Research Article Comprehensive Bioinformatics Analysis of Toll-Like Receptors (TLRs) in Pan-Cancer
Wei Ping,1 Senyuan Hong D,2 Yang Xun 1,2 and Cong Li DD2
1Department of Thoracic Surgery, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, No. 1095 Jiefang Avenue, 430030 Wuhan, China
2Department of Urology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, No. 1095 Jiefang Avenue, 430030 Wuhan, China
Correspondence should be addressed to Cong Li; licongtjm@163.com
Received 19 April 2022; Revised 20 June 2022; Accepted 14 July 2022; Published 28 July 2022
Academic Editor: Mujeeb Zafar Banday
Copyright @ 2022 Wei Ping et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background. To conduct a comprehensive bioinformatics analysis on the transcriptome signatures of Toll-like receptors (TLRs) in pan-cancer. Materials and methods. A total of 11,057 tissues consisting of 33 types of carcinoma in The Cancer Genome Atlas (TCGA) were retrieved, and then we further explored the correlation between TLRs’ expression with tumorigenesis, immune infiltration, and drug sensitivity. We conducted a comprehensive bioinformatics analysis on TLR1 to 10 in pan-cancer, including differential expression analysis between normal and tumor tissues, differential immune subtype correlation, survival analysis, tumor immune infiltration estimating, stemness indices correlation, and drug responses correlation. Results. TLR2 was highly expressed in most types of tumors. TLR9 was hardly expressed compared to other TLR genes, which lead to TLR9 showing less correlation with both immune-estimate scores and stromal-estimate scores. All the TLRs were related with immune subtype of tumor samples that all of them were differentially expressed in differential immune subtype samples. The expression of TLRs was positively related with immune-estimate scores and stromal-estimate scores in almost all types of tumor. The expression of TLRs was negatively correlated with mRNA expression-based stemness scores (RNAss) in nearly almost type of tumors except kidney renal clear cell carcinoma (KIRC) and also negatively correlated with DNA methylation- based stemness scores (DNAss) in many types of tumors except adrenocortical carcinoma (ACC), cholangiocarcinoma (CHOL), KIRC, acute myeloid leukemia (LAML), low-grade glioma (LGG), testicular germ cell tumors (TGCT), thyroid carcinoma (THCA), thymoma (THYM), and uveal melanoma (UVM). The expression of TLR9 was significantly positively correlated with the drug sensitivity of fluphenazine, alectinib, carmustine, and 7-hydroxystaurosporine. TLR7 was significantly positively correlated with the drug sensitivity of alectinib. Conclusions. Our study reveals the significant role of TLRs family in pan-cancer and provides potential therapeutic strategies of cancer.
1. Introduction
Toll-like receptors (TLRs) are a family of transmembrane pattern recognition receptors that play essential roles in innate immunity for the detection of and defense against microbial pathogens [1]. TLRs are the first-line protective immune sentries that can recognize pathogen-associated molecular patterns (PAMPs), which typically include unmethylated double-stranded DNA (CpG), single- stranded RNA (ssRNA), lipoproteins, lipopolysaccharide (LPS), and flagellin [2]. They have been widely studied as
the main mediators of innate immunity in animals, from insects to humans [3-5]. The discovery of TLRs as compo- nents that recognize the conserved structures in pathogens has greatly promoted the understanding of how the body perceives pathogen invasion, triggers innate immune responses, and initiates antigen-specific adaptive immu- nity [6].
It was reported that Drosophila strains with mutants of the Toll gene were highly susceptible to fungal infection, which was the first indication of the innate immune function of TLRs [7]. A human Toll homologue, now called TLR4,
was then identified [8]. Currently, a total of 10 TLR family members have been identified in humans, and at least 13 have been discovered in mice. These are usually expressed by various immune cells, such as dendritic cells (DCs), macrophages, T-cell subsets, and B-cells. They are also expressed in nonimmune cells (e.g., epithelial cells and fibroblasts) in humans [9]. All TLRs include an N- terminal domain characterized by multiple leucine-rich repeats and a carboxyl-terminal TIR domain that interacts with TIR-containing adapters. Nucleic acid-sensing TLRs (TLR3, TLR7, TLR8, and TLR9) are located in the endo- plasmic chamber, whereas the remaining TLRs are present on the plasma membrane [10, 11].
In recent years, TLRs have gained great interest in can- cer research because of their role in tumor progression, and many therapeutic interventions for TLR have been developed or studied. Some studies have explored in detail the role of TLR regulation in cancer development [12-14]. Compared to that in normal patients, the expression of TLR1, 2, 4, and 8 mRNA was increased in patients with colorectal cancer [15]. TLRs have also been associated with prostate cancer, but they may be a double-edged sword in prostate tumorigenesis because they can both promote malignant transformation of epithelial cells thereby enhancing tumor growth and induce apoptosis, thus, inhi- biting tumor progression [16]. In addition, the regulation of TLRs not only increases the susceptibility to infection from some microorganisms but also contributes to the development of cancer by altering the microbiota resulting in inflammation [17]. On one hand, TLRs play an essen- tial role in tumor immunity by activating a variety of cells, such as DCs, T-cell subsets, and even tumor cells; on the other hand, the activation of TLRs can also lead to inflam- mation that results in tumor promotion [18].
However, the characteristics of TLRs differ, and differ- ent homologous types may have different effects on differ- ent tumor types. In addition, to date, no bioinformatics study has systematically investigated the transcriptional levels of each TLR across multiple cancers. Therefore, it is of great significance to study the expression patterns of TLRs in cancer tissues and to develop potential TLR- targeted drugs for treatment of tumors with differentially expressed TLRs. In this study, we analyzed the expression characteristics of TLR1 to TLR 10 in various cancer tissues using a variety of bioinformatics methods, comprehen- sively analyzed TLRs, and found that the transcriptional levels of TLRs were associated with stemness, tumor purity, and drug sensitivity in cancer tissues included in The Cancer Genome Atlas (TCGA).
2. Materials and Methods
2.1. Data Sources. The transcriptome profile, clinical phe- notype information, survival information, immune subtype profile, and DNA and RNA stemness profiles of 33 types of tumors were downloaded from the Genomic Data Commons (GDC) TCGA sets or TCGA pan-cancer sets in the UCSC Xena database (http://xena.ucsc.edu/) on November 15, 2020. Transcriptome profiles containing
both tumor and normal adjacent tumor (NAT) tissues yielded a total of 11,057 samples, coded as fragments per kilobase per million (FPKM).
2.2. Expression Status of TLRs across Multiple Cancer Types. We first extracted and visualized the pan-cancer expression of TLRs. We then selected the five most highly expressed TLRs for further differential expression analysis. We sorted the expression profiles for cancer types whose expression profiles retained the expression profile of NAT tissues, and they were BLCA, BRCA, CHOL, COAD, ESCA, GBM, HNSC, KICH, KIRC, KIRP, LIHC, LUAD, LUSC, PRAD, READ, STAD, THCA, and UCEC. We then extracted the expression of the 5 most highly expressed TLRs in these can- cer types and performed differential expression analysis between tumors and NAT using the Wilcoxon test. In addi- tion, for all the TLRs, we calculated the log2 fold change (logFC) of each TLR in these cancer types and presented it in a heatmap. Subsequently, we applied a correlation test to explore the coexpression of the 10 TLRs according to their expression profiles.
2.3. Prognostic Value of TLRs across Multiple Cancer Types. For each TLR gene and tumor type, we separately performed log-rank survival analysis (grouped by the medium expres- sion of the TLR in each cancer type) and univariate Cox regression to explore the pan-cancer prognostic value of TLRs. We then visualized the survival curves with significant differences and drew a forest plot of the resulting hazard ratios (HRs) and their 95% confidence interval.
2.4. Immune Subtype Correlations, Stemness Indices Correlations, and Tumor Microenvironment (TME) Estimations. Based on the immune subtype profile of each TCGA sample downloaded from the UCSC Xena, we explored the differential expression status of TLRs in differ- ent immune subtypes using the Wilcoxon test. We further probed the correlation between the expression of TLRs and the stemness index of the tissue samples containing DNA methylation-based stemness scores (DNAss) and mRNA expression-based stemness scores (RNAss) across multiple cancer types using the Spearman’s correlation test. In addi- tion, we applied the ESTIMATE method to analyze the immune-estimate score and stromal-estimate score of each sample and then performed the Spearman’s correlation test to examine the correlation between the expression of TLRs and these two scores.
2.5. Drug Sensitivity Analysis of TLRs across Multiple Cancer Types. Data including both expression of TLRs and drug sensitivity were retrieved from the CellMiner database ((https://discover.nci.nih.gov/cellminer/), which collects genomic and pharmacologic information for investigators to determine the correlation between gene expression and drug sensitivity in the NCI-60 cell line sets. Thus, we extracted the expression values of TLRs in NCI-60 cell lines and their corresponding drug sensitivities to different drugs and conducted a Pearson correlation test between the expression of TLRs and drug sensitivity to explore the drug sensitivity in patients.
UCSC XENA database
RNA-seq data
Phenotype data survival data stemness scores immune subtype data
Expression of TLR family in pan-cancer
Differential analysis
Co-Expression analysis
Clinical analysis
Immune subtype analysis
Stemness score analysis
Tumor purity analysis
Drug Senditivity
Stage analysis Survival analysis Cox proportional hazard model
RNAss DNAss
Stromal score Immune score
2.6. TLRs in KIRC. Finally, as TLR expression performed well in predicting the overall survival for KIRC, we further explored the significance of TLRs in KIRC. We separately investigated the differential expression of TLRs among dif- ferent immune subtypes, the correlation between TLR expression and stemness indices, and the correlation between TLR expression and ESTIMATE scores in KIRC samples. In addition, we explored the differential expres- sion status of TLRs between stages I and IV to determine whether TLRs could serve as biomarkers of survival and progression in KIRC.
2.7. Statistical Analysis. All statistical analyses were con- ducted using the R software (version 4.0.2). Statistical signif- icance was set at p < 0.05.
3. Results
3.1. Differential Expression Analysis of TLRs between Tumor and NAT Tissues. The flowchart of the study is summarized in Figure 1, and the abbreviations of the 33 tumor types in TCGA are provided in Table 1. The pan-cancer gene expres- sion of TLR1 to TLR10 is displayed in Figure 2(a), and it seems that the expression of TLR9 was low compared to that of the other TLR genes. In addition, differential expression analysis with the Wilcoxon test was performed on the 10 TLR family genes between tumor and NAT tissues. Further- more, the five most highly expressed genes, TLR1 to TLR5, were selected to show the differential expression status. TLR1 expression was significantly low in most types of can- cers, except CHOL, GBM, and KIRC (Figure 2(b)). TLR2
was significantly expressed in most tumor types, except BRCA, LIHC, LUAD, LUSC, and PRAD (Figure 2(c)). TLR3 expression was significantly low in most type of tumors, except GBM and KIRC (Figure 2(d)). TLR4 expres- sion was significantly low in most type of tumors, except GBM and KIRC (Figure 2(e)). TLR5 expression was signifi- cantly low in most type of tumors, except CHOL, GBM, and LIHC (Figure 2(f)).
3.2. Coexpression Analysis of TLRs across Multiple Cancer Types and Log-Rank Survival Analysis. More detailed infor- mation about the differential expression status, including log2FC, is shown in Figure 3(a). It was obvious that TLR2 was highly expressed in most types of cancer, and TLR fam- ily members were least expressed in LUSC and LUAD. In addition, coexpression analysis of TLRs suggested that all TLRs were positively correlated with each other, except TLR3, which was negatively correlated with TLR9 (Figure 3(b)). We then employed Kaplan-Meier methods to plot survival curves and performed a log-rank analysis to investigate the prognostic value of TLRs for the 33 TCGA cancers. The prognostic values of TLRs with cancer type and p value are shown in Table 2. We then selected KIRC to plot the survival curves for the four TLR genes with prognostic values for KIRC, TLR1 (Figure 3(c)), TLR3 (Figure 3(d)), TLR4 (Figure 3(e)), and TLR9 (Figure 3(f)). Among these, low expression of TLR1, TLR3, and TLR4 was significantly associated with poor overall survival, while high expression of TLR9 was significantly associated with poor overall sur- vival in KIRC.
| Abbreviation | Tumor type |
|---|---|
| ACC | Adrenocortical carcinoma |
| BLCA | Bladder urothelial carcinoma |
| BRCA | Breast invasive carcinoma |
| CESC | Cervical squamous cell carcinoma and endocervical adenocarcinoma |
| CHOL | Cholangiocarcinoma |
| COAD | Colon adenocarcinoma |
| DLBC | Lymphoid neoplasm diffuse large B-cell lymphoma |
| ESCA | Esophageal carcinoma |
| GBM | Glioblastoma multiforme |
| HNSC | Head and neck squamous cell carcinoma |
| KICH | Kidney chromophobe |
| KIRC | Kidney renal clear cell carcinoma |
| KIRP | Kidney renal papillary cell carcinoma |
| LAML | Acute myeloid leukemia |
| LGG | Brain lower grade glioma |
| LIHC | Liver hepatocellular carcinoma |
| LUAD | Lung adenocarcinoma |
| LUSC | Lung squamous cell carcinoma |
| MESO | Mesothelioma |
| OV | Ovarian serous cystadenocarcinoma |
| PAAD | Pancreatic adenocarcinoma |
| PCPG | Pheochromocytoma and paraganglioma |
| PRAD | Prostate adenocarcinoma |
| READ | Rectum adenocarcinoma |
| SARC | Sarcoma |
| SKCM | Skin cutaneous melanoma |
| STAD | Stomach adenocarcinoma |
| TGCT | Testicular germ cell tumors |
| THCA | Thyroid carcinoma |
| THYM | Thymoma |
| UCEC | Uterine corpus endometrial carcinoma |
| UCS | Uterine carcinosarcoma |
| UVM | Uveal melanoma |
3.3. Cox Regression and Immune Subtype Analysis. Univari- ate Cox proportional hazard regression was performed to explore the prognostic values of TLRs for the 33 types of cancer. Genes were considered a risk factor if the HR was >1 or a protective factor if the HR was <1. According to the forest plot (Figure 4(a)), we found that TLRs play a com- plex role in cancer prognosis, which is risky in some types of tumors but protective in the remaining types of tumors. In addition, we performed a Kruskal test on the expression of TLRs in the six immune subtypes across the 33 TCGA can- cer types (Figure 4(b)). Interestingly, all TLRs were differen- tially expressed in the different immune subtype samples. Among them, TLR1, TLR2, TLR3, TLR5, TLR6, TLR7, and TLR8 showed the highest expression in C6 immune subtype samples, whereas TLR4 and TLR10 showed the highest expression in the C5 immune subtype.
3.4. TLRs and TME across Multiple Cancer Types. Immune- estimate scores and stromal-estimate scores of samples were calculated using the R package “ESTIMATE” [19], and Spearman’s correlation test was used to explore the correla- tion between TLR expression and the TME. For the immune score, expression of TLRs was positively correlated with immune scores in almost all types of cancer, except TLR1 in UVM, TLR3, 4, and 5 in THYM, and TLR10 in DLBC (Figure 5(a)). In addition, for the stromal scores, the expres- sion of TLRs was positively correlated with stromal scores in almost all types of cancer, except TLR1 in UVM and TLR3 in ACC, LAML, MESO, and READ (Figure 5(b)). TLR9 showed low correlation with both immune and stromal scores, which may be due to the low expression of TLR9 in all the tumor samples.
3.5. TLRs and Stemness Indices across Multiple Cancer Types. We downloaded the stemness indices for all the samples from the UCSC Xena database, which were calculated using the one-class logistic regression (OCLR) as proposed by Malta et al. [20]. Two types of stemness indices were assessed: DNAss and RNAss. Interestingly, the expression of TLRs was negatively correlated with RNAss in nearly all types of cancer, except KIRC (Figure 5(c)), and negatively correlated with DNAss in many types of cancer, except ACC, CHOL, KIRC, LAML, LGG, TGCT, THCA, THYM, and UVM (Figure 5(d)). Among the DNAss scores, nearly all TLRs, except for TLR7 and TLR9, were positively corre- lated with DNAss in THYM samples.
3.6. TLRs and Drug Responses across Multiple Cancer Types. The expression profile of NCI-60 cancer cell lines and their drug sensitivity were downloaded from the CellMiner data- base; the Pearson correlation test was then performed to fur- ther analyze the correlation between the expression and the response to 263 antineoplastic drugs. All results with signif- icant correlation between TLRs and drug sensitivity are dis- played in Supplementary Table (available here), and the 25 most significant results with the smallest p value are shown as scatter plots ranked by p value (Figure 5(e)). Among them, the five most significant correlations were as follows: the expression of TLR9 had a significant positive correlation with the response to fluphenazine (coefficient = 0.680, p < 0.001), alectinib (coefficient = 0.637, p < 0.001), carmustine (coefficient = 0.598, p < 0.001), and 7-hydroxystaurosporine (coefficient = 0.550, p < 0.001), while TLR7 had a significant positive correlation with alectinib (coefficient = 0.595, p < 0.001).
3.7. TLRs in KIRC. Finally, we explored TLRs in KIRC by comparing the transcriptional expression of TLRs at differ- ent stages of KIRC, comparing the differential expression of TLRs in different immune subtypes, and investigating the correlation between TLRs and stemness indices or tumor purity in KIRC. TLR2, TLR3, TLR4, and TLR10 were significantly differentially expressed between stages I and IV (p <0.05) (Figure 6(a)), and TLR1, TLR3, TLR4, TLR7, TLR8, and TLR10 were significantly differentially expressed between the C1 and C6 immune subtypes
TLR1 expression
0
1
2
3
4
5
⁎
Tumor
Normal
Type
BLCA
BRCA
CHOL
⁎
Gene expression
0
1
2
3
4
5
COAD
ESCA
TLR1
GBM
I
TLR2
FIGURE 2: Continued.
HNSC
TLR3
…
KICH
TLR4
(b)
Cancer type
KIRC
⁎⁎
(a)
TLR5
KIRP
TLR6
LIHC
TLR7
⁎
LUAD
⁎⁎⁎
TLR8
LUSC
**
⁎⁎⁎
TLR9
PRAD
TLR10
READ
STAD
..
THCA
…
UCEC
TLR3 expression
TLR2 expression
0
2
4
6
0
2
4
6
Tumor
Normal
Type
Type
BLCA
**
Tumor
Normal
BLCA
**
BRCA
BRCA
CHOL
CHOL
**
COAD
COAD
..
ESCA
ESCA
GBM
1
GBM
FIGURE 2: Continued.
HNSC
HNSC
KICH
KICH
**
(d)
Cancer type
KIRC
(c)
Cancer type
KIRC
KIRP
KIRP
⁎
LIHC
LIHC
..
.. ..
LUAD
LUAD
LUSC
LUSC
PRAD
PRAD
READ
READ
STAD
STAD
THCA
THCA
UCEC
UCEC
pan-cancer. (f) Differential expression of TLR5 in pan-cancer.
(f)
Differential expression of TLR2 in pan-cancer. (d) Differential expression of TLR3 in pan-cancer. (e) Differential expression of TLR4 in FIGURE 2: Expression status of TLRs. (a) Expression of TLRs in pan-cancer. (b) Differential expression of TLR1 in pan-cancer. (c)
TLR5 expression
0
1
2
3
4
5
Type
Tumor
Normal
Type
Tumor
Normal
BLCA
BRCA
CHOL
0
COAD
ESCA
GBM
0
**
HNSC
KICH
Cancer type
KIRC
(e)
KIRP
LIHC
LUAD
LUSC
PRAD
READ
…
STAD
⁎
THCA
UCEC
TLR4 expression
0
2
4
6
BLCA
BRCA
CHOL
**
COAD
ESCA
GBM
+
⁎
HNSC
KICH
⁎
Cancer type
KIRC
KIRP
LIHC
LUAD
LUSC
PRAD
**
READ
STAD
THCA
⁎
…
UCEC
CHOL
1.5
GBM
KIRC
LUAD
1
LUSC
KICH
0.5
COAD
READ
BRCA
0
LIHC
PRAD
-0.5
ESCA
STAD
THCA
-1
BLCA
UCEC
HNSC
-1.5
KIRP
TLR1
TLR2
TLR3
TLR4
TLR5
TLR6
TLR7
TLR8
TLR9
TLR10
(a)
TLR1
TLR2
TLR3
TLR4
TLR5
TLR6
TLR7
TLR8
TLR9
TLR10
1
TLR1
TLR2
0.66
☒
☒
☒
☒
☒
0.8
☒
0.6
TLR3
0.4
0.33
☒
☒
☒
☒
TLR4
0.56
0.42
0.34
☒
☒
☒
0.4
0.2
TLR5
0.44
0.41
0.28
0.17
☒
0
TLR6
0.79
0.59
0.16
0.4
0.36
-0.2
TLR7
0.72
0.54
0.4
0.66
0.4
0.52
-0.4
TLR8
0.72
0.65
0.28
0.59
0.29
0.58
0.69
-0.6
TLR9
0.24
0.24
-0.09
0.16
0.07
0.32
0.13
0.22
-0.8
TLR10
0.55
0.39
0.13
0.37
0.26
0.57
0.53
0.49
0.22
-1
(b)
Cancer: KIRC
1.00
Overall survival
0.75
0.50
0.25
p = 0.035
0.00
0
2
4
6
8
10
12
Time (years)
TLR1 levels
High
265
192
117
54
25
8
1
Low
266
168
101
45
16
5
0
0
2
4
6
8
10
12
Time (years)
TLR1 levels
+ High
+ Low
(c)
Cancer: KIRC
1.00
Overall survival
0.75
0.50
0.25
P < 0.001
0.00
0
2
4
6
8
10
12
Time (years)
TLR3 levels
High
265
195
122
Low
63
25
9
1
266
165
96
36
16
4
0
0
2
4
6
8
10
12
Time (years)
TLR3 levels
+
High
+
Low
(d)
Cancer: KIRC
1.00
Overall survival
0.75
0.50
0.25
p = 0.007
0.00
0
2
4
6
8
10
12
Time (years)
TLR4 levels
High
265
182
114
53
27
8
1
Low
266
178
104
46
14
5
0
0
2
4
6
8
10
12
Time (years)
TLR4 levels
+
High
+
Low
(e)
Cancer: KIRC
1.00
Overall survival
0.75
0.50
0.25
p = 0.020
0.00
0
2
4
6
8
10
12
Time (years)
TLR9 levels
High
265
174
92
31
13
3
0
Low
266
186
126
68
28
10
1
0
2
4
6
8
10
12
Time (years)
TLR9 levels
+
High
+ Low
(f)
(p <0.001) (Figure 6(b)). For RNAss in the KIRC samples, TLR5 and TLR9 had significant negative correlations (correlation coefficient = - 0.12, p=0.042 and correlation coefficient = - 0.23, p <0.001, respectively), but TLR1, TLR2, and TLR3 had significant positive correlations (correlation coefficient = 0.11, p=0.048; correlation coefficient = 0.14, p=0.014; and correlation coefficient = 0.14; p=0.013, respectively). For DNAss in the KIRC samples, it was interesting that all the TLRs were nega- tively correlated in KIRC patients, among which TLR1, TLR2, TLR6, TLR7, TLR8, and TLR10 were significant at p <0.05. In addition, all the TLRs had significant positive correlations with the immune scores, stromal scores, and ESTIMATE scores. Among them, TLR1, TLR2, TLR4, TLR5, TLR6, TLR7, TLR8, and TLR10 were positively cor- related with stromal scores (p <0.05); TLR1, TLR2, TLR4, TLR5, TLR6, TLR7, TLR8, TLR9, and TLR10 were posi- tively correlated with immune scores (p<0.05); and TLR1, TLR2, TLR4, TLR5, TLR6, TLR7, TLR8, and TLR10 were positively correlated with the ESTIMATE scores (p <0.05) (Figure 6(c)).
4. Discussion
Many studies have demonstrated that several cellular and molecular mechanisms can help tumors escape the body’s natural immune response [21, 22]. The importance of immune regulation in cancer progression can be explained by the increase in the number of immunosuppressive factors and cells and the lack of immune system-activating signals
in the TME. TLRs are important receptors that activate immune cells and have been reported to play an important role in cancers, such as bladder cancer and colorectal cancer [23, 24]. This makes TLRs suitable targets for ligand drug discovery strategies to establish new therapeutics for cancer [25]. Hence, it is worthwhile to explore the role of TLRs in tumor development. TLRs can upregulate the expression of costimulatory molecules, such as CD40, CD80, and CD86, and cytokines, such as IL-12, thus stimulating other immune cells, including T lymphocytes [26, 27]. However, TLR expression can lead to tumor growth by stimulating other cells, including cancer cells [28].
In this study, we explored the relationship between TLR transcriptional expression and TCGA tumor characteristics, including the TME, clinical significance, immune subtypes, stem cells, and drug response. We found that TLR isotypes have a significant effect on tumorigenesis. First, we analyzed the differential expression of 33 TCGA cancer types in 11,057 samples (including 10,327 tumor samples and 730 paracancerous samples). Through multidimensional analy- sis, we found significant differences in TLR expression levels among different cancer types. Survival and Cox proportional hazard regression analyses were also performed. For some types of cancers, we found a statistically significant differ- ence in survival between patients with high and low TLR expression, suggesting that TLRs may be a potential prog- nostic indicator for clinical applications. Furthermore, we performed drug response analysis to explore the relationship between drug sensitivity and TLRs. This is expected to pro- vide insights for new cancer therapies.
| Gene | Cancer type | p value |
|---|---|---|
| TLR1 | KIRC | 0.034688969 |
| TLR1 | LGG | 0.000296933 |
| TLR1 | SARC | 0.026721888 |
| TLR1 | SKCM | 0.000697683 |
| TLR1 | UVM | 0.002732005 |
| TLR2 | LGG | 1.65E-05 |
| TLR2 | LUAD | 0.008433019 |
| TLR2 | MESO | 0.017243009 |
| TLR2 | SKCM | 2.17E-06 |
| TLR2 | TGCT | 0.018921076 |
| TLR2 | THYM | 0.009423895 |
| TLR3 | KIRC | 2.94E-07 |
| TLR3 | KIRP | 0.004130991 |
| TLR3 | LGG | 0.000104245 |
| TLR3 | MESO | 0.002692585 |
| TLR3 | PAAD | 0.024365165 |
| TLR3 | SARC | 0.009139017 |
| TLR3 | SKCM | 0.000167396 |
| TLR3 | TGCT | 0.042432124 |
| TLR3 | UCEC | 0.031991718 |
| TLR4 | ACC | 0.007177616 |
| TLR4 | KIRC | 0.007164204 |
| TLR4 | LAML | 0.044171562 |
| TLR4 | LUAD | 0.028500171 |
| TLR4 | SKCM | 7.46E-05 |
| TLR4 | TGCT | 0.021867869 |
| TLR4 | THYM | 0.020130819 |
| TLR4 | UCEC | 0.00545563 |
| TLR5 | ACC | 0.01419818 |
| TLR5 | ESCA | 0.040248748 |
| TLR5 | LGG | 0.014075805 |
| TLR5 | OV | 0.036628212 |
| TLR5 | SKCM | 0.022197327 |
| TLR5 | STAD | 0.008021708 |
| TLR5 | THYM | 0.005545537 |
| Gene | Cancer type | p value |
|---|---|---|
| TLR6 | BLCA | 0.036456726 |
| TLR6 | ESCA | 0.01912187 |
| TLR6 | KIRP | 0.008085447 |
| TLR6 | LGG | 0.003399361 |
| TLR6 | SKCM | 0.003736075 |
| TLR7 | DLBC | 0.032371891 |
| TLR7 | LAML | 0.01921935 |
| TLR7 | LGG | 0.00593611 |
| TLR7 | LUAD | 0.000486804 |
| TLR7 | SARC | 0.016297397 |
| TLR7 | SKCM | 0.001124964 |
| TLR7 | UVM | 0.03400155 |
| TLR8 | LAML | 0.032090719 |
| TLR8 | LGG | 0.003214408 |
| TLR8 | SKCM | 1.26E-06 |
| TLR8 | THYM | 0.020533224 |
| TLR8 | UVM | 0.004248326 |
| TLR9 | KIRC | 0.020215421 |
| TLR9 | LAML | 0.038214202 |
| TLR9 | UCEC | 2.97E-05 |
| TLR10 | CESC | 0.02579765 |
| TLR10 | COAD | 0.041270017 |
| TLR10 | HNSC | 0.013737092 |
| TLR10 | LGG | 0.003769908 |
| TLR10 | LUAD | 0.000369848 |
| TLR10 | READ | 0.028148853 |
| TLR10 | SARC | 0.000659968 |
| TLR10 | SKCM | 1.31E-05 |
| TLR10 | UCEC | 0.003987076 |
TLR1
TLR2
TLR3
TLR4
TLR5
TLR6
TLR7
TLR8
TLR9
TLR10
ACC
BLCA
BRCA
CESC
CHOL
COAD
DLBC
ESCA
GBM
HNSC
KICH
KIRC
KIRP
LAML
LGG
LIHC
LUAD
LUSC
MESO
OV
PAAD
PCPG
PRAD
READ
SARC
SKCM
STAD
TGCT
THCA
THYM
UCEC
UCS
UVM
0.01 0.1
1
10
100
0.1
1
10
100 0.001
0.1
1
10 100 0.1
1
10
100 1000 0.01 0.1
1 10
1001e-040.01 1 10
1000 0.001 0.1
10
100 0.001
0.1
10
1001e-040.01 1 10
1000 le-040.01 1 10 1000
Hazard Ratio
(a)
TLR1 ***
TLR2 ***
TLR3 ***
TLR4 ***
TLR5 ***
TLR6 ***
TLR7 ***
TLR8 ***
TLR9 ***
TLR10 ***
6
Gene expression
4
2
0
C1C2C3C4C5C6 C1C2C3C4C5C6 C1C2C3C4C5C6 C1C2C3C4C5C6 C1C2C3C4C5C6 C1C2C3C4C5C6 C1C2C3C4C5C6 C1C2C3C4C5C6 C1C2C3C4C5C6 C1C2C3C4C5C6
Immune subtype
| Immune subtype | |
| C1 | C4 |
| C2 | C5 |
| C3 | C6 |
(b)
FIGURE 4: Cox regression and immune subtype analysis in pan-cancer. (a) Univariate Cox regression for each TLR gene in pan-cancer. (b) Differential expression of TLRs in differential immune subtype.
BioMed Research International
15
Immune score
UVM
1
UCEC
UCS
THYM
THCA
READ
STAD
TGCT
SKCM
SARC
PCPG
PRAD
OV
LUSC
MESO
LUAD
LIHC
PAAD
LAML
KIRC
GBM
KIRP
LGG
KICH
HNSC
COAD
ESCA
CHOL
DLBC
CESC
BRCA
ACC
BLCA
TLR1
☒
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
·
0.8
TLR2
☒
☒
☐
☒
☐
☒
☐
☐
☒
☐
☐
☐
☐
☒
☐
☐
☒
☐
0.6
TLR3
·
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
.
O
☐
☐
☐
☐
☐
.
☐
☐
☐
☐
☐
☐
☐
☐
O
0.4
.
.
.
.
.
TLR4
☒
☐
☒
☐
☐
☐
☐
O
.
0.2
O
.
O
TLR5
.
☒
.
☒
O
☐
☐
☐
.
0
.
.
.
.
.
.
TLR6
.
·
☐
e
☐
☐
☐
TLR7
☒
☐
☐
☐
☐
☐
☐
☐
-0.2
·
☐
☐
☐
☐
TLR8
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
0.4
-0.6
TLR9
.
.
.
C
C
.
O
.
.
·
.
.
.
☐
O
☐
.
☐
.
.
-0.8
.
.
.
.
TLR10
☐
☐
☐
-1
(a)
Stromal score
1
UVM
UCEC
UCS
THYM
THCA
TGCT
READ
STAD
SKCM
SARC
PRAD
PCPG
LUSC
PAAD
LUAD
OV
MESO
LIHC
LAML
LGG
KIRC
GBM
KIRP
KICH
HNSC
DLBC
ESCA
COAD
CHOL
CESC
ACC
BLCA
BRCA
TLR1
☒
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
·
0.8
O
TLR2
☒
☐
☐
O
☒
☒
☐
☒
·
☒
☐
☐
☒
☐
☐
☒
☐
☐
☐
☒
☐
☐
0.6
TLR3
.
·
☐
☐
.
☐
☐
.
.
.
☐
☐
☐
☐
☐
0
☐
☐
☐
·
☐
☐
☐
·
.
O
.
TLR4
☒
☐
☒
☒
☐
☒
O
O
☐
☐
☐
☒
☒
☐
☒
.
☒
☐
☐
☐
☒
0.4
0.2
TLR5
O
.
☒
O
☐
☐
☐
☐
☐
☐
.
☐
O
·
.
O
☐
☐
☐
☐
6
·
0
.
.
TLR6
☐
☐
.
O
☐
☐
.
☐
☐
☐
☐
☐
☐
☐
0
☐
C
-0.2
TLR7
☒
☐
☐
☐
.
☐
☐
☐
☐
☐
☒
☐
☐
☐
☐
☐
☐
·
☐
☐
TLR8
☒
☐
☒
☐
☒
☐
☒
☐
.
☐
☐
☐
0.4
-0.6
TLR9
·
0
☐
☐
.
O
.
☐
☐
O
·
☐
☐
·
0
☐
O
☐
☐
O
☐
O
☐
O
·
O
☐
0
·
.
O
TLR10
☒
☐
☐
☐
☐
☐
☐
☐
☐
O
☐
☐
-0.8
O
-1
.
O
(b)
RNAss
1
UVM
UCEC
UCS
THYM
THCA
STAD
SKCM
TGCT
READ
SARC
PRAD
OV
PCPG
PAAD
MESO
LIHC
LUAD
LUSC
LAML
LGG
GBM
HNSC
KIRC
KIRP
COAD
KICH
ESCA
CHOL
DLBC
CESC
ACC
BLCA
BRCA
TLR1
O
☒
☐
O
☐
☒
☒
0
☒
☐
☒
☐
☐
☒
☒
☒
☐
☐
3.
☐
U
0.8
O
-
O
O
TLR2
☒
A
☐
☐
C
C
☐
☐
☐
☐
O
·
☐
☐
☐
.
☐
☐
☐
.
☐
☐
☐
☐
☐
☐
-
0.6
TLR3
.
☐
☐
.
☐
.
☐
C
☐
O
☐
.
.
.
☐
.
☒
☐
☐
☐
0
.
☒
☐
O
☐
☒
.
.
.
0.4
0
O
TLR4
☐
☐
☒
☐
☐
☐
.
.
☐
☐
☒
☐
☒
.
☐
☐
☐
☐
☐
☐
☐
·
O
TLR5
☐
.
.
☐
☒
☒
.
☐
☐
☒
☒
☐
☐
☐
☐
☐
☐
☐
☐
☒
☐
☐
.
☐
0.2
0
TLR6
.
☐
0
.
☐
☒
☐
☐
☒
☒
☐
☒
☐
.
☐
☐
☐
☐
☐
0
·
☐
☐
☒
O
☐
☐
-0.2
.
·
.
TLR7
☐
☐
☐
☐
☐
☐
☐
☐
0
.
☐
☐
☐
☐
☐
☐
☐
.
☐
.
☐
.
☐
☐
-0.4
.
TLR8
☐
.
O
☐
.
☐
☐
☐
☒
·
☐
☐
☒
☐
☐
☐
.
☐
☐
☐
☐
O
☐
☐
☐
☐
.
☐
☐
-0.6
O
TLR9
C
☐
C
☐
.
☐
·
☐
☐
☐
☐
.
☐
☐
☐
.
☐
☐
O
☐
☐
☐
.
☐
.
☐
☐
.
.
☐
-0.8
O
TLR10
O
☐
☐
O
·
C
O
☐
·
☐
☐
☐
O
☐
☐
☐
·
☐
☐
☐
☐
☐
☐
☐
☐
☐
☐
-1
.
(c)
DNAss
1
UVM
UCS
UCEC
THCA
THYM
SKCM
READ
TGCT
SARC
STAD
PCPG
PRAD
PAAD
OV
LUAD
LUSC
MESO
LIHC
LAML
GBM
LGG
KIRC
KICH
CHOL
HNSC
KIRP
ESCA
DLBC
CESC
COAD
BRCA
BLCA
ACC
TLR1
0
☒
O
.
☐
.
☐
O
☒
.
.
.
☐
☒
☒
·
.
.
·
·
.
O
.
0.8
TLR2
·
☒
·
☐
.
☐
☐
☐
1.
☐
O
.
☐
·
☐
O
.
☐
☐
0
☐
O
0
.
☐
0
☒
.
0
0.6
.
TLR3
·
☐
.
☐
☐
C
.
.
☒
☐
☐
☐
☐
C
☐
☒
.
O
O
0.4
TLR4
☒
☒
·
·
.
.
☐
.
.
·
.
.
.
O
☐
·
.
.
.
☒
O
.
0.2
TLR5
.
14
.
.
.
☒
.
.
.
O
☒
·
·
.
0
O
☒
☐
☐
C
.
.
0
.
TLR6
.
·
.
·
.
.
.
.
.
.
.
.
☒
.
C
.
.
.
☐
.
.
-0.2
TLR7
·
.
.
.
.
·
·
·
.
.
.
.
.
.
.
.
·
.
.
-0.4
TLR8
.
☒
·
.
.
.
·
.
.
.
.
.
.
.
·
.
.
.
.
.
-0.6
A
TLR9
.
.
.
.
·
.
0
.
·
.
.
.
·
.
.
.
.
.
.
.
·
.
.
0
-0.8
TLR10
O
☒
C
.
.
.
.
.
.
0
.
.
.
.
.
.
.
.
.
.
-1
(d)
FIGURE 5: Continued.
TLR9, Fluphenazine
TLR9, Alectinib
TLR9, Carmustine
TLR7, Alectinib
TLR9, 7-Hydroxystaurosporine Cor = 0.550, p < 0.001
Cor = 0.680, p < 0.001
Cor = 0.637, p <0.001
Cor = 0.598, p < 0.001
Cor = 0.595, p < 0.001
6
3
2
4
4
2
4
1
2
1
2
0
2
0
0
0
0
-1
0
1
2
3
4
-2
0
1
2
3
4
0
1
2
3
4
0.0
0.2
0.4
0.6
0
1
2
3
4
TLR9, Etoposide Cor = 0.548, p < 0.001
TLR9, Imexon
TLR7, Denileukin diftitox ontak
TLR9, Irofulven Cor = - 0.539, p < 0.001
TLR9, Dimethylaminoparthenolide Cor = 0.537, p < 0.001
Cor = 0.545, p < 0.001
Cor = 0.541, p < 0.001
3
2
6
2
3
1
2
4
0
2
1
0
1
-2
0
2
0
-1
-2
-1
0
-4
-1
0
1
2
3
4
0
1
2
3
4
0.0
0.2
0.4
0.6
0
1
2
3
4
0
1
2
3
4
TLR9, Hydroxyurea Cor = 0.535, p < 0.001
TLR9, Nelarabine Cor = 0.535, p < 0.001
TLR9, Ifosfamide
TLR7, Irofulven
TLR9, LDK-378 Cor = 0.527, p < 0.001
Cor = 0.534, p < 0.001
Cor = - 0.528, p < 0.001
3
2
6
2
2
0
4
1
4
1
2
-2
2
0
0
-1
0
-1
-4
0
0
1
2
3
4
0
1
2
3
4
0
1
2
3
4
0.0
0.2
0.4
0.6
0
1
2
3
4
TLR9, Estramustine
TLR9, Chlorambucil
TLR9, Pipobroman Cor = 0.512, p < 0.001
TLR9, Cyclophosphamide Cor = 0.511, p < 0.001
TLR7, Fluphenazine Cor = 0.506, p < 0.001
Cor = 0.515, p < 0.001
Cor = 0.515, p < 0.001
4
2
2
3
6
2
1
1
2
4
0
0
1
0
2
-1
-1
0
0
0
1
2
3
4
0
1
2
3
4
0
1
2
3
4
0
1
2
3
4
0.0
0.2
0.4
0.6
TLR9, XK-469
TLR7, Isotretinoin
TLR9, Lomustine
TLR9, Melphalan
TLR9, Denileukin Diftitox Ontak Cor = 0.475, p < 0.001
Cor = 0.497, p < 0.001
Cor = 0.495, p < 0.001
Cor = 0.477, p < 0.001
Cor = 0.477, p < 0.001
2
4
2
6
3
4
1
1
4
2
2
0
0
2
1
0
0
-1
-1
0
-1
0
1
2
3
4
0.0
0.2
0.4
0.6
0
1
2
3
4
0
1
2
3
4
0
1
2
3
4
(e)
In our study, TLR2 was highly expressed in most cancer types. This result is similar to that of most previous studies [29-31]. Gergen et al. [32] reported that TLR2 activation induces the proliferation of lung adenocarcinoma cells by activating NF-KB. As a special link between lung cancer cells and mesenchymal stem cells in the TME, TLR2 promotes crosstalk and ultimately promotes changes in the tumor- supporting phenotype of mesenchymal cells [33]. Further- more, the expression of TLR2 protein was shown to be upregulated in colon cancer and significantly correlated with a low overall survival rate of patients with colon cancer [34, 35]. Thus, the TLR2 signaling pathway may be an important potential therapeutic target in cancer.
In our study, we found that TLR9 was hardly expressed compared to the other TLR genes, which led to TLR9 show- ing less correlation with both immune and stromal scores. However, several studies have reported that TLR9 is associ- ated with the development of cancers, especially gynecologic cancer [36, 37]. The activation of TLR9 on DCs and plasma- cytoid DCs promotes the secretion of a large amount of type I IFN, which has both direct (tumor cell inhibitory effect) and indirect (antitumor immune responses) effects on can- cer cells and is most evident in the early stages of antitumor immune responses [38].
Thorsson et al. [39] identified the immune landscape of cancer in the C1-C6 immune subtypes. In our study, we clas- sified tumor samples by representative immune signatures
and detected the RNA-seq levels of TLR 1-10 in C1 to C6. Interestingly, all TLRs were differentially expressed in differ- ent immune subtype samples. The TME, including the extra- cellular matrix, tumor vascular system, and tumor cell types, is closely related to immune functions and has an important impact on treatment response and clinical prognosis [40]. TLRs are expressed in the TME [41]. We further confirmed this information by extracting data on the fractions of stro- mal and immune cells in tumor samples from the 33 TCGA cancer types by calculating stromal scores, immune scores, and ESTIMATE scores. TLR expression was positively cor- related with immune and stromal scores in almost all cancer types. On one hand, TLRs are expressed during pro- grammed cell death induced by TME; on the other hand, they trigger the release of cytokines and chemokines in the TME and recruit immune cells to further release proinflam- matory cytokines, angiogenic factors, and growth factors, such as TGF ß, IL-8, CXCR4, ICAM-1, and VEGF. TLRs can repair the antitumor function and apoptotic response of antigen-presenting cells and effector T-cells [42, 43]. TLR signaling pathways play an essential role in controlling tumor progression, metastasis, recurrence, and chemother- apy tolerance through inappropriate immune enhancement and antitumor immunity [44].
Stemness was used to distinguish the stem cell-like char- acteristics of the tumor, such as self-renewal and dedifferen- tiation [45]. Two types of stemness indices were assessed:
Cancer: KIRC
TLR1
TLR2 **
TLR3*
TLR4 ***
TLR5
TLR6
TLR7
TLR8
TLR9
TLR10*
6
4
Gene expression
2
0
Stage I
Stage II
Stage III
Stage IV
Stage I
Stage II
Stage III
Stage IV
Stage I
Stage II
Stage III
Stage IV
Stage I
Stage II
Stage III
Stage IV
Stage I
Stage II
Stage III
Stage IV
Stage I
Stage II
Stage III
Stage IV
Stage I
Stage II
Stage III
Stage IV
Stage I
Stage II
Stage III
Stage IV
Stage I
Stage II
Stage III
Stage IV
Stage I
Stage II
Stage III
Stage IV
Stage
Stage
Stage I
Stage III
Stage II
Stage IV
(a)
Cancer: KIRC
TLR1 ***
TLR2
TLR3 ***
TLR4 ***
TLR5
TLR6
TLR7 ***
TLR8 ***
TLR9
TLR10 ***
6
4
Gene expression
2
0
C1C2C3C4C5C6
C1C2C3C4C5C6
C1C2C3C4C5C6
C1C2C3C4C5C6
C1C2C3C4C5C6
C1C2C3C4C5C6
C1C2C3C4C5C6
C1C2C3C4C5C6
C1C2C3C4C5C6
C1C2C3C4C5C6
Immune subtype
Immune subtype
C1
C4
C2
C5
C3
C6
(b)
DNAss
StromalScore
Immune score
ESTIMATEScore
Gene expression
(c)
| Cancer: KIRC | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| TLR1 | TLR2 | TLR3 | TLR4 | TLR5 | TLR6 | TLR7 | TLR8 | TLR9 | TLR10 |
| RNAss | |||||||||
FIGURE 6: TLRs in KIRC. (a) Differential expression of TLRs between stage I and stage IV in KIRC. (b) Differential expression of TLRs in different immune subtype in KIRC. (c) Correlation between the expression of TLRs and stemness indices, tumor microenvironment.
DNAss and RNAss [46]. We found that the expression of TLRs was negatively correlated with RNAss in nearly all types of cancers, except KIRC, and negatively correlated with DNAss in many types of cancers, except ACC, CHOL, KIRC, LAML, LGG, TGCT, THCA, THYM, and UVM. TLR3 activation facilitates the expression of stemness- associated genes, including OCT3/4, NANOG, and SOX2 [47]. TLR4 expression in HCC is associated with increased stem-like properties [48]. NF -? B, activated by TLR signaling, was closely aligned with proliferation, invasion, and tumor- igenesis [49].
Our study also found that the transcriptional expression levels of TLR7 and TLR9 were associated with drug response. Among them, the expression of TLR9 had a signif- icant positive correlation with drug sensitivity to fluphen- azine, alectinib, carmustine, and 7-hydroxystaurosporine. There was a significant positive correlation between TLR7 and the drug sensitivity of alectinib. These results have clin- ical relevance for guiding selection of antitumor therapies.
Finally, we explored the relationship between TLRs and KIRC. TLR2, TLR3, TLR4, and TLR10 were significantly dif- ferentially expressed between stages I and IV. TLR1, TLR3, TLR4, TLR7, TLR8, and TLR10 were significantly differen- tially expressed between C1 and C6 immune subtypes. All TLRs were positively correlated with immune, stromal, and ESTIMATE scores. Morikawa et al. [50] reported that TLR3 was overexpressed in KIRC, suggesting that the
TLR3 pathway may be a novel therapeutic target in KIRC. Moreover, the expression of TLR9 is an independent prog- nostic marker of KIRC, and the loss of TLR9 expression is related to poor prognosis of KIRC [51]. Our results provide guidance for further exploration of the role of TLRs in KIRC.
Although this is the first study to multidimensionally analyze TLRs across multiple cancer types, it has some lim- itations. First, our results have not been verified using other independent databases; thus, it is necessary to validate the conclusions by generating our own data and using other public databases in the future. Second, this was a dry lab study [52], and we have not explored the underlying mech- anisms behind the bioinformatics analyses through molecu- lar and animal experiments. Finally, we studied the relationship between TLR family members and various com- binatorial data. However, biometric correlations may not clarify the mechanisms of interaction and regulation directly; thus, further studies are needed to verify these potential mechanisms via laboratory-based molecular exper- iments. Further investigation is needed to determine the potential of TLRs and their coactivators as therapeutic tar- gets in cancer.
5. Conclusions
TLRs were expressed differently in different cancer types and different immune subtype tissue and were positively
correlated with immune-estimate scores and stromal- estimate scores. The expression of TLR9 had a significant positive correlation with the drug sensitivities to fluphen- azine, alectinib, carmustine, and 7-hydroxystaurosporine. TLR7 had a significant positive correlation with alectinib sensitivity. We demonstrated the significant pan-cancer role of the TLR family and potential therapeutic strategies for cancer. However, further laboratory studies are required to confirm our results.
Abbreviations
| TLRs: | Toll-like receptors |
| RNAss: | mRNA expression-based stemness scores |
| KIRC: | Kidney renal clear cell carcinoma |
| DNAss: | DNA methylation-based stemness scores |
| ACC: | Adrenocortical carcinoma |
| HOL: | Cholangiocarcinoma |
| LAML: | Acute myeloid leukemia |
| LGG: | Low-grade glioma |
| TGCT: | Testicular germ cell tumors |
| THCA: | Thyroid carcinoma |
| THYM: | Thymoma |
| UVM: | Uveal melanoma |
| PAMPs: | Pathogen-associated molecular patterns |
| ssRNA: | Single-stranded RNA |
| LPS: | Lipopolysaccharide |
| DCs: | Dendritic cells |
| TCGA: | The Cancer Genome Atlas |
| HR: | Hazard ration |
| OCLR: | One-class logistic regression. |
Data Availability
Source data of this study were derived from the public repos- itories, as indicated in the section of “Materials and Methods” of the manuscript. And all data that support the findings of this study are available from the corresponding author upon reasonable request.
Ethical Approval
This study was not applicable for ethical approval, and source data of this study were derived from the public repositories.
Disclosure
A preprint has previously been published [53].
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Authors’ Contributions
PW and LC contributed to the design, analysis and interpre- tation of data, drafting of the manuscript, and critical revi- sion of the manuscript; PW and HSY carried out the statistical analysis; PW, XY, HSY, and LC carried out the
methodology; LC performed the project administration; PW and XY performed the writing (original draft); PW, HSY, XY, and LC performed the writing (review and editing).
Acknowledgments
We thank all the R programming package developer.
Supplementary Materials
Supplementary Table. Detailed information about the drug sensitivity of TLRs.13. (Supplementary Materials)
References
[1] J. J. Brennan and T. D. Gilmore, “Evolutionary origins of Toll- like receptor signaling,” Molecular Biology and Evolution, vol. 35, no. 7, pp. 1576-1587, 2018.
[2] K. Newton and V. M. Dixit, “Signaling in innate immunity and inflammation,” Cold Spring Harbor Perspectives in Biology, vol. 4, no. 3, article a006049, 2012.
[3] S. Akira, “Pathogen recognition by innate immunity and its signaling,” Proceedings of the Japan Academy. Series B, Physi- cal and Biological Sciences, vol. 85, no. 4, pp. 143-156, 2009.
[4] Y. Li, X. Sun, X. Hu et al., “Scallop genome reveals molecular adaptations to semi-sessile life and neurotoxins,” Nature Com- munications, vol. 8, no. 1, p. 1721, 2017.
[5] S. A. Lindsay and S. A. Wasserman, “Conventional and non- conventional _Drosophila_ Toll signaling,” Developmental and Comparative Immunology, vol. 42, no. 1, pp. 16-24, 2014.
[6] T. Kawai and S. Akira, “The role of pattern-recognition recep- tors in innate immunity: update on Toll-like receptors,’ Nature Immunology, vol. 11, no. 5, pp. 373-384, 2010.
[7] B. Lemaitre, E. Nicolas, L. Michaut, J. M. Reichhart, and J. A. Hoffmann, “The dorsoventral regulatory gene cassette spät- zle/Toll/cactus controls the potent antifungal response in Dro- sophila adults,” Cell, vol. 86, no. 6, pp. 973-983, 1996.
[8] R. Medzhitov, P. Preston-Hurlburt, and C. A. Janeway Jr., “A human homologue of the _Drosophila_ Toll protein signals activation of adaptive immunity,” Nature, vol. 388, no. 6640, pp. 394-397, 1997.
[9] S. Akira, S. Uematsu, and O. Takeuchi, “Pathogen recognition and innate immunity,” Cell, vol. 124, no. 4, pp. 783-801, 2006.
[10] A. L. Blasius and B. Beutler, “Intracellular Toll-like receptors,” Immunity, vol. 32, no. 3, pp. 305-315, 2010.
[11] A. F. McGettrick and L. A. O’Neill, “Localisation and traffick- ing of Toll-like receptors: an important mode of regulation,” Current Opinion in Immunology, vol. 22, no. 1, pp. 20-27, 2010.
[12] S. Rakoff-Nahoum and R. Medzhitov, “Toll-like receptors and cancer,” Nature Reviews. Cancer, vol. 9, no. 1, pp. 57-63, 2009.
[13] J. P. Pradere, D. H. Dapito, and R. F. Schwabe, “The Yin and Yang of Toll-like receptors in cancer,” Oncogene, vol. 33, no. 27, pp. 3485-3495, 2014.
[14] E. Y. So and T. Ouchi, “The application of Toll like receptors for cancer therapy,” International Journal of Biological Sci- ences, vol. 6, no. 7, pp. 675-681, 2010.
[15] C. C. Lu, H. C. Kuo, F. S. Wang, M. H. Jou, K. C. Lee, and J. H. Chuang, “Upregulation of TLRs and IL-6 as a marker in
human colorectal cancer,” International Journal of Molecular Sciences, vol. 16, pp. 159-177, 2015.
[16] S. Zhao, Y. Zhang, Q. Zhang, F. Wang, and D. Zhang, “Toll- like receptors and prostate cancer,” Frontiers in Immunology, vol. 5, p. 352, 2014.
[17] A. A. Khan, Z. Khan, and S. Warnakulasuriya, “Cancer-associ- ated toll-like receptor modulation and insinuation in infection susceptibility: association or coincidence?,” Annals of Oncol- ogy, vol. 27, no. 6, pp. 984-997, 2016.
[18] N. Javaid and S. Choi, “Toll-like receptors from the perspective of cancer treatment,” Cancers, vol. 12, no. 2, p. 297, 2020.
[19] K. Yoshihara, M. Shahmoradgoli, E. Martínez et al., “Inferring tumour purity and stromal and immune cell admixture from expression data,” Nature Communications, vol. 4, no. 1, p. 2612, 2013.
[20] T. M. Malta, A. Sokolov, A. J. Gentles et al., “Machine learning identifies stemness features associated with oncogenic dedif- ferentiation,” Cell, vol. 173, no. 2, pp. 338-354.e15, 2018.
[21] G. P. Dunn, A. T. Bruce, and H. Ikeda, “Cancer immunoedit- ing: from immunosurveillance to tumor escape,” Nature Immunology, vol. 3, no. 11, pp. 991-998, 2002.
[22] D. Pardoll, “Does the immune system see tumors as foreign or self?,” Annual Review of Immunology, vol. 21, no. 1, pp. 807- 839, 2003.
[23] P. Behzadi, “The role of Toll-like receptor (TLR) polymor- phisms in urinary bladder cancer,” in Genetic Polymorphism and cancer susceptibility, pp. 281-317, Springer, Singapore, 2021.
[24] A. S. Sameer and S. Nissar, “Toll-like receptors (TLRs): struc- ture, functions, signaling, and role of their polymorphisms in colorectal cancer susceptibility,” BioMed Research Interna- tional, vol. 2021, Article ID 1157023, 14 pages, 2021.
[25] P. Behzadi, H. A. García-Perdomo, and T. M. Karpiński, “Toll- like receptors: general molecular and structural biology,” Jour- nal of Immunology Research, vol. 2021, Article ID 9914854, 21 pages, 2021.
[26] A. Iwasaki and R. Medzhitov, “Toll-like receptor control of the adaptive immune responses,” Nature Immunology, vol. 5, no. 10, pp. 987-995, 2004.
[27] J. E. Boudreau, A. Bonehill, K. Thielemans, and Y. Wan, “Engi- neering dendritic cells to enhance cancer immunotherapy,” Molecular Therapy, vol. 19, no. 5, pp. 841-853, 2011.
[28] M. Fukata and M. T. Abreu, “Role of toll-like receptors in gas- trointestinal malignancies,” Oncogene, vol. 27, no. 2, pp. 234- 243, 2008.
[29] M. A. Lanki, H. E. Seppänen, H. K. Mustonen et al., “Toll-like receptor 2 and Toll-like receptor 4 predict favorable prognosis in local pancreatic cancer,” Tumour Biology, vol. 40, no. 9, p. 1010428318801188, 2018.
[30] Y. D. Liu, L. Yu, L. Ying et al., “Toll-like receptor 2 regulates metabolic reprogramming in gastric cancer via superoxide dis- mutase 2,” International Journal of Cancer, vol. 144, no. 12, pp. 3056-3069, 2019.
[31] Y. D. Liu, C. B. Ji, S. B. Li et al., “Toll-like receptor 2 stimula- tion promotes colorectal cancer cell growth via PI3K/Akt and NF-KB signaling pathways,” International Immunophar- macology, vol. 59, pp. 375-383, 2018.
[32] A. K. Gergen, P. D. Kohtz, A. L. Halpern et al., “Activation of toll-like receptor 2 promotes proliferation of human lung ade- nocarcinoma cells,” Anticancer Research, vol. 40, no. 10, pp. 5361-5369, 2020.
[33] X. Li, S. Wang, R. Zhu, H. Li, Q. Han, and R. C. Zhao, “Lung tumor exosomes induce a pro-inflammatory pheno- type in mesenchymal stem cells via NFKB-TLR signaling pathway,” Journal of Hematology & Oncology, vol. 9, no. 1, p. 42, 2016.
[34] K. Paarnio, S. Väyrynen, K. Klintrup et al., “Divergent expres- sion of bacterial wall sensing Toll-like receptors 2 and 4 in colorectal cancer,” World Journal of Gastroenterology, vol. 23, no. 26, pp. 4831-4838, 2017.
[35] F. A. Scheeren, A. H. Kuo, L. J. van Weele et al., “A cellintrinsic role for TLR2- MYD88 in intestinal and breast epithelia and oncogenesis,” Nature Cell Biology, vol. 16, no. 12, pp. 1238- 1248, 2014.
[36] S. Tian, L. Zhang, T. Yang et al., “The associations between toll-like receptor 9 gene polymorphisms and cervical cancer susceptibility,” Mediators of Inflammation, vol. 2018, Article ID 9127146, 11 pages, 2018.
[37] E. Fehri, E. Ennaifer, R. Bel Haj Rhouma, L. Guizani-Tabbane, I. Guizani, and S. Boubaker, “The role of toll-like receptor 9 in gynecologic cancer,” Current Research in Translational Medi- cine, vol. 64, no. 3, pp. 155-159, 2016.
[38] L. Karapetyan, J. J. Luke, and D. Davar, “Toll-like receptor 9 agonists in cancer,” Oncotargets and Therapy, vol. 13, pp. 10039-10061, 2020.
[39] V. Thorsson, D. L. Gibbs, S. D. Brown et al., “The immune landscape of cancer,” Immunity, vol. 48, no. 4, pp. 812- 830.e14, 2018.
[40] M. R. Junttila and F. J. de Sauvage, “Influence of tumour micro-environment heterogeneity on therapeutic response,” Nature, vol. 501, no. 7467, article BFnature12626, pp. 346- 354, 2013.
[41] S. Gonzalez-Reyes, L. Marín, L. González et al., “Study of TLR3, TLR4 and TLR9 in breast carcinomas and their associ- ation with metastasis,” BMC Cancer, vol. 10, no. 1, p. 665, 2010.
[42] S. D. Killeen, J. H. Wang, E. J. Andrews, and H. P. Redmond, “Exploitation of the Toll-like receptor system in cancer: a doubled-edged sword?,” British Journal of Cancer, vol. 95, no. 3, pp. 247-252, 2006.
[43] M. Zhou, M. M. McFarland-Mancini, H. M. Funk, N. Husseinzadeh, T. Mounajjed, and A. F. Drew, “Toll-like receptor expression in normal ovary and ovarian tumors,” Cancer Immunology, Immunotherapy, vol. 58, no. 9, pp. 1375-1385, 2009.
[44] D. J. Connolly and L. A. O’Neill, “New developments in Toll- like receptor targeted therapeutics,” Current Opinion in Phar- macology, vol. 12, no. 4, pp. 510-518, 2012.
[45] D. Friedmann-Morvinski and I. M. Verma, “Dedifferentiation and reprogramming: origins of cancer stem cells,” EMBO Reports, vol. 15, no. 3, pp. 244-253, 2014.
[46] R. Huang, J. Zhang, M. Li et al., “The role of peroxisome proliferator-activated receptors (PPARs) in pan-cancer,” PPAR Research, vol. 2020, Article ID 6527564, 19 pages, 2020.
[47] D. Jia, W. Yang, L. Li et al., “ß-Catenin and NF-KB co- activation triggered by TLR3 stimulation facilitates stem cell- like phenotypes in breast cancer,” Cell Death and Differentia- tion, vol. 22, no. 2, pp. 298-310, 2015.
[48] W. T. Liu, Y. Y. Jing, G. F. Yu et al., “Toll like receptor 4 facil- itates invasion and migration as a cancer stem cell marker in hepatocellular carcinoma,” Cancer Letters, vol. 358, no. 2, pp. 136-143, 2015.
[49] D. W. Yeh, L. R. Huang, Y. W. Chen, C. F. Huang, and T. H. Chuang, “Interplay between inflammation and stemness in cancer cells: the role of Toll-like receptor signaling,” Journal of Immunology Research, vol. 2016, Article ID 4368101, 14 pages, 2016.
[50] T. Morikawa, A. Sugiyama, H. Kume et al., “Identification of Toll-like receptor 3 as a potential therapeutic target in clear cell renal cell carcinoma,” Clinical Cancer Research, vol. 13, no. 19, pp. 5703-5709, 2007.
[51] H. Ronkainen, P. Hirvikoski, S. Kauppila et al., “Absent Toll- like receptor-9 expression predicts poor prognosis in renal cell carcinoma,” Journal of Experimental & Clinical Cancer Research, vol. 30, no. 1, p. 84, 2011.
[52] R. Ranjbar, P. Behzadi, A. Najafi, and R. Roudi, “DNA micro- array for rapid detection and identification of food and water borne bacteria: from dry to wet lab,” Open Microbiology Jour- nal, vol. 11, no. 1, pp. 330-338, 2017.
[53] W. Ping, S. Hong, Y. Xun, and C. Li, Comprehensive bioinfor- matics analysis of Toll-like receptors (TLRs) in pan-cancer, Research Square, 2021.