENK) as a Novel Biomarker for Kidney Function. The journal of applied laboratory medicine (2017) 2:400–412. doi.10.1373/jalm.2017.023598
W.J. Wang, R. Huang, T. Zheng, Q. Du, M.N. Yang et al., Genome-Wide Placental Gene Methylations in Gestational Diabetes Mellitus, Fetal Growth and Metabolic Health Biomarkers in Cord Blood. Frontiers in endocrinology (2022) 13:875180. doi.10.3389/fendo.2022.875180
J.R. Shannon, N.L. Flattem, J. Jordan, G. Jacob, B.K. Black et al., Orthostatic intolerance and tachycardia associated with norepinephrine-transporter deficiency. The New England journal of medicine (2000) 342:541-9. doi.10.1056/nejm200002243420803
P. Ksiazek, K. Buraczynska, M. Buraczynska, Norepinephrine transporter gene (NET) polymorphism in patients with type 2 diabetes. Kidney & blood pressure research (2006) 29:338 – 43. doi.10.1159/000097356
A. Sheu, Y. Chan, A. Ferguson, M.B. Bakhtyari, W. Hawke et al., A proinflammatory CD4(+) T cell phenotype in gestational diabetes mellitus. Diabetologia (2018) 61:1633–1643. doi.10.1007/s00125-018-4615-1
L. Schober, D. Radnai, J. Spratte, A. Kisielewicz, E. Schmitt et al., The role of regulatory T cell (Treg) subsets in gestational diabetes mellitus. Clinical and experimental immunology (2014) 177:76–85. doi.10.1111/cei.12300

Download PDF

Editorial decision: Reject
10 Feb, 2024
Reviewers agreed at journal
29 Oct, 2023
Reviewers invited by journal
10 Oct, 2023
Editor assigned by journal
03 Oct, 2023
First submitted to journal
01 Oct, 2023

You are reading this latest preprint version

Identification of oxidative stress-related diagnostic markers from gestational diabetes mellitus and immune infiltration by machine learning

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Introduction

Materials and methods

Data download and processing

Identification of DEGs

Enrichment Analysis

Screening hub genes by machine learning

Gene Set Enrichment Analysis (GSEA)

Immune Infiltration analysis

Drugs screening

Statistical Analysis

Results

Identification of oxidative stress related differentially expressed genes (OSRDEGs)

Identification and evaluation of GDM diagnostic biomarkers

Immune infiltration analysis

Drugs From the DrugBank

Discussion

Conclusion

Declarations

References

Status:

Version 1