Keywords
atherosclerosis -
PF4
- platelet - RNA sequencing - WGCNA
Introduction
Atherosclerosis (AS) is a lipid-driven chronic progressive inflammatory disease that
affects mainly large- and medium-sized arteries.[1] It is characterized by the formation of fibrous or atheromatous plaques in the intima
of blood vessels, leading to stenosis of the lumen and weakening of the elasticity
of the wall, which affects blood perfusion in the corresponding tissues and organs.[1]
[2] In addition, plaque rupture and subsequent occlusive thrombosis may be the major
direct triggers of most acute myocardial ischemic events, such as unstable angina
and acute myocardial infarction.[3]
[4]
The currently available literature has revealed that platelets play an important role
in the pathogenesis and progression of AS.[5] First, as a major risk factor for AS, hyperlipidemia has long been considered primarily
responsible for the development of AS.[6]
[7] Research has shown that hyperlipidemia significantly increases the biogenesis, turnover,
and activity of platelets.[8] Second, AS is a chronic inflammatory disease, and the slow formation process of
AS lesions involves the participation of many bone marrow and immune cells (including
neutrophils, eosinophils, monocytes, and lymphocytes)[1]; moreover, platelets seem to play a key role in the recruitment of these inflammatory
effector cells, and the crosstalk between platelets and these inflammatory cells regulates
the progression of inflammation.[9]
[10]
[11]
[12] Finally, the clinical application of antiplatelet therapy can effectively prevent
the formation of new lesions and stabilize existing lesions in patients with AS.[13]
[14]
The current diagnostic workflow for AS heavily relies on traditional markers such
as lipid profiles and carotid artery ultrasound imaging. However, these methods, while
important, do not fully capture the complexity of AS pathophysiology. This limitation
has hindered the identification of highly sensitive biomarkers that could enable early
detection and more accurate risk stratification. Therefore, the aim of this study
was to explore novel platelet biomarkers that could enhance the diagnostic accuracy
for AS, potentially leading to better clinical management of the disease. While previous
studies have demonstrated that platelet characteristics play a role in the development
of AS, the mechanisms remain unclear. Research has shown that platelet activation
and aggregation are crucial in AS progression, yet these processes are not sufficiently
integrated into the current diagnostic approach.[15]
[16] By focusing on RNA-seq of platelets from AS patients, we aim to uncover gene expression
signatures that could serve as early diagnostic markers, improving upon current diagnostic
practices. In the present study, we found that the expression levels of the ITGA2B, TGFB1, PF4, and GP9 genes were significantly correlated with the intima-media thickness (IMT), suggesting
potential avenues for the future prediction or diagnosis of AS.
While previous studies have demonstrated that platelet characteristics play a role
in the development of AS, the mechanisms remain unclear. Research has shown that platelet
activation and aggregation are crucial in AS progression, yet these processes are
not sufficiently integrated into the current diagnostic approach (Zhu et al., 2018[3]; Wang and Tang, 2020[7]). By focusing on RNA-seq of platelets from AS patients, we aim to uncover gene expression
signatures that could serve as early diagnostic markers, improving upon current diagnostic
practices.
The aim of this study was to investigate the differential gene expression in platelets
between these two groups. To achieve this, we recruited healthy subjects and AS patients
and applied parametric transcriptome sequencing. Our findings may provide valuable
insights for clinicians and researchers in elucidating the role of platelets in AS
pathogenesis at the global transcriptional level, as well as identifying potential
targets for effective interventions.
Methods
General Information
Eighty-five consecutive patients with AS admitted to the outpatient clinic and wards
of Xi Yuan Hospital, China Academy of Chinese Medical Sciences, from September 2019
to September 2020, were retrospectively divided into a normal group (N) and an AS
group. The study was approved by the Ethics Committee of Xi Yuan Hospital, China Academy
of Chinese Medical Sciences (2020XLA057).
Diagnostic Criteria
According to the 2010 American College of Cardiology Foundation/American Heart Association
(ACCF/AHA) guidelines and the Chinese Guidelines for the Diagnosis of Carotid Atherosclerosis,[17] the diagnosis of AS was based on the measurement of IMT using carotid ultrasound.
An IMT of less than 1.0 mm was considered normal, an IMT between 1.0 and 1.4 mm was
classified as thickening, and an IMT greater than 1.5 mm or the presence of plaques
was diagnostic of carotid plaque formation.
Inclusion Criteria
The inclusion criterion for AS patients was as follows: Age between 20 and 65 years
(including both boundary ages and applicable to both sexes); IMT >1.0 mm or visible
plaque who without prior use of lipid-lowering drugs, antiplatelet drugs, and other
medications that could potentially affect the experimental results. The inclusion
criteria for Group N were the same as those for the AS group (age between 20 and 65
years [including both boundary ages and applicable to both sexes]), with the addition
of general physical examination, liver and kidney function tests, blood lipid analysis,
blood tests, urine tests, and routine stool tests—all yielding normal results. No
participant had been involved in any other drug clinical trials within the past month.
All the subjects participated voluntarily and signed informed consent forms.
Exclusion Criteria
Those who had received lipid-lowering or anti-AS medications within the past 3 months,
were currently receiving drug treatment for other medical conditions, or had medication-induced
dyslipidemia, as well as those who had known conditions that could affect IMT measurements,
such as hypertension, coronary artery disease, chronic kidney disease, or diabetes,
were excluded from the study because of their inability to fully cooperate in completing
all examinations and scales.
Physical Examination and Laboratory Tests
In the present study, we collected comprehensive demographic information, disease
history, medication history, diet and daily living conditions, gastrointestinal status,
daily exercise, and other relevant information from the subjects. Blood pressure was
measured by the doctor via a standard mercury sphygmograph after the patient had been
seated for 5 minutes. Blood samples were collected in the fasting state at 8 a.m.
After allowing the peripheral blood samples to stand at room temperature for half
an hour, the serum samples were centrifuged at 3,000 rpm for 5 minutes before being
extracted, packaged, and stored at −80 °C. Samples were promptly sent to our hospital's
clinical laboratory department to assess the following clinical laboratory parameters:
Routine blood analysis, lipid profile assessment, liver and kidney function evaluation,
glucose measurements, and platelet aggregation tests. Additional types of clinical
information, such as defecation status, dietary habits, daily living conditions, gastrointestinal
status, disease history, and medication history, were obtained through face-to-face
questionnaire surveys conducted by visiting physicians.
Platelet Aggregation Test
Blood was collected from healthy and AS individuals via venipuncture and stored in
vacuum-anticoagulated tubes containing lithium heparin. The whole blood was immobilized
at 37 °C for 10 minutes before platelet-rich plasma extraction was performed. Anticoagulated
whole blood was centrifuged at 300 × g for 7 minutes to obtain platelet-rich plasma.
Platelet aggregation was induced by adding activators at 37 °C with a platelet aggregator
(Chrono-Log Corp, Havertown) with sample agitation set at 1,000 rpm. Human platelet-rich
plasma was activated with ADP (2.5 μM) and collagen (5 μg/mL). Gel-filtered platelets
were activated with collagen (10 μg/mL) and thrombin (0.5 U). Additionally, the effects
of human recombinant apolipoprotein A-IV on platelet aggregation induced by different
agonists were measured after incubation for 2 minutes. Changes in transmittance resulting
from platelet aggregation were detected and recorded for a minimum of 10 minutes.
Carotid Ultrasound Examination
The proximal end of the common carotid artery was initially examined, followed by
scanning of the internal and external carotid arteries along the vessel. The arterial
IMT was measured, and any local protrusions were observed. The maximum size of each
plaque was documented. Carotid ultrasonography was performed with a color Doppler
ultrasound diagnostic instrument (Mindray M7, China) with a probe frequency of 10 MHz.
The procedure involved two individuals; one conducted the examination, while the other
recorded and entered the inspection results.
RNA-Seq of Platelets
Total RNA was extracted from control and AS platelet samples, and RNA sequencing was
performed via OEBiotech (Shanghai, China). Firstly, multivariate statistical analysis
was conducted on the obtained gene expression data to elucidate intragroup repeatability
and intergroup differences, as well as to ensure the reliability of sequencing results.
The orthogonal partial least squares-discriminant analysis (OPLS-DA) method was employed
in this study. Differentially expressed genes (DEGs) were screened and identified
mainly through log2 (fold change) and p-values. Finally, the DEGs were characterized via Gene Set Enrichment Analysis (GSEA),
Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis, and protein–protein
interaction (PPI) network analysis.
Weighted Correlation Network Analysis and Pearson's Correlation Analysis
The gene coexpression network was constructed using OECloud tools, and hierarchical
clustering analysis was performed based on weighted correlation. The resulting clusters
were segmented according to predefined criteria to obtain distinct gene modules, which
were visually represented by branches and different colors in the cluster tree. Subsequently,
the modules with higher correlation coefficients were identified by calculating the
correlation between gene modules and IMT.
Statistics
SPSS 25.0 software (IBM) was used for data analysis and processing. Count data were
expressed as the number of cases and percentages, and the χ2 test was used for comparisons between groups. Measurement data that conformed to
a normal distribution were expressed as the mean ± standard deviation (x), and independent-samples
t-tests were used for comparisons between two groups. A non-parametric test was used
for non-normally distributed data. General linear bivariate Pearson's linear correlation
analysis was used to examine the associations. Differences were considered statistically
significant when p < 0.05.
Results
Baseline Characteristics of the Study Participants
A total of 41 normal individuals (N) and 40 AS patients were included in the study.
There were no significant differences in age, sex, body mass index, systolic blood
pressure, or diastolic blood pressure between the two groups (p > 0.05; [Table 1]).
Table 1
Baseline characteristics of study participants
|
Variable
|
N (n = 41)
|
AS (n = 40)
|
p-Value
|
|
Age, years
|
48.27 ± 7.88
|
51.48 ± 8.49
|
0.082
|
|
Male, number (%)
|
10 (24.4%)
|
18(45.0%)
|
0.051
|
|
Body mass index, kg/m2
|
23.94 ± 2.25
|
23.90 ± 2.17
|
0.924
|
|
Systolic pressure (mm Hg)
|
120.00 (10.00)
|
126.50 (19.50)
|
0.143
|
|
Diastolic pressure (mm Hg)
|
80.00 (0)
|
80.00 (5.00)
|
0.417
|
Abbreviations: AS, atherosclerosis group; N, control group.
Carotid Ultrasonography Revealed Significant Differences in IMT and Flow-Mediated
Dilation between the Two Groups
Among the 40 patients included in the AS group, visible plaques were detected in 21
patients. In this study, none of the 40 AS patients included had a carotid ultrasound
report showing an IMT >1.5 mm, although the ultrasound results did report the presence
and size of plaques (detailed data can be found in [Supplementary Table S1]). The echocardiography results revealed arterial stiffening and decreased vascular
elasticity and compliance in the AS group. Specifically, carotid IMT was significantly
greater (p < 0.01) and vascular endothelial function (flow-mediated dilation; FMD) was significantly
lower (p < 0.05) in the experimental group than in the N group. There were no significant
differences observed between the two groups regarding elasticity parameters, arterial
compliance, arterial index, stiffness index beta, or aortic index (p > 0.05; [Table 2]).
Table 2
Ultrasound assessment of carotid arteries
|
Variable
|
N (n = 41)
|
AS (n = 40)
|
p-Value
|
|
Left
|
|
IMT (mm)
|
0.56 (0.14)
|
1.04 (0.55)
|
0.005[a]
|
|
Ep (N/m2)
|
95.0 (41.50)
|
99.00 (105.00)
|
0.501
|
|
AC (mL/mm Hg)
|
0.72 (0.41)
|
0.71 (0.32)
|
0.951
|
|
AI
|
12.10 (14.90)
|
13.7 (13.60)
|
0.781
|
|
PMVβ (m/s)
|
5.90 (1.15)
|
6.10 (3.50)
|
0.501
|
|
β (cm/s)
|
6.90 (2.60)
|
6.70 (5.20)
|
0.801
|
|
Right
|
|
IMT (mm)
|
0.56 (0.14)
|
1.04 (0.49)
|
<0.001[b]
|
|
Ep (N/m2)
|
91.00 (47.00)
|
98.00 (40.75)
|
0.143
|
|
AC (mL/mm Hg)
|
0.81 ± 0.29
|
0.77 ± 0.26
|
0.591
|
|
AI
|
12.11 ± 9.04
|
10.20 ± 11.08
|
0.476
|
|
PMVβ (m/s)
|
12.20 (11.25)
|
11.75 (12.28)
|
0.971
|
|
β (cm/s)
|
5.60 (1.35)
|
6.05 (1.33)
|
0.062
|
|
FMD (%)
|
10.9 (6.81)
|
5.48 (6.33)
|
0.024[c]
|
Abbreviations: AC, arterial compliance; AI, arterial index; AS, atherosclerosis group;
Ep, elasticity parameters; FMD, flow-mediated dilation; IMT, intima-media thickness;
N, control group; PMVβ, aortic index; β, stiffness index beta.
a
p < 0.1.
b
p < 0.001.
c
p < 0.5.
Comparison of Laboratory Test Results between the Two Groups
Blood lipids, blood glucose, liver and kidney function, coagulation function, and
platelet aggregation test results were compared between the two groups. The results
of the experiment showed that the levels of triglyceride, total cholesterol, low-density
lipoprotein cholesterol (LDL-C), apolipoprotein B, red blood cells (RBCs), and hemoglobin
concentration (HGB) in the AS group were significantly greater than those in the N
group (p < 0.05), whereas no significant differences were detected in the other test results
between the two groups (p > 0.05; [Table 3]).
Table 3
Comparison of blood fat, blood glucose, renal and liver function tests, coagulation
function test, and platelet aggregation test
|
Variable
|
N (n = 41)
|
AS (n = 40)
|
p-Value
|
|
GLU (mmol/L)
|
5.36 (0.67)
|
5.52 (0.71)
|
0.385
|
|
WBC (×109/L)
|
5.47 ± 1.41
|
5.95 ± 1.69
|
0.160
|
|
RBC (×1012/L)
|
4.25 (0.44)
|
4.54 (0.50)
|
0.012[a]
|
|
HGB (g/L)
|
137.00 (13.75)
|
146.50 (17.50)
|
0.045[b]
|
|
LY (%)
|
33.58 ± 6.52
|
33.51 ± 8.53
|
0.964
|
|
NEUT (%)
|
58.96 ± 6.86
|
58.51 ± 8.50
|
0.794
|
|
PLT (×109/L)
|
248.34 ± 45.46
|
244.90 ± 54.00
|
0.757
|
|
MONO (%)
|
5.41 ± 1.00
|
5.13 ± 0.93
|
0.196
|
|
TG (mmol/L)
|
0.91 (0.56)
|
1.15 (1.18)
|
0.024[b]
|
|
TC (mmol/L)
|
4.67 (1.23)
|
5.11 (1.08)
|
0.007[a]
|
|
HLDL-C (mmol/L)
|
1.30 (0.31)
|
1.19 (0.38)
|
0.388
|
|
LDL-C (mmol/L)
|
2.81 (1.01)
|
3.41 (1.16)
|
<0.001[c]
|
|
VLDL (mmol/L)
|
0.52 (0.21)
|
0.45 (0.31)
|
0.362
|
|
ApoA (g/L)
|
1.12 ± 0.13
|
1.10 ± 0.12
|
0.357
|
|
ApoB (g/L)
|
0.80 (0.19)
|
0.92 (0.19)
|
<0.001[c]
|
|
LP (A) (mg/dL)
|
87.21 (123.75)
|
115.57 (179.32)
|
0.134
|
|
ALT (U/L)
|
13.45 (5.80)
|
16.05 (11.50)
|
0.250
|
|
AST (U/L)
|
17.55 (6.70)
|
18.70 (8.88)
|
0.250
|
|
γ-GT (U/L)
|
15.09 (9.33)
|
18.45 (21.14)
|
0.050
|
|
UREA (mmol/L)
|
4.64 ± 1.05
|
5.03 ± 1.19
|
0.129
|
|
BUN (mg/dL)
|
12.75 ± 3.28
|
14.09 ± 3.33
|
0.076
|
|
CREA (μmol/L)
|
61.50 (9.75)
|
74.50 (16.50)
|
<0.001[c]
|
|
UA (μmol/L)
|
272.46 ± 62.99
|
300.13 ± 79.15
|
0.092
|
|
PT (seconds)
|
11.10(0.475)
|
11.10(0.525)
|
0.877
|
|
PTA (%)
|
107.00 (6.00)
|
107.00(6.65)
|
0.657
|
|
INR
|
0.96 (0.04)
|
0.96 (0.04)
|
0.709
|
|
APTT (seconds)
|
27.85 (2.90)
|
27.90 (2.08)
|
0.795
|
|
TT (seconds)
|
17.20 (0.65)
|
17.20 (0.90)
|
0.661
|
|
Fbg (g/L)
|
2.69 (0.49)
|
2.81 (0.62)
|
0.378
|
|
PAg-AA (%)
|
80.10 (25.28)
|
78.88 (16.62)
|
0.988
|
|
PAg-ADP (%)
|
61.16 (38.78)
|
64.42 (46.05)
|
0.937
|
|
PAg-COL (%)
|
75.97 (31.54)
|
75.07 (31.33)
|
0.988
|
Abbreviations: ALT, alanine aminotransferase; ApoA, apolipoprotein A; ApoB, apolipoprotein
B; APTT, activated partial thromboplastin time; AS, atherosclerosis group; AST, aspartate
aminotransferase; BUN, blood urea nitrogen; CREA, creatinine; FBG, fibrinogen; GLU,
blood glucose; HGB, hemoglobin; HLDL-C, high-density lipoprotein cholesterol; INR,
prothrombin time ratio; LDL-C, low-density lipoprotein; LP (A), lipoprotein A; LY,
lymphocyte; MONO, monocyte; N, control group; NEUT, neutrophil; PAg-AA, platelet maximal
aggregation rate–arachidonic acid; PAg-ADP, platelet maximal aggregation rate–phosphate
adenosine; PAg-COL, platelet maximal aggregation rate–collagen; PLT, platelet; PT,
prothrombin time; PTA, prothrombin activity; RBC, red blood cell; TC, total cholesterol;
TG, triglyceride; TT, prothrombin time; UA, uric acid; UREA, urea; VLDL, very low-density
lipoprotein; WBC, white blood cell count; γ-GT, γ-glutamyl transpeptidase.
a
p < 0.1.
b
p < 0.5.
c
p < 0.001.
Multivariate Statistical Analysis
The transcriptome sequencing results of the two groups were subjected to multivariate
statistical analysis, yielding an OPLS-DA score plot, as depicted in [Fig. 1A]. To assess the potential overfitting of the model, a permutation test was conducted
on the OPLS-DA model. The grouping markers for each sample were randomly shuffled
prior to modeling and prediction. A robust supervised model requires that the intercept
between the regression line (dashed line) at Q2 and the y-axis is less than 0, which
indicates satisfactory model quality. As shown in [Fig. 1B], the OPLS-DA model employed in this study did not exhibit overfitting and effectively
revealed the differences between the two groups of samples.
Fig. 1
(A) OPLS-DA results of the two groups. (B) Permutation test of OPLS-DA. (C) Volcano map of the DEGs. (D) DEGs between the two groups. (E) Heatmap of the top 50 DEGs between the groups. DEG, differentially expressed gene;
OPLS-DA, orthogonal partial least squares-discriminant analysis.
Comparative Analysis and Enrichment Assessment of Platelet Transcriptome Findings
between the Two Groups
A total of 784 DEGs, consisting of 141 downregulated genes and 643 upregulated genes,
were identified via the DESeq tool ([Fig. 1C], [D]). The cluster heatmap visually represented the top 50 DEGs with fold changes between
the two groups ([Fig. 1E]). Gene Ontology (GO) enrichment analysis and KEGG signaling pathway enrichment analysis
were subsequently performed to further elucidate the specific biological processes
associated with these DEGs ([Fig. 2]).
Fig. 2
(A) GO analysis of upregulated DEGs. (B) GO analysis of downregulated DEGs. (C) KEGG analysis of upregulated DEGs. (D) KEGG analysis of downregulated DEGs. DEG, differentially expressed gene; GO, Gene
Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes.
The GO biological function enrichment analysis indicated that the upregulated DEGs
were involved in processes such as blood coagulation, proton motive force-driven mitochondrial
ATP synthesis, and positive regulation of platelet activation. Additionally, these
genes were significantly enriched in biological processes, including the regulation
of chaperone-mediated autophagy (CMA) and megakaryocyte development. Furthermore,
the KEGG pathway enrichment analysis revealed that 26 KEGG pathways, including but
not limited to extracellular matrix–receptor interaction, platelet activation, focal
adhesion, neutrophil extracellular trap formation, fluid shear stress, and AS, were
significantly enriched among the upregulated genes.
The downregulated genes were subjected to GO enrichment analysis, which revealed significant
associations with negative regulation of receptor recycling, endothelial cell activation,
and inhibition of hormone secretion, as well as positive regulation of cytokines and
the Janus Kinase–Signal Transducer and Activator (JAK-STAT) signaling pathway. Additionally,
KEGG pathway enrichment analysis revealed nine significantly enriched pathways among
the downregulated genes, including cytokine–cytokine receptor interactions, cell adhesion
molecules, the Toll-like receptor signaling pathway, and the chemokine signaling pathway.
Hub Genes' Identification through the Integration of Weighted Gene Coexpression Network
Analysis
We computed the module modules to represent the overall gene expression levels of
each module, which were then clustered on the basis of their correlation patterns.
A total of 12 modules were identified and assigned unique colors for labeling purposes
([Fig. 3A], [B]). A heatmap was subsequently generated to visualize the relationships between modules
and traits via Spearman's correlation coefficients, allowing assessment of the association
between IMT and the disease under investigation ([Fig. 3C]). Five modules were identified as significantly correlated with IMT. Biological
process enrichment analysis of genes across different modules revealed that the blue
module is associated with biological processes such as wound healing and megakaryocyte
differentiation, whereas the light cyan module pertains to T-cell activation and lymphocyte
activation. The purple module is linked to leukocyte activation, whereas the dark
module is related to the innate immune response. Finally, the tan module is associated
with cytoplasmic translation ([Fig. 4]).
Fig. 4 A: GO analysis of the blue module; B: GO analysis of the light cyan module; C: GO
analysis of the purple module; D: GO analysis of the dark module; (E) GO biological process analysis of the blue, light cyan, purple, dark, and tan modules.
GO, Gene Ontology.
Fig. 3
(A) Clustering dendrogram of genes, utilizing dissimilarity metrics derived from topological
overlay, along with the corresponding assigned module colors. (B) Heatmap of intergene correlation based on topological overlapping matrix. (C) Correlation and significance between different gene modules and IMT. IMT, intima-media
thickness.
Among these modules, two (referred to as “purple” and “blue”) exhibited significant
associations with AS and were consequently selected as IMT-related modules (purple
module: r = − 0.47, p = 2.22E-05; blue module: r = 0.41, p = 0.000260805; [Supplementary Table S2]). Notably, while the blue module displayed a positive correlation with IMT encompassing
214 genes, the purple module demonstrated a negative correlation with IMT involving
1,228 genes. These specific genes were retained for subsequent analyses.
The DEGs identified earlier were subsequently intersected with the enriched blue and
purple module genes. The Venn diagram revealed that there were 111 intersection genes
between the DEGs and the blue module, whereas 38 intersection genes were found between
the DEGs and the purple module ([Fig. 5A]). Furthermore, a PPI network was constructed from these intersecting genes, leading
to the identification of hub genes such as ITGA2B, TGFB1, PF4, GP9, and GATA1 ([Fig. 5B]). Pearson correlation analysis was subsequently conducted to examine the relationships
between the hub genes and IMT, thereby validating the findings. The correlation coefficients
and p-values for the five hub genes related to IMT are presented in [Table 4]. ITGA2B (r = 0.327), TGFB1 (r = 0.362), PF4 (r = 0.240), and GP9 (r = 0.302) exhibited statistically significant moderate correlations with IMT (p < 0.05).
Table 4
Pearson's correlation test between hub genes and intima-media thickness
|
Pearson correlation coefficient
|
p-Value
|
|
PAg-AA
|
−0.027
|
0.817
|
|
PAg-ADP
|
−0.06
|
0.612
|
|
PAg-COL
|
0.023
|
0.844
|
|
ITGA2B
|
0.327
|
0.004[a]
|
|
TGFB1
|
0.362
|
0.001[a]
|
|
PF4
|
0.240
|
0.038[b]
|
|
GP9
|
0.302
|
0.008[a]
|
|
GATA1
|
0.184
|
0.114
|
Abbreviations: PAg-AA, platelet maximal aggregation rate–arachidonic acid; PAg-ADP,
platelet maximal aggregation rate–phosphate adenosine; PAg-COL, platelet maximal aggregation
rate–collagen.
a
p < 0.1.
b
p < 0.5.
Fig. 5
(A) Venn diagram of DEGs with purple module and blue module genes. (B) PPI network analysis of intersection genes. DEG, differentially expressed gene; PPI,
protein–protein interaction.
Discussion
The prevailing understanding of platelets has been confined to their fundamental role
in hemostasis and thrombosis.[5]
[18]
[19]
[20] The utilization of antiplatelet medications such as thromboxane A2 inhibitors and
adenosine diphosphate purinergic receptor P2Y, G-Protein coupled, 12 (P2Y12) inhibitors
for the prevention and treatment of AS and thrombosis has resulted in promising outcomes.[21]
[22]
[23] Over the past decade, our understanding of platelet pathophysiology has undergone
significant updates[24]
[25]; however, the precise involvement of platelets in AS pathogenesis remains unknown.
Intriguingly, this study did not reveal any notable statistical disparity in platelet
aggregation parameters between the two groups, implying that current platelet aggregation
detection methods may not be effective for the discovery of AS. Consequently, we used
transcriptome sequencing technology to provide a comprehensive understanding of the
role of platelets in AS development, thereby offering novel insights into the development
of more sensitive platelet biomarkers and safer antiplatelet therapies.
The analysis of DEGs revealed 141 downregulated genes and 643 upregulated genes in
patients with AS. GO and KEGG analyses revealed that these DEGs were not only enriched
in traditional platelet-related biological processes such as blood coagulation and
platelet activation but also significantly enriched in pathways related to Toll-like
receptor signaling, chemokine signaling, cytokine–cytokine receptor interaction, and
neutrophil extracellular trap formation. Similarly, among the five gene modules identified
by Weighted Gene Coexpression Network Analysis (WGCNA) that were significantly associated
with IMT, the enrichment of genes in three modules indicates T-cell activation (light
cyan module), lymphocyte activation (light cyan module), and biological processes
such as leukocyte activation (purple module) and the innate immune response (dark
module). These findings suggest a close association between the role of platelets
in AS pathogenesis and inflammation status. As previous research has shown, AS is
fundamentally characterized by chronic inflammation involving various effector cells,
including neutrophils, monocytes, and lymphocytes.[26]
[27] On one hand, activated platelets secrete various cytokines and chemokines to facilitate
immune cell recruitment to the lesion site.[28]
[29]
[30]
[31] On the other hand, there is a complex interplay between platelets and diverse inflammatory
effector cells that collectively regulate the inflammatory response, thereby exerting
control over the onset and progression of AS.[9]
[10]
[11]
[12]
[28]
[32]
The correlations between the identified central genes and the severity of IMT were
analyzed, revealing positive associations of ITGA2B (r = 0.327, p = 0.004), TGFB1 (r = 0.362, p = 0.001), PF4 (r = 0.240, p = 0.038), and GP9 (r = 0.302, p = 0.008) with IMT. Platelet Factor 4 (PF4), also referred to as CXCL4 (C-X-C motif chemokine ligand 4), encodes a member of
the CXC chemokine family.[33] In the present study, PF4 was upregulated in patients with AS, which is consistent with prior publications.
This chemokine is released from activated platelet alpha granules in the form of a
homotetramer with a strong affinity for heparin and plays a role in platelet aggregation.[34] Moreover, this protein has chemotactic effects on various cell types and acts as
an inhibitor of hematopoiesis, angiogenesis, and T-cell function.[35] Previous studies have demonstrated the proatherogenic effects of PF4 through its ability to form heterodimers and oligomers with CCL5, leading to CXCL4-induced
monocyte binding to endothelial cells and subsequent monocyte exudation into the subendothelial
space.[36]
[37]
[38] Additionally, CXCL4 influences monocyte differentiation by inducing a specific macrophage
phenotype known as M4, a proinflammatory marker associated with plaque instability.[36] Furthermore, PF4 serves as an important mediator in regulating T-cell differentiation related to platelet
function and participates in thrombosis progression.[39] Combined with our study, PF4 may be a potential platelet biomarker for AS.
GP9 represents another differentially expressed hub gene encoding a small membrane glycoprotein
located on the surface of human platelets.[40] It forms a non-covalent complex with glycoprotein Ib, which is part of a platelet
surface membrane glycoprotein complex that serves as a receptor for von Willebrand
factor.[41] The complete receptor complex comprises the non-covalent association of alpha and
beta subunits with the protein encoded by this gene, along with platelet glycoprotein
V.[42] Currently, research concerning GP9 primarily focuses on Bernard–Soulier syndrome,[43]
[44]
[45] and no association with AS status has been documented. A study conducted by Burkard
et al. demonstrated that glycoprotein VI exacerbates lipopolysaccharide-induced acute
lung injury in mice through stimulation of neutrophil extracellular trap formation.[46] In the present study, KEGG pathway analysis revealed significant enrichment of upregulated
genes within the neutrophil extracellular trap formation pathway, potentially providing
insiPghts for further investigation into the mechanisms underlying platelet involvement
in AS development.
Similarly, TGFB1 also showed a high correlation with IMT. The involvement of TGFβ
in various biological processes, including cell proliferation, differentiation, migration,
adhesion, and extracellular matrix production, has been extensively acknowledged.[47] Notably, the TGFβ signaling signature and the expression profiles of TGFβ ligands,
receptors, and diverse Smad proteins have been documented in atherosclerotic plaques.[48]
[49] Chen et al. identified endothelial TGFβ signaling as one of the primary drivers
of vascular inflammation associated with AS.[50] Platelets contain substantial amounts of TGFβ1[51]; however, there are currently no published studies detailing the specific mechanisms
by which platelet-derived TGFβ1 regulates the pathogenesis of AS.
In the present study, we identified associations between carotid IMT and the hub genes
ITGA2B, TGFB1, PF4, and GP9 through WGCNA and correlation analysis, and these findings are novel. These hub genes
may serve as platelet biomarkers that could aid in the prediction and diagnosis of
AS; however, further validation in a larger cohort is necessary.
Furthermore, our study presents several compelling new findings. The upregulated DEGs
were significantly enriched in the regulation of CMA. While autophagy has been demonstrated
to occur in platelets, this finding contrasts with conclusions drawn from previous
studies.[52]
[53] Qiao et al. reported that CMA function is compromised during the progression of
AS, leading to increased activation of the NOD-like receptor family pyrin domain containing
3 (NLRP3) inflammasome and secretion of IL-1β, thereby promoting vascular inflammation
and advancing AS.[54]
[55] However, these observations may be influenced by varying stages of AS or differences
in model species, necessitating further validation through more rigorous experimental
designs.
In addition to the aforementioned platelet-related information, our laboratory examination
of the two cohorts revealed significant between-group differences in RBC and HGB levels,
which constitutes a noteworthy finding. Prior studies have shown that the structure
and function of RBCs are influenced by circulating plasma LDL-C levels[56]; furthermore, abnormal collisions of RBCs with arterial walls can lead to endothelial
damage, lipid retention, and localized hemolysis, resulting in the release of toxic
heme iron—an aspect contributing to the complex pathogenesis of AS.[57] While our findings aligned with those of previous reports, we observed only variations
in RBC count and HGB levels between the two groups without delving into their underlying
implications.
In recent years, increasing attention has been focused on the role of platelets in
the development of AS. While traditional AS biomarkers, such as lipid profiles and
imaging techniques, remain widely used, emerging evidence suggests that platelet activation
and the identification of platelet-related biomarkers may offer more accurate diagnostic
and prognostic capabilities. In addition to the present study, several other studies
have explored the role of platelet biomarkers and platelet receptors in the progression
and prognosis of AS. For example, a study identified specific platelet receptors associated
with the risk of atherosclerotic plaque rupture.[58] Similarly, another demonstrated the inflammatory role of platelet-derived chemokines
in acute coronary syndrome and their prognostic value.[59] These findings further support the potential of platelet biomarkers in enhancing
the early diagnostic accuracy and therapeutic efficacy of AS.
Conclusion
In conclusion, our study provides new insights into the role of platelet biomarkers
in the diagnosis and progression of AS. Through RNA sequencing of platelets from AS
patients, we identified key genes, including ITGA2B, TGFB1, PF4, and GP9, whose expression levels were significantly correlated with IMT, suggesting their
potential as novel diagnostic biomarkers for AS. While current diagnostic methods,
such as lipid profiles and carotid artery ultrasounds, are widely used, our findings
highlight the limitations of these traditional techniques and the need for more sensitive
biomarkers. The results of this study not only contribute to the understanding of
platelet-related mechanisms in AS but also open potential avenues for future clinical
applications in early diagnosis and personalized treatment strategies. Further research
is needed to validate these biomarkers in larger patient cohorts and explore their
clinical utility in real-world settings.
Acknowledgment of Limitations
Acknowledgment of Limitations
This study uses carotid IMT thickening as the diagnostic criterion for AS, based on
the fact that IMT has been recognized as a reliable biomarker for early AS in both
international and domestic guidelines (such as the ACCF/AHA 2010 guidelines and the
Chinese Guidelines for the Diagnosis of Carotid Atherosclerosis). IMT is closely associated
with pathological changes in the vessel wall, such as lipid deposition and inflammatory
responses. However, several limitations should be noted.
Although IMT thickening is an early marker of AS, it primarily reflects diffuse thickening
of the vessel wall. In contrast, advanced imaging techniques, such as contrast-enhanced
angiography (e.g., CTA/MRA), serve as the gold standard for more comprehensive assessment
of plaque morphology, calcification, and lumen narrowing, which are characteristics
of more advanced lesions. As a result, this study may underestimate cases predominantly
characterized by non-thickened plaques (such as vulnerable plaques) or overlook subclinical
lesions with blood flow restriction despite no significant IMT thickening.
Furthermore, while IMT measurement has the advantages of being non-invasive and repeatable,
its diagnostic performance is highly dependent on the operator's technical skills
and equipment consistency, which may introduce measurement bias. Notably, although
all 40 AS patients included in the study had plaques detected via ultrasound, none
had an IMT exceeding the 1.5-mm threshold. This suggests that the study population
may be concentrated in the early stages of plaque formation, lacking patients with
more advanced lesions characterized by significant IMT thickening. While the presence
of plaques itself aligns with the diagnostic criteria in the guidelines, the distribution
of IMT values in the population may reflect potential recruitment bias. For example,
early-stage patients are more likely to be included in clinical practice, or region-specific/population-specific
factors (such as younger age or lower cardiovascular risk) may limit the extent of
IMT thickening. Such characteristics of the study population could limit the generalizability
of the results to patients with significantly thickened IMT or complex plaques.
On the methodological level, the risk of blood clotting after sample collection could
impact the quality of platelet-rich plasma, potentially interfering with platelet
aggregation results. Additionally, technical variations during platelet extraction
(such as differences in centrifugation conditions) may affect sample consistency.
Interobserver variability in ultrasound imaging interpretation could also challenge
the accuracy of IMT measurements.
Furthermore, the integrity of RNA in samples stored at −80 °C depends on strict temperature
control; fluctuations in storage temperature could lead to RNA degradation, affecting
the reliability of RNA-seq data. Finally, individual patient factors (such as diet,
lifestyle, and comorbidities) could introduce confounding effects on platelet aggregation
and clinical parameters like IMT, while self-reported data in the questionnaires may
be subject to reporting bias. Future studies should systematically control for these
factors and increase the sample size to include subgroups representing different stages
of IMT progression, thus improving the robustness and generalizability of the findings.
What Is Known About This Topic?
The utilization of antiplatelet medications such as thromboxane A2 inhibitors and
adenosine diphosphate P2Y12 inhibitors for the prevention and treatment of AS and
thrombosis has resulted in promising outcomes; however, the specific mechanism of
platelet involvement in the pathogenesis of AS is still ambiguous.
What Does This Paper Add?
We conducted an RNA-seq analysis to examine the differences in platelet gene expression
between healthy individuals and patients with AS. By utilizing bioinformatics tools,
we successfully identified four potential platelet biomarkers that could potentially
aid in the diagnosis and monitoring of AS progression. These biomarkers include ITGA2B (r = 0.327, p = 0.004), TGFB1 (r = 0.362, p = 0.001), PF4 (r = 0.240, p = 0.038), and GP9 (r = 0.302, p = 0.008).
Bibliographical Record
Zhanfei Tan, Fan Guo, Jiaming Gao, Lanlan Li, Shujuan Xu, Yehao Zhang, Jianhua Fu,
Jianxun Liu. Platelet RNA-Seq Reveals Genes Associated with Carotid Intima-Media Thickness:
A Cross-Sectional Study. TH Open 2025; 09: a26616472.
DOI: 10.1055/a-2661-6472
