DNA Hypomethylation of DOCK1 Leading to High Expression Correlates with Neurologic Deterioration and Poor Function Outcomes after Spontaneous Intracerebral Hemorrhage.

OBJECTIVE: Spontaneous intracerebral hemorrhage (ICH) is a blood clot arising in the brain parenchyma in the absence of trauma or surgery and accounts for 10% to 15% of all strokes, leading to higher rates of mortality and morbidity than either ischemic stroke or subarachnoid hemorrhage. We sought to investigate the potential association of DOCK1 with neurological deficits and outcomes in patients with spontaneous ICH.
METHODS: Identification of methylation-regulated differentially expressed genes (MeDEGs) between ICH patients and matched controls was performed by analyzing the raw data from the GSE179759 and GSE125512 datasets deposited in the Gene Expression Omnibus. A total of 114 patients who were admitted to our hospital for spontaneous ICH were retrospectively analyzed, with 108 healthy volunteers who had received physical examinations at the same period as controls. The mRNA expression of DOCK1 was determined by quantitative real-time polymerase chain reaction (qRT-PCR). The hematoma volume was calculated according to the Coniglobus formula. The severity of neurological deficits was evaluated using National Institutes of Health Stroke Scale (NIHSS) scores and function outcomes were evaluated by modified Rankin Scale (mRS) scores.
RESULTS: A total of 15 MeDEGs between ICH patients and matched controls were identified. The mRNA expression of DOCK1 was remarkably higher in the serum samples of patients with spontaneous ICH than in the healthy controls. According to hematoma volume after ICH attack, small (<10 mL), medium (10 to 30 mL), and large (>30 mL) groups were arranged. The proportions of male patients and patients aged ≥60 years were significantly higher in the large group than in the small and medium groups (P < 0.05). The mRNA expression of DOCK1 was significantly higher in the large group than in the small and medium groups (P < 0.05). According to NIHSS scores, mild (NIHSS scores ≤15), moderate (NIHSS scores from 16 to 30), and severe (NIHSS scores from 31 to 45) groups were classified. It was observed that the severe group had higher proportions of male patients and patients aged ≥60 years than the mild and moderate groups (P < 0.05). The severe group exhibited a higher mRNA expression of DOCK1 than the mild and moderate groups (P < 0.05). According to mRS scores, higher proportions of male patients and patients aged ≥60 years were observed in the unfavorable group than the favorable group (P < 0.05). The patients in the unfavorable group showed an elevated DOCK1 mRNA expression compared to those in the favorable group (P < 0.05).
CONCLUSION: The study provided evidence that male gender, older age, and higher DOCK1 mRNA expression were related to higher admission hematoma volume, neurologic deterioration, and poor function outcomes in patients with spontaneous ICH.

1. Introduction

Intracerebral hemorrhage (ICH) refers to the hemorrhage caused by nontraumatic rupture of blood vessels in brain parenchyma, which includes primary ICH and secondary ICH [1]. Primary ICH is characterized by nonsingle definite potential lesion and mostly caused by hypertensive and cerebral amyloidosis, accounting for 78% to 88% of cases of ICH. Secondary ICH is induced by various definite factors, such as the presence of vascular malformations or aneurysms, vasculitis, brain trauma, and coagulation disorders [2, 3]. ICH is considered to be the second common stroke subtype after ischemic stroke, with about 10% to 20% of all strokes, leading to severe disability or death [4]. A variety of potential factors, such as groups above 45 years of age, male gender, Asians, and low- and middle-income countries, have been identified related to the occurrence of ICH [5, 6]. Additionally, arterial hypertension, heavy smoking, excessive drinking, obesity, and diabetes [7] as well as oral anticoagulants [8] have been associated with an increased risk of ICH. Among them, degenerative changes of small artery perforating branches caused by hypertension are considered as the primary risk factor of ICH [9]. ICH patients benefit from targeted neurointensive and surgical interventions, early control of elevated blood pressure, rapid reversal of abnormal coagulation, active management goals with avoidance of early palliative care orders, and well-coordinated specialist stroke unit care [10]. At present, computed tomography of the head (CTH) is the first choice for the diagnosis of ICH. Although the parameters presented in CTH together with clinical symptoms have been used to predict hematoma enlargement and nervous system deterioration, no specific intervention can improve the clinical outcome after ICH [11]. Therefore, identification of pathogenesis of ICH is particularly important for the prevention and treatment of ICH.

Dedicator of cytokinesis 1 (DOCK1, also known as DOCK180) is a prototypical member of a new family of guanine nucleotide exchange factors (GEFs) which have the ability to temporally and spatially regulate the activation of Rho GTPases, such as Ras-related C3 botulinum toxin substrate 1 (Rac1) as one of regulators of actin cytoskeleton dynamics and endothelial contractions [12]. Not surprisingly, DOCK family members have been engaged in a variety of pathologies, including carcinogenesis and impairments in the central nervous and immune systems [13]. Recent literatures showed that DOCK1 overexpression may participate in tumor development and metastasis, including breast cancer [14], bladder cancer [15], and acute myeloid leukemia [16]. More specifically, DOCK1 could promote the IL8-induced chemotaxis and mesenchymal transition of glioma cells [17].

It has been reported that genetic background is responsible for about 37.9% of stroke pathogenesis [18, 19]. The frequency of susceptible genetic variants in stroke patients ranges from 5 to 10%, indicating the presence of cryptic genetic changes beyond DNA sequences to explain stroke pathogenesis, such as DNA methylation. Accordingly, we firstly performed integrative analysis of DNA methylation profiles between ICH patients and healthy controls and found DNA hypomethylation of DOCK1 after ICH attack. In the subsequent analyses, the association between DOCK1 expression levels, hematoma volume, neurological impairments, and clinical outcome in ICH was clarified.

2. Materials and Methods

2.1. Microarray Datasets

The DNA methylation profiles deposited by Zhang et al. with accession no. GSE179759, including peripheral blood samples from 30 patients with ICH and 34 healthy controls, were downloaded from the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo). GSE179759 was generated on the GPL23976 platform and detected by Illumina Infinium Human Methylation 850 BeadChip (Illumina, Inc., San Diego, CA, USA). The genome-wide expression profiles deposited by Walsh et al. (accession no. GSE125512), containing 11 pairs of peripheral blood samples within 24 hours of symptom onset or 72 hours following the first collected from 11 patients with ICH, were also downloaded from the GEO. GSE125512 was generated on the GPL15433 platform and detected by Illumina HiSeq 1000 (Homo sapiens).

2.2. Differential Expression Analysis

The raw data of GSE179759 were imported into R version 3.2.2 (http://www.R-project.org/), and site-level analysis was performed based on the Illumina Methylation Analyzer package (Illumina, Inc.) to screen the differentially methylated CpG sites (DMSs) between ICH peripheral blood samples and matched control samples, with |log2fold change (FC)| > 1.5 and false discovery rate (FDR) < 0.05 as thresholds. DMSs were mapped to the corresponding genes (DMGs) and genomic regions based on the full annotation file of the microarray. For the gene expression dataset, the raw data were also introduced into R version, normalized via the affy package, and differentially analyzed by the limma package to obtain differentially expressed genes (DEGs) with the criteria of FC)| > 1.5 and FDR < 0.05.

2.3. Venn Analysis

Venn intersect function was used to reveal overlapping genes between GSE179759 and GSE125512 dataset after differential analysis. Hypomethylation-high expression genes were obtained following intersection of hypomethylated genes of GSE179759 and upregulated genes of GSE125512. Likewise, hypermethylation-low expression genes stood out after superimposition of hypermethylated genes of GSE179759 and downregulated genes of GSE125512. The hypermethylation-low expression genes and hypomethylation-high expression genes were identified as methylation-regulated differentially expressed genes (MeDEGs).

2.4. Functional Enrichment Analysis

Mapping a set of genes to the associated biological annotation is very important for gene functional analysis of any high-throughput data. Functional enrichment of hypermethylation-low expression genes and hypomethylation-high expression genes was evaluated by Gene Ontology (GO), which is a common tool to characterize biological properties for transcriptome data or high-throughput genomes. GO functional annotation refers to three fields, biological process (BP), cellular component (CC), and molecular function (MF). Each field encompasses various gene items. The biological functions of the DEGs are associated with their enrichments in GO items. GO enrichment analysis was performed by using the clusterProfiler package to analyze differentially expressed genes at the functional level, with p value < 0.05 as cutoff value.

2.5. Study Subjects

A total of 114 patients who were admitted to our hospital for spontaneous ICH between January 2019 and May 2021 were retrospectively analyzed, including 69 males and 45 females with age of 58.43 ± 7.71 years (ranging from 40 to 72 years). The time of symptom onset to hospital admission was 4.03 ± 1.11 h (ranging from 1.5 to 7 h). All included ICH patients must meet the Chinese guidelines for diagnosis and treatment of acute intracerebral hemorrhage (2019) with confirmation by CTH, be within 24 h of symptom onset to hospital admission, have age above 18 years, and sign an informed consent. The patients were excluded if they had cerebral infarction, nonspontaneous ICH attributable to trauma, metastatic hemorrhagic cerebral lesions, venous sinus thrombosis and underlying vascular lesions, administration of immunosuppressive agents, hormones or anti-infection drugs within 3 months, malignant tumors, serious infection or acute/chronic inflammatory disease, or previous history of mental and neurological diseases. Healthy controls (n = 108) were those who had received physical examinations at the same period, including 65 males and 43 females and age of 56.82 ± 7.11 years (ranging from 38 to 70 years). Two groups of study subjects were comparable, since no significant difference existed in gender and age (P > 0.05).

2.6. Blood Sample Collection and Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)

Peripheral blood (3 ml) was collected from ICH patients within 2 h after their admission and fasting healthy controls in the morning during their physical examinations and then the patients underwent a centrifugation (3000 r/m, 15 min and 4°C), with the supernatant collected. The serum was lysed using a TRIzol kit (Thermo Fisher Scientific, USA) to extract total RNA. The synthesis of cDNA from 400 ng total RNA was achieved by using a PrimeScript RT Reagent kit (Takara, Dalian, China) (42°C and 15 min and 85°C and 5 s) with a 20 μL reaction system including template RNA 1.0 μL, 5 × PrimeScript Buffer 24.0 μL, PrimeScript RT Enzyme Mix 1.0 μL, RT Primer Mix 1.0 μL, and solution for genome removal 10.0 μL, supplemented with RNase-free DEPC. Subsequently, qRT-PCR was performed with SYBR® Premix Ex Taq™ II (Tli RNase H Plus) kit (Takara) using an ABI PRISM®7500 System (Applied Biosystems, Foster City, CA, USA). The thermocycling conditions were as follows: predenaturation at 94°C for 10 min, 45 cycles of 97°C for 15 s, 64°C for 45 s, 97°C for 15 s, 64°C for 1 min, 94°C for 15 s, and a final extension at 75°C for 10 min. The primer sequences (DOCK1: forward, 5′-CCGCCGCAAACTTTTTCCTC-3′, and reverse, 5′-AGATGTGCACAGTGTCTCCG-3′; GAPDH: forward, 5′-TTCGACAGTCAGCCGCATCTT-3′, and reverse, 5′-CCCAATACGACCAAATCCGTT-3′) were designed and synthesized by Invitrogen (USA). Relative mRNA expression of DOCK1 was determined by the 2−ΔΔCt method and normalized to GAPDH, a house-keeping gene.

2.7. Baseline Characteristics and Outcome Measures

Demographics, medical history, and baseline radiological and clinical parameters were recorded. All patients underwent CTH at hospital admission. Hematoma volume was calculated according to the Coniglobus formula; that is, V = a × b × c × 1/2 (a indicates the longest diameter of hematoma at the level of maximum hematoma area, b indicates the longest diameter perpendicular to the longest diameter at the level of maximum hematoma area, and c indicates the number of layers with bleeding in CT images). Small-volume hematoma was defined as less than <10 mL, medium-volume hematoma was defined as 10 to 30 mL, and large-volume hematoma was defined as >30 mL. The severity of neurological deficits was evaluated using National Institutes of Health Stroke Scale (NIHSS) that contains 15 items of measures to evaluate the impact of acute stroke on different areas encompassing level of consciousness, neglect, motor strength, facial palsy, ataxia, dysarthria, and sensory loss. The severity of neurological deficits was classified into mild (NIHSS scores ≤15), moderate (NIHSS scores from 16 to 30), and severe (NIHSS scores from 31 to 45) categories. Function outcomes were evaluated by modified Rankin Scale (mRS) which is a clinician-reported, 6-level outcome measure of global disability or dependence in the daily activities of stroke patients or other causes of neurological disability. The scores of this scale have been recently reported as ranging from 0, indicating a lack of symptoms, to 6, indicating death [20]. Good function outcomes were defined for those patients who had mRS scores ≤2 at 90-day follow-up.

2.8. Statistical Processing

Data analysis was performed using the GraphPad Prism 8.0 software (IBM Corp., Armonk, NY, USA). Data were shown as manners of mean ± standard deviation (for measurement data) and ratio (for enumeration data). The comparison of measurement data between ICH patients and healthy controls and between good function outcome and poor function outcome was analyzed by t-test; the comparison of enumeration data between ICH patients and healthy controls was analyzed by chi-square test. The three groups were compared in pairs. A difference level less than 0.05 was indicative of statistical significance.

3. Results

3.1. Identification of Methylated Genes and Expression Genes between ICH Patients and Matched Controls

After analyzing the raw DNA methylation data of the GSE179759 dataset, we identified 119 hypermethylated genes and 561 hypomethylated genes between peripheral blood samples from 30 ICH patients and 34 matched controls with |log2FC| > 1.5 and FDR < 0.05 as thresholds, as shown by the volcano plot (Figure 1(a)). The top 20 in log2FC of hypermethylated and hypomethylated CpG positions are shown by the heatmap in Figure 1(b). After analyzing the raw genome-wide expression data of GSE125512 dataset, we identified 347 upregulated DEGs and 126 downregulated DEGs between 11 pairs of peripheral blood samples within 24 hours of symptom onset or 72 hours following the first collected from 11 ICH patients with |log2FC| > 1.5 and FDR < 0.05 as thresholds (Figures 1(c) and 1(d)).

Figure 1

Differential analysis of methylated genes and expression genes between ICH patients and matched controls. (a) The volcano plot showing 119 hypermethylated genes and 561 hypomethylated genes between peripheral blood samples from 30 ICH patients and 34 matched controls by analyzing the GSE179759 dataset. (b) The heatmap showing the top 20 in log2FC of hypermethylated and hypomethylated CpG positions. (c) The volcano plot showing 347 upregulated DEGs and 126 downregulated DEGs between 11 pairs of peripheral blood samples within 24 hours of symptom onset or 72 hours following the first collected from 11 ICH patients by analyzing the GSE125512 dataset. (d) The heatmap showing the top 20 DEGs.

3.2. Identification of MeDEGs between ICH Patients and Matched Controls and Functional Enrichment Analysis

Venn intersect function was used to reveal overlapping genes between GSE179759 and GSE125512 datasets, and a total of 15 MeDEGs, 1 hypermethylation low-expression gene and 14 hypomethylation high-expression genes, were identified (Figure 2(a) and Table 1). In order to explore the functional roles of the MeDEGs between ICH patients and matched controls, GO enrichment analysis was conducted and significantly enriched GO terms of MeDEGs are listed in Figure 2(b). In view of the enrichment results of the BP category, we found that MeDEGs were significantly enriched in phagocytosis (GO: 0006909), positive regulation of cell-substrate adhesion (GO: 0010811), leukocyte chemotaxis (GO: 0030595), cell chemotaxis (GO: 0060326), etc. In terms of the enrichment results of the CC category, the MeDEGs were mainly enriched in extrinsic component of cytoplasmic side of plasma membrane (GO: 0031234), integral component of plasma membrane (GO: 0005887), intrinsic component of plasma membrane (GO: 0031226), cytoplasmic side of plasma membrane (GO: 0009898), extrinsic component of plasma membrane (GO: 0019897), etc. With regard to the enrichment results of the MF category, the MeDEGs were mostly enriched in iron ion binding (GO: 0005506), G protein-coupled receptor activity (GO: 0004930), etc.

Figure 2

Identification of MeDEGs between ICH patients and matched controls and functional enrichment analysis. (a) Venn intersection analysis of hypermethylation low-expression genes and hypomethylation high-expression genes; a total of 15 MeDEGs were identified. (b) GO term enrichment analysis of MeDEGs.

Table 1

A full list of 15 MeDEGs consisting of 1 hypermethylation low-expression gene and 14 hypomethylation high-expression genes between ICH patients and matched controls.

Gene name	Pattern	log2FC	P value
CEBPE	Hypomethylation high-expression	0.638870963	0.026565526
C1QTNF7	Hypomethylation high-expression	0.694955007	0.002048426
GPR146	Hypomethylation high-expression	0.598434508	0.027390936
CNR2	Hypomethylation high-expression	0.783539904	0.000223684
DOCK1	Hypomethylation high-expression	0.792833816	0.002136729
ADGRE1	Hypomethylation high-expression	0.745167319	0.000016
ALOX15	Hypomethylation high-expression	1.149227101	0.000288933
GUCY1A3	Hypomethylation high-expression	0.74496542	0.0000366
TRHDE	Hypomethylation high-expression	1.712021654	0.0000485
SLC14A1	Hypomethylation high-expression	0.652210606	0.013680268
TMC5	Hypomethylation high-expression	0.611357429	0.000111768
ZNRF1	Hypomethylation high-expression	0.770871066	0.0000259
TRIP10	Hypomethylation high-expression	0.644199551	0.040645351
CYP7B1	Hypomethylation high-expression	1.030957021	0.013598049
SEPT7P9	Hypermethylation low-expression	−0.728686684	0.006511196

3.3. Increased mRNA Expression of DOCK1 in the Serum of ICH Patients

Among 15 MeDEGs between ICH patients and matched controls, DOCK1 gene was selected for further analysis. We performed qRT-PCR to determine the mRNA expression of DOCK1 in the serum samples collected from 114 patients with spontaneous ICH and 108 healthy controls. As shown in Figure 3, the mRNA expression of DOCK1 was remarkably higher in the serum samples of patients with spontaneous ICH than in the healthy controls (0.58 ± 0.29 versus 0.20 ± 0.10, P < 0.001). These data showed that DOCK1 was upregulated in patients with ICH attack.

Figure 3

The mRNA expression of DOCK1 in the serum samples collected from 114 patients with spontaneous ICH and 108 healthy controls was determined by qRT-PCR.

3.4. Gender, Age, and DOCK1 mRNA Expression Were Associated with Hematoma Volume after ICH Attack

Each patient with spontaneous ICH underwent HCT at admission, 41 patients with small volume of hematoma (<10 mL) were arranged into small group, 48 patients with medium volume of hematoma (10 to 30 mL) were arranged into medium group, and 25 patients with large volume of hematoma (>30 mL) were arranged into large group. As shown in Table 2, the proportions of male patients and patients aged ≥60 years were significantly higher in the large group than in the small and medium groups (P < 0.05); the medium group also had a higher proportion of patients aged ≥60 years compared to the small group (P < 0.05); no significant difference was noted among the three groups concerning the time from symptom onset to hospital admission (P > 0.05); it was revealed that the mRNA expression of DOCK1 was significantly higher in the large group than in the small and medium groups (P < 0.05). No significant difference was indicated between the small and medium groups with regard to the proportions of male patients and DOCK1 mRNA expression. These data suggested that gender, age, and DOCK1 expression may be associated with the hematoma volume in ICH patients.

Table 2

Patient's gender, age, and DOCK1 mRNA expression were associated with hematoma volume after ICH attack.

Group	n	Male gender (%)	Age ≥ 60 years (%)	Symptom onset to hospital admission (h)	DOCK1 mRNA expression
Small group	39	17 (43.59%)∗	11 (28.21%)∗#	3.97 ± 1.13	0.53 ± 0.25∗
Medium group	50	30 (60.00%)∗	26 (52.00%)∗	4.07 ± 1.09	0.56 ± 0.27∗
Large group	25	22 (88.00%)	20 (80.00%)	4.02 ± 1.15	0.72 ± 0.34

P < 0.05 compared to the large group; #P < 0.05 compared to the medium group.

3.5. Association between Baseline Characteristics, DOCK1 mRNA Expression, and Neurological Deficits after ICH Attack

The severity of neurological deficits was evaluated by NIHSS scores. Among 114 patients with spontaneous ICH, there were 55 cases in the mild group (NIHSS scores ≤15), 38 cases in the moderate group (NIHSS scores from 16 to 30), and 21 cases in the severe group (NIHSS scores from 31 to 45). It was observed that the severe group had higher proportions of male patients and patients aged ≥60 years than the mild and moderate groups (P < 0.05). The moderate group had a higher proportion of patients aged ≥60 years than the mild group (P < 0.05). With regard to the time from symptom onset to hospital admission, significant differences were only noted between the mild and severe groups (P < 0.05). The severe group exhibited a higher mRNA expression of DOCK1 than the mild and moderate groups (P < 0.05). The moderate group demonstrated an elevated DOCK1 mRNA expression compared to the mild group (P < 0.05). The detailed data are listed in Table 3. These results uncovered that gender, age, time of symptom onset to hospital admission, and DOCK1 expression may be associated with the severity of neurological deficits in ICH patients.

Table 3

Patient's gender, age, time of symptom onset to hospital admission, and DOCK1 mRNA expression were associated with neurological deficits after ICH attack.

Group	n	Male gender (%)	Age > 60 years (%)	Symptom onset to hospital admission (h)	DOCK1 mRNA expression
Mild group	55	28 (50.91%)∗	18 (32.73%)∗#	3.81 ± 1.20∗	0.49 ± 0.24∗#
Moderate group	38	23 (60.53%)∗	21 (60.53%)∗	4.10 ± 0.95	0.61 ± 0.26∗
Severe group	21	18 (85.71%)	18 (85.71%)	4.45 ± 1.03	0.78 ± 0.36

P < 0.05 compared to the severe group; #P < 0.05 compared to the moderate group.

3.6. Gender, Age, and DOCK1 mRNA Expression Were Associated with Function Outcomes of Patients with Spontaneous ICH

Function outcomes of patients with spontaneous ICH were evaluated by the mRS scores at 90-day follow-up. According to their mRS scores, 72 patients were arranged into the favorable group (mRS scores ≤2) and the unfavorable group (mRS scores >2). As listed in Table 4, higher proportions of male patients and patients aged ≥60 years were observed in the unfavorable group than in the favorable group (P < 0.05). No significant difference was noted among the three groups concerning the time from symptom onset to hospital admission (P > 0.05). The patients in the unfavorable group showed an elevated DOCK1 mRNA expression compared to those in the favorable group (P < 0.05). It was uncovered that male gender, older age, and increased DOCK1 expression may be risk factors influencing function outcomes of patients with spontaneous ICH.

Table 4

Gender, age, and DOCK1 mRNA expression were associated with function outcomes of patients with spontaneous ICH.

Group	n	Male gender (%)	Age > 60 years (%)	Symptom onset to hospital admission (h)	DOCK1 mRNA expression
Favorable group	72	38 (52.78%)∗	30 (41.67%)∗	4.01 ± 1.20	0.53 ± 0.24∗
Unfavorable group	42	31 (73.81%)	27 (64.29%)	4.05 ± 0.94	0.68 ± 0.34

P < 0.05 compared to the unfavorable group.

4. Discussion

When ICH occurs, blood leaks into the brain parenchyma, leading to primary and secondary brain injury. Mass effect and tissue damage are the main reasons for the occurrence of primary ICH. Secondary ICH is mainly caused by inflammatory reaction and erythrolysis [21]. The occurrence of ICH is related to the two main factors including aging [22] and hypertension [23], which are uncontrollable. Although the risk of ICH increases significantly with age, there are still a considerable proportion of people under the age of 50 [24]. In addition, an increasing number of cases are relevant with the greater use of anticoagulant or antiplatelet medications [25]. The prevalence of ICH has brought about a huge burden to patients, caregivers, family members, and society; unfortunately, the prognosis after ICH has not improved significantly in the past few decades [26]. At present, no interventional surgical or pharmacologic care has been proven effective in reducing the mortality or incidence rate of ICH. There has been a growing interest in exploring pathogenesis of ICH which might provide new intervention care for ICH patients.

Numerous studies have proven that DNA methylation is involved in many different types of cancer [27, 28]. DNA methylation was found more than 80 years ago, which usually refers to the direct chemical modification of DNA [29]. Now, its key functions in many biological processes such as transcription regulation, transposable factor silencing, genomic imprinting, and X chromosome inactivation have been fully confirmed [30]. However, little is known about the mechanism of DNA methylation in ICH. In this study, peripheral blood samples from 30 ICH patients and 34 healthy individuals were collected from GSE179759, as well as 11 ICH patients from GSE125512, who developed symptoms within 24 or 72 hours, so as to analyze the correlations between ICH prognosis and DNA methylation.

A total of 15 MeDEGs, 1 hypermethylation low-expression gene and 14 hypomethylation high-expression genes, were identified in the present study. In order to further study the interaction between these MeDEGs, GO term analysis was carried out. It was found that MeDEGs were enriched in some functions including cellular functions, immune functions, and metabolism functions. These findings were similar to previous study, suggesting that cell adhesion dysfunction, immune response, inflammatory response, and cytokine receptor interaction were associated with the pathogenesis of ischemic stroke [31]. Furthermore, DOCK1 gene selected from 15 MeDEGs was further analyzed. The qRT-PCR assays indicated that, compared to healthy individuals, expressions of DOCK1 were significantly higher in the ICH patients. DOCK proteins belong to the atypical Rho guanine nucleotide exchange factor family, which is involved in many cellular processes, such as cytoskeleton organization, cell adhesion, and migration [32]. DOCK1 has been shown to play a role in cell movement, myoblast fusion, and Rac1 mediated phagocytosis [13]. Other studies have revealed that high expression of DOCK1 is associated with poor prognosis and low survival rate in breast cancer patients [14, 33]. The present study demonstrated that DOCK1 plays an essential role in the pathogenesis of ICH. The reason is that DOCK1 might regulate the development of ICH through activation of the RhoA/Rho-kinase signaling pathway. RhoA is involved in the regulation of smooth muscle tone and activates many downstream kinases. RhoA and its main downstream kinase Rho-kinase have been confirmed related to various diseases, such as vascular disease, hypertension, diabetes, heart attack, stroke, and cancer [34, 35]. In order to identify other high risk factors associated with ICH, this study assigned ICH patients into 3 groups based on volume of hematoma. The results showed that male patients and patients above 60 years of age were more in the group with hematoma volume larger than 30 mL; additionally, the patients above 60 years of age accounted for a larger proportion in the group with hematoma volume from 10 to 30 mL compared to the group with volume of hematoma less than 10 mL. In the research of intracerebral hemorrhage in young and elderly patients, Yang et al. pointed out the elderly were prone to suffer from hypertension, leading to the increased risk of ICH and poor functional outcomes [36]. Jolink et al. also confirmed that hypertension was associated with the high risk of ICH. Male gender was more likely to be attacked by ICH which might be due to excessive drinking [37]. A systematic review and meta-analysis of intracerebral hemorrhage have indicated that the incidence of ICH increased with age, especially in patients above 45 years of age, and the incidence rate of males is higher than that of the female patients [5]. In this study, the group with hematoma volume larger than 30 mL revealed higher expression of DOCK1 compared to the other two groups as well. These outcomes we found suggested that male gender, elderly, and DOCK1 expression might be related to the hematoma volume in ICH patients. This study evaluated the severity of neurological deficits by NIHSS scores and divided the ICH patients into 3 groups. It is observed that, compared to the other two groups, the proportion of male patients and elderly patients above 60 years of age was much higher in the severe group. The patients in the moderate group had more patients above 60 years of age than the mild group. Moreover, the higher expression of DOCK1 was positively associated with the NIHSS scores. The function outcomes showed that the male gender, the elderly patients above 60 years of age, and increased expression of DOCK1 were related to the increased risk of adverse prognosis of ICH. These outcomes were similar to another study, which indicated large hematoma volume, male gender, deep bleeding site, and age between 51 and 60 years were the related factors of poor short-term prognosis in ICH [38]. A recent study reported by Al-Khaled et al. indicated that the elderly and the patients with severe neurological injury showed lower survival rate, and the gender, history of hypertension, and bleeding site were relevant to prognosis of ICH [39].

In summary, our results uncovered that upregulated DOCK1 expression after ICH attack may be attributed to its DNA hypomethylation, which was associated with the development of ICH, including admission hematoma volume, neurologic deterioration, and poor function outcomes. In addition to that, male gender and older age were related to higher admission hematoma volume, neurologic deterioration, and poor function outcomes in patients with spontaneous ICH. However, further investigations are required due to some limitations involved in the study, such as small sample size and the absence of research on bleeding site.