- Email: [email protected]
Increased expression of TTC21A in lung adenocarcinoma infers favorable prognosis and high immune infiltrating level
International Immunopharmacology 78 (2020) 106077
Contents lists available at ScienceDirect
International Immunopharmacology journal homepage: www.elsevier.com/locate/intimp
Increased expression of TTC21A in lung adenocarcinoma infers favorable prognosis and high immune infiltrating level Wei Wanga,b,1, Shiqi Rena,b,1, Ziheng Wanga,b,1, Chenlin Zhangd, Jianfei Huanga,c,
T
⁎
a
Department of Clinical Biobank, Nantong University Affiliated Hospital, Nantong, PR China Department of Medicine, Nantong University Xinling College, Nantong, PR China c Institute of Oncology, Affiliated Hospital of Nantong University, Nantong 226001, Jiangsu, China d Department of Spine, Wuxi Hospital Affiliated to Nanjing University of Chinese Medicine, Wuxi 214000, PR China b
A R T I C LE I N FO
A B S T R A C T
Keywords: Lung adenocarcinoma Biomarker Immune infiltration TTC21A Prognosis Lymphocyte
Background: Lung adenocarcinoma (LUAD) is a crucial pathological type of lung cancer. Immune-infiltration of the tumor microenvironment positively associated with overall survival in LUAD. TTC21A is a gene has not reported in cancer, and the mechanism behind it is still unclear. Our study assesses TTC21A role in LUAD, via TCGA data. Methods: GEPIA was utilized to analyze the expression of TTC21A in LUAD. We evaluated the influence of TTC21A on survival of LUAD patients by survival module. Then, data sets of LUAD were downloaded from TCGA. The correlations between clinical information and TTC21A expression were analyzed using logistic regression. Clinicopathologic characteristics associated with overall survival in TCGA patients using Cox regression. In addition, we explored the correlation between TTC21A and cancer immune infiltrates using CIBERSORT and “Correlation” module of GEPIA. Results: The univariate analysis using logistic regression, wherein TTC21A expression served as a categorical dependent variable (with a median expression value of 2.5), indicated that increased TTC21A expression is significantly correlated with pathological stage, tumor status and lymph nodes. Moreover, multivariate analysis revealed that the up-regulated TTC21A expression, negative results of pathological stage and distant metastasis are independent prognostic factors for good prognosis. Specifically, a positive correlation between increased TTC21A expression and immune infiltrating level of B cells, Neutrophils, Mast cells and T cells was established using CIBERSORT analysis. Furthermore, we confirmed it in “correlation” module of GEPIA. Conclusion: Together with all these findings, increased TTC21A expression correlates with favorable prognosis and increased proportion of immune cells, such as B cells, Neutrophils, Mast cells and T cells in LUAD. These conclusions indicate that TTC21A could serve as a potential biomarker to assess prognosis and immune infiltration level in LUAD.
1. Introduction Lung cancer is a type of malignant tumor in characteristic of uncontrolled cell growth in the lung and bronchus. Aside from prostate cancer and Breast cancer, lung cancer is the second most frequent diagnosed cancer in males and females. It is also the most common inducement of cancer death in people and take up more than 25% of all cancer deaths [1]. In addition, 5-year relative survival for the overall age of lung cancer is respectively 15.5% in men and 20.3% in women [2]. Therefore, further researches are necessary for the development of lung cancer treatment, including the study of the immunophenotypes of
tumor-immune communications, the identification of promising biomarkers and novel immune-related therapeutic targets for lung cancer patients. Non-small cell lung cancer (NSCLC) takes up about 85% of lung cancer cases [3]. According to SEER Cancer Statistics Review, the latest statistics show that the average 5-year survival rate of non-small cell lung cancer in men is 19.2% and in women is 26.8%. The major histological phenotypes are classified as squamous cell carcinoma, adenocarcinoma, and large cell carcinoma. However, lung adenocarcinoma, which accounts for 50% of NSCLC, is the crucial pathological type [4]. Immune-infiltration of the tumor microenvironment was
⁎
Corresponding author. E-mail address: [email protected] (J. Huang). 1 These authors contributed equally to this study. https://doi.org/10.1016/j.intimp.2019.106077 Received 3 June 2019; Received in revised form 27 October 2019; Accepted 21 November 2019 1567-5769/ © 2019 Elsevier B.V. All rights reserved.
International Immunopharmacology 78 (2020) 106077
W. Wang, et al.
2. Materials and methods
associated with survival for a multitude of patients with solid tumors. Many examples can be given, for example, CD8+ TILs have been proved to be a good prognostic factor in lung adenocarcinoma. In this tumor, it has been disclosed that the CD8t TILs can specifically kill tumor cells. A large amount of researches in tumor-infiltration in the last few years have found that high expression of B cell and T cell, such as breast CD8 T Cells, adenocarcinoma B Cells, melanoma LCK and so on, predicts perfect overall survival across many cancer types including lung adenocarcinoma [5]. However, the precise prognostic correlation of Immune-infiltration in lung adenocarcinoma remains unclear. TTC21A (Tetratricopeptide Repeat Domain 21A) is a Protein Coding gene. TTC21A contains several tetratricopeptide repeat (TPR) domains that frequently exist in IFT proteins and seem important for ciliary function. The deduced 1320-amino acid protein contains 10 N-terminal tetratricopeptide repeats of about 34 amino acids each. An important paralog of the gene is TTC21B. TTC21A, encoding a protein of the TPR family, is expressed in various types of tissues and organs. According to the previous research, quite a few members of the TPR family were linked with ciliary function of non-human model organisms. For instance, TTC10 has been associated with cilium biogenesis [6]. Studies of receiver biases suggest that ciliary function plays a key role in the adjustment of cell signaling pathways. In addition, upregulation of function of primary cilia and the existence of structural defects in a great amount of mankind cancer tissue samples manifesting interrelation of function of primary cilia with cancer [7]. No previous study has reported that TTC21A involves in development of cancer. However, the definite correlation of TTC21A with the prognosis of LUAD was established in our present study. Presently, we downloaded the data from the public domain, The Cancer Genome Atlas (TCGA). Meanwhile, the correlation of TTC21A with prognosis of lung adenocarcinoma was identified via Gene Expression Profiling Interactive Analysis (GEPIA) and COX regression analysis. Moreover, CIBERSORT, a recent metagene approach, was used to examine the relative proportions of different kinds of Tumor- infiltrating immune cells (TIICs) in different tumor microenvironments to study the relationship of TTC21A with tumor-infiltrating immune cells [8]. The findings in this study help us enhance the understanding of the possible positive effects of TTC21A in lung adenocarcinoma and shed light on a potential correlation as well as a possible mechanism between TTC21A and tumor-immune interactions. Thus, TTC21A has the potential to become a novel predictor to evaluate prognosis and immune infiltration for LUAD patients. The workflow of our study is presented in Fig. 1.
2.1. Data acquisition LUAD patient datasets, with gene expression profiles and paired clinical information, were downloaded from the publicly available TCGA, and included 538 tumor tissues. Subsequent processing excluded cases with insufficient or missing data on age, overall survival time, local invasion, lymph node metastasis, distant metastasis and TNM stage. Finally, 476 cases with eligible clinical information were devoted into Cox regression analysis. In order to study the influences of TTC21A expression on immune microenvironment, 535 tumor tissues were retained to perform CIBERSORT analysis. 2.2. Survival and expression analysis by GEPIA Gene Expression Profiling Interactive Analysis (GEPIA) (http:// gepia.cancer-pku.cn/index.html), an online database was used to evaluate the correlation between TTC21A expression and clinicopathologic information in lung adenocarcinoma. GEPIA [9] is a web server for analyzing the RNA sequencing expression data of 9736 tumors and 8587 normal samples from the TCGA and the GTEx projects, using a standard processing pipeline. “Survival” module of GEPIA allowed us to evaluate the correlation of TTC21A expression with prognosis of LUADs. Meanwhile, boxplots using disease state (Tumor or Normal) as variable was graphed to visualize differential expression of TTC21A between tumor and normal tissues. In addition, clinical staging boxplots using pathological stage as variable was graphed to compare TTC21A expression in different pathological stages. 2.3. Evaluation of tumor-infiltrating immune cells CIBERSORT (http://cibersort.stanford.edu/), a deconvolution algorithm based on gene expression can evaluate the changes in the expression of one set of genes relative to all other genes in the sample. TIIC concentration can therefore be precisely estimated via this process. CIBERSORT’s consistent performance encouraged a growing focus on cell heterogeneity studies [10–12]. Our current analysis gauged the proportions of 22 TIICs in LUAD via CIBERSORT, so as to assess its correlation with survival and molecular subpopulation. In short, gene expression datasets were set out using standard annotation files and uploaded to CIBERSORT web portal, with the algorithm running with its default signature matrix at one thousand permutations. CIBERSORT estimated a P-value for deconvolution via Monte Carlo sampling, establishing a measure of confidence in the results. In order to assess the influence of TTC21A expression, we used 535 samples from the TCGA, where the full range of genes are represented. The p-value < 0.05 was set as the criterion to select the lymphocyte possibly effected by the
Fig. 1. Workflow of this study. 2
International Immunopharmacology 78 (2020) 106077
W. Wang, et al.
3.3. Relationship between TTC21A expression and tumor-infiltrating immune cells
expression of TTC21A. To detect the correlation between 22 types of immune cells, we conducted the correlation heat map, a chart of the correlation between every two different immune cells in samples. In addition, the “correlation” module of GEPIA was used to further confirm relationships between expression of TTC21A and possibly gene markers of tumor-infiltrating immune cells. The gene markers included markers of CD8+ T cells, T cells (general), B cells, neutrophils, natural killer (NK) cells, T-helper 1 (Th1) cells, T-helper 2 (Th2) cells, follicular helper T (Tfh) cells, T-helper 17 (Th17) cells, exhausted T cells and Mast cells. Prior studies provide a reference for the gene markers [13–15]. The correlation module helped chart expression scatter plots of a user-defined gene pair from a specific cancer type, alongside the Spearman’s R and projected statistical significance. P-value < 0.01 was determined as the threshold. To detect the correlation between 22 types of immune cells, we conducted the correlation heat map, a chart of the correlation between every two different immune cells in samples.
Previous analyses suggest tumor-infiltrating lymphocytes as independent predictors of sentinel lymph node status and survival in cancer patients [16]. Therefore, we tried to find whether TTC21A expression relates to immune infiltration in lung adenocarcinoma. Among 535 LUAD samples, samples with the top 1/3 and the lowest 1/3 TTC21A expression were included into high expression group and low expression group, respectively. An established computational resource (CIBERSORT) was used to explore gene expression profiles of downloaded samples to infer the fraction of 22 types of immune cells in high and low TTC21A expression groups. Finally, 161 samples of high expression group and 168 samples of low expression group met screening criterion. The results of CIBERSORT were exhibited in Fig. 4. The proportions of 22 subpopulations of immune cells were clearly rendered on it. As shown in Fig. 4A, B cells naive, B cells memory, T cells CD8, T cells CD4 memory resting, T cells CD4 memory activated, NK cells activated, Mast cells resting, Mast cells activated and Neutrophils are main immune cells affected by TTC21A expression. Among them, B cells naive (p = 0.035), T cells CD4 memory resting (p < 0.001), Mast cells resting (p = 0.044) share a higher proportion in high expression group compared with low expression group. In contrast, the proportion of T cells CD8 (p = 0.049), NK cells activated (p = 0.008) are apparently lower. In addition, correlation heatmap (Fig. 4B) revealed that the proportions of different TIICs subpopulations were weakly to moderately correlated. “Correlation” module of GEPIA helped us to analyze the link between TTC21A expression and gene markers of different types of tumorinfiltrating immune cells, including CD8+ T cells, T cells (general), B cells, neutrophils, Mast cells and DCs, as also different functional T cells, namely Th1, Th2, Tfh, Th17, Tregs, and exhausted T cells (Table 3). Results confirmed that TTC21A expression is correlated with over half of the marker sets of different immune cells in LUAD. The gene markers effected by TTC21A expression include CD3E of T cell (general), CD19 and CD79A of B cell, CD66b, CD11b, CCR7 of Neutrophils, T-bet, STAT4, TNF-α of Th1, GATA3, STAT6, STAT5A, IL13 of Th2, BCL6 of Tfh, STAT3 of Th17, as well as CTLA4, LAG3 of T cell exhaustion. Correlations were evaluated using Spearman correlation coefficient. Correlation results between TTC21A and markers of B cells, Neutrophils, Mast cells and T cells were similar to CIBERSORT. Thus, these findings suggest that TTC21A may play an important role in regulating the abundance of B cells, Neutrophils, Mast cells and T cells. Further studies need to be done to explore whether TTC21A is a significant factor that relate to immune infiltration of NK cells.
2.4. Statistical analysis Statistical acquired from TCGA were all conducted by R-3.5.3. The correlations between clinical characteristics and TTC21A expression were analyzed using logistic regression. The COX regression analysis was conducted to identify overall survival-related clinical characteristics in the TCGA patients. P value lower than 0.05 was considered statistically significant in this study. The gene expression correlation was accessed by Spearman’s R and statistical significance. The absolute value of R greater than 0.1 was considered to be relevant and Pvalue < 0.01 was considered statistically significant.
3. Results 3.1. Survival outcomes and multivariate analysis As shown in Fig. 2, reduced expression of TTC21A is significantly correlated with poor overall survival (Fig. 1A, P = 0.002) and advanced pathological stage (Fig. 1B, P < 0.001). In addition, TTC21A expression in tumor samples is obviously higher than in normal (Fig. 1C, Log2FC < 2, P-value < 0.01). As shown in Table 1a, Univariate analysis using Cox regression revealed that some factors, including pathological stage (HR = 1.667, P-value < 0.001), tumor status (HR = 1.525, P-value < 0.001), distant metastasis (HR = 1.990, P = 0.001), lymph node status (HR = 1.725, Pvalue = 0.000) along with the expression of TTC21A (HR = 0.794, Pvalue < 0.046) are significantly associated with overall survival. In multivariate analysis (Table 1b, Fig. 3), the up-regulated TTC21A expression, lower pathological stage and negative distant metastasis are independent prognostic factors of favorable prognosis.
4. Discussion TTC21A is a gene only reported in male infertility in prior studies. Here, we find that variations in TTC21A expression level relate to prognosis in LUAD. An up-regulated TTC21A expression is an independent prognostic factor for favorable prognosis. Meanwhile, increased levels of TTC21A expression is significantly associated with several clinical characteristics, including pathological stage, tumor status and lymph node status. In addition, this study revealed that diverse immune marker sets and immune infiltration levels are correlated with TTC21A expression in LUAD. Thus, the outcome of previous study suggested that TTC21A may have a potential influence on tumor immunology and could serve as a promising cancer biomarker. In this study, we found the correlation of TTC21A expression with prognosis of LUAD patients using GEPIA, an online database. The upregulated expression of TTC21A is associated with the favorable prognosis. We also observed that TTC21A expression differs in normal and tumor tissues of LUAD. To further study the underlying mechanisms and relationships of TTC21A expression in cancer, we downloaded datasets from TCGA. Statistical analysis by R-3.5.3 revealed that
3.2. Association between TTC21A expression and clinicopathologic variables The underlying mechanism of TTC21A expression in cancer requires further study, hence we analyzed and correlated it with certain clinical aspects in cases of lung adenocarcinoma. LUAD cases with eligible clinical information were analyzed by R-3.5.3. As shown in Table 2, univariate analysis using logistic regression with TTC21A expression as a categorical dependent variable (based on median expression value of 2.5) revealed that increased expression of TTC21A correlated significantly with the pathological stage (IIvsI, p = 0.014; III vs I, p = 0.001; III&IV vs I&II, p = 0.006), tumor status (T2 vs T1, p = 0.001; T3 vs T1; p = 0.043), and lymph node (N1 vs N0, p = 0.035; N2 vs N0, p = 0.000; N1&N2&N3 vs N0, p = 0.001).
3
International Immunopharmacology 78 (2020) 106077
W. Wang, et al.
Fig. 2. Survival outcome and expression difference analyzed by GEPIA. A. Increased TTC21A expression is associated with favorable outcome. B. Differential expression of TTC21A in different disease state (Tumor or Normal). C. Differential expression of TTC21A in different pathological stage.
different immune cells and TTC21A expression implicated the significant meaning of TTC21A in regulating tumor immune microenvironment of LUAD. First, using CIBERSORT algorithm, we found that the proportions of B cells, several types of T cells, Mast cells resting were apparently increased in high expression group compared with the low expression group. In addition, we used “Correlation” module of GEPIA to confirm the discovery. CD19 and CD79A of B cell showed positive correlations with TTC21A expression. This result showed us the potential regulating role of TTC21A in abundance of tumor-associated B cells. Th1, Th2, Tfh, Th17 are different functional T cells. The increase in TTC21A expression positively correlated with their markers, such as T-bet, STAT4, TNF-α, GATA3, STAT6, STAT5A, IL13, BCL6, STAT3. These correlations could indicate a possible mechanism where TTC21A regulates T cell functions in LUAD. Furthermore, significant correlations were observed between markers of Neutrophils and Mast cells and TTC21A. As per the preceding reports, Neutrophils are most prevalent, and dominate immune cell composition in non-small cell lung cancer [17]. High expression of T-cell and B-cell signatures correlates to prolonged overall survival, predicted in several types of tumors, including lung cancer [5]. Together these findings show that TTC21A plays a pivotal role in regulation and recruitment of immune infiltrating cells in LUAD. Studies on the role of TIICs (tumor-infiltrating immune cells) in human tumors typically focus on T-cells, with several reporting their response to immune checkpoint inhibition, and survival thereof [18–21]. This study appends the evolving body of literature ascertaining T-cells as a positive prognostic aspect. Another preceding study suggested that tumor-infiltrating B cells are significant regulators of lung adenocarcinoma development [22]. Tumor-infiltrating B lymphocytes (TIBs) can be observed in overall process of lung cancer development, and their concentration differs between histological
Table 1 The results of Cox regression analysis. A. Univariate analysis using Cox regression revealed that some factors, including pathological stage (HR = 1.667, P-value < 0.001), tumor status (HR = 1.525, P-value < 0.001), distant metastasis (HR = 1.990, P = 0.001), lymph node status (HR = 1.725, Pvalue = 0.000) along with the expression of TTC21A (HR = 0.794, Pvalue < 0.046) are significantly associated with overall survival. B. The upregulated TTC21A expression, lower pathological stage and negative distant metastasis are independent prognostic factors of favorable prognosis. Clinicopathologic variable
HR (95%CI)
p-Value
A. Age Gender Stage T (Tumor status) M (distant metastasis) N (lymph node) TTC21A
1.010 1.044 1.667 1.525 1.990 1.725 0.794
(0.994–1.026) (0.762–1.430) (1.438–1.931) (1.251–1.861) (1.328–2.984) (1.444–2.060) (0.633–0.996)
0.233 0.789 0.000 0.000 0.001 0.000 0.046
B. Stage M (distant metastasis) TTC21A
1.358 (1.029–1.194) 1.675 (1.014–2.768) 0.783 (0.623–0.984)
0.031 0.044 0.036
TTC21A expression relates to several clinical characteristic including the pathological stage, tumor status, and lymph node status. Multivariate analysis showed that TTC21A expression is an independent prognostic factor of LUAD patient’s prognosis. A noteworthy conclusion in our study concerned the correlation of TTC21A expression with immune infiltration levels in LUAD. CIBERSORT analysis indicated a substantial explicit connection of TTC21A expression with infiltration levels of B cells, Neutrophils, Mast cells and T cells in LUAD. Likewise, the relationships between gene markers of 4
International Immunopharmacology 78 (2020) 106077
W. Wang, et al.
Fig. 3. Multivariate Cox analysis of TTC21A expression and other clinicopathological factors. The up-regulated TTC21A expression, lower pathological stage and negative distant metastasis are independent prognostic factors of favorable prognosis.
TIB exert anti-tumor immunity through secretion of tumor-specific antibodies, promoting T cell responses, and maintaining the structure and function of TLS, all of which are associated with beneficial outcomes for lung cancer [26–29]. Thus, the positive influence of TTC21A on LUAD are consistence with the role of higher abundance of T cells and B cells, which indicates a possible mechanism where TTC21A effects the overall survival of LUAD. Several previous studies provided potential mechanisms which account for why TTC21A expression relates to T cells immune infiltration, as well as favorable prognosis. TTC10, IFT proteins disruption of which has been characterized in the microenvironment of the cancer, is an important paralog of TTC21A. Genetic analysis meanwhile indicates that IFT motor protein Kif3a disruption is needed for Hedgehog signaling, upstream of direct targets, and a step downstream of the Hedgehog receptor, Patched-1 [30]. TTC21B is another important paralog of the TTC21A. Somatic TTC21B mutations have been indicated, when cancer sets in. Nonetheless, loss-of-function TTC21B mutations upregulated cilia dependent processes in mice [31], and are causal mutations in primary cilia malfunction or ciliopathy-driven human developmental diseases [32]. It was further observed that primary cilia reduced subsequent to Hh pathway inhibitor treatment. Primary cilia have been associated with Hh pathway signaling; their ablation in SMO-activated cells therefore impedes tumor growth [33]. The Hh signaling pathway diminishes TCR signal strength in mature peripheral T cells [34,35], whilst the pathway repression progressively activates T-cell [36]. These observations indicate the Hh pathway
Table 2 Association between TTC21A expression and clinicopathologic variables using logistic regression. Increased expression of TTC21A significantly related to the pathological stage, tumor status, and lymph node status. Clinical characteristic
Total (N)
Odds ratio in TTC21A expression
P-Value
Age (continuous) Gender (male vs female) Stage(II vs I) Stage(III vs I) Stage (IV vsI) Stage (III&IV vs I&II) Tumor status (T2 vs T1) Tumor status (T3 vs T1) Tumor status (T4 vs T1) Tumor status (T3&T4 vs T1& T2) Distant metastasis (M1 vs M0) Lymph node (N1 vs N0) Lymph node (N2 vs N0) Lymph node (N1&N2&N3 vs N0)
522 531 395 359 300 505 444 215 187 510
1.14 0.72 0.58 0.41 0.56 0.54 0.50 0.50 0.57 0.81
(0.78–1.67) (0.51–1.03) (0.38–0.89) (0.25–0.68) (0.23–1.32) (0.35–0.84) (0.33–0.74) (0.26–0.98) (0.21–1.48) (0.48–1.36)
0.50 0.07 0.01 0.00 0.19 0.01 0.00 0.04 0.24 0.43
369 425 404 501
0.78 0.61 0.39 0.51
(0.34–1.75) (0.38–0.96) (0.22–0.65) (0.35–0.75)
0.54 0.04 0.00 0.00
*Categorical dependent variable, greater or less than the median expression level.
subtypes and stage, suggesting an important role for B cells in lung cancer development [23–25]. Driven by signals within the tumor microenvironment, B cells infiltrate, proliferate, and develop in tumors. 5
International Immunopharmacology 78 (2020) 106077
W. Wang, et al.
Fig. 4. TTC21A-related immune infiltration alteration. A. B cells naive (p = 0.035), T cells CD4 memory resting (p < 0.001), Mast cells resting (p = 0.044) share a higher proportion in high expression group compared with low expression group. In contrast, the proportion of T cells CD8 (p = 0.049), NK cells activated (p = 0.008) are apparently lower. B. The proportions of different TIICs subpopulations were weakly to moderately correlated.
Acknowledgement
inhibitor directly affects peripheral T cells, thereby activating adaptive immune responses. Therefore, interaction between TTC21A and Hedgehog signaling pathway could be a potential mechanism for upregulated immune infiltration level in LUAD. In summary, increased TTC21A expression correlates to favorable prognosis. Meanwhile, changed TTC21A expression is associated with different proportions of immune cells, such as B cells, Neutrophils, Mast cells and T cells in LUAD. Therefore, TTC21A likely has an essential influence on immune infiltration and has the potential to serve as a prognostic biomarker of LUAD.
We would like to thank Professor Lihua Shen from College of Foreign Languages, Nantong University for language polishing.
Declaration of Competing Interest The authors declared that there is no conflict of interest.
6
International Immunopharmacology 78 (2020) 106077
W. Wang, et al.
and flagella, J. Cell Biol. 151 (3) (2000) 709–718. [7] B. Yasar, K. Linton, C. Slater, R. Byers, Primary cilia are increased in number and demonstrate structural abnormalities in human cancer, J. Clin. Pathol. 70 (7) (2017) 571–574. [8] A.M. Newman, C.L. Liu, M.R. Green, et al., Robust enumeration of cell subsets from tissue expression profiles, Nat. Meth. 12 (5) (2015) 453–457. [9] Z. Tang, C. Li, B. Kang, G. Gao, C. Li, Z. Zhang, GEPIA: a web server for cancer and normal gene expression profiling and interactive analyses, Nucl. Acids Res. 45 (Web Server issue) (2017). [10] H.R. Ali, L. Chlon, P.D. Pharoah, F. Markowetz, C. Caldas, Patterns of immune infiltration in breast cancer and their clinical implications: a gene-expression-based retrospective study, PLoS Med. 13 (12) (2016) e1002194. [11] X. Liu, S. Wu, Y. Yang, M. Zhao, G. Zhu, Z. Hou, The prognostic landscape of tumorinfiltrating immune cell and immunomodulators in lung cancer, Biomed. Pharmacother. 95 (2017) 55–61. [12] R.D. Bense, C. Sotiriou, M.J. Piccartgebhart, et al., Relevance of tumor-infiltrating immune cell composition and functionality for disease outcome in breast cancer, J. Natl. Cancer Inst. 109 (1) (2016) djw192. [13] N.O. Siemers, J.L. Holloway, H. Chang, Genome-wide association analysis identifies genetic correlates of immune infiltrates in solid tumors, PLoS ONE 12 (7) (2017) e0179726. [14] P. Danaher, S. Warren, L. Dennis, et al., Gene expression markers of tumor infiltrating leukocytes, J. ImmunoTher. Cancer 5 (1) (2017) 18. [15] S. Sousa, J. Määttä, The role of tumour-associated macrophages in bone metastasis, J. Bone Oncol. 5 (3) (2016) 135–138. [16] F. Azimi, R.A. Scolyer, P. Rumcheva, et al., Tumor-infiltrating lymphocyte grade is an independent predictor of sentinel lymph node status and survival in patients with cutaneous melanoma, J. Clin. Oncol. 30 (21) (2012) 2678–2683. [17] J. Kargl, S.E. Busch, G.H. Yang, et al., Neutrophils dominate the immune cell composition in non-small cell lung cancer, Nat. Commun. 8 (2017) 14381. [18] T. Powles, J.P. Eder, G.D. Fine, et al., MPDL3280A (anti-PD-L1) treatment leads to clinical activity in metastatic bladder cancer, Nature 194 (4) (2014) 558–562. [19] E.I. Buchbinder, D.F. Mcdermott, Cytotoxic T-lymphocyte antigen-4 blockade in melanoma, Clin. Ther. 37 (4) (2015) 755–763. [20] F. Massari, M. Santoni, C. Ciccarese, et al., PD-1 blockade therapy in renal cell carcinoma: current studies and future promises, Cancer Treat. Rev. 41 (2) (2015) 114–121. [21] S.M.A. Mahmoud, E.C. Paish, D.G. Powe, et al., Tumor-infiltrating CD8+ lymphocytes predict clinical outcome in breast cancer, J. Clin. Oncol. 29 (15) (2011) 1949–1955. [22] S.S. Wang, W. Liu, D. Ly, H. Xu, L. Qu, L. Zhang, Tumor-infiltrating B cells: their role and application in anti-tumor immunity in lung cancer, Cell. Mol. Immunol. 16 (1) (2019) 6–18. [23] M.C. Dieu-Nosjean, J. Goc, N.A. Giraldo, C. Sautès-Fridman, W.H. Fridman, Tertiary lymphoid structures in cancer and beyond, Trends Immunol. 35 (11) (2014) 571–580. [24] G.A. Banat, A. Tretyn, S.S. Pullamsetti, et al., Immune and inflammatory cell composition of human lung cancer stroma, PLoS One 10 (9) (2015) e0139073. [25] Y. Kurebayashi, K. Emoto, Y. Hayashi, et al., Comprehensive immune profiling of lung adenocarcinomas reveals four immunosubtypes with plasma cell subtype a negative indicator, Cancer Immunol. Res. 4 (3) (2016) 234–247. [26] M.J. Campa, M.A. Moody, R. Zhang, H.X. Liao, E.B. Gottlin, E.F. Patz, Interrogation of individual intratumoral B lymphocytes from lung cancer patients for molecular target discovery, Cancer Immunol. Immunother. 65 (2) (2016) 171–180. [27] P. Costantino, G.W. Jones, B. Michele, S.A. Jones, Ectopic lymphoid-like structures in infection, cancer and autoimmunity, J. Nat. Rev. Immunol. 14 (7) (2014) 447–462. [28] T.C. Bruno, P.J. Ebner, B.L. Moore, et al., Antigen-presenting intratumoral B cells affect CD4+ TIL phenotypes in non-small cell lung cancer patients, Cancer Immunol. Res. 5 (10) (2017) 898. [29] W. Zhu, C. Germain, Z. Liu, et al., A high density of tertiary lymphoid structure B cells in lung tumors is associated with increased CD4+ T cell receptor repertoire clonality, Oncoimmunology 4 (12) (2015). [30] D. Huangfu, A. Liu, A.S. Rakeman, N.S. Murcia, L. Niswander, K.V. Anderson, Hedgehog signalling in the mouse requires intraflagellar transport proteins, Nature 426 (6962) (2003) 83–87. [31] P. Tran, C. Haycraft, T. Besschetnova, et al., THM1 negatively modulates mouse sonic hedgehog signal transduction and affects retrograde intraflagellar transport in cilia, Nat. Genet. 40 (4) (2008) 403–410. [32] E.E. Davis, Z. Qi, L. Qin, et al., TTC21B contributes both causal and modifying alleles across the ciliopathy spectrum, Nat. Genet. 43 (3) (2011) 189–196. [33] S.Y. Wong, A.D. Seol, P.L. So, et al., Primary cilia can both mediate and suppress Hedgehog pathway–dependent tumorigenesis, Nat. Med. 15 (9) (2009) 1055–1061. [34] N.J. Rowbotham, A.L. Hager-Theodorides, M. Cebecauer, et al., Activation of the Hedgehog signaling pathway in T-lineage cells inhibits TCR repertoire selection in the thymus and peripheral T-cell activation, Blood 109 (9) (2007) 3757–3766. [35] N.J. Rowbotham, A.L. Hager-Theodorides, A.L. Furmanski, C. Tessa, A novel role for Hedgehog in T-cell receptor signaling: implications for development and immunity, Cell Cycle 6 (17) (2007) 2138–2142. [36] N.J. Rowbotham, A.L. Furmanski, A.L. Hager-Theodorides, et al., Repression of hedgehog signal transduction in T-lineage cells increases TCR-induced activation and proliferation, Cell Cycle 7 (7) (2008) 904–908.
Table 3 Correlation analysis between TTC21A expression and gene markers of CD8+ T cells, T cells (general), B cells, neutrophils, natural killer (NK) cells, T-helper 1 (Th1) cells, T-helper 2 (Th2) cells, follicular helper T (Tfh) cells, T-helper 17 (Th17) cells, exhausted T cells and Mast cells via “correlation” module of GEPIA. Description
Gene markers
LUAD Tumor
CD8+ T cell T cell (general) B cell Natural killer cell
Neutrophils
Th1
Th2
Tfh Th17 T cell exhaustion
Mast cells
CD8A CD8B CD2 CD3E CD19 CD79A KIR2DL1 KIR2DL3 KIR2DL4 KIR3DL1 KIR3DL2 KIR3DL3 KIR2DS4 CD66b CD11b CCR7 T-bet STAT4 TNF-α GATA3 STAT6 STAT5A IL13 BCL6 STAT3 IL17A PD-1 CTLA4 LAG3 TIM-3 TPSB2 TPSAB1 CPA3 MS4A2 HDC
Normal
R
P
R
P
0.019 −0.0066 0.091 0.13 0.21 0.13 0.034 0.014 −0.086 0.11 0.039 −0.054 0.028 0.15 0.21 0.3 0.25 0.17 0.18 0.14 0.41 0.27 0.19 0.39 0.2 −0.074 0.094 0.17 0.15 −0.023 0.21 0.21 0.13 0.25 0.42
0.68 0.88 0.045 * *** * 0.45 0.76 0.058 0.017 0.4 0.23 0.54 ** *** *** *** ** *** * *** *** *** *** *** 0.11 0.039 ** ** 0.61 *** *** * *** ***
−0.22 −0.27 −0.029 −0.062 0.18 0.036 −0.033 −0.014 −0.13 −0.066 −0.17 0.066 −0.0065 0.12 −0.0084 −0.0024 0.11 0.15 0.2 0.24 0.35 0.12 −0.26 0.24 0.2 0.14 −0.14 0.13 0.47 −0.35 −0.15 −0.23 −0.26 0.045 0.18
0.089 0.039 0.83 0.64 0.18 0.79 0.8 0.92 0.31 0.62 0.19 0.62 0.96 0.36 0.95 0.99 0.4 0.25 0.14 0.066 * 0.36 0.044 0.073 0.12 0.29 0.28 0.32 ** * 0.26 0.084 0.044 0.73 0.17
Tumor, correlation analysis in LUAD tumor tissue of TCGA. Normal, correlation analysis in LUAD normal tissue of TCGA. *P < 0.01, **P < 0.001, ***P < 0.0001.
Funding Training Program of Innovation and Entrepreneurship for College Students in Jiangsu (NO. 201913993001Y and 201913993021H). References [1] R.L. Siegel, K.D. Miller, A. Jemal, Cancer statistics, 2016, Ca A Cancer J. Clin. 60 (5) (2010) 277–300. [2] A. Eberle, L. Jansen, F. Castro, et al., Lung cancer survival in Germany: a population-based analysis of 132,612 lung cancer patients, Lung Cancer 90 (3) (2015) 528–533. [3] J.R. Molina, P. Yang, S.D. Cassivi, S.E. Schild, A.A. Adjei, -small cell lung cancer: epidemiology, risk factors, treatment, and survivorship, Mayo Clin. Proc. 83 (5) (2008) 584–594. [4] Z. Chen, C.M. Fillmore, P.S. Hammerman, C.F. Kim, K.K. Wong, Non-small-cell lung cancers: a heterogeneous set of diseases, Nat. Rev. Cancer 14 (8) (2014) 535–546. [5] M.D. Iglesia, J.S. Parker, K.A. Hoadley, J.S. Serody, C.M. Perou, B.G. Vincent, Genomic analysis of immune cell infiltrates across 11 tumor types, J. Natl. Cancer Inst. 108 (11) (2016) djw144. [6] G.J. Pazour, B.L. Dickert, Y. Vucica, et al., Chlamydomonas IFT88 and Its mouse homologue, polycystic kidney disease gene Tg737, are required for assembly of cilia
7
Contents lists available at ScienceDirect
International Immunopharmacology journal homepage: www.elsevier.com/locate/intimp
Increased expression of TTC21A in lung adenocarcinoma infers favorable prognosis and high immune infiltrating level Wei Wanga,b,1, Shiqi Rena,b,1, Ziheng Wanga,b,1, Chenlin Zhangd, Jianfei Huanga,c,
T
⁎
a
Department of Clinical Biobank, Nantong University Affiliated Hospital, Nantong, PR China Department of Medicine, Nantong University Xinling College, Nantong, PR China c Institute of Oncology, Affiliated Hospital of Nantong University, Nantong 226001, Jiangsu, China d Department of Spine, Wuxi Hospital Affiliated to Nanjing University of Chinese Medicine, Wuxi 214000, PR China b
A R T I C LE I N FO
A B S T R A C T
Keywords: Lung adenocarcinoma Biomarker Immune infiltration TTC21A Prognosis Lymphocyte
Background: Lung adenocarcinoma (LUAD) is a crucial pathological type of lung cancer. Immune-infiltration of the tumor microenvironment positively associated with overall survival in LUAD. TTC21A is a gene has not reported in cancer, and the mechanism behind it is still unclear. Our study assesses TTC21A role in LUAD, via TCGA data. Methods: GEPIA was utilized to analyze the expression of TTC21A in LUAD. We evaluated the influence of TTC21A on survival of LUAD patients by survival module. Then, data sets of LUAD were downloaded from TCGA. The correlations between clinical information and TTC21A expression were analyzed using logistic regression. Clinicopathologic characteristics associated with overall survival in TCGA patients using Cox regression. In addition, we explored the correlation between TTC21A and cancer immune infiltrates using CIBERSORT and “Correlation” module of GEPIA. Results: The univariate analysis using logistic regression, wherein TTC21A expression served as a categorical dependent variable (with a median expression value of 2.5), indicated that increased TTC21A expression is significantly correlated with pathological stage, tumor status and lymph nodes. Moreover, multivariate analysis revealed that the up-regulated TTC21A expression, negative results of pathological stage and distant metastasis are independent prognostic factors for good prognosis. Specifically, a positive correlation between increased TTC21A expression and immune infiltrating level of B cells, Neutrophils, Mast cells and T cells was established using CIBERSORT analysis. Furthermore, we confirmed it in “correlation” module of GEPIA. Conclusion: Together with all these findings, increased TTC21A expression correlates with favorable prognosis and increased proportion of immune cells, such as B cells, Neutrophils, Mast cells and T cells in LUAD. These conclusions indicate that TTC21A could serve as a potential biomarker to assess prognosis and immune infiltration level in LUAD.
1. Introduction Lung cancer is a type of malignant tumor in characteristic of uncontrolled cell growth in the lung and bronchus. Aside from prostate cancer and Breast cancer, lung cancer is the second most frequent diagnosed cancer in males and females. It is also the most common inducement of cancer death in people and take up more than 25% of all cancer deaths [1]. In addition, 5-year relative survival for the overall age of lung cancer is respectively 15.5% in men and 20.3% in women [2]. Therefore, further researches are necessary for the development of lung cancer treatment, including the study of the immunophenotypes of
tumor-immune communications, the identification of promising biomarkers and novel immune-related therapeutic targets for lung cancer patients. Non-small cell lung cancer (NSCLC) takes up about 85% of lung cancer cases [3]. According to SEER Cancer Statistics Review, the latest statistics show that the average 5-year survival rate of non-small cell lung cancer in men is 19.2% and in women is 26.8%. The major histological phenotypes are classified as squamous cell carcinoma, adenocarcinoma, and large cell carcinoma. However, lung adenocarcinoma, which accounts for 50% of NSCLC, is the crucial pathological type [4]. Immune-infiltration of the tumor microenvironment was
⁎
Corresponding author. E-mail address: [email protected] (J. Huang). 1 These authors contributed equally to this study. https://doi.org/10.1016/j.intimp.2019.106077 Received 3 June 2019; Received in revised form 27 October 2019; Accepted 21 November 2019 1567-5769/ © 2019 Elsevier B.V. All rights reserved.
International Immunopharmacology 78 (2020) 106077
W. Wang, et al.
2. Materials and methods
associated with survival for a multitude of patients with solid tumors. Many examples can be given, for example, CD8+ TILs have been proved to be a good prognostic factor in lung adenocarcinoma. In this tumor, it has been disclosed that the CD8t TILs can specifically kill tumor cells. A large amount of researches in tumor-infiltration in the last few years have found that high expression of B cell and T cell, such as breast CD8 T Cells, adenocarcinoma B Cells, melanoma LCK and so on, predicts perfect overall survival across many cancer types including lung adenocarcinoma [5]. However, the precise prognostic correlation of Immune-infiltration in lung adenocarcinoma remains unclear. TTC21A (Tetratricopeptide Repeat Domain 21A) is a Protein Coding gene. TTC21A contains several tetratricopeptide repeat (TPR) domains that frequently exist in IFT proteins and seem important for ciliary function. The deduced 1320-amino acid protein contains 10 N-terminal tetratricopeptide repeats of about 34 amino acids each. An important paralog of the gene is TTC21B. TTC21A, encoding a protein of the TPR family, is expressed in various types of tissues and organs. According to the previous research, quite a few members of the TPR family were linked with ciliary function of non-human model organisms. For instance, TTC10 has been associated with cilium biogenesis [6]. Studies of receiver biases suggest that ciliary function plays a key role in the adjustment of cell signaling pathways. In addition, upregulation of function of primary cilia and the existence of structural defects in a great amount of mankind cancer tissue samples manifesting interrelation of function of primary cilia with cancer [7]. No previous study has reported that TTC21A involves in development of cancer. However, the definite correlation of TTC21A with the prognosis of LUAD was established in our present study. Presently, we downloaded the data from the public domain, The Cancer Genome Atlas (TCGA). Meanwhile, the correlation of TTC21A with prognosis of lung adenocarcinoma was identified via Gene Expression Profiling Interactive Analysis (GEPIA) and COX regression analysis. Moreover, CIBERSORT, a recent metagene approach, was used to examine the relative proportions of different kinds of Tumor- infiltrating immune cells (TIICs) in different tumor microenvironments to study the relationship of TTC21A with tumor-infiltrating immune cells [8]. The findings in this study help us enhance the understanding of the possible positive effects of TTC21A in lung adenocarcinoma and shed light on a potential correlation as well as a possible mechanism between TTC21A and tumor-immune interactions. Thus, TTC21A has the potential to become a novel predictor to evaluate prognosis and immune infiltration for LUAD patients. The workflow of our study is presented in Fig. 1.
2.1. Data acquisition LUAD patient datasets, with gene expression profiles and paired clinical information, were downloaded from the publicly available TCGA, and included 538 tumor tissues. Subsequent processing excluded cases with insufficient or missing data on age, overall survival time, local invasion, lymph node metastasis, distant metastasis and TNM stage. Finally, 476 cases with eligible clinical information were devoted into Cox regression analysis. In order to study the influences of TTC21A expression on immune microenvironment, 535 tumor tissues were retained to perform CIBERSORT analysis. 2.2. Survival and expression analysis by GEPIA Gene Expression Profiling Interactive Analysis (GEPIA) (http:// gepia.cancer-pku.cn/index.html), an online database was used to evaluate the correlation between TTC21A expression and clinicopathologic information in lung adenocarcinoma. GEPIA [9] is a web server for analyzing the RNA sequencing expression data of 9736 tumors and 8587 normal samples from the TCGA and the GTEx projects, using a standard processing pipeline. “Survival” module of GEPIA allowed us to evaluate the correlation of TTC21A expression with prognosis of LUADs. Meanwhile, boxplots using disease state (Tumor or Normal) as variable was graphed to visualize differential expression of TTC21A between tumor and normal tissues. In addition, clinical staging boxplots using pathological stage as variable was graphed to compare TTC21A expression in different pathological stages. 2.3. Evaluation of tumor-infiltrating immune cells CIBERSORT (http://cibersort.stanford.edu/), a deconvolution algorithm based on gene expression can evaluate the changes in the expression of one set of genes relative to all other genes in the sample. TIIC concentration can therefore be precisely estimated via this process. CIBERSORT’s consistent performance encouraged a growing focus on cell heterogeneity studies [10–12]. Our current analysis gauged the proportions of 22 TIICs in LUAD via CIBERSORT, so as to assess its correlation with survival and molecular subpopulation. In short, gene expression datasets were set out using standard annotation files and uploaded to CIBERSORT web portal, with the algorithm running with its default signature matrix at one thousand permutations. CIBERSORT estimated a P-value for deconvolution via Monte Carlo sampling, establishing a measure of confidence in the results. In order to assess the influence of TTC21A expression, we used 535 samples from the TCGA, where the full range of genes are represented. The p-value < 0.05 was set as the criterion to select the lymphocyte possibly effected by the
Fig. 1. Workflow of this study. 2
International Immunopharmacology 78 (2020) 106077
W. Wang, et al.
3.3. Relationship between TTC21A expression and tumor-infiltrating immune cells
expression of TTC21A. To detect the correlation between 22 types of immune cells, we conducted the correlation heat map, a chart of the correlation between every two different immune cells in samples. In addition, the “correlation” module of GEPIA was used to further confirm relationships between expression of TTC21A and possibly gene markers of tumor-infiltrating immune cells. The gene markers included markers of CD8+ T cells, T cells (general), B cells, neutrophils, natural killer (NK) cells, T-helper 1 (Th1) cells, T-helper 2 (Th2) cells, follicular helper T (Tfh) cells, T-helper 17 (Th17) cells, exhausted T cells and Mast cells. Prior studies provide a reference for the gene markers [13–15]. The correlation module helped chart expression scatter plots of a user-defined gene pair from a specific cancer type, alongside the Spearman’s R and projected statistical significance. P-value < 0.01 was determined as the threshold. To detect the correlation between 22 types of immune cells, we conducted the correlation heat map, a chart of the correlation between every two different immune cells in samples.
Previous analyses suggest tumor-infiltrating lymphocytes as independent predictors of sentinel lymph node status and survival in cancer patients [16]. Therefore, we tried to find whether TTC21A expression relates to immune infiltration in lung adenocarcinoma. Among 535 LUAD samples, samples with the top 1/3 and the lowest 1/3 TTC21A expression were included into high expression group and low expression group, respectively. An established computational resource (CIBERSORT) was used to explore gene expression profiles of downloaded samples to infer the fraction of 22 types of immune cells in high and low TTC21A expression groups. Finally, 161 samples of high expression group and 168 samples of low expression group met screening criterion. The results of CIBERSORT were exhibited in Fig. 4. The proportions of 22 subpopulations of immune cells were clearly rendered on it. As shown in Fig. 4A, B cells naive, B cells memory, T cells CD8, T cells CD4 memory resting, T cells CD4 memory activated, NK cells activated, Mast cells resting, Mast cells activated and Neutrophils are main immune cells affected by TTC21A expression. Among them, B cells naive (p = 0.035), T cells CD4 memory resting (p < 0.001), Mast cells resting (p = 0.044) share a higher proportion in high expression group compared with low expression group. In contrast, the proportion of T cells CD8 (p = 0.049), NK cells activated (p = 0.008) are apparently lower. In addition, correlation heatmap (Fig. 4B) revealed that the proportions of different TIICs subpopulations were weakly to moderately correlated. “Correlation” module of GEPIA helped us to analyze the link between TTC21A expression and gene markers of different types of tumorinfiltrating immune cells, including CD8+ T cells, T cells (general), B cells, neutrophils, Mast cells and DCs, as also different functional T cells, namely Th1, Th2, Tfh, Th17, Tregs, and exhausted T cells (Table 3). Results confirmed that TTC21A expression is correlated with over half of the marker sets of different immune cells in LUAD. The gene markers effected by TTC21A expression include CD3E of T cell (general), CD19 and CD79A of B cell, CD66b, CD11b, CCR7 of Neutrophils, T-bet, STAT4, TNF-α of Th1, GATA3, STAT6, STAT5A, IL13 of Th2, BCL6 of Tfh, STAT3 of Th17, as well as CTLA4, LAG3 of T cell exhaustion. Correlations were evaluated using Spearman correlation coefficient. Correlation results between TTC21A and markers of B cells, Neutrophils, Mast cells and T cells were similar to CIBERSORT. Thus, these findings suggest that TTC21A may play an important role in regulating the abundance of B cells, Neutrophils, Mast cells and T cells. Further studies need to be done to explore whether TTC21A is a significant factor that relate to immune infiltration of NK cells.
2.4. Statistical analysis Statistical acquired from TCGA were all conducted by R-3.5.3. The correlations between clinical characteristics and TTC21A expression were analyzed using logistic regression. The COX regression analysis was conducted to identify overall survival-related clinical characteristics in the TCGA patients. P value lower than 0.05 was considered statistically significant in this study. The gene expression correlation was accessed by Spearman’s R and statistical significance. The absolute value of R greater than 0.1 was considered to be relevant and Pvalue < 0.01 was considered statistically significant.
3. Results 3.1. Survival outcomes and multivariate analysis As shown in Fig. 2, reduced expression of TTC21A is significantly correlated with poor overall survival (Fig. 1A, P = 0.002) and advanced pathological stage (Fig. 1B, P < 0.001). In addition, TTC21A expression in tumor samples is obviously higher than in normal (Fig. 1C, Log2FC < 2, P-value < 0.01). As shown in Table 1a, Univariate analysis using Cox regression revealed that some factors, including pathological stage (HR = 1.667, P-value < 0.001), tumor status (HR = 1.525, P-value < 0.001), distant metastasis (HR = 1.990, P = 0.001), lymph node status (HR = 1.725, Pvalue = 0.000) along with the expression of TTC21A (HR = 0.794, Pvalue < 0.046) are significantly associated with overall survival. In multivariate analysis (Table 1b, Fig. 3), the up-regulated TTC21A expression, lower pathological stage and negative distant metastasis are independent prognostic factors of favorable prognosis.
4. Discussion TTC21A is a gene only reported in male infertility in prior studies. Here, we find that variations in TTC21A expression level relate to prognosis in LUAD. An up-regulated TTC21A expression is an independent prognostic factor for favorable prognosis. Meanwhile, increased levels of TTC21A expression is significantly associated with several clinical characteristics, including pathological stage, tumor status and lymph node status. In addition, this study revealed that diverse immune marker sets and immune infiltration levels are correlated with TTC21A expression in LUAD. Thus, the outcome of previous study suggested that TTC21A may have a potential influence on tumor immunology and could serve as a promising cancer biomarker. In this study, we found the correlation of TTC21A expression with prognosis of LUAD patients using GEPIA, an online database. The upregulated expression of TTC21A is associated with the favorable prognosis. We also observed that TTC21A expression differs in normal and tumor tissues of LUAD. To further study the underlying mechanisms and relationships of TTC21A expression in cancer, we downloaded datasets from TCGA. Statistical analysis by R-3.5.3 revealed that
3.2. Association between TTC21A expression and clinicopathologic variables The underlying mechanism of TTC21A expression in cancer requires further study, hence we analyzed and correlated it with certain clinical aspects in cases of lung adenocarcinoma. LUAD cases with eligible clinical information were analyzed by R-3.5.3. As shown in Table 2, univariate analysis using logistic regression with TTC21A expression as a categorical dependent variable (based on median expression value of 2.5) revealed that increased expression of TTC21A correlated significantly with the pathological stage (IIvsI, p = 0.014; III vs I, p = 0.001; III&IV vs I&II, p = 0.006), tumor status (T2 vs T1, p = 0.001; T3 vs T1; p = 0.043), and lymph node (N1 vs N0, p = 0.035; N2 vs N0, p = 0.000; N1&N2&N3 vs N0, p = 0.001).
3
International Immunopharmacology 78 (2020) 106077
W. Wang, et al.
Fig. 2. Survival outcome and expression difference analyzed by GEPIA. A. Increased TTC21A expression is associated with favorable outcome. B. Differential expression of TTC21A in different disease state (Tumor or Normal). C. Differential expression of TTC21A in different pathological stage.
different immune cells and TTC21A expression implicated the significant meaning of TTC21A in regulating tumor immune microenvironment of LUAD. First, using CIBERSORT algorithm, we found that the proportions of B cells, several types of T cells, Mast cells resting were apparently increased in high expression group compared with the low expression group. In addition, we used “Correlation” module of GEPIA to confirm the discovery. CD19 and CD79A of B cell showed positive correlations with TTC21A expression. This result showed us the potential regulating role of TTC21A in abundance of tumor-associated B cells. Th1, Th2, Tfh, Th17 are different functional T cells. The increase in TTC21A expression positively correlated with their markers, such as T-bet, STAT4, TNF-α, GATA3, STAT6, STAT5A, IL13, BCL6, STAT3. These correlations could indicate a possible mechanism where TTC21A regulates T cell functions in LUAD. Furthermore, significant correlations were observed between markers of Neutrophils and Mast cells and TTC21A. As per the preceding reports, Neutrophils are most prevalent, and dominate immune cell composition in non-small cell lung cancer [17]. High expression of T-cell and B-cell signatures correlates to prolonged overall survival, predicted in several types of tumors, including lung cancer [5]. Together these findings show that TTC21A plays a pivotal role in regulation and recruitment of immune infiltrating cells in LUAD. Studies on the role of TIICs (tumor-infiltrating immune cells) in human tumors typically focus on T-cells, with several reporting their response to immune checkpoint inhibition, and survival thereof [18–21]. This study appends the evolving body of literature ascertaining T-cells as a positive prognostic aspect. Another preceding study suggested that tumor-infiltrating B cells are significant regulators of lung adenocarcinoma development [22]. Tumor-infiltrating B lymphocytes (TIBs) can be observed in overall process of lung cancer development, and their concentration differs between histological
Table 1 The results of Cox regression analysis. A. Univariate analysis using Cox regression revealed that some factors, including pathological stage (HR = 1.667, P-value < 0.001), tumor status (HR = 1.525, P-value < 0.001), distant metastasis (HR = 1.990, P = 0.001), lymph node status (HR = 1.725, Pvalue = 0.000) along with the expression of TTC21A (HR = 0.794, Pvalue < 0.046) are significantly associated with overall survival. B. The upregulated TTC21A expression, lower pathological stage and negative distant metastasis are independent prognostic factors of favorable prognosis. Clinicopathologic variable
HR (95%CI)
p-Value
A. Age Gender Stage T (Tumor status) M (distant metastasis) N (lymph node) TTC21A
1.010 1.044 1.667 1.525 1.990 1.725 0.794
(0.994–1.026) (0.762–1.430) (1.438–1.931) (1.251–1.861) (1.328–2.984) (1.444–2.060) (0.633–0.996)
0.233 0.789 0.000 0.000 0.001 0.000 0.046
B. Stage M (distant metastasis) TTC21A
1.358 (1.029–1.194) 1.675 (1.014–2.768) 0.783 (0.623–0.984)
0.031 0.044 0.036
TTC21A expression relates to several clinical characteristic including the pathological stage, tumor status, and lymph node status. Multivariate analysis showed that TTC21A expression is an independent prognostic factor of LUAD patient’s prognosis. A noteworthy conclusion in our study concerned the correlation of TTC21A expression with immune infiltration levels in LUAD. CIBERSORT analysis indicated a substantial explicit connection of TTC21A expression with infiltration levels of B cells, Neutrophils, Mast cells and T cells in LUAD. Likewise, the relationships between gene markers of 4
International Immunopharmacology 78 (2020) 106077
W. Wang, et al.
Fig. 3. Multivariate Cox analysis of TTC21A expression and other clinicopathological factors. The up-regulated TTC21A expression, lower pathological stage and negative distant metastasis are independent prognostic factors of favorable prognosis.
TIB exert anti-tumor immunity through secretion of tumor-specific antibodies, promoting T cell responses, and maintaining the structure and function of TLS, all of which are associated with beneficial outcomes for lung cancer [26–29]. Thus, the positive influence of TTC21A on LUAD are consistence with the role of higher abundance of T cells and B cells, which indicates a possible mechanism where TTC21A effects the overall survival of LUAD. Several previous studies provided potential mechanisms which account for why TTC21A expression relates to T cells immune infiltration, as well as favorable prognosis. TTC10, IFT proteins disruption of which has been characterized in the microenvironment of the cancer, is an important paralog of TTC21A. Genetic analysis meanwhile indicates that IFT motor protein Kif3a disruption is needed for Hedgehog signaling, upstream of direct targets, and a step downstream of the Hedgehog receptor, Patched-1 [30]. TTC21B is another important paralog of the TTC21A. Somatic TTC21B mutations have been indicated, when cancer sets in. Nonetheless, loss-of-function TTC21B mutations upregulated cilia dependent processes in mice [31], and are causal mutations in primary cilia malfunction or ciliopathy-driven human developmental diseases [32]. It was further observed that primary cilia reduced subsequent to Hh pathway inhibitor treatment. Primary cilia have been associated with Hh pathway signaling; their ablation in SMO-activated cells therefore impedes tumor growth [33]. The Hh signaling pathway diminishes TCR signal strength in mature peripheral T cells [34,35], whilst the pathway repression progressively activates T-cell [36]. These observations indicate the Hh pathway
Table 2 Association between TTC21A expression and clinicopathologic variables using logistic regression. Increased expression of TTC21A significantly related to the pathological stage, tumor status, and lymph node status. Clinical characteristic
Total (N)
Odds ratio in TTC21A expression
P-Value
Age (continuous) Gender (male vs female) Stage(II vs I) Stage(III vs I) Stage (IV vsI) Stage (III&IV vs I&II) Tumor status (T2 vs T1) Tumor status (T3 vs T1) Tumor status (T4 vs T1) Tumor status (T3&T4 vs T1& T2) Distant metastasis (M1 vs M0) Lymph node (N1 vs N0) Lymph node (N2 vs N0) Lymph node (N1&N2&N3 vs N0)
522 531 395 359 300 505 444 215 187 510
1.14 0.72 0.58 0.41 0.56 0.54 0.50 0.50 0.57 0.81
(0.78–1.67) (0.51–1.03) (0.38–0.89) (0.25–0.68) (0.23–1.32) (0.35–0.84) (0.33–0.74) (0.26–0.98) (0.21–1.48) (0.48–1.36)
0.50 0.07 0.01 0.00 0.19 0.01 0.00 0.04 0.24 0.43
369 425 404 501
0.78 0.61 0.39 0.51
(0.34–1.75) (0.38–0.96) (0.22–0.65) (0.35–0.75)
0.54 0.04 0.00 0.00
*Categorical dependent variable, greater or less than the median expression level.
subtypes and stage, suggesting an important role for B cells in lung cancer development [23–25]. Driven by signals within the tumor microenvironment, B cells infiltrate, proliferate, and develop in tumors. 5
International Immunopharmacology 78 (2020) 106077
W. Wang, et al.
Fig. 4. TTC21A-related immune infiltration alteration. A. B cells naive (p = 0.035), T cells CD4 memory resting (p < 0.001), Mast cells resting (p = 0.044) share a higher proportion in high expression group compared with low expression group. In contrast, the proportion of T cells CD8 (p = 0.049), NK cells activated (p = 0.008) are apparently lower. B. The proportions of different TIICs subpopulations were weakly to moderately correlated.
Acknowledgement
inhibitor directly affects peripheral T cells, thereby activating adaptive immune responses. Therefore, interaction between TTC21A and Hedgehog signaling pathway could be a potential mechanism for upregulated immune infiltration level in LUAD. In summary, increased TTC21A expression correlates to favorable prognosis. Meanwhile, changed TTC21A expression is associated with different proportions of immune cells, such as B cells, Neutrophils, Mast cells and T cells in LUAD. Therefore, TTC21A likely has an essential influence on immune infiltration and has the potential to serve as a prognostic biomarker of LUAD.
We would like to thank Professor Lihua Shen from College of Foreign Languages, Nantong University for language polishing.
Declaration of Competing Interest The authors declared that there is no conflict of interest.
6
International Immunopharmacology 78 (2020) 106077
W. Wang, et al.
and flagella, J. Cell Biol. 151 (3) (2000) 709–718. [7] B. Yasar, K. Linton, C. Slater, R. Byers, Primary cilia are increased in number and demonstrate structural abnormalities in human cancer, J. Clin. Pathol. 70 (7) (2017) 571–574. [8] A.M. Newman, C.L. Liu, M.R. Green, et al., Robust enumeration of cell subsets from tissue expression profiles, Nat. Meth. 12 (5) (2015) 453–457. [9] Z. Tang, C. Li, B. Kang, G. Gao, C. Li, Z. Zhang, GEPIA: a web server for cancer and normal gene expression profiling and interactive analyses, Nucl. Acids Res. 45 (Web Server issue) (2017). [10] H.R. Ali, L. Chlon, P.D. Pharoah, F. Markowetz, C. Caldas, Patterns of immune infiltration in breast cancer and their clinical implications: a gene-expression-based retrospective study, PLoS Med. 13 (12) (2016) e1002194. [11] X. Liu, S. Wu, Y. Yang, M. Zhao, G. Zhu, Z. Hou, The prognostic landscape of tumorinfiltrating immune cell and immunomodulators in lung cancer, Biomed. Pharmacother. 95 (2017) 55–61. [12] R.D. Bense, C. Sotiriou, M.J. Piccartgebhart, et al., Relevance of tumor-infiltrating immune cell composition and functionality for disease outcome in breast cancer, J. Natl. Cancer Inst. 109 (1) (2016) djw192. [13] N.O. Siemers, J.L. Holloway, H. Chang, Genome-wide association analysis identifies genetic correlates of immune infiltrates in solid tumors, PLoS ONE 12 (7) (2017) e0179726. [14] P. Danaher, S. Warren, L. Dennis, et al., Gene expression markers of tumor infiltrating leukocytes, J. ImmunoTher. Cancer 5 (1) (2017) 18. [15] S. Sousa, J. Määttä, The role of tumour-associated macrophages in bone metastasis, J. Bone Oncol. 5 (3) (2016) 135–138. [16] F. Azimi, R.A. Scolyer, P. Rumcheva, et al., Tumor-infiltrating lymphocyte grade is an independent predictor of sentinel lymph node status and survival in patients with cutaneous melanoma, J. Clin. Oncol. 30 (21) (2012) 2678–2683. [17] J. Kargl, S.E. Busch, G.H. Yang, et al., Neutrophils dominate the immune cell composition in non-small cell lung cancer, Nat. Commun. 8 (2017) 14381. [18] T. Powles, J.P. Eder, G.D. Fine, et al., MPDL3280A (anti-PD-L1) treatment leads to clinical activity in metastatic bladder cancer, Nature 194 (4) (2014) 558–562. [19] E.I. Buchbinder, D.F. Mcdermott, Cytotoxic T-lymphocyte antigen-4 blockade in melanoma, Clin. Ther. 37 (4) (2015) 755–763. [20] F. Massari, M. Santoni, C. Ciccarese, et al., PD-1 blockade therapy in renal cell carcinoma: current studies and future promises, Cancer Treat. Rev. 41 (2) (2015) 114–121. [21] S.M.A. Mahmoud, E.C. Paish, D.G. Powe, et al., Tumor-infiltrating CD8+ lymphocytes predict clinical outcome in breast cancer, J. Clin. Oncol. 29 (15) (2011) 1949–1955. [22] S.S. Wang, W. Liu, D. Ly, H. Xu, L. Qu, L. Zhang, Tumor-infiltrating B cells: their role and application in anti-tumor immunity in lung cancer, Cell. Mol. Immunol. 16 (1) (2019) 6–18. [23] M.C. Dieu-Nosjean, J. Goc, N.A. Giraldo, C. Sautès-Fridman, W.H. Fridman, Tertiary lymphoid structures in cancer and beyond, Trends Immunol. 35 (11) (2014) 571–580. [24] G.A. Banat, A. Tretyn, S.S. Pullamsetti, et al., Immune and inflammatory cell composition of human lung cancer stroma, PLoS One 10 (9) (2015) e0139073. [25] Y. Kurebayashi, K. Emoto, Y. Hayashi, et al., Comprehensive immune profiling of lung adenocarcinomas reveals four immunosubtypes with plasma cell subtype a negative indicator, Cancer Immunol. Res. 4 (3) (2016) 234–247. [26] M.J. Campa, M.A. Moody, R. Zhang, H.X. Liao, E.B. Gottlin, E.F. Patz, Interrogation of individual intratumoral B lymphocytes from lung cancer patients for molecular target discovery, Cancer Immunol. Immunother. 65 (2) (2016) 171–180. [27] P. Costantino, G.W. Jones, B. Michele, S.A. Jones, Ectopic lymphoid-like structures in infection, cancer and autoimmunity, J. Nat. Rev. Immunol. 14 (7) (2014) 447–462. [28] T.C. Bruno, P.J. Ebner, B.L. Moore, et al., Antigen-presenting intratumoral B cells affect CD4+ TIL phenotypes in non-small cell lung cancer patients, Cancer Immunol. Res. 5 (10) (2017) 898. [29] W. Zhu, C. Germain, Z. Liu, et al., A high density of tertiary lymphoid structure B cells in lung tumors is associated with increased CD4+ T cell receptor repertoire clonality, Oncoimmunology 4 (12) (2015). [30] D. Huangfu, A. Liu, A.S. Rakeman, N.S. Murcia, L. Niswander, K.V. Anderson, Hedgehog signalling in the mouse requires intraflagellar transport proteins, Nature 426 (6962) (2003) 83–87. [31] P. Tran, C. Haycraft, T. Besschetnova, et al., THM1 negatively modulates mouse sonic hedgehog signal transduction and affects retrograde intraflagellar transport in cilia, Nat. Genet. 40 (4) (2008) 403–410. [32] E.E. Davis, Z. Qi, L. Qin, et al., TTC21B contributes both causal and modifying alleles across the ciliopathy spectrum, Nat. Genet. 43 (3) (2011) 189–196. [33] S.Y. Wong, A.D. Seol, P.L. So, et al., Primary cilia can both mediate and suppress Hedgehog pathway–dependent tumorigenesis, Nat. Med. 15 (9) (2009) 1055–1061. [34] N.J. Rowbotham, A.L. Hager-Theodorides, M. Cebecauer, et al., Activation of the Hedgehog signaling pathway in T-lineage cells inhibits TCR repertoire selection in the thymus and peripheral T-cell activation, Blood 109 (9) (2007) 3757–3766. [35] N.J. Rowbotham, A.L. Hager-Theodorides, A.L. Furmanski, C. Tessa, A novel role for Hedgehog in T-cell receptor signaling: implications for development and immunity, Cell Cycle 6 (17) (2007) 2138–2142. [36] N.J. Rowbotham, A.L. Furmanski, A.L. Hager-Theodorides, et al., Repression of hedgehog signal transduction in T-lineage cells increases TCR-induced activation and proliferation, Cell Cycle 7 (7) (2008) 904–908.
Table 3 Correlation analysis between TTC21A expression and gene markers of CD8+ T cells, T cells (general), B cells, neutrophils, natural killer (NK) cells, T-helper 1 (Th1) cells, T-helper 2 (Th2) cells, follicular helper T (Tfh) cells, T-helper 17 (Th17) cells, exhausted T cells and Mast cells via “correlation” module of GEPIA. Description
Gene markers
LUAD Tumor
CD8+ T cell T cell (general) B cell Natural killer cell
Neutrophils
Th1
Th2
Tfh Th17 T cell exhaustion
Mast cells
CD8A CD8B CD2 CD3E CD19 CD79A KIR2DL1 KIR2DL3 KIR2DL4 KIR3DL1 KIR3DL2 KIR3DL3 KIR2DS4 CD66b CD11b CCR7 T-bet STAT4 TNF-α GATA3 STAT6 STAT5A IL13 BCL6 STAT3 IL17A PD-1 CTLA4 LAG3 TIM-3 TPSB2 TPSAB1 CPA3 MS4A2 HDC
Normal
R
P
R
P
0.019 −0.0066 0.091 0.13 0.21 0.13 0.034 0.014 −0.086 0.11 0.039 −0.054 0.028 0.15 0.21 0.3 0.25 0.17 0.18 0.14 0.41 0.27 0.19 0.39 0.2 −0.074 0.094 0.17 0.15 −0.023 0.21 0.21 0.13 0.25 0.42
0.68 0.88 0.045 * *** * 0.45 0.76 0.058 0.017 0.4 0.23 0.54 ** *** *** *** ** *** * *** *** *** *** *** 0.11 0.039 ** ** 0.61 *** *** * *** ***
−0.22 −0.27 −0.029 −0.062 0.18 0.036 −0.033 −0.014 −0.13 −0.066 −0.17 0.066 −0.0065 0.12 −0.0084 −0.0024 0.11 0.15 0.2 0.24 0.35 0.12 −0.26 0.24 0.2 0.14 −0.14 0.13 0.47 −0.35 −0.15 −0.23 −0.26 0.045 0.18
0.089 0.039 0.83 0.64 0.18 0.79 0.8 0.92 0.31 0.62 0.19 0.62 0.96 0.36 0.95 0.99 0.4 0.25 0.14 0.066 * 0.36 0.044 0.073 0.12 0.29 0.28 0.32 ** * 0.26 0.084 0.044 0.73 0.17
Tumor, correlation analysis in LUAD tumor tissue of TCGA. Normal, correlation analysis in LUAD normal tissue of TCGA. *P < 0.01, **P < 0.001, ***P < 0.0001.
Funding Training Program of Innovation and Entrepreneurship for College Students in Jiangsu (NO. 201913993001Y and 201913993021H). References [1] R.L. Siegel, K.D. Miller, A. Jemal, Cancer statistics, 2016, Ca A Cancer J. Clin. 60 (5) (2010) 277–300. [2] A. Eberle, L. Jansen, F. Castro, et al., Lung cancer survival in Germany: a population-based analysis of 132,612 lung cancer patients, Lung Cancer 90 (3) (2015) 528–533. [3] J.R. Molina, P. Yang, S.D. Cassivi, S.E. Schild, A.A. Adjei, -small cell lung cancer: epidemiology, risk factors, treatment, and survivorship, Mayo Clin. Proc. 83 (5) (2008) 584–594. [4] Z. Chen, C.M. Fillmore, P.S. Hammerman, C.F. Kim, K.K. Wong, Non-small-cell lung cancers: a heterogeneous set of diseases, Nat. Rev. Cancer 14 (8) (2014) 535–546. [5] M.D. Iglesia, J.S. Parker, K.A. Hoadley, J.S. Serody, C.M. Perou, B.G. Vincent, Genomic analysis of immune cell infiltrates across 11 tumor types, J. Natl. Cancer Inst. 108 (11) (2016) djw144. [6] G.J. Pazour, B.L. Dickert, Y. Vucica, et al., Chlamydomonas IFT88 and Its mouse homologue, polycystic kidney disease gene Tg737, are required for assembly of cilia
7