Abstract
The immunosuppressive microenvironment is associated with poor prognosis in papillary thyroid cancer (PTC); however, the molecular mechanisms involved are unknown. Among the triggering receptors expressed on myeloid cell (TREM) family, we found that TREM1 expression in PTC was significantly higher than that in normal tissues. TREM1 overexpression was associated with BRAFV600E profiles and advanced tumor stages. Furthermore, TREM1 mRNA expression was negatively correlated with promoter methylation status. Specifically, hypomethylation of CpG site cg06196379 in the TREM1 promoter was related with poor patient disease-free survival (DFS) and a high PTC recurrence rate. Mechanistically, TREM1 was mainly expressed in malignant epithelial cells but not in macrophages in PTC by single-cell analysis. PTC tissues with high TREM1 levels had enhanced infiltration of regulatory T cells (Tregs) and decreased infiltration of CD8+ T cells. Our study confirms that hypomethylation-mediated overexpression of TREM1 in PTC cells promotes an immunosuppressive microenvironment by enhancing Treg infiltration. We recommend the future use of therapeutic strategy targeting TREM1 for the treatment of PTC.
Introduction
Thyroid cancer is the most prevalent endocrine malignancy and comprises a variety of histotypes with different behaviors and distinct prognosis. There is a dramatically increasing trend in thyroid cancer prevalence worldwide, and almost all the rising data have been assigned to the increased incidence of papillary thyroid cancer (PTC) (Powers et al. 2019). PTC generally has a good prognosis if completely eradicated by surgery alone or in combination with adjuvant radioiodine therapy; yet 10% of patients develop invasive disease and 20% of the patients relapse at some point during their lifetime (Haugen et al. 2015). The recurrence risk of PTC can be evaluated by the 2015 American Thyroid Association Risk Stratification System (van Velsen et al. 2019). However, many prognostic variables in this system are obtained during the follow-up period; thus, additional factors for earlier prediction of disease recurrence are required.
The tumor immune microenvironment (TIM) is a complex milieu in which cancer cells substantially communicate with tumor-infiltrating immune cells (TIICs) to regulate tumor progression and metastasis (Binnewies et al. 2018). Several studies have revealed that the clinical outcomes of PTC are associated with immune cell infiltrations (Galdiero et al. 2016, Na & Choi 2018). Infiltration of suppressive immune cells, particularly myeloid lineages, is associated with a poor prognosis of PTC (Menicali et al. 2020). A high density of tumor-associated macrophages (TAMs), with an M2-TAM predominance in PTC, positively correlates with lymph node metastasis and decreased survival (Qing et al. 2012). Increased infiltration of immature dendritic cells (iDCs) with impaired antigen presentation ability was reported in PTC; however, its association with tumor prognosis is somewhat controversial (Bergdorf et al. 2019). Regarding lymphoid lineages, PTC patients with an advanced phenotype display decreased enrichment of CD8+ cytotoxic T cells, as well as enhanced infiltration of regulatory T cells (Tregs) that block antitumor immune responses by expressing immune checkpoint regulators (Gogali et al. 2012). Although great progress has been made in mapping the immune landscape of PTC, its molecular mechanisms that are critical for the development of immunotherapeutic strategies remain largely unknown.
Triggering receptors expressed on myeloid cells (TREMs) are immunoglobulin superfamily receptors expressed primarily by myeloid cells. Six transcriptional members of the family, including TREM1, TREM2, and TREM-like transcripts −1, −2, −3, and −4 (TREML1, TREML2, TREML3, and TREML4), have been identified in the human genome. The TREM family functions as an essential regulator of myeloid cell differentiation and thus participates in innate immune responses. Recent studies have established their contributions to various pathophysiological processes, including metabolic disease, autoimmune disease, and cancers (Tammaro et al. 2017, Deczkowska et al. 2020). Regarding cancer, emerging evidence suggests the involvement of TREMs in the inflammation-suppressive tumor microenvironment (Qinchuan et al. 2019, Cioni et al. 2020). However, their major role in thyroid carcinoma is unknown. In our study, TREM expression and its prognostic value in PTC were explored. TREM1 was demonstrated to be epigenetically overexpressed and associated with poor clinical outcomes in PTC. Moreover, its overexpression in tumor epithelial cells participated in PTC progression at least partially through regulating the infiltration of immune cells, specifically Tregs, in the PTC immune microenvironment.
Materials and methods
TREM expression and its clinical contribution
The Oncomine database was used for analyzing the mRNA expression of TREMs in human cancers. Primary filters for ‘thyroid gland carcinoma‘ and ‘cancer vs normal analysis’ were used. ‘Human Genome U133 Array platform’ was chosen as the screening condition in order to ensure the reliability of the screening results. The thresholds were set as a P -value of 0.001 and a fold change of 2. The levels of TREMs mRNA were further evaluated using Gene Expression Profiling Interactive Analysis (GEPIA) with a cut-off P -value of 0.01 in thyroid cancer tissues (n = 512) and normal thyroid tissues (n = 337).
The clinical information and mutation data of thyroid cancer (THCA) samples were retrieved from The Cancer Genome Atlas (TCGA); and the PTC-integrated genomic characterization of BRAFV600E-RAS score, ERK score, and differentiation score were downloaded from TCGA Research Network (Agrawal et al. 2014). The association of TREMs mRNA expression with clinicopathological characteristics was validated.
DNA methylation analysis
The relationship between the mRNA expression and DNA methylation levels of TREMs in PTC samples (n = 496) was assessed through the cBioPortal database. MEXPRESS (https://mexpress.be/) was used to integrate the expression, DNA methylation, and clinical TCGA data of TREMs in THCA.
To assess the prognostic value of the TREMs methylation status, we determined the association between TREMs methylation levels and recurrence events. The methylation status of specific CpG sites in the TREMs promoter, described as the β-value, was downloaded from the DNA methylation (HumanMethylation450) section of TCGA. The recurrence of PTC was defined as a new tumor event after initial treatment, including biochemical evidence of disease, distant metastasis, locoregional recurrence, and new primary tumor (Xing et al. 2015).
Single-cell transcriptomic analysis
To explore the heterogeneous expression of TREM1 in PTC, single-cell RNA-seq (scRNA-seq) data processing seven female and four male primary PTC samples (ID: GSE158291) were obtained from the Gene Expression Omnibus (GEO) (Peng et al. 2021). After filtering out low-quality cells, a total of 12,757 cells were included in the analyses. The scRNA-seq data were quality controlled and statistically analyzed by Seurat V3 (Stuart et al. 2019). Principal component analysis was performed to reduce dimensionality with the 2000 most variable genes. The differentially expressed genes (DEGs) were defined as those with a false discovery rate adjusted P -value <0.05 and log2 (fold change) >1. The t-distributed stochastic neighbor embedding (tSNE) algorithm was applied to classify and annotate different cell clusters according to the previously identified expression signatures.
Functional enrichment analysis of TREM1 in PTC
TCGA raw mRNA expression data of THCA (n = 512) were divided into the TREM1 high group and TREM1 low group according to the median value of TREM1 expression. DEGs were identified between those two groups, with a P -value <0.05 and absolute fold change ≥1.0. Then, the clusterProfiler package was used to annotate and enrich the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways for upregulated DEGs (Yu et al. 2012). Moreover, the main biological function of TREM1 in PTC was explored by Gene Set Enrichment Analysis (GSEA) software (version 4.1.0). The 19,580 genes from the gene expression RNAseq data of THCA samples in TCGA were analyzed by their relative TREM1 expression levels (TREM1 high group vs TREM1 low group).
Immune cell infiltration analysis of TREM1
Three different algorithms, namely, CIBERSORT, TISIDB, and ssGSEA, were used to explore the relationship between TREM1 expression and TIICs. The fractions of the TIICs in PTC and normal tissues were calculated by CIBERSORT (Vinayak et al. 2015). Relative immune cell fractions in different TREM1 mRNA expression statuses were also calculated by CIBERSORT. Moreover, the GSVA (gene set variation analysis) package of R was used to validate the relationship between TREM1 and Tregs. The association of TREM1 with lymphocytes and chemokines in THCA was depicted through the TISIDB database (http://cis.hku.hk/TISIDB/index.php).
Kaplan–Meier plotter
The Shiny Methylation Analysis Resource Tool (SMART) is a website for analyzing the DNA methylation of TCGA projects. Kaplan–Meier (KM) survival curves illustrating the correlation between TREM-associated CpG sites and patient disease-free survival (DFS) were generated using SMART (bioinfo-zs.com/smartapp); and the survival analysis based on TREM1 mRNA expression in the enriched Tregs and decreased Tregs groups was evaluated by the KM-plotter tool. The tools ‘auto-select best cut-off’ and ‘follow-up 240 months’ were used.
Ethics statement and patient samples
With approval of the Ethics Review Committee of the First Affiliated Hospital of Xi’an Jiaotong University, postsurgical PTC tissues of 52 untreated patients were collected. All pathological diagnoses were made according to World Health Organization criteria by two senior pathologists. Informed consent was obtained from all patients.
Quantitative RT-PCR
Total RNA extraction and quantitative RT-PCR (qRT-PCR) analyses were proceeded as per our previous work (Fu et al. 2013). The primer sequences were TREM1, 5’-GAACTCCGAGCTGCAACTAAA-3’ (sense) and 5’-TCTAGCGTGTAGTCACATTTCAC-3’ (antisense); GAPDH, 5’-ACTCCACTCACGGCAAATTC- 3’ (sense) and 5’-TCTCCATGGTGGTGAAGACA- 3’ (antisense).
Immunohistological analysis of thyroid sections and image quantification
The surgical resection specimens (n = 52) were paraffin-embedded. For immunohistochemistry (IHC), 52 sections were incubated with anti-TREM1 (Novus, MM0583-9T3, Centennial, CO, USA; 1:100), and 40 sections were stained with anti-p-ERK (Cell Signaling Technology, #4370, 1:200), anti-FOXP3 (Abcam, ab20034, 1:500), and anti-CD8α (Abcam, ab17147, 1:100) antibodies overnight at 4°C. For image analysis, TREM1 and other protein levels were measured by the immunoreactive score (IRS), calculated by the product of the percentage positive cells (4, >80%; 3, 51–80%; 2, 10–50%; 1, <10%; 0, no staining) and intensity of staining (3, strong; 2, moderate; 1, mild; 0, no staining) (Remmele & Stegner 1987). For clinicopathological analysis, TREM1 and p-ERK were categorized into a high-expression (≥median IRS) group or a low-expression (<median IRS) group (Werner et al. 2016, 2018). For immunofluorescence, all sections were incubated with anti-human TREM1 antibody (Novus, MM0583-9T3, 1:40). For double staining, specimens were stained with anti-human EPCAM antibody (Santa Cruz, sc-66020, 1:100) and anti-human KRT19 antibody (Santa Cruz, sc-6278, 1:100).
Statistical analysis
SPSS 24.0 was applied to statistical analyses. All values are presented as the mean ± s.d., and the threshold of significance was defined as a P -value <0.05. Statistical analysis of TREM1 expression levels between different groups was analyzed using t-tests or Mann–Whitney U tests. Clinical pathological correlations were evaluated using the χ2 test or Fisher’s exact test, if appropriate. Cox univariate proportional hazards regression models were used to analyze tumor biological factors and clinical parameters associated with patient survival.
Results
Elevated mRNA expression of TREM1 and TREM2 in thyroid cancer
To assess the role of TREMs in thyroid cancer, we utilized the Oncomine database to perform a meta-analysis of public transcriptomic profiling data. In Fig. 1A, eight analyses including 135 specimens with 24 normal tissues and 111 tumor tissues were enrolled, and a remarkable elevation of TREM1 and TREM2 in thyroid cancer tissues was observed. Detailed TREM1 and TREM2 mRNA expression data in thyroid cancer are shown in Supplementary Table 1 (see section on supplementary materials given at the end of this article). Consistently, the mRNA expression of both TREM1 and TREM2 was notably upregulated in thyroid cancer tissues (n = 512) compared to the controls (n = 337) by GEPIA (Fig. 1B). However, there were no significant changes in TREML1-4 mRNA expression in thyroid cancer.
Differential expression levels of TREMs in thyroid cancer and other human cancers. (A) TREMs expression in eight datasets of thyroid cancer compared with thyroid normal tissues in Oncomine. (B) GEPIA analysis of TREMs mRNA expression in thyroid cancer tissues (T, n = 512) compared to the non-cancerous (N, n = 337) tissues. (C) TREMs profiles among different cancers compared with normal tissues based on the Oncomine database. **P < 0.01.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Differential expression levels of TREMs in thyroid cancer and other human cancers. (A) TREMs expression in eight datasets of thyroid cancer compared with thyroid normal tissues in Oncomine. (B) GEPIA analysis of TREMs mRNA expression in thyroid cancer tissues (T, n = 512) compared to the non-cancerous (N, n = 337) tissues. (C) TREMs profiles among different cancers compared with normal tissues based on the Oncomine database. **P < 0.01.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Differential expression levels of TREMs in thyroid cancer and other human cancers. (A) TREMs expression in eight datasets of thyroid cancer compared with thyroid normal tissues in Oncomine. (B) GEPIA analysis of TREMs mRNA expression in thyroid cancer tissues (T, n = 512) compared to the non-cancerous (N, n = 337) tissues. (C) TREMs profiles among different cancers compared with normal tissues based on the Oncomine database. **P < 0.01.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
We further explored the role of TREMs in human carcinogenesis. TREM1 and TREM2 mRNA levels were significantly higher in head and neck cancers, as well as in some other cancer types including brain, colorectal, esophageal, gastric, breast, kidney, pancreatic, and prostate cancer, compared with levels in respective normal tissue, whereas they were decreased in lung cancer and leukemia. TREML1, TREML2, and TREML4 mRNA levels were mostly decreased in breast, kidney, lung, leukemia, and myeloma, and TREML3 mRNA expression was relatively high in melanoma (Fig. 1C). Detailed TREM1 and TREM2 mRNA expression data in different types of cancers are shown in Supplementary Tables 2 and 3.
Association of TREM1 and TREM2 expression with PTC clinical features
In deciphering the association between TREMs expression and PTC histotypes, we observed that TREM1 mRNA expression in classical PTC (n = 366) was dramatically upregulated compared with that in follicular variant PTC (n = 102, 6.52 ± 1.82 vs 4.66 ± 2.26, P < 0.001; Fig. 2A). Higher mRNA expression of TREM1 was also observed in tall cell PTC (n = 36, 7.45 ± 1.45 vs 4.66 ± 2.26, P < 0.001). In addition, a similar association was found for TREM2. As shown in Fig. 2B, TREM2 mRNA expression in classical PTC (n = 366, 7.35 ± 1.43, P < 0.001) and tall cell PTC (n = 36, 7.58 ± 1.16, P < 0.05) was significantly higher than that in follicular variant PTC (n = 102, 6.30 ± 1.90).
Relationship between TREM1 and TREM2 expressions and clinical features of PTC in TCGA. (A and B) The association between mRNA expression levels of TREM1 (A) and TREM2 (B) with histologic subtypes. (C and D) The expression of TREM1 (C) and TREM2 (D) in BRAFV600E-like PTCs (n = 272) compared to that in RAS-like PTCs (n = 119). (E) PTC (n = 391) were ranked by BRAFV600E-RAS score (BRS), with BRAFV600E -like and RAS-like samples having negative (−1 to 0) and positive scores (0 to 1), respectively. TREM1 and TREM2 mRNA were associated with BRAF-RAS score, ERK score, and differentiation score. (F and G) The correlation of mRNA expression levels of TREM1 (F) and TREM2 (G) with tumor stages. *P < 0.05, ***P < 0.001.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Relationship between TREM1 and TREM2 expressions and clinical features of PTC in TCGA. (A and B) The association between mRNA expression levels of TREM1 (A) and TREM2 (B) with histologic subtypes. (C and D) The expression of TREM1 (C) and TREM2 (D) in BRAFV600E-like PTCs (n = 272) compared to that in RAS-like PTCs (n = 119). (E) PTC (n = 391) were ranked by BRAFV600E-RAS score (BRS), with BRAFV600E -like and RAS-like samples having negative (−1 to 0) and positive scores (0 to 1), respectively. TREM1 and TREM2 mRNA were associated with BRAF-RAS score, ERK score, and differentiation score. (F and G) The correlation of mRNA expression levels of TREM1 (F) and TREM2 (G) with tumor stages. *P < 0.05, ***P < 0.001.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Relationship between TREM1 and TREM2 expressions and clinical features of PTC in TCGA. (A and B) The association between mRNA expression levels of TREM1 (A) and TREM2 (B) with histologic subtypes. (C and D) The expression of TREM1 (C) and TREM2 (D) in BRAFV600E-like PTCs (n = 272) compared to that in RAS-like PTCs (n = 119). (E) PTC (n = 391) were ranked by BRAFV600E-RAS score (BRS), with BRAFV600E -like and RAS-like samples having negative (−1 to 0) and positive scores (0 to 1), respectively. TREM1 and TREM2 mRNA were associated with BRAF-RAS score, ERK score, and differentiation score. (F and G) The correlation of mRNA expression levels of TREM1 (F) and TREM2 (G) with tumor stages. *P < 0.05, ***P < 0.001.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Since BRAFV600E-like and RAS-like PTCs have distinct signaling and differentiation characteristics (Agrawal et al. 2014), we next assessed the association of TREMs expression levels between the two distinct driver groups in PTC. The results showed that TREM1 and TREM2 mRNA expression in BRAFV600E-like PTCs (n = 272) was significantly higher than that in RAS-like PTCs (n = 119, 6.88 ± 1.60 vs 4.50 ± 2.25, P < 0.001 for TREM1; 7.59 ± 1.13 vs 6.04 ± 1.93, P < 0.001 for TREM2; Fig. 2C and D). Consistently, PTCs with higher expression of TREM1 and TREM2 had relatively lower BRAF-RAS score, higher ERK score, as well as lower differentiation score (Fig. 2E). To further determine the relationship between TREMs expression and the genetic background of BRAFV600E-like PTCs, we compared the expression levels of TREM1 and TREM2 in PTCs with BRAFV600E, BRAF fusions, and RET/PTCs. We did not observe any significant differences except for a higher TREM1 expression in RET/PTCs than that in BRAFV600E PTCs (Supplementary Fig. 1).
To further explore the clinical contribution of TREM1 and TREM2 in PTC, we analyzed the association between cancer stage and gene expression. A significantly higher mRNA expression of TREM1 in high stages (stages III–IV, n = 170) than that in low stages (stages I–II, n = 341) was observed (6.76 ± 1.98 vs 5.94 ± 2.06, P < 0.001; Fig. 2F). However, this association was not found for TREM2 (7.22 ± 1.47 vs 7.19 ± 1.61, P > 0.05; Fig. 2G). These data implied that TREM1 has a more important role in PTC progression than TREM2.
Hypomethylation of TREM1 has prognostic significance in PTC
DNA methylation is essential for the epigenetic regulation of gene expression. We thus used MEXPRESS to determine the role of DNA methylation in TREMs gene regulation. The anticorrelation between promoter methylation and mRNA expression was confirmed for both TREM1 (Fig. 3A) and TREM2 (Supplementary Fig. 2A) by Pearson correlation coefficients. Using cBioPortal, we also revealed a strong negative correlation between TREM1 methylation and its expression in thyroid cancer (R = −0.71, P < 0.001; Fig. 3B). Moreover, a similar but slightly weaker correlation was also observed between TREM2 mRNA expression and its methylation (R = −0.33, P < 0.001; Supplementary Fig. 2B).
Relationship of TREM1 methylation with its mRNA expression and clinical features of PTC in TCGA. (A) Association of TREM1 expression with DNA methylation levels on different CpG sites. (B) The relationship between TREM1 methylation and its mRNA expression. (C) Disease-free survival (DFS) curves of methylation levels of CpG site cg06196379 in TREM1 promoter. (D) Comparison of methylation levels of CpG site cg06196379 between recurrent (n = 35) and no-recurrent (n = 439) PTC patients. **P < 0.01.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Relationship of TREM1 methylation with its mRNA expression and clinical features of PTC in TCGA. (A) Association of TREM1 expression with DNA methylation levels on different CpG sites. (B) The relationship between TREM1 methylation and its mRNA expression. (C) Disease-free survival (DFS) curves of methylation levels of CpG site cg06196379 in TREM1 promoter. (D) Comparison of methylation levels of CpG site cg06196379 between recurrent (n = 35) and no-recurrent (n = 439) PTC patients. **P < 0.01.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Relationship of TREM1 methylation with its mRNA expression and clinical features of PTC in TCGA. (A) Association of TREM1 expression with DNA methylation levels on different CpG sites. (B) The relationship between TREM1 methylation and its mRNA expression. (C) Disease-free survival (DFS) curves of methylation levels of CpG site cg06196379 in TREM1 promoter. (D) Comparison of methylation levels of CpG site cg06196379 between recurrent (n = 35) and no-recurrent (n = 439) PTC patients. **P < 0.01.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Next, the prognostic prediction values of the methylation status of particular CpG sites in the TREM1 promoter (cg03843170, cg10981439, cg21328082, cg09310966, cg06196379, and cg18505453) and in the TREM2 promoter (cg20095587, cg01980222, cg10725937, and cg25748868) were assessed. Hypomethylation of CpG site cg06196379 in the TREM1 promoter was identified to be connected with poor patient DFS in PTC (hazard ratio = 0.27, P < 0.05, Fig. 3C). No significant associations of the methylation status of other CpG sites in the TREM1 promoter or the TREM2 promoter with patient survival were observed (Supplementary Figs 2C and 3). We also determined the relationship between PTC recurrence and the methylation status of CpG site cg06196379 in the TREM1 promoter. As shown in Fig. 3D, the methylation level of CpG site cg06196379 in the recurrence group (n = 35) was obviously lower than that in the no-recurrence group (n = 439, P < 0.01). Taken together, the above data indicate that hypomethylation of TREM1, instead of TREM2, participates in PTC progression.
Validation of TREM1 expression and its clinical significance in PTC
As shown in Fig. 4A, TREM1 mRNA expression was dramatically elevated in PTCs compared to levels in the noncancerous controls using qRT-PCR (8.24 ± 7.90 vs 1.42 ± 1.22, P < 0.001). TREM1 protein levels were then evaluated in 52 paired PTC and their adjacent normal tissues through IHC. An increased TREM1 level was observed in PTC compared to that in normal thyroid tissues (3.60 ± 2.50 vs 1.35 ± 1.03, P < 0.001; Fig. 4B and D). Furthermore, the TREM1 protein was upregulated in stages III–IV PTC (n = 9) than that in stages I–II PTC (n = 43) according to IHC staining (7.12 ± 2.43 vs 2.87 ± 1.79, P < 0.001; Fig. 4C and D).
IHC analysis of TREM1 expression and its correlation with tumor stages in PTC. (A) The mRNA expression levels of TREM1 in PTCs (T) compared to matched noncancerous tissues (N) by qRT-PCR assay (n = 32). (B) TREM1 protein level was significantly higher in PTC tissues (T, n = 52) compared to adjacent noncancerous tissues (N, n = 52). (C) Comparison of TREM1 level in high (stages III–IV, n = 9) and low (stages I–II, n = 43) stage PTCs. (D) Representative IHC photograph of TREM1 in PTC tissues (stages I–II and III–IV) and adjacent normal tissues (N). (E) Representative IHC staining using TREM1 and phospho-ERK in PTC specimens. Positive correlation of TREM1 protein levels and phosphorylation levels of ERK in IHC. Images are presented at 20× magnification. ***P < 0.001.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
IHC analysis of TREM1 expression and its correlation with tumor stages in PTC. (A) The mRNA expression levels of TREM1 in PTCs (T) compared to matched noncancerous tissues (N) by qRT-PCR assay (n = 32). (B) TREM1 protein level was significantly higher in PTC tissues (T, n = 52) compared to adjacent noncancerous tissues (N, n = 52). (C) Comparison of TREM1 level in high (stages III–IV, n = 9) and low (stages I–II, n = 43) stage PTCs. (D) Representative IHC photograph of TREM1 in PTC tissues (stages I–II and III–IV) and adjacent normal tissues (N). (E) Representative IHC staining using TREM1 and phospho-ERK in PTC specimens. Positive correlation of TREM1 protein levels and phosphorylation levels of ERK in IHC. Images are presented at 20× magnification. ***P < 0.001.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
IHC analysis of TREM1 expression and its correlation with tumor stages in PTC. (A) The mRNA expression levels of TREM1 in PTCs (T) compared to matched noncancerous tissues (N) by qRT-PCR assay (n = 32). (B) TREM1 protein level was significantly higher in PTC tissues (T, n = 52) compared to adjacent noncancerous tissues (N, n = 52). (C) Comparison of TREM1 level in high (stages III–IV, n = 9) and low (stages I–II, n = 43) stage PTCs. (D) Representative IHC photograph of TREM1 in PTC tissues (stages I–II and III–IV) and adjacent normal tissues (N). (E) Representative IHC staining using TREM1 and phospho-ERK in PTC specimens. Positive correlation of TREM1 protein levels and phosphorylation levels of ERK in IHC. Images are presented at 20× magnification. ***P < 0.001.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
We next divided the 52 PTC samples into a TREM1 high group (n = 28) and a TREM1 low group (n = 24) and determined the correlation of TREM1 levels with patients’ clinicopathological characteristics. As shown in Table 1, a high TREM1 level was statistically correlated with lymph node metastasis (P < 0.01) and tumor stages (P < 0.01). Our analyses also confirmed the positive correlation between ERK phosphorylation levels and TREM1 protein levels by IHC staining (Fig. 4E, P < 0.05).
Association of TREM1 level with clinicopathological characteristics in PTCs.
| Variable | n | TREM1 expression | P-value | |
|---|---|---|---|---|
| Low level | High level | |||
| Age | ||||
| ≤55 | 37 | 19 | 18 | 0.238 |
| >55 | 15 | 5 | 10 | |
| Gender | ||||
| Male | 19 | 11 | 8 | 0.198 |
| Female | 33 | 13 | 20 | |
| Invasion | ||||
| Yes | 1 | 0 | 1 | 1.000 |
| No | 51 | 24 | 27 | |
| Lymph node metastasis | ||||
| Yes | 29 | 8 | 21 | 0.003 |
| No | 23 | 16 | 7 | |
| TNM stages | ||||
| I, II | 43 | 24 | 19 | 0.007 |
| III, IV | 9 | 0 | 9 | |
Bold indicates statistical significance, P < 0.05.
Characterization of TREM1 expression in PTC cells
TREMs are a family of cell surface receptors expressed primarily on myeloid cells. However, we observed by IHC that TREM1 signals were mostly detected in PTC cells (Fig. 4D). We thus reanalyzed previous PTC scRNA-seq data to identify the heterogeneous expression of TREM1 in PTC (Peng et al. 2021). A total of 12,757 cells from four male patients with PTC and seven female patients with PTC were included in our analysis. Cells in PTC samples were successfully classified into 19 clusters (Fig. 5A). Differential expression analysis was performed, and a total of 11,422 marker genes from all 19 clusters were identified (Supplementary Table 4). According to the expression patterns of the marker genes, these clusters were classified into ten major cell types (Fig. 5B) based on known markers. Interestingly, consistent with our IHC data, TREM1 was mainly expressed in PTC cells rather than other cell types, such as macrophages (Fig. 5C). To further verify TREM1 expression in PTC cells, immunofluorescent staining for TREM1 and the malignant epithelial markers EPCAM and KRT19 was performed on PTC tissues. We observed colocalization of TREM1 (red) with EPCAM and KRT19 (green) on merged images (Fig. 5D and E).
TREM1 expression in PTC cells. (A) Dot plot of cell markers in PTC single cell 19 clusters; sizes of dots represent abundance, while color represents expression level. (B) tSNE plot visualization showed cell fractions in ten groups based on cell marker genes. (C) Violin plots of the TREM1 expression distribution in ten cell groups. TREM1 was mainly expressed in PTC tumor cells. (D) Immunofluorescent staining for TREM1 (red) and malignant epithelial markers EPCAM and KRT19 (green) on tumor section. Merged image showing colocalization of TREM1 (red) and EPCAM or KRT19 (green) in PTC. Images are presented at 40× magnification.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
TREM1 expression in PTC cells. (A) Dot plot of cell markers in PTC single cell 19 clusters; sizes of dots represent abundance, while color represents expression level. (B) tSNE plot visualization showed cell fractions in ten groups based on cell marker genes. (C) Violin plots of the TREM1 expression distribution in ten cell groups. TREM1 was mainly expressed in PTC tumor cells. (D) Immunofluorescent staining for TREM1 (red) and malignant epithelial markers EPCAM and KRT19 (green) on tumor section. Merged image showing colocalization of TREM1 (red) and EPCAM or KRT19 (green) in PTC. Images are presented at 40× magnification.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
TREM1 expression in PTC cells. (A) Dot plot of cell markers in PTC single cell 19 clusters; sizes of dots represent abundance, while color represents expression level. (B) tSNE plot visualization showed cell fractions in ten groups based on cell marker genes. (C) Violin plots of the TREM1 expression distribution in ten cell groups. TREM1 was mainly expressed in PTC tumor cells. (D) Immunofluorescent staining for TREM1 (red) and malignant epithelial markers EPCAM and KRT19 (green) on tumor section. Merged image showing colocalization of TREM1 (red) and EPCAM or KRT19 (green) in PTC. Images are presented at 40× magnification.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Functional enrichment analysis of TREM1 in PTC
To find the biological mechanism of TREM1 in PTC, GO and KEGG pathways were applied to annotate and enrich the DEGs between the low and the high TREM1 expression matrices. As shown in Fig. 6A, the following tumor-related pathways were associated with TREM1: positive regulation of cell adhesion, negative regulation of the apoptotic signaling pathway, activation of MAPK activity, PI3K-AKT signaling pathway, and transcriptional misregulation in cancer. Other enriched GO and KEGG pathways related to the TIM were also involved, including immune cell migration (myeloid leukocyte migration, granulocyte migration, neutrophil migration, macrophage migration, and leukocyte transendothelial migration) and immunity regulation (cytokine–cytokine receptor interaction, macrophage activation, and negative regulation of immune response, details in Supplementary Table 5). In addition, by using GSEA, TREM1 was connected with the activation of T cells and macrophages; the chemotaxis of granulocytes, leukocytes, monocytes, and lymphocytes; and the migration of granulocytes, neutrophils, T cells, leukocytes, mononuclear cells, and lymphocytes (Fig. 6B). These data suggest that TREM1 expressed in PTC cells may participate in the communication between cancer cells and TIICs to promote tumor progression.
Functional enrichment analysis of TREM1 in PTC. (A) GO and KEGG pathways for DEGs upregulated by TREM1. (B) GSEA analysis for biological processes related with TREM1 expression. TREM1 was connected with the activation of T cell and macrophage; the migration of granulocyte, neutrophil, T cell, leukocyte, mononuclear cell, and lymphocyte; the chemotaxis of granulocyte, leukocyte, monocyte, and lymphocyte.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Functional enrichment analysis of TREM1 in PTC. (A) GO and KEGG pathways for DEGs upregulated by TREM1. (B) GSEA analysis for biological processes related with TREM1 expression. TREM1 was connected with the activation of T cell and macrophage; the migration of granulocyte, neutrophil, T cell, leukocyte, mononuclear cell, and lymphocyte; the chemotaxis of granulocyte, leukocyte, monocyte, and lymphocyte.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Functional enrichment analysis of TREM1 in PTC. (A) GO and KEGG pathways for DEGs upregulated by TREM1. (B) GSEA analysis for biological processes related with TREM1 expression. TREM1 was connected with the activation of T cell and macrophage; the migration of granulocyte, neutrophil, T cell, leukocyte, mononuclear cell, and lymphocyte; the chemotaxis of granulocyte, leukocyte, monocyte, and lymphocyte.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Enhanced infiltration of Tregs in PTC with high TREM1 expression
To determine the relationship between TREM1 expression and immune infiltration in PTC, we first evaluated 22 immune signatures in PTC and their adjacent normal tissues by CIBERSORT (Supplementary Fig. 4A). Protumor immune cells, such as Tregs (P < 0.01), M2 macrophages (P < 0.001), and resting natural killer cells (P < 0.01), were significantly increased, while antitumor immune cells, such as CD8+ T cells (P < 0.001) and M1 macrophages (P < 0.001), were dramatically decreased in PTC. We then used TISIDB to investigate the TIICs that are regulated by TREM1 (Supplementary Fig. 4B) and found that Tregs (R = 0.534, P < 0.001), myeloid-derived suppressor cells (R = 0.532, P < 0.001), and iDCs (R = 0.342, P < 0.001) were positively correlated with TREM1 expression (Fig. 7A).
Enhanced infiltration of Tregs in PTC with TREM1 high expression. (A) Tregs, myeloid-derived suppressor cells (MDSC), and immature dendritic cells (iDC) were positively correlated with TREM1 expression identified by TISIDB. (B) Differential analysis of tumor-infiltrating immune cells (TIICs) between TREM1 high and TREM1 low expression group in PTC using CIBERSORT. (C) The positive correlation of TREM1 expression and Tregs immune infiltration by ssGSEA. (D) TREM1 mRNA expression was significantly correlated with FOXP3 mRNA expression in TCGA analyzed by cBioPortal. (E) TREM1 mRNA expression was positively correlated with CCR4, CCL17, and CCL22 by cBioPortal and TISIDB. (F) Relapse-free survival curves of high and low expression TREM1 levels in PTC based on the abundance of tumor-infiltrating Tregs.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Enhanced infiltration of Tregs in PTC with TREM1 high expression. (A) Tregs, myeloid-derived suppressor cells (MDSC), and immature dendritic cells (iDC) were positively correlated with TREM1 expression identified by TISIDB. (B) Differential analysis of tumor-infiltrating immune cells (TIICs) between TREM1 high and TREM1 low expression group in PTC using CIBERSORT. (C) The positive correlation of TREM1 expression and Tregs immune infiltration by ssGSEA. (D) TREM1 mRNA expression was significantly correlated with FOXP3 mRNA expression in TCGA analyzed by cBioPortal. (E) TREM1 mRNA expression was positively correlated with CCR4, CCL17, and CCL22 by cBioPortal and TISIDB. (F) Relapse-free survival curves of high and low expression TREM1 levels in PTC based on the abundance of tumor-infiltrating Tregs.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Enhanced infiltration of Tregs in PTC with TREM1 high expression. (A) Tregs, myeloid-derived suppressor cells (MDSC), and immature dendritic cells (iDC) were positively correlated with TREM1 expression identified by TISIDB. (B) Differential analysis of tumor-infiltrating immune cells (TIICs) between TREM1 high and TREM1 low expression group in PTC using CIBERSORT. (C) The positive correlation of TREM1 expression and Tregs immune infiltration by ssGSEA. (D) TREM1 mRNA expression was significantly correlated with FOXP3 mRNA expression in TCGA analyzed by cBioPortal. (E) TREM1 mRNA expression was positively correlated with CCR4, CCL17, and CCL22 by cBioPortal and TISIDB. (F) Relapse-free survival curves of high and low expression TREM1 levels in PTC based on the abundance of tumor-infiltrating Tregs.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Furthermore, differential analysis of TIICs was performed between the TREM1 high and TREM1 low expression groups in PTC using CIBERSORT. A significantly higher proportion of immunosuppressive cells and Tregs was observed in the TREM1 high group of PTC (P < 0.001; Fig. 7B). The positive correlation of TREM1 expression and Treg immune infiltration was further confirmed by ssGSEA (Fig. 7C). Consistently, TREM1 mRNA expression was significantly correlated with the Treg-specific transcription factor FOXP3 in TCGA analyzed by cBioPortal (Fig. 7D). In addition, the expression levels of C-C chemokine receptor 4 (CCR4, R = 0.440, P < 0.001) and its ligands C-C chemokine ligand 17 (CCL17, R= 0.612, P < 0.001) and 22 (CCL22, R = 0.574, P < 0.001), which contribute to Treg recruitment (Kohli et al. 2021), were also positively correlated with TREM1 mRNA expression (Fig. 7E). We further performed a prognostic analysis of TREM1 expression in the immune cell subgroup. TREM1 overexpression predicted poor prognosis in patients with enriched Tregs but not in those with decreased Tregs (Fig. 7F). The results indicated that TREM1 was associated with Treg infiltration in PTC.
To validate the role of TREM1 in immune infiltration, we conducted IHC in 40 PTC tissues. As shown in Fig. 8A, PTC with high TREM1 expression tended to have enhanced Treg infiltration (4.15 ± 1.88 vs 2.15 ± 1.34, P < 0.001), as reflected by the positive association between the TREM1 and FOXP3 protein levels (R = 0.647, P < 0.001, Fig. 8B). Additionally, PTC with high TREM1 expression had less CD8+ T cell infiltration (3.00 ± 1.06 vs 4.34 ± 1.64, P < 0.01), as reflected by the negative association between TREM1 and CD8A (R = −0.362, P < 0.05, Fig. 8B). This result was consistent with our previous knowledge that Treg infiltration suppresses tumor-specific CD8+ T cell cytotoxicity in thyroid cancer (Ferrari et al. 2019). Accordingly, the above data suggest that hypomethylation-mediated overexpression of TREM1 contributed to PTC progression, at least partially, owing to the enhanced infiltration of Tregs.
FOXP3 and CD8A expression in TREM1 high and TREM1 low groups of PTC by IHC. (A) Comparison of FOXP3 and CD8A expression between TREM1 high (n = 21) and TREM1 low groups (n = 19) in the PTC. (B) Correlation of TREM1 expression with FOXP3 and CD8A expression in the PTC. (C) Possible schematics depicting that hypomethylation-mediated overexpression of TREM1 recruits Treg infiltration in PTC by secreting CCL17 and CCL22. Images are presented at 20× magnification. **P < 0.01, ***P < 0.001.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
FOXP3 and CD8A expression in TREM1 high and TREM1 low groups of PTC by IHC. (A) Comparison of FOXP3 and CD8A expression between TREM1 high (n = 21) and TREM1 low groups (n = 19) in the PTC. (B) Correlation of TREM1 expression with FOXP3 and CD8A expression in the PTC. (C) Possible schematics depicting that hypomethylation-mediated overexpression of TREM1 recruits Treg infiltration in PTC by secreting CCL17 and CCL22. Images are presented at 20× magnification. **P < 0.01, ***P < 0.001.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
FOXP3 and CD8A expression in TREM1 high and TREM1 low groups of PTC by IHC. (A) Comparison of FOXP3 and CD8A expression between TREM1 high (n = 21) and TREM1 low groups (n = 19) in the PTC. (B) Correlation of TREM1 expression with FOXP3 and CD8A expression in the PTC. (C) Possible schematics depicting that hypomethylation-mediated overexpression of TREM1 recruits Treg infiltration in PTC by secreting CCL17 and CCL22. Images are presented at 20× magnification. **P < 0.01, ***P < 0.001.
Citation: Endocrine-Related Cancer 29, 2; 10.1530/ERC-21-0297
Discussion
Despite the indolent clinical course, one-third of PTC cases relapse during a 30-year follow-up (Dong et al. 2019), and molecular predictors for PTC recurrence remain to be defined. While prior studies have shown that certain immune cells in the PTC immune microenvironment have prognostic value (Bergdorf et al. 2019, Menicali et al. 2020), the potential mechanisms underlying this immune landscape are yet to be uncovered. In this study, we showed that TREM1 and TREM2 are overexpressed in PTC and are related to BRAFV600E profiles. However, hypomethylation of TREM1, instead of TREM2, affects patient prognosis. In particular, the methylation status of CpG site cg06196379 in the TREM1 promoter is correlated with PTC recurrence. Mechanistically, we found that TREM1 is mainly expressed in PTC cells and participates in PTC progression at least partially through the regulation of Treg infiltration.
TREM1 was initially recognized as a major amplifier in immune responses associated with infectious inflammatory disease (Bouchon et al. 2000). Recent studies, however, have established its contributions to various pathophysiological processes in noninfectious inflammatory diseases, including atherosclerosis, inflammatory bowel disease, gout, and cancers (Tammaro et al. 2017). Here, we demonstrated that the expression of TREM1 was elevated in PTC and was related to advanced tumor stage, indicating its oncogenic role. Moreover, TREM1 mRNA expression was negatively correlated with the promoter methylation status. In particular, we identified that a hypomethylated DNA methylation probe of TREM1, that is, cg06196379, was able to predict poor prognosis of patients with PTC. These results were consistent with a recent study that identified TREM1 as one of the methylation-regulated genes with prognostic prediction value in thyroid cancer (Lv et al. 2020). Thus, overexpression of TREM1 is hypomethylation-mediated and is involved in PTC progression.
Global hypomethylation, which is frequently seen in BRAFV600E-PTCs, is a major event for PTC development and progression (Klein Hesselink et al. 2018). Our analysis revealed that TREM1 mRNA expression in BRAFV600E-like PTCs was significantly higher than that in RAS-like PTCs. Consistently, our data showed that PTCs with higher TREM1 is associated with BRAFV600E profiles, including lower BRAF-RAS score, higher ERK score, lower differentiation levels, as well as enriched classical and tall cell histology (Agrawal et al. 2014). Compared with RAS-like PTCs, BRAFV600E-like PTCs have enhanced MAPK/ERK signaling because of the failure of the response to feedback inhibition of RAF signaling (Pratilas et al. 2009). These results suggested that TREM1 overexpression may depend on ERK signaling. As expected, TREM1 mRNA and protein expression levels were demonstrated to be significantly positively correlated with ERK signaling by bioinformatics and IHC analysis, respectively. However, on analysis of the genetic background of BRAFV600E-like PTCs, we observed a higher TREM1 expression in RET/PTC group than that in BRAFV600E group. We currently do not have a good explanation of this since nearly all of the RET/PTCs were weakly BRAFV600E-like. As an immune-related gene, increased TREM1 has been recently shown to be induced by UV irradiation and gamma irradiation (Qiang et al. 2017). In light of the high prevalence of radiation exposure history in PET/PTCs patients, this elevation may be related to radiation exposure. Of course, further studies are needed to clarify the role of TREM1 in PTCs with different genetic variations.
The development of PTC has been closely linked to inflammation in light of its frequent coexistence with Hashimoto’s thyroiditis and remarkable lymphatic infiltration even without a preexisting autoimmune process (Ehlers & Schott 2014). TREM1 has been shown to be expressed by immune cells, particularly myeloid lineages, and functions as an amplifier of inflammatory responses through the activation of myeloid cells with increased secretion of cytokines such as TNF-α, IL-6, and IL-1 (Bouchon et al. 2001). However, by reanalyzing scRNA-seq data, we found that TREM1 was mainly expressed in malignant epithelial cells in PTC that were characterized by the epithelial marker EPCAM (Peng et al. 2021) and specific tumor cell markers KRT19, ECM1, and CHI3L1 (Cheung et al. 2001, Qiu et al. 2018, Wang et al. 2019). Moreover, our immunofluorescence data further confirmed TREM1 protein expression in PTC cells. Although two recent investigations have established the TREM1 expression in nonimmune cells, including gastric epithelial cells (Schmausser et al. 2008) and hepatocellular carcinoma cells (Duan et al. 2015), our study represents the first report of TREM1 protein expression in PTC cells.
To further uncover the tumor-promoting mechanism of TREM1, we first performed functional enrichment analysis, exhibiting the involvement of TREM1 in the activation, migration, and chemotaxis of immune cells in PTC. In our analysis of immune infiltration in PTC, we found that TREM1 expression was markedly correlated with the infiltration of Tregs, which was further confirmed by its positive correlation with the Treg marker FOXP3. The enrichment of Tregs has been shown to be correlated with more aggressive PTC behaviors, such as lymph node metastasis, recurrent events, and extrathyroidal extension (French et al. 2012, Ryu et al. 2014, Liu et al. 2015). We observed that in patients with enriched Tregs, high expression of TREM1 predicted poor prognosis. Compared to BRAFwt-PTCs, BRAFV600E-PTC tumors display a more suppressive immune landscape and a lower CD8+/FOXP3+ cell ratio (Angell et al. 2014). Consistently, our data showed a negative correlation between TREM1 expression and CD8+ T cell infiltration in PTC. In liver cancer, TREM1 in TAMs has been shown to induce CD8+ T cell exhaustion through CCL20-mediated recruitment of CCR6+ Tregs (Qinchuan et al. 2019). In our study, TREM1 expression was significantly correlated with the mRNA expression of CCR4 and its ligands CCL22 and CCL17 (Duan et al. 2015). Thus, the immunosuppressive effects of TREM1 in PTC may at least be partially attributed to CCL17- and CCL22-mediated recruitment of CCR4+FOXP3+ Treg cells, resulting in CD8+ T cell dysfunction (Fig. 8C). The therapeutic potential of targeting TREM1 in high-risk PTCs is promising.
Our above results have demonstrated that hypomethylation of TREM1 is associated with BRAFV600E profiles and participates in PTC progression by regulating the infiltration of Tregs. Consistently, immune infiltration of thyroid cancer has been shown to be related to the BRAFV600E mutation (Bergdorf et al. 2019, Menicali et al. 2020). However, one major limitation of this study is that these data do not clarify how BRAF-driven ERK activation induces the hypomethylation of TREM1 in PTC cells. TREM1 was identified as a hypoxia-inducible gene in human DCs and macrophages (Bosco et al. 2011). Hypoxic stress has recently been shown to induce TREM1 activation in TAMs through hypoxia-inducible factor-1α (HIF1α) signaling in liver cancer (Qinchuan et al. 2019). The BRAFV600E mutation is known to promote the hypoxic tumor environment and to enhance the expression of HIF1α, which is heavily implicated in facilitating tumor growth and metastasis (Zerilli et al. 2010). Thus, it is possible that the overexpression of TREM1 is driven by enhanced HIF1α signaling in a hypoxic environment caused by the BRAFV600E mutation in PTC. However, this hypothesis remains to be clarified by experimental studies. In addition, we cannot exclude the involvement of other mechanisms, since increased TREM1 on TAMs has been shown to be induced by androgen receptor signaling in prostate cancer (Yuan et al. 2014) and by the cyclooxygenase pathway in lung cancer (Cioni et al. 2020).
Compared to TREM1, TREM2 is expressed in a wider range of human myeloid-derived cells, including osteoclasts, microglia, and macrophages in adipose tissue, adrenal gland, and placenta, and is involved in various pathologies (Molgora et al. 2020). Recent evidence suggests that TREM2 participates in carcinogenesis (Katzenelenbogen et al. 2020, Xiong et al. 2020) and might be a potential target for enhancing checkpoint immunotherapy. In our study, we determined that hypomethylation of TREM2 is associated with BRAF-driven ERK activation. However, TREM2 elevation is not related to the clinical outcome of patients with PTC. TREM2 has been shown to be elevated in liver cancer but plays a protective role in hepatocarcinogenesis via different pleiotropic effects (Tang et al. 2019, Esparza-Baquer et al. 2021). Thus, the exact role of TREM2 in PTC tumorigenesis remains to be clarified.
Conclusion
In summary, we presented the first report of TREM1 expression in PTC cells and clarified that hypomethylation of TREM1 is correlated with poor prognosis of PTC. Mechanistically, TREM1 is associated with BRAFV600E profiles and participates in PTC progression at least partially by regulating Treg infiltration. Our results suggest the risk prediction value of TREM1 methylation status in PTC and open up the possibility for the further use of TREM1 as a therapeutic target for PTC.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/ERC-21-0297.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
The work was supported by grants from National Science Foundation of China (grant no. 81500597), Natural Science Basic Research Plan in Shaanxi Province of China (grant No. 2017JM8051), and the Fundamental Research Funds for the Central Universities (sxjh012019062).
Acknowledgements
The authors thank Dr Ying Zhao (Xi’an Fourth Hospital) and Dr Shu Liu (The First Affiliated Hospital of Xi’an Jiaotong University) for scoring the tissue immunohistochemical staining. The authors also thank the TCGA databases and GEO (ID: GSE158291) for providing PTC gene expression profiles.
References
Angell TE, Lechner MG, Jang JK, Correa AJ, LoPresti JS & Epstein AL 2014 BRAFV600E in papillary thyroid carcinoma is associated with increased programmed death ligand 1 expression and suppressive immune cell infiltration. Thyroid 24 1385–1393. (https://doi.org/10.1089/thy.2014.0134
Agrawal N, Akbani R, Aksoy BA, Ally A, Arachchi H, Asa SL, Auman JT, Balasundaram M, Balu S & Baylin SB et al.2014 Integrated genomic characterization of papillary thyroid carcinoma. Cell 159 676–690. (https://doi.org/10.1016/j.cell.2014.09.050
Bergdorf KN, Ferguson DC, Mehrad M, Ely K, Stricker T & Weiss VL 2019 Papillary thyroid carcinoma behavior: clues in the tumor microenvironment. Endocrine-Related Cancer 26 601–614. (https://doi.org/10.1530/ERC-19-0074
Binnewies M, Roberts EW, Kersten K, Chan V, Fearon DF, Merad M, Coussens LM, Gabrilovich DI, Ostrand-Rosenberg S & Hedrick CC et al.2018 Understanding the tumor immune microenvironment (TIME) for effective therapy. Nature Medicine 24 541–550. (https://doi.org/10.1038/s41591-018-0014-x
Bosco MC, Pierobon D, Blengio F, Raggi F, Vanni C, Gattorno M, Eva A, Novelli F, Cappello P & Giovarelli M et al.2011 Hypoxia modulates the gene expression profile of immunoregulatory receptors in human mature dendritic cells: identification of TREM-1 as a novel hypoxic marker in vitro and in vivo. Blood 117 2625–2639. (https://doi.org/10.1182/blood-2010-06-292136
Bouchon A, Dietrich J & Colonna M 2000 Cutting edge: inflammatory responses can be triggered by TREM-1, a novel receptor expressed on neutrophils and monocytes. Journal of Immunology 164 4991–4995. (https://doi.org/10.4049/jimmunol.164.10.4991
Bouchon A, Facchetti F, Weigand MA & Colonna M 2001 TREM-1 amplifies inflammation and is a crucial mediator of septic shock. Nature 410 1103–1107. (https://doi.org/10.1038/35074114
Cheung CC, Ezzat S, Freeman JL, Rosen IB & Asa SL 2001 Immunohistochemical diagnosis of papillary thyroid carcinoma. Modern Pathology 14 338–342. (https://doi.org/10.1038/modpathol.3880312
Cioni B, Zaalberg A, van Beijnum JR, Melis MHM, van Burgsteden J, Muraro MJ, Hooijberg E, Peters D, Hofland I & Lubeck Y et al.2020 Androgen receptor signalling in macrophages promotes TREM-1-mediated prostate cancer cell line migration and invasion. Nature Communications 11 4498. (https://doi.org/10.1038/s41467-020-18313-y
Deczkowska A, Weiner A & Amit I 2020 The physiology, pathology, and potential therapeutic applications of the TREM2 signaling pathway. Cell 181 1207–1217. (https://doi.org/10.1016/j.cell.2020.05.003
Dong W, Horiuchi K, Tokumitsu H, Sakamoto A, Noguchi E, Ueda Y & Okamoto T 2019 Time-varying pattern of mortality and recurrence from papillary thyroid cancer: lessons from a long-term follow-up. Thyroid 29 802–808. (https://doi.org/10.1089/thy.2018.0128
Duan M, Wang ZC, Wang XY, Shi JY, Yang LX, Ding ZB, Gao Q, Zhou J & Fan J 2015 TREM-1, an inflammatory modulator, is expressed in hepatocellular carcinoma cells and significantly promotes tumor progression. Annals of Surgical Oncology 22 3121–3129. (https://doi.org/10.1245/s10434-014-4191-7
Ehlers M & Schott M 2014 Hashimoto’s thyroiditis and papillary thyroid cancer: are they immunologically linked? Trends in Endocrinology and Metabolism 25 656–664. (https://doi.org/10.1016/j.tem.2014.09.001
Esparza-Baquer A, Labiano I, Sharif O, Agirre-Lizaso A, Oakley F, Rodrigues PM, Zhuravleva E, O’Rourke CJ, Hijona E & Jimenez-Aguero R et al.2021 TREM-2 defends the liver against hepatocellular carcinoma through multifactorial protective mechanisms. Gut 70 1345–1361. (https://doi.org/10.1136/gutjnl-2019-319227
Ferrari SM, Fallahi P, Galdiero MR, Ruffilli I, Elia G, Ragusa F, Paparo SR, Patrizio A, Mazzi V & Varricchi G et al.2019 Immune and inflammatory cells in thyroid cancer microenvironment. International Journal of Molecular Sciences 20 4413. (https://doi.org/10.3390/ijms20184413
French JD, Kotnis GR, Said S, Raeburn CD, McIntyre RC Jr, Klopper JP & Haugen BR 2012 Programmed death-1+ T cells and regulatory T cells are enriched in tumor-involved lymph nodes and associated with aggressive features in papillary thyroid cancer. Journal of Clinical Endocrinology and Metabolism 97 E934–E943. (https://doi.org/10.1210/jc.2011-3428
Fu J, Lv H, Guan H, Ma X, Ji M, He N, Shi B & Hou P 2013 Metallothionein 1G functions as a tumor suppressor in thyroid cancer through modulating the PI3K/Akt signaling pathway. BMC Cancer 13 462. (https://doi.org/10.1186/1471-2407-13-462
Galdiero MR, Varricchi G & Marone G 2016 The immune network in thyroid cancer. OncoImmunology 5 e1168556. (https://doi.org/10.1080/2162402X.2016.1168556
Gogali F, Paterakis G, Rassidakis GZ, Kaltsas G, Liakou CI, Gousis P, Neonakis E, Manoussakis MN & Liapi C 2012 Phenotypical analysis of lymphocytes with suppressive and regulatory properties (Tregs) and NK cells in the papillary carcinoma of thyroid. Journal of Clinical Endocrinology and Metabolism 97 1474–1482. (https://doi.org/10.1210/jc.2011-1838
Haugen BR, Alexander EK, Bible KC, Doherty G, Mandel SJ, Nikiforov YE, Pacini F, Randolph G, Sawka A & Schlumberger M 2015 American Thyroid Association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association guidelines task force on thyroid nodules and differentiated thyroid cancer. Thyroid 2015 2165. (https://doi.org/10.1089/thy.2015.0020
Katzenelenbogen Y, Sheban F, Yalin A, Yofe I, Svetlichnyy D, Jaitin DA, Bornstein C, Moshe A, Keren-Shaul H & Cohen M et al.2020 Coupled scRNA-seq and intracellular protein activity reveal an immunosuppressive role of TREM2 in cancer. Cell 182 872 .e19–885.e19. (https://doi.org/10.1016/j.cell.2020.06.032
Klein Hesselink EN, Zafon C, Villalmanzo N, Iglesias C, van Hemel BM, Klein Hesselink MS, Montero-Conde C, Buj R, Mauricio D & Peinado MA et al.2018 Increased global DNA hypomethylation in distant metastatic and dedifferentiated thyroid cancer. Journal of Clinical Endocrinology and Metabolism 103 397–406. (https://doi.org/10.1210/jc.2017-01613
Kohli K, Pillarisetty VG & Kim TS 2021 Key chemokines direct migration of immune cells in solid tumors. Cancer Gene Therapy [epub]. (https://doi.org/10.1038/s41417-021-00303-x
Liu Y, Yun X, Gao M, Yu Y & Li X 2015 Analysis of regulatory T cells frequency in peripheral blood and tumor tissues in papillary thyroid carcinoma with and without Hashimoto’s thyroiditis. Clinical and Translational Oncology 17 274–280. (https://doi.org/10.1007/s12094-014-1222-6
Lv L, Cao L, Hu G, Shen Q & Wu J 2020 Methylation-driven genes identified as novel prognostic indicators for thyroid carcinoma. Frontiers in Genetics 11 294. (https://doi.org/10.3389/fgene.2020.00294
Menicali E, Guzzetti M, Morelli S, Moretti S & Puxeddu E 2020 Immune landscape of thyroid cancers: new insights. Frontiers in Endocrinology 11 637826. (https://doi.org/10.3389/fendo.2020.637826
Molgora M, Esaulova E, Vermi W, Hou J, Chen Y, Luo J, Brioschi S, Bugatti M, Omodei AS & Ricci B et al.2020 TREM2 modulation remodels the tumor myeloid landscape enhancing anti-PD-1 immunotherapy. Cell 182 886 .e17–900.e17. (https://doi.org/10.1016/j.cell.2020.07.013
Na KJ & Choi H 2018 Immune landscape of papillary thyroid cancer and immunotherapeutic implications. Endocrine-Related Cancer 25 523–531. (https://doi.org/10.1530/ERC-17-0532
Peng M, Wei G, Zhang Y, Li H, Lai Y, Guo Y, Chen Y, Liu L, Xiao H & Guan H et al.2021 Single‐cell transcriptomic landscape reveals the differences in cell differentiation and immune microenvironment of papillary thyroid carcinoma between genders. Cell and Bioscience 11 39. (https://doi.org/10.1186/s13578-021-00549-w
Powers AE, Marcadis AR, Lee M, Morris LGT & Marti JL 2019 Changes in trends in thyroid cancer incidence in the United States, 1992 to 2016. JAMA 322 2440–2441. (https://doi.org/10.1001/jama.2019.18528
Pratilas CA, Taylor BS, Ye Q, Viale A, Sander C, Solit DB & Rosen N 2009 (V600E)BRAF is associated with disabled feedback inhibition of RAF-MEK signaling and elevated transcriptional output of the pathway. PNAS 106 4519–4524. (https://doi.org/10.1073/pnas.0900780106
Qiang L, Sample A, Shea CR, Soltani K, Macleod KF & He YY 2017 Autophagy gene ATG7 regulates ultraviolet radiation-induced inflammation and skin tumorigenesis. Autophagy 13 2086–2103. (https://doi.org/10.1080/15548627.2017.1380757
Qinchuan, Wuhua, Zhou, Shengyong, Yin, Yuan, Tianchi, Chen, Junjie & Qian 2019 Blocking triggering receptor expressed on myeloid Cells-1-positive tumor-associated macrophages induced by hypoxia reverses immunosuppression and anti-programmed cell death ligand 1 resistance in liver cancer. Hepatology 70 198–214. (https://doi.org/10.1002/hep.30593
Qing W, Fang WY, Ye L, Shen LY, Zhang XF, Fei XC, Chen X, Wang WQ, Li XY & Xiao JC et al.2012 Density of tumor-associated macrophages correlates with lymph node metastasis in papillary thyroid carcinoma. Thyroid 22 905–910. (https://doi.org/10.1089/thy.2011.0452
Qiu QC, Wang L, Jin SS, Liu GF, Liu J, Ma L, Mao RF, Ma YY, Zhao N & Chen M et al.2018 CHI3L1 promotes tumor progression by activating TGF-β signaling pathway in hepatocellular carcinoma. Scientific Reports 8 15029. (https://doi.org/10.1038/s41598-018-33239-8
Remmele W & Stegner HE 1987 Recommendation for uniform definition of an immunoreactive score (IRS) for immunohistochemical estrogen receptor detection (ER-ICA) in breast cancer tissue. Pathologe 8 138–140.
Ryu HS, Park YS, Park HJ, Chung YR, Yom CK, Ahn SH, Park YJ, Park SH & Park SY 2014 Expression of indoleamine 2, 3-dioxygenase and infiltration of FOXP3+ regulatory T cells are associated with aggressive features of papillary thyroid microcarcinoma. Thyroid 24 1232–1240. (https://doi.org/10.1089/thy.2013.0423
Schmausser B, Endrich S, Beier D, Moran AP, Burek CJ, Rosenwald A, Rieckmann P, Müller‐Hermelink HK & Eck M 2008 Triggering receptor expressed on myeloid cells‐1 (TREM‐1) expression on gastric epithelium: implication for a role of TREM‐1 in helicobacter pylori infection. Clinical and Experimental Immunology 152 88–94. (https://doi.org/10.1111/j.1365-2249.2008.03608.x
Stuart T, Butler A, Hoffman P, Hafemeister C, Papalexi E, Mauck WM III, Hao Y, Stoeckius M, Smibert P & Satija R 2019 Comprehensive integration of single-cell data. Cell 177 1888 .e21–1902.e21. (https://doi.org/10.1016/j.cell.2019.05.031
Tammaro A, Derive M, Gibot S, Leemans JC, Florquin S & Dessing MC 2017 TREM-1 and its potential ligands in non-infectious diseases: from biology to clinical perspectives. Pharmacology and Therapeutics 177 81–95. (https://doi.org/10.1016/j.pharmthera.2017.02.043
Tang W, Lv B, Yang B, Chen Y, Yuan F, Ma L, Chen S, Zhang S & Xia J 2019 TREM2 acts as a tumor suppressor in hepatocellular carcinoma by targeting the PI3K/Akt/β-catenin pathway. Oncogenesis 8 9. (https://doi.org/10.1038/s41389-018-0115-x
van Velsen EFS, Stegenga MT, van Kemenade FJ, Kam BLR, van Ginhoven TM, Visser WE & Peeters RP 2019 Evaluating the 2015 American Thyroid Association risk stratification system in high risk papillary and follicular thyroid cancer patients. Thyroid 29 1073–1079. (https://doi.org/10.1089/thy.2019.0053
Vinayak S, Newman A, Adams S, Afghahi A, Jensen KC, Badve SS, Ford JM, Alizadeh AA & Telli ML 2015 Abstract P5-04-03: Deconvoluting immune cell populations using ‘in silico flow cytometry’ with CIBERSORT: Association with neoadjuvant therapy response and genomic instability in TNBC. Cancer Research 75 (9_Suppl) Abstract P5-04-03. (https://doi.org/10.1158/1538-7445.SABCS14-P5-04-03
Wang X, Xu X, Peng C, Qin Y, Gao T, Jing J & Zhao H 2019 BRAFV600E-induced KRT19 expression in thyroid cancer promotes lymph node metastasis via EMT. Oncology Letters 18 927–935. (https://doi.org/10.3892/ol.2019.10360
Werner TA, Tamkan-Olcek Y, Dizdar L, Riemer JC, Wolf A, Cupisti K, Verde PE, Knoefel WT & Krieg A 2016 Survivin and XIAP: two valuable biomarkers in medullary thyroid carcinoma. British Journal of Cancer 114 427–434. (https://doi.org/10.1038/bjc.2016.5
Werner TA, Nolten I, Dizdar L, Riemer JC, Schutte SC, Verde PE, Raba K, Schott M, Knoefel WT & Krieg A 2018 IAPs cause resistance to TRAIL-dependent apoptosis in follicular thyroid cancer. Endocrine-Related Cancer 25 295–308. (https://doi.org/10.1530/ERC-17-0479
Xing M, Alzahrani AS, Carson KA, Shong YK, Kim TY, Viola D, Elisei R, Bendlová B, Yip L & Mian C et al.2015 Association between BRAF V600E mutation and recurrence of papillary thyroid cancer. Journal of Clinical Oncology 33 42–50. (https://doi.org/10.1200/JCO.2014.56.8253
Xiong D, Wang Y & You M 2020 A gene expression signature of TREM2 hi macrophages and γδ T cells predicts immunotherapy response. Nature Communications 11 1.
Yu G, Wang LG, Han Y & He QY 2012 clusterProfiler: an R package for comparing biological themes among gene clusters. Omics 16 284–287. (https://doi.org/10.1089/omi.2011.0118
Yuan Z, Mehta HJ, Mohammed K, Nasreen N, Roman R, Brantly M & Sadikot RT 2014 TREM-1 is induced in tumor associated macrophages by cyclo-oxygenase pathway in human non-small cell lung cancer. PLoS ONE 9 e94241. (https://doi.org/10.1371/journal.pone.0094241
Zerilli M, Zito G, Martorana A, Pitrone M, Cabibi D, Cappello F, Giordano C & Rodolico V 2010 BRAF(V600E) mutation influences hypoxia-inducible factor-1alpha expression levels in papillary thyroid cancer. Modern Pathology 23 1052–1060. (https://doi.org/10.1038/modpathol.2010.86
