Abstract
Purpose
Tuberculosis is common infectious diseases, characterized by
infectivity, concealment and chronicity, and the early diagnosis is
helpful to block the spread of tuberculosis and reduce the resistance
of Mycobacterium tuberculosis to anti-tuberculosis drugs. At present,
there are obvious limitations in the application of clinical detection
methods used for the early diagnosis of tuberculosis. RNA sequencing
(RNA-Seq) has become an economical and accurate gene sequencing method
for quantifying transcripts and detecting unknown RNA species.
Methods
A peripheral blood mRNA sequencing was used to screen the
differentially expressed genes between healthy people and tuberculosis
patients. A protein-protein interaction (PPI) network of differentially
expressed genes was constructed through Search Tool for the Retrieval
of Interacting Genes/Proteins (STRING) database. The potential
diagnostic targets of tuberculosis were screened by the calculation of
degree, betweenness and closeness in Cytoscape 3.9.1 software. Finally,
the functional pathways and the molecular mechanism of tuberculosis
were clarified in combination of the prediction results of key gene
miRNAs, and by Gene Ontology (GO) enrichment analysis and the Kyoto
Encyclopedia Genes and Genomes (KEGG) pathway annotation analysis.
Results
556 Differential genes of tuberculosis were screened out by mRNA
sequencing. Six key genes (AKT1, TP53, EGF, ARF1, CD274 and PRKCZ) were
screened as the potential diagnostic targets for tuberculosis by
analyzing the PPI regulatory network and using three algorithms. Three
pathways related to the pathogenesis of tuberculosis were identified by
KEGG pathway analysis, and two key miRNAs (has-miR-150-5p and
has-miR-25-3p) that might participate in the pathogenesis of
tuberculosis were screened out by constructing a miRNA-mRNA pathway
regulatory network.
Conclusion
Six key genes and two important miRNAs that could regulate them were
screened out by mRNA sequencing. The 6 key genes and 2 important miRNAs
may participate in the pathogenesis of infection and invasion of
Mycobacterium tuberculosis through herpes simplex virus 1 infection,
endocytosis and B cell receptor signaling pathways.
Keywords: mRNA sequencing, tuberculosis, diagnostic target, regulatory
network, mechanism of action
1. Introduction
Tuberculosis (TB) is a chronic infectious disease caused by
Mycobacterium tuberculosis ([33]1). According to the World Health
Organization (WHO), there were 5.8 million new cases and 1.5 million TB
deaths in 2021 ([34]2), and the long-term latency and drug resistance
of Mycobacterium tuberculosis have not been solved yet ([35]3). It is
necessary to take adequate prevention and treatment measures to reduce
the pain and economic loss of patients with pulmonary tuberculosis. The
traditional single gene screening for targets of anti-tuberculosis
drugs can not meet the needs of clinical medicine ([36]4), and the
existing early diagnosis technology of tuberculosis can not keep up
with the development of the times ([37]5, [38]6). Screening new drug
targets and diagnostic targets for tuberculosis by mRNA sequencing
technology may provide a new breakthrough for our existing research.
RNA sequencing (RNA-Seq) has become an economical and accurate gene
sequencing method for quantifying transcripts and detecting unknown RNA
species ([39]7). So far, it has been proven that the application of RNA
sequencing can be used to predict relevant clinical outcomes, and has
been increasingly used in the diagnosis of diseases to achieve a
personalized treatment, bringing a new insight into the transcriptional
biology at the same time ([40]8, [41]9). In this study, differentially
expressed genes of tuberculosis patients and healthy people were
screened by peripheral blood mRNA sequencing technology, potential
diagnostic targets of tuberculosis were screened by constructing PPI
regulatory network and establishing various algorithms ([42]10), and
miRNAs regulated by them were predicted and miRNA-mRNA regulatory
network was constructed to clarify the possible mechanism of
tuberculosis, which was attempted to lay a foundation for the early
clinical diagnosis of tuberculosis and the development of new
anti-tuberculosis drugs.
2. Materials and methods
2.1. Data source and grouping
A total of 20 blood samples was collected, including those from 10
patients with pulmonary tuberculosis ([43]Supplementary Material table
1) and 10 healthy controls ([44] Table 1 ). The 20 volunteers were all
over 20 years old, without a history of previous malignancies, and all
of them signed the written informed consent approved by the Ethics
Committee of the Affiliated Hospital of Beihua University abd the
committee also approved this study.
Table 1.
Basic information for 10 healthy individuals.
Health groups Sex Tuberculin test
1 male negative
2 female negative
3 female negative
4 male negative
5 female negative
6 female negative
7 female negative
8 female negative
9 male negative
10 female negative
[45]Open in a new tab
2.2. mRNA sequencing
The whole blood samples of 20 volunteers were obtained by antecubital
venipuncture. Total RNA was extracted from whole blood with the PAXgene
Blood RNA Kit. The total RNA of whole blood samples was purified using
Agilent Bioanalyzer 2100 (Agilent Technologies, USA) for quality
control. The RNA with RIN ≥ 7.0 (RNA Integrity Number) was used to
ensure the construction of the downstream high-quality total RNA SEQ
library. The sequencing conditions were: RNA sample concentration ≥ 100
ng/µL, total amount > 2 μg, OD260/280 values between 1.8 and 2.2, and
OD260/230 ≥ 2.0. Finally, the library was sequenced using 2 × 150 bp
double ended sequencing strategy on an Illumina Hiseq 2500
high-throughput sequencing platform.
2.3. Construction of PPI regulatory network
STING database ([46]http://www.string-db.org/) is a protein interaction
database that can be used to search for known and predicted
protein-protein interactions ([47]11). In this study, the
differentially expressed genes of the screened tuberculosis patients
and healthy controls were entered into the STING database and the study
species were selected as human (“Homo sapiens”) and the free nodes were
removed to obtain the differential gene PPI network. The PPI network
was visualized using the CytoNCA plug-in for network centrality
analysis in Cytoscape 3.9.1 software ([48]12).
2.4. Screening of potential key genes
The degree betweenness and closeness of differential genes in the PPI
network were calculated by three algorithms: degree centrality (DC)
that judges the importance or influence of nodes in the network,
betweenness centrality (BC) that represents the degree of interaction
between nodes and closeness centrality (CC) that measures the average
shortest distance between nodes and other points in their connected
components ([49]13, [50]14). The calculation formulas are as follows:
2.4.1. Degree centrality:
[MATH: DCi=kiN−1 :MATH]
2.4.2. Betweeness centrality:
[MATH: BCi=∑s≠i≠tnstigst
:MATH]
2.4.3. Closeness centrality:
[MATH: di=1N−1∑j=1Ndij <
/mo> CCi=1di :MATH]
2.5. GO functional clustering analysis and KEGG pathway enrichment analysis
The screened differentially expressed genes were submitted to the
Database for Annotation, Visualization and Integrated Discovery (DAVID)
([51]https://david.ncifcrf.gov/tools.jsp) ([52]15), and Homo sapiens
was taken as the research object. Gene Ontology (GO) ([53]16)
clustering analysis and Kyoto Encyclopedia of genes and genomes (KEGG)
([54]17) pathway analysis were performed on the screened differential
genes using the functional annotation module in DAVID.
2.6. Statistical analysis
GraphPadPrism 8 software was used for mapping and t-test. P< 0.05 was
considered to be statistically significant.
3. Results
3.1. mRNA sequencing and data analysis
The screened 556 differentially expressed genes ([55]Supplementary
Material table 2) between healthy people and tuberculosis patients were
screened by peripheral blood mRNA sequencing with P value< 0.005 and
|log FC| ≥ 2 as the setting conditions, of which 256 genes were
up-regulated and 300 genes were down-regulated ([56] Figures 1 , [57]2
).
Figure 1.
[58]Figure 1
[59]Open in a new tab
Hotspot map of mRNA sequencing results. Case: tuberculosis group;
Control: healthy control group; Red: up-regulated differential genes;
Blue: down-regulated differential genes.
Figure 2.
[60]Figure 2
[61]Open in a new tab
Volcano map of mRNA sequencing results. Red: up-regulated differential
genes; Green: down-regulated differential genes.
3.2. Differential gene PPI network
The screened 556 differentially expressed genes were entered into the
STRING database, and free genes were removed through hide disconnected
nodes in the network to further obtain 430 important genes. The PPI
regulatory network was visualized using Cytoscape 3.9.1 software ([62]
Figure 3 ), and there were 430 nodes and 909 edges in the PPI
regulatory network.
Figure 3.
[63]Figure 3
[64]Open in a new tab
PPI regulatory network diagram. The network color is the progressive
distribution from green to red, and the node size is the progressive
arrangement from small to large. The redder the node and the larger,
the greeter the degree of the node, and the greener the node, the
smaller the degree of the node.
3.3. Screening of potential key genes of tuberculosis
The key genes from 430 differential genes in the PPI regulatory network
were screened out by three different algorithms, DC, BC and CC. The top
ten genes screened by each algorithm were set as the core genes ([65]
Table 2 ) and the intersection screening of them was performed by using
Venn diagram ([66] Figure 4 ), and finally, six key genes (AKT1, TP53,
EGF, ARF1, CD274 and PRKCZ) were obtained. These six key genes were
considered more likely to be the potential diagnostic targets of
tuberculosis ([67] Table 3 ).
Table 2.
Top 10 differential genes ranked by degree, betweenness and closeness
values.
key nodes Degrree key nodes Betweeness key nodes Closeness
AKT1 69 TP53 60851.12161 AKT1 0.398698885
TP53 59 AKT1 56959.60582 TP53 0.395027624
EGF 22 ARF1 15621.89434 EGF 0.34047619
NCK1 21 EGF 10192.68202 ARF1 0.33411215
ARF1 19 NCK1 10033.49694 PRKCZ 0.332300542
ITGB1 16 RAP1B 9468.955752 RHOC 0.331786543
CD274 16 PRKCZ 8071.668674 ITGB1 0.331274131
PRKCZ 15 EIF4G1 7495.967994 PPP2R5C 0.33
RHOC 15 CD274 7490.923055 CD274 0.328735632
FGFR1 13 RRM2 7213.222312 CYFIP2 0.324263039
[68]Open in a new tab
Figure 4.
[69]Figure 4
[70]Open in a new tab
Wayne diagram of intersection of top ten key nodes by DC, BC and CC
algorithms.Blue: top ten key nodes by DC algorithm; Red: top ten key
nodes by BC algorithm; Green: top ten key nodes by CC algorithm.
Table 3.
Intersection data results of top ten key nodes by DC, BC and CC
algorithms.
Name Total Elements
Betweenness and Closeness and Degree 6 AKT1、TP53、EGF、ARF1、CD274、PRKCZ
Betweenness and Degree 1 NCK1
Closeness and Degree 2 ITGB1、RHOC
Degree 1 SEC13
Betweenness 3 RAP1B、RRM2、EIF4G1
Closeness 2 PPP2R5C、CYFIP2
[71]Open in a new tab
3.4. GO clustering analysis and KEGG gene pathway enrichment analysis
3.4.1. GO cluster pathway analysis
GO clustering analysis including biological process (BP), cellular
component (CC) and molecular function (MF) was performed on 556
differential genes using DAVID database, and a total of 166 pathways
were found. Among them, 79 pathways were related to BP, and the
pathways with the highest significance of P-value were regulation of
transcription and DNA-templated, 50 pathways were correlated with CC
and the pathway with the highest P-value was the nucleus, and 37
pathways were associated with MF and the pathway with the highest
significance of P-value was protein binding ([72] Figure 5 ).
Figure 5.
[73]Figure 5
[74]Open in a new tab
GO enrichment analysis.
3.5. KEGG enrichment pathway analysis
The enrichment pathways of 556 differential genes were also analyzed
through KEGG database, and 14 pathways were found, which were arranged
from small to large according to their significance (P value) ([75]
Figure 6 ). Among them, the pathway with the highest significance was
the herpes simplex virus 1 infection pathway, closely related to the
infection mechanism of tuberculosis, and the endocytosis pathway was
the second most significant pathway, closely related to the invasive
mechanism of tuberculosis. There were three signaling pathways in these
pathways, of which B cell receptor signaling pathway was an
immune-related pathway with a higher significance, while AMPK signaling
pathway and sphingolipid signaling pathway had no significance ([76]
Table 4 ).
Figure 6.
[77]Figure 6
[78]Open in a new tab
KEGG pathway enrichment analysis.
Table 4.
KEGG enrichment pathways with significance.
Pathway Gene number PValue
Herpes simplex virus 1 infection 29 0.000664519
Endocytosis 15 0.016634411
Central carbon metabolism in cancer 7 0.016965457
Transcriptional misregulation in cancer 12 0.027251953
Pyrimidine metabolism 6 0.027791341
B cell receptor signaling pathway 7 0.033930065
Osteoclast differentiation 9 0.035590242
Nucleotide metabolism 7 0.039437744
PD-L1 expression and PD-1 checkpoint pathway in cancer 7 0.047620204
[79]Open in a new tab
3.6. Construction of miRNA-mRNA-pathway regulatory network
To further elucidate the regulatory mechanism of key genes, miRNAs
associated with six key genes (AKT1, TP53, EGF, ARF1, CD274 and PRKCZ)
were searched through mirWalk database
([80]http://miRWalk.umm.uni-heidelberg.de/). MiRNAs that could be
annotated on Targetscan database, miRDB database and mirTarBASE
database at the same time were directly screened by using mirWalk
database as our research objects. It was found that two key miRNAs
(has-mir-150-5p and has-mir-25-3p) were closely related to these six
genes ([81]Supplementary Material table 3).
The six key genes and two corresponding miRNAs, other mRNAs in the PPI
regulatory network related to these six key genes and the three
pathways where these six genes were located were used together to
construct the miRNA-mRNA-pathway regulatory network ([82] Figure 7 ).
Figure 7.
[83]Figure 7
[84]Open in a new tab
miRNA-mRNA-pathway regulatory network diagram.Red: up-regulated
expression genes; Green: down-regulated expression genes; Blue: related
pathways; Round: mRNA, the larger circle represents the 6 key mRNAs,
the diamond represents the miRNA, and the square represents the related
pathway. mRNA-miRNA and mRNA-pathway was emphasized with yellow lines.
It was found in the miRNA-mRNA-pathway regulatory network that the two
miRNAs has-miR-150-5p and has-miR-25-3p could regulate gene TP53 and
ARF1, respectively, and the expression of gene TP53 and ARF1 was
up-regulated because the expression of these two miRNAs was inhibited.
The up-regulated expression of gene TP53 and ARF1 could activate the
expression of AKT1, EGF and CD274 and activate the pathway herpes
simplex virus 1 infection and B cell receptor signaling pathway. At the
same time, the up-regulated expression of TP53 and ARF1 suppressed the
expression of PRKCZ and the endocytosis pathway.
4. Discussion
Pulmonary tuberculosis is an infectious disease of respiratory system
caused by Mycobacterium tuberculosis, and the main cause of infectious
death in the world, caused 1.5 million deaths in 2021 ([85]2). The
current clinical diagnosis and treatment technology of tuberculosis
have not kept up with the development of the times, so it is urgent for
us to need innovative diagnosis methods that can diagnose it rapidly at
its early stage and new effective anti-tuberculosis drugs that can
shorten the treatment time and improve the prevention and treatment
effect ([86]6). In this study, the potential diagnostic targets of
tuberculosis were screened out through mRNA sequencing technology,
which may provide a theoretical basis for the early diagnosis of
tuberculosis in clinics and the development of specific new drugs used
for the prevention and treatment of tuberculosis.
First, 556 differentially expressed genes of tuberculosis were screened
out, then the PPI regulatory network was constructed and the key nodes
of the PPI regulatory network through were screened out by the
combination of three algorithms (DC, BC and CC), and finally we found
six key genes (AKT1, TP53, EGF, ARF1, CD274 and PRKCZ).
Among the six key genes, the protein expressed by AKT1 gene is the key
enzyme in controlling the intracellular growth in tuberculosis, and
some studies have pointed out that it may be a well candidate for
genetic markers for assessing tuberculosis risks ([87]18). TP53 gene
can regulate the immune function of human body to evade attacks,
eventually leading to the existing of Mycobacterium tuberculosis in a
non-proliferative state in the host through its impact on the host
immune system to cause the delay of the disease course and the repeated
and persistent infection ([88]19). EGF gene is the growth factor of
pathogenic mycobacteria in granuloma tissue and macrophages, and may
improve the growth rate of intracellular and extracellular mycobacteria
at the infection site ([89]20). Moreover, it has been demonstrated that
EGF combined with the other four tuberculosis specific biomarkers can
accurately predict 90.9-100% of active pulmonary tuberculosis ([90]21).
ARF1 gene can affect the rearrangement of actin cytoskeleton in
epithelial cells, and the initial activation of ARF1 can promote the
invasion of mycobacteria ([91]22). CD274, also known as programmed cell
death ligand 1 (PD-L1), is associated with the suppression of the
immune system ([92]23). It has been found that Mycobacterium
tuberculosis can promote the expansion of regulatory T cells by
inducing CD274 gene in dendritic cells ([93]24). In addition, it has
been shown that 10 biomarkers, including CD274 and EGF, can be used for
the diagnosis of tuberculosis, the differential diagnosis of latent
tuberculosis infection/active tuberculosis, and the risk of progression
to active tuberculosis ([94]25). PRKCZ is a member of the PKC family,
and the gene level of the PKC family can be used as a predictor of
response to immunotherapy ([95]26). The above analysis indicates that
these six genes (AKT1, TP53, EGF, ARF1, CD274 and PKCZ) are all related
to the immune system of the host and involved in the pathogenesis of
tuberculosis, EGF and CD274 have been used as biomarkers for the
diagnosis of tuberculosis ([96]21, [97]25). Therefore, we suggest that
EGF and CD274 are potential drug targets and diagnostic targets for TB,
and AKT1, TP53, ARF1 and PRKCZ may be new potential drug targets and
diagnostic targets for TB.
The GO enrichment analysis showed that the enrichment of regulation of
transcription and DNA-templated was the most significant in biological
process (BP), which may be because the transcription process is the
first step of protein biosynthesis, and DNA-templated is one of the
indispensable conditions in the transcription process; the enrichment
of the nucleus was the most significant in the cellular component (CC),
and it was found in the study that the proteins contained in
Mycobacterium tuberculosis can ectopic to the host nucleus to inhibit
the production of nitric oxide ([98]27), which may be because the
nucleus is the control center of cells genetic and metabolic
activities; the enrichment of protein binding was the most significant
in molecular function (MF). The results of GO enrichment analysis
indicate that Mycobacterium tuberculosis may affect the protein binding
function through the first step of protein biosynthesis, and then cause
the effector protein of Mycobacterium tuberculosis to enter the host
nucleus to regulate the host gene expression ([99]28).
In the KEGG pathway analysis of differential genes, 14 pathways were
screened out, of which the most significant pathway was the herpes
simplex virus 1 infection pathway, the herpes virus infection pathway.
Some studies have found that patients with pulmonary tuberculosis have
a higher active infection of herpes simplex virus ([100]29), so this
pathway may also be related to the pathogenesis of tuberculosis.
Endocytosis pathway was the second most significant pathway, and
Mycobacterium tuberculosis often invades cells through endocytosis, so
this pathway is closely related to the invasion mechanism of
Mycobacterium tuberculosis ([101]30). B cell receptor signaling
pathway, an important immune pathway, can regulate the survival,
proliferation and differentiation of B lymphocytes ([102]31, [103]32).
In addition, we also found that among the six key genes previously
screened, TP53 and AKT1 genes were in the herpes simplex virus 1
infection pathway, genes ARF1 and PRKCZ were in the endocytosis
pathway, and gene AKT1 was in the B cell receptor signaling pathway,
indicating that these four genes all play important regulatory roles in
the pathway. Therefore, it was speculated that genes AKT1, TP53, EGF,
ARF1, CD274 and PRKCZ may cause the infection and invasion of
Mycobacterium tuberculosis through herpes simplex virus 1 infection,
endocytosis and B cell receptor signaling pathways.
A miRNA-mRNA-pathway regulatory network was constructed by using the
six key genes (AKT1, TP53, EGF, ARF1, CD274 and PRKCZ), two miRNAs
(has-mir-150-5p and has-mir-25-3p) and three pathways (herpes simplex
virus 1 infection, endocytosis and B CLL receptor signaling pathway)
together. The analysis of the miRNA-mRNA-pathway regulatory network
showed that two miRNAs could regulate genes TP53 and ARF1,
respectively. MiR-150, a developmental regulator of several immune cell
types, can regulate the differentiation of B cells/T cells/natural
killer cells ([104]33). In addition, by reviewing the literature, we
found that previous studies have proved that has-miR-25-3p is
down-regulated in tuberculosis patients by RT-PCR experiments
([105]34).
Furthermore, we also found that gene TP53 was in the herpes simplex
virus 1 infection pathway, and gene AKT1 could play a regulatory role
not only in the herpes simplex virus 1 infection pathway but also in
the B cell receptor signaling pathway, suggesting genes TP53 and AKT1
may play a key role in regulating the immunity and the intracellular
growth of Mycobacteria tuberculosis. In addition to the key genes TP53
and AKT1 in the herpes simplex virus 1 infection pathway, we found that
the up-regulated target gene HLA-DPA1 corresponding to both EGF and
TP53 could also play a regulatory role in the herpes simplex virus 1
infection pathway. So it is speculated that the up-regulated genes
TP53, AKT1 and HLA-DPA1 may activate the herpes simplex virus 1
infection pathway in the body, leading to the occurrence of
tuberculosis. Besides genes ARF1 and PRKCZ, it was found in the
endocytosis pathway that the up-regulated gene ARFGAP2 corresponding to
ARF1 and the down-regulated genes RABEP1, DNM3 and GIT1 were also in
the endocytosis pathway, so it is speculated that endocytosis’s cell
surface receptor internalization and signal regulation functions may be
disordered by the down-regulated genes PRKCZ, RABEP1, DNM3 and GIT1,
and The up-regulated expression of ARF1 and ARFGAP2 may cause the
endocytosis mechanism of host cells to be hijacked by the pathogen,
leading to the invasion of Mycobacterium tuberculosis into the body.
Therefore, we speculate that in tuberculosis patients, the expression
of genes AKT1 and TP53 is up-regulated, and the herpes simplex virus 1
infection pathway is activated at the same time, so that Mycobacterium
tuberculosis is rapidly infected, and in addition, gene AKT1 may
activate the B cell receptor signaling pathway, that is, activating the
human immune system, and at the same time, gene PRKCZ may inhibit the
endocytosis pathway to inhibit the endocytosis, so that the infection
of Mycobacterium tuberculosis cannot be suppressed by endocytosis,
thereby accelerating the pathogenesis of tuberculosis.
5. Conclusion
Six key genes and 2 miRNAs that regulate them were screened out by mRNA
sequencing, and PPI regulatory network construction and analysis. They
may regulate the pathogenesis of infection and invasion of
Mycobacterium tuberculosis through the herpes simplex virus 1
infection, endocytosis and B cell receptor signaling pathway, which may
lay a foundation for the early clinical diagnosis and research of
tuberculosis and the development of new anti-tuberculosis drugs.
Data availability statement
The datasets presented in this study can be found in online
repositories. The names of the repository/repositories and accession
number(s) can be found below:
[106]https://www.ncbi.nlm.nih.gov/bioproject/PRJNA876021. SRA data:
PRJNA876021.
Ethics statement
The studies involving human participants were reviewed and approved by
medical Ethics Committee of Beihua University Affiliated Hospital. The
patients/participants provided their written informed consent to
participate in this study.
Author contributions
YY wrote the manuscript and derived the datasets, YF and SS revised the
manuscript, CJ and XP performed data analysis. GX designed the study.
All authors contributed to the article and approved the submitted
version.
Funding Statement
This research was funded by the National Natural Science Foundation of
China (81973371), Jilin Provincial Department of Education Science and
Technology Research Project (JJKH20220065KJ), Innovation and
Entrepreneurship Training Program for College Students in Jilin
Province (202210201034).
Conflict of interest
The authors declare that the research was conducted in the absence of
any commercial or financial relationships that could be construed as a
potential conflict of interest.
Publisher’s note
All claims expressed in this article are solely those of the authors
and do not necessarily represent those of their affiliated
organizations, or those of the publisher, the editors and the
reviewers. Any product that may be evaluated in this article, or claim
that may be made by its manufacturer, is not guaranteed or endorsed by
the publisher.
Supplementary material
The Supplementary Material for this article can be found online at:
[107]https://www.frontiersin.org/articles/10.3389/fimmu.2023.1038647/fu
ll#supplementary-material
[108]Click here for additional data file.^ (100.7KB, xlsx)
References