†These authors contributed equally.
Academic Editor: Yoshihiro Noda
Background: Overexposure to manganese (Mn) can lead to
neurodegenerative damage, resulting in manganism with similar syndromes to
Parkinson’s disease (PD). However, little is known about changes in
transcriptomics induced by the toxicological level of Mn. In this study, we
conducted RNA-seq to explore the candidate genes and signaling pathways included
by Mn in human SH-SY5Y neuroblastoma cells. Methods: The differentially
expressed genes (DEGs) between the Mn-treated group and the control group were
screened, and weighted gene co-expression network analysis (WGCNA) was employed
to identify hub genes. Then, pathway enrichment analyses for those candidate
genes were performed in Gene Ontology (GO) and Kyoto Encyclopedia of Genes and
Genomes (KEGG). We further validated the concentration- and time-response effects
of Mn exposure (0–500
Manganese (Mn) is an essential nutrient metal used in various physiological processes, including lipid, protein, and carbohydrate metabolism, especially in neurodevelopment [1]. However, the excessive accumulation of Mn in the central nervous system (CNS) can lead to manganism that features symptomatology similar to Parkinson’s disease (PD) [2]. The biological mechanisms of neurotoxic effects induced by Mn include oxidative stress, mitochondria dysfunction, protein folding abnormalities, endoplasmic reticulum stress, autophagy disorders, apoptosis, and impaired metabolism pathways [3, 4]. Although mitochondria are the main targets of Mn, the specific mechanism of its neurotoxicity is not known [5, 6].
The previous study has shown endoplasmic reticulum unfolded protein response
(UPR
To fully explore the molecular participation in Mn neurotoxicity, the
transcriptional changes in neurocytes were detected. The appropriate maximum
exposure concentration of Mn was selected by cell viability. After the screening
of differentially expressed genes (DEGs), weighted gene coexpression network
analysis (WGCNA) was used to identify critical modules and candidate genes to
reveal potential biomarkers. The potential biological pathways were explored in
Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases.
Furthermore, we validated the concentration- and time-response effects of Mn on
the expression of candidate genes by qRT-PCR in SH-SY5Y cells induced by Mn
(0–500
The human neuroblastoma cell line SH-SY5Y was purchased from the Cell Resource
Center of Peking Union Medical College. The cells were cultured [13] in the high
glucose (4.5 g/L D-Glucose) DMEM (Gicbo Inc, 11960044, Grand Island, NE, USA),
medium containing 10% fetal bovine serum, 2% L-Glutamine, and 1% sodium
pyruvate at 37 °C, 5% CO
CCK-8 kit (ZOMANBIO Inc., ZP328, Beijing, China) was used to measure the
viability of SH-SY5Y cells. A total of 1
According to ISO 10,993–5, the cell viability threshold for the test is
To eliminate differences between samples, we used a robust multichip average
algorithm (RMA), including background correction (removes array
auto-fluorescence), quantile normalization (makes all intensity distributions
identical), and probe set summarization (calculates one representative value per
probe set). After gene expression data processing and normalizing, we used the
“limma” package in R (version 4.0.3, https://www.r-project.org/) to screen
differentially expressed mRNA,
We use the “WGCNA” package in R (version 4.0.3) to analyze
the genes [14, 15]. The scale-free fitting index of various soft threshold powers
(power-value = 0.85) was analyzed by the coexpression network. The nodes with
high correlation were placed into a single module, and the genes were clustered
together with the module eigengene and intramodular. Then the clustering tree was
cut into different modules using dynamic shearing. The hub genes in a module are
the genes with the maximum connectedness (KWithin) under the module. The MM value
can be obtained by correlation analysis between the expression amount of the gene
and the first principal component of the module. At the same time, correlation
analysis was conducted between the expression level of this gene and the
corresponding phenotypic value, and the final correlation coefficient value was
GS. Hub gene is defined as the gene of
GO enrichment analysis was used to determine target molecular functions (MF),
biological processes (BP), and cellular components (CC) of DEGs. Based on the
KEGG database, we performed pathway enrichment analysis for DEGs. For both
analyses, Benjamini-Hochberg statistical method, p
Based on the candidate genes, and previous research, we validated the expression
of ATF3, CCL2, CHOP, CLPP, HSP10, HSP60, HSP90, and LONP1 by qRT-PCR. We isolated
the total RNA from SH-SY5Y cells by the Trizol method (Solarbio Inc., 15596026,
Beijing, China). cDNA was synthesized from 1
As shown in Fig. 1, the normalized data were distributed in a symmetrical
median, indicating relatively high reliability of the experimental design and
data. Among the 19,991 total mRNAs in SH-SY5Y cells exposed to Mn, the Mn-treated
group had 860 DEGs (
Differentially expressed genes. The X-axis shows the FC
(log-scaled), and Y-axis indicates p-values (log-scaled). Red and blue
dots represent up-regulated and down-regulated genes, respectively. Grey dots
represent non-DEGs. Under the criterion of
The first 50% median of absolute deviation was incorporated into the WGCNA. The dynamic tree cutting method was used to identify each module, and the correlation degree was more than 0.75 modules consolidated into 18 modules (Fig. 2A). The merged module correlation diagram was redrawn, and the correlation coefficients between the modules and phenotype were visualized (Fig. 2B). There were 1396 hub genes clustered with the most significant positive correlation, while the “turquoise” module was the most significant negative correlation, with 2843 hub genes aggregated (Fig. 2C,D). Intersection analysis with DEGs showed that after Mn exposure, the expression level of 823 candidate genes changed significantly, with 160 up-regulated genes and 663 down-regulated genes (Fig. 2E). These genes were highly correlated with their corresponding modules and relative traits.
Visualization of mRNAs expression hierarchical clustering, and
gene module partitioning. (A) Clustering of mRNAs, cutting the clustering tree
by dynamic shearing into different modules. The cut height was set as 0.75 to
merge similar modules. (B) The correlation coefficients between models and
phenotype. Red and blue represent positive correlation and negative correlation,
respectively. (C) The module with the most significant positive correlation. (D)
The module with the most significant negative correlation. Under the criterions
of
To clarify the function of the candidate genes, we performed the GO enrichment analysis. The results showed that differentially expressed candidate genes were mainly involved in 10 terms related to neurotoxicity (Fig. 3A), which enriched 73 mRNAs (Fig. 3B,C). By GO enrichment analysis of these genes, the potential biological functions of nerve cell injury were further discussed. The results showed that Mn mainly affected the regulation of neuron death, and mitochondrial functions including regulation of reactive oxygen species biosynthetic process, mitochondrial outer membrane permeabilization, the release of cytochrome c from mitochondria, and apoptotic signaling pathway (Fig. 3D). As mitochondria are the major target of Mn, KEGG pathway enrichment analysis was further performed for genes enriched in mitochondrial-related functions. These genes participated in the MAPK signaling pathway, UPR, longevity regulating pathway, inflammatory bowel disease, and mitophagy (Fig. 3E).
Identification and functional enrichment of candidate genes. (A) The biological process, hub genes enriched, includes 10 GO terms related to neurotoxicity. The X-axis shows the negative logarithm of the p-value, Y-axis shows the name of the GO terms. (B) The hub genes are mainly enriched in biological processes related to neurotoxicity. (C) Heatmap of candidate genes related to neurotoxicity. Red and blue represent positive correlation and negative correlation, respectively. (D) GO enrichment analysis of candidate genes. (E) KEGG enrichment analysis of candidate genes.
Recent studies showed UPR is a new mechanism
for neurodegenerative diseases, therefore the expressions of UPR
The expressions of UPR
Mn is widely used in industrial production, such as mining, metal smelting,
welding, and other fields. Mn can cause neurotoxicity in vitro and in vivo,
leading to manganism, Parkinson-like symptoms, pathologically characterized by
the loss of dopaminergic neurons [16]. In this study, we found the expressions of
ATF3 and CCL2 increased, while the expressions of CHOP, CLPP, and LONP1 decreased
in a concentration- and time-dependent manner after Mn exposure in SH-SY5Y cells.
Our results revealed that the UPR
Manganism is most commonly associated with occupational or environmental exposure to Mn, which exhibits neurotoxicity similar to PD and is characterized by cognitive and motor dysfunction. PD is a progressive movement disorder characterized by selective neurodegeneration of dopaminergic neurons in the substantia nigra [17]. Under neurotoxic conditions, Mn has been shown to accumulate in the subcortical structures of the basal ganglia, particularly in the substantia nigra pallidum and striatum. Positron emission tomography imaging in the striatum of brains exposed to Mn has revealed impaired DA transmission, a feature of PD [18]. Proteins related to PD pathogenesis were accumulated in Mn treated SH-SY5Y cells [19, 20]. The human neuroblastoma cell line, SH-SY5Y, is a common model in studies related to neurotoxicity and neurodegenerative diseases [13, 21]. Although the SH-SY5Y cell line exhibits multiple genetic aberrations due to its cancer origin, each PD-related gene has at least one copy and the major PD pathways were intact in the SH-SY5Y genome [22]. However, the difference between their transcriptome and that of healthy neurons affects the neurotoxicity of Mn is worth exploring.
PERK, IRE1
Using comprehensive bioinformatics methods and qRT-PCR, we also found ATF3 and
CCL2 increased in a concentration- and time-response manner. Both ATF3 and CCL2
participated in regulating the mitochondrial integrated stress response. CCL2 is
involved in immunomodulatory and inflammatory processes as a chemokine.
Endothelial cell injury upregulates downstream molecules of UPR
Our results indicated an interesting phenomenon that Mn can simultaneously
activate and inhibit gene expression of UPR
Our results showed Mn could trigger and inhibit UPR
AD, Alzheimer’s disease; ATF3, activating transcription factor 3; CCL2, C-C
motif chemokine ligand 2; CHOP, C/EBP homologous protein; CLPP, caseinolytic
protease P; CNS, the central nervous system; DEGs, differentially expressed
genes; FC, fold change; GO, Gene Ontology; GS, gene significance; HSP10, heat
shock protein 10; HSP60, heat shock protein 60; HSP70, heat shock protein 70;
HSP90, heat shock protein 90; KEGG, Kyoto Encyclopedia of Genes and Genomes;
LONP1, Lon protease 1; MAPK, mitogen-activated protein kinases; MM, module
member-ship; Mn, manganese; PD, Parkinson’s disease; UPR, unfolded protein
response; UPR
These should be presented as follows: PN, JL, SZ and LC designed the research study. SZ and LC performed the research. YZ, JM, HJ provided help and advice on validation. SZ, LC and JL analyzed the data. SZ, JL and LC wrote the manuscript. TC, JM, CG and ZX contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.
Not applicable.
Not applicable.
This research was funded by the National Natural Science Foundation of China, grant number 81973007.
The authors declare no conflict of interest. JL is serving as one of the Guest editors of this journal. We declare that JL had no involvement in the peer review of this article and has no access to information regarding its peer review. Full responsibility for the editorial process for this article was delegated to YN.
Publisher’s Note: IMR Press stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.