Skip to main content
NCBI home page
Translational Oncology logoLink to Translational Oncology
. 2023 Aug 5;36:101754. doi: 10.1016/j.tranon.2023.101754

Multi-Omics Analysis Elucidates The Immune And Intratumor Microbes Characteristics Of Ubiquitination Subtypes In Lung Adenocarcinoma

Siqi Wang a,b, Pei Liu a,b, Jie Yu a,b, Tongxiang Liu a,b,⁎
PMCID: PMC10423929  PMID: 37549605

Highlights

  • •

    Established a ubiquitination scoring model that can accurately predict the survival rate of patients with LUAD.

  • •

    Dividing LUAD patients into three ubiquitination subtypes through unsupervised clustering.

  • •

    Revealing the landscape of intratumor microbes of patients with LUAD.

  • •

    Evaluated the role of ubiquitination scoring model in immunotherapy.

Keywords: Ubiquitination, Lung adenocarcinoma, Prognosis, Immune microenvironment, Microorganisms, Immunotherapy

Abstract

Ubiquitination modification is closely related to cancer and participates in the regulation of tumor microenvironment. However, the role of ubiquitination modification in the immune response and prognosis of lung adenocarcinoma has not been elucidated. This study aims to establish a disease classification associated with ubiquitination and reveal the landscape of intratumor microbes in patients with lung adenocarcinoma for the first time. A total of 1314 patients with lung adenocarcinoma in the GEO and TCGA databases were included in our study. We constructed a ubiquitination scoring model using WGCNA and constructed ubiquitination subtypes using unsupervised clustering, analyzed the clinical characteristics, immune characteristics, and intratumor microbes characteristics, and screened out the relevant gene signatures, which were verified by RT-qPCR in human cancer cells. The results showed that the high ubiquitination subtype had poor prognosis, low degree of immune infiltration, high index of tumor stemness, and poor effect of immunotherapy. The subtypes with lower ubiquitination scores have better prognosis, higher tumor microenvironment score and better immunotherapy effect. The C2 subtype has high level of immune infiltration, lower intratumor microbes diversity and abundance, and good prognosis. The C3 subtype has low level of immune infiltration, higher intratumor microbes diversity and abundance, and poor prognosis. The C1 subtype has characteristics between C2 and C3. In summary, this paper constructs a scoring system and several subtypes based on ubiquitination genes, and analyzed the characteristics, which can help provide new methods for clinical treatment.

Introduction

Lung cancer, as one of the most dangerous malignant tumors to human health and life, has the fastest growth in incidence rate and mortality [1]. The etiology of lung cancer is still not completely clear. A large number of data show that long-term smoking is closely related to the occurrence of lung cancer [2]. Lung adenocarcinoma is a kind of lung cancer, belonging to non small cell carcinoma, accounting for 40% - 50% of the total number of lung cancer. Most of them originate from bronchial mucosal epithelium, and a few originate from mucous glands of the bronchus. Lung adenocarcinoma is more likely to occur in women and smokers. At present, the treatment methods of lung adenocarcinoma mainly include surgery, chemotherapy and radiotherapy, immunotherapy [[3], [4]] and targeted drug therapy [5], Immunotherapy is a novel cancer treatment method that activates or enhances the patients' immune system to combat cancer cells [6]. Immune checkpoint inhibitors (ICIs), as a method of immunotherapy, can block or weaken the function of immune checkpoints such as PD-1 [7], thereby activating the patients' immune system. However, some patients may develop tolerance to treatment, leading to weakened or ineffective treatment effects. Therefore, new biomarkers and prognosis models are urgently needed to improve the poor prognosis of lung adenocarcinoma patients.

Ubiquitination modification refers to the process in which ubiquitin (a kind of low molecular weight protein) molecules, under the action of a series of special enzymes, classify proteins in cells, select target protein molecules from them, and specifically modify target proteins. These special enzymes include ubiquitin-activating enzymes, ubiquitin-conjugation enzymes, ubiquitin-protein ligase [8] and degrading enzymes. After ubiquitination, proteins are recognized and degraded by proteasomes located in the cytoplasm and nucleus. Endogenous proteins are generally degraded through the ubiquitin-proteasome system (UPS) [9]. The abnormal ubiquitination modification will affect the process of cell cycle regulation, cell growth, proliferation, apoptosis, DNA repair and other cell signal transduction, which is closely related to the occurrence and development of malignant tumors, cardiovascular diseases, neurodegenerative diseases and other diseases. For example, P53 is a typical tumor suppressor protein, which is activated in the case of DNA damage or oncogene activation to induce cell cycle arrest and apoptosis, thereby inhibiting the growth and proliferation of cancer cells, and also promoting DNA damage repair [10]. Ubiquitin proteasome system (UPS) has become one of the targets for the treatment of cancer [11]. Inhibiting the ubiquitin proteasome system and regulating the related substrate proteins have become a new direction of drug research and development.

Tumor microbiome is a new research field. Recent research shows that microorganisms in tumors may have an impact on the development and treatment of cancer [12]. These microorganisms may promote tumor development through a variety of mechanisms, such as activating cancer cell signaling pathways, producing metabolites, interfering with the immune system, etc. [13]. However, more research is still needed to understand the exact role of these microorganisms in the development of cancer. In addition, tumor microbiome may also affect the effectiveness of tumor treatment. Some studies have shown that changes in microbiome may affect the sensitivity of tumor cells to chemotherapy and immunotherapy [14]. Therefore, the analysis of microbial composition may contribute to the development of personalized cancer treatment.

Methods

Data collection

Use the “GEO query” package to obtain the gene expression data and clinical data of GSE42127. the “TCGAbiolink” package [15] in the R to obtain the RNA sequencing (RNA-seq) data of TCGA-LUAD through the TCGA database, At the same time, clinical data, somatic mutation data, copy number variation (CNV) data, methylation data were obtained. Downloading the gene set related to ubiquitination from the Msigdb (https://www.gsea-msigdb.org/gsea/msigdb) database (Supplement Table 1). Downloading gencode.v41.annotation.gtf.gz data in the GENCODE database (www.gencodegenes.org/) to perform gene ID conversion, change Ensemble ID to gene Symbol. Obtaining the genes expression data and clinical data of LUAD on the ICGC online website (https://dcc.icgc.org/) as an external validation set. GSE91061 cohort are used for subsequent immunotherapy analysis. GSE91061 cohort gets from online website (https://www.ncbi.nlm.nih.gov/geo/).

Establishment of a Ubiquitination Scoring Model

The gene sets downloaded from MSigDB database are scored by GSVA analysis with “GSVA” R package, Carrying out WGCNA [16] (Weighted correlation network analysis) on the score results to find genes related to ubiquitination of proteins in LUAD, TCGA-LUAD samples are divided into training sets and test sets in a ratio of 7:3, and then in the train set, screening the gene markers related to ubiquitination through univariate cox and lasso regression analysis, and score LUAD patients according to the following formula. Ubiquitination Score = Lasso.coefgene[1]*Expgene[1]+Lasso.coefgene[2]*Expgene[2]+Lasso.coefgene[3]*Expgene[3]……+Lasso.coefgene[n]*Expgene[n]. And then dividing the patients into high and low ubiquitination groups according to the critical value of ROC (receiver operating characteristic) curve. The test set uses the same formula to calculate the ubiquitination score, and uses the same cut-off value as the training set to divide into high and low ubiquitination groups

Establishment of the nomogram

Based on the ubiquitination score and clinical information, we constructed a nomogram to predict the one-year, three-year and five-year survival rates of LUAD patients through “rms” R package, and evaluated the accuracy of the model according to the ROC curve and calibration curve

Unsupervised cluster analysis

Nonnegative Matrix Factorization (NMF) method and Consensus Clustering method [17] can be used for dimension reduction analysis and belongs to unsupervised clustering analysis. In biomedical and chemical research, it is often necessary to use computers to analyze and process experimental data [18]. NMF algorithm also provides a new efficient and fast way for processing these data. “NMF” R package and “Consensus Clustering” R package are used to cluster gene expression matrix related to ubiquitination.

Analysis of stemness index

We first used the human stem cell data in the Progenitor cell biology consortium (PCBC https://progenitorcells.org/) database as the training data, and used the one-class logistic regression (OCLR) machine learning algorithm to quantify the tumor sample stemness [19]. Extract the expression profile and weight of intersection genes between training data and LUAD samples, and use Spearman to calculate the correlation between weight and expression profile as the sample stemness index. The higher the stemness index, the lower the degree of differentiation.

Analysis of tumor microenvironment (TME)

MCPCOUNT, CIBERSORT, EPIC and XCELL were used for scoring, and the correlation with ubiquitination score was calculated. In the tumor microenvironment, immune cells and stromal cells are two main types of lung tumor cells. We use the “estimate” package to analyze the gene expression data, so as to predict the score of stromal cells and immune cells, and then predict the content of these two types of cells; Finally, we can calculate the tumor purity in each tumor sample

Immune checkpoint (IC), as the regulator of the immune system, is crucial to maintain the autoimmune tolerance and regulate the duration and range of the immune response of peripheral tissues [20]. We collected and sorted out 16 immune checkpoints from the literature [21] and analyzed the differential expression of immune checkpoints in the high and low ubiquitination groups. The TIP website provides seven signature sets related to Cancer-Immunity Cycle [22], we scored the seven steps with GSVA, and analyzed the difference between the scores in the high and low ubiquitination groups. TISIDB database is a database for tumor immunoassay, we obtained the gene signatures of 28 tumor-infiltrating lymphocytes from TISIDB database [23], and performed ssGSEA score.

Acquisition and processing of microbiome data

The microbiome sequencing data of LUAD was obtained from Poore et al., who used the microbiome analysis of blood and tissue samples as a cancer diagnosis method [24]. using the vegan R package for alpha diversity analysis, using the ggraph R package to draw the network map, and the pheatmap R package to draw the heat map.

Prediction of chemotherapy and immunotherapy

“oncoPredict” package [25] was used to predict the IC50 value of drugs, and calculate the correlation with intratumor microbes through the spearson correlation coefficient. Connectivity Map (cMap https://clue.io/) is a gene expression database created by researchers such as Harvard University [26]. It uses the gene expression differences after treating human cells with different disruptors (including small molecules) to establish a biological application database for the correlation of disruptors, gene expression and diseases. The cohorts of GSE91061 and Imvigor-210 are used to predict the immunotherapeutic effect of ubiquitination score. GSE91061 is a cohort of melanoma patients treated with immune checkpoint inhibitors. Imvigor-210 is a bladder cancer cohort receiving immunotherapy. We used TIDE to predict the immune response of the TCGA cohort, predict the probability of immune escape, and score immune dysfunction. IPS data is obtained from TCIA (https://tcia.at/home) to predict the treatment of patients with CTLA-4 and PD-1 immune checkpoint inhibitors.

Validation in vitro

A549 cells, BEAS-2B cells and PC9 cells are all provided by the School of Pharmacy of the Minzu University of China. RNA was extracted from lung adenocarcinoma cell A549 and human normal epithelial cell BEAS-2B with SG high purity total RNA extraction kit, and reverse transcription was performed using Thermo First cDNA Synthesis Kit, and then 2 × SG Green qPCR Mix (with ROX) reagent for Real-time Quantitative PCR. See Supplement Table 2 for primer sequence.

Statistical analysis

The Log-rank test is used to estimate the survival difference between different subtypes. Using bilateral test, p value less than 0.05 is considered statistically significant. Spearman correlation analysis is used to judge the correlation between two continuous variables with abnormal distribution. Kruskal-Wallis test, as a non-parametric test method, is used for the difference analysis of three or more samples. Wilcox test is used to analyze the difference between two samples with abnormal distribution.

Results

Construction of ubiquitination subtypes in GSE42127 cohort

We conducted univariate cox on multiple pathways including ubiquitination in GSE72094, GSE42127, GSE31210 and TCGA-LUAD cohorts, and found that ubiquitination was a risk pathway in these cohorts (Fig. 1A). We used the ubiquitination gene set to conduct unsupervised cluster analysis on the GSE42127 cohort. The patients were divided into three subtypes (Fig. 1B), The expression of ubiquitination gene set of the three subtypes was different (Fig. 1G). and the survival probability of Cluster.1 was low (Fig. 1D). PCA analysis showed that the overlap rate between the three subtypes was low and the resolution was good (Fig. 1C). The number of patient deaths in Cluster1 was high (Fig. 1E), and the score of ubiquitination pathway was also high (Fig. 1F).

Fig. 1.

Figure 1

Open in a new tab

(A) Univariate cox of pathways related to lung cancer. (B) Consensus cluster analysis of GSE42127 cohort. (C) PCA analysis of three subtypes of GSE42127 cohort. (D) Survival analysis of three subtypes of GSE42127 cohort. (E) Bar-plot of patient status for each subtype in GSE42127 cohort. (F) GSVA score of each subtype in GSE42127 cohort. (G) Heatmap shows the expression of ubiquitination pathway genes.

Establishment of the Ubiquitination Scoring Model

The genes related to ubiquitination in lung adenocarcinoma samples were obtained through WGCNA. Drawing sample cluster dendrogram (Supplement Fig. 1D) and gene cluster dendrogram (Supplement Fig. 1B), The soft threshold value of 0.85 was selected as the parameter for scale-free network construction, soft-thresholding powers is 4 (Supplement Fig. 1A) and the genes were allocated to a total of 21 modules (Supplement Fig. 1C), of which the genes of the brown module were the most relevant to the ubiquitination scoring model (Supplement Fig. 1E). After that, the brown module gene was analyzed by lasso regression, and the lambda-min gene was selected to obtain 15 gene signatures (Supplement Fig. 1F-G). Then the ubiquitination score was calculated according to the formula in the method. The higher the ubiquitination score, the more active the ubiquitin-proteasome system of the patient was. According to the critical value of 0.332 of the ROC curve (Supplement Fig. 1H). We applied the ubiquitination scoring model to the training set, test set and the ICGC validation set. The survival analysis showed that the median survival of the high ubiquitination subtype was lower than that of the low ubiquitination subtype (Fig. 2A,D,G), and with the increase of the ubiquitination score, the number of deaths of patients increased (Fig. 2B,E,H). The ROC curve and calibration curves shows that the Ubiquitination Scoring Model has a better ability to predict the prognosis of LUAD patients (Fig. 2C,F,I).

Fig. 2.

Figure 2

Open in a new tab

(A) Survival analysis of high and low ubiquitination subtypes in training set. (B) Ubiquitination score distribution of patients in training set. (C) ROC curve of Ubiquitination Scoring Model of training set. (D) Survival analysis of high and low ubiquitination subtypes in test set. (E) Ubiquitination score distribution of patients in test set. (F) ROC curve of Ubiquitination Scoring Model of test set. (G) Survival analysis of high and low ubiquitination subtypes in ICGC validation set. (H) Ubiquitination score distribution of patients in ICGC validation set. (I) ROC curve of Ubiquitination Scoring Model of ICGC validation set.

Clinical characteristics of high and low ubiquitination subtypes

We found that the ubiquitination score of clinical stage III and IV was significantly higher than that of clinical stage I (Fig. 3A), which is consistent with our previously predicted poor prognosis. There was no difference in the ubiquitination score in gender and age groups (Fig. 3B,D). In tumor grading, the ubiquitination score of T2 group was significantly higher than that of T1 group (Fig. 3C). The ubiquitination score of C3 in NMF group was significantly higher than that of C1 and C2 (Fig. 3E). High copy variation number (CNV) and high tumor mutational burden (TMB) were more likely to get higher ubiquitination score (Fig. 3F,G). Clinical analysis also shows that mutations in some key genes are also closely related to ubiquitination modification. We collected the five genes most prone to mutation in LUAD and divided the samples into wild type and mutant type. The ubiquitination score of TP53, FAT4 and SETBP1 mutant type was significantly higher than that of wild type (Fig. 3H-L). Whether there had been malignant tumors before did not affect the ubiquitination score (Fig. 3M). The treatment effect of high ubiquitination score was better than low ubiquitination score (Fig. 3N). The more serious the smoking, the higher the ubiquitination score (Fig. 3O). In order to further evaluate the role of ubiquitination modification in clinical practice, we conducted a univariate and multivariate cox analysis of the ubiquitination scoring model. The results showed that the ubiquitination scoring model was a significant risk factor (Fig. 3P-Q). We have also constructed a nomogram to predict the survival probability of patients for one year, three years, and five years (Fig. 3R). The results indicate that the survival probability of the high ubiquitination subtype is low, which is also consistent with our previous analysis. The ROC curve indicates that the nomogram has good prediction ability (Fig. 3S, Supplement Fig. 2A-C) .

Fig. 3.

Figure 3

Open in a new tab

(A) Boxplot of ubiquitination score about clinical stage group. (B) Boxplot of ubiquitination score about gender group. (C) Boxplot of ubiquitination score about tumor grade. (D) Boxplot of ubiquitination score about age group. (E) Boxplot of ubiquitination score about NMF subtype. (F) Boxplot of ubiquitination score about CNV subtype. (G) Boxplot of ubiquitination score about TMB subtype. (H) Boxplot of ubiquitination score about TP53 group. (I) Boxplot of ubiquitination score about KRAS group. (J) Boxplot of ubiquitination score about FAT4 group. (K) Boxplot of ubiquitination score about EGFR group. (L) Boxplot of ubiquitination score about SETBP1 group. (M) Boxplot of ubiquitination score about prior malignancy. (N) Boxplot of ubiquitination score about treatment group. (O) Boxplot of ubiquitination score about degree of smoking. (P) Univariate cox of clinical factors. (Q) Multivariate cox of clinical factors. (R) Construction of nomogram about clinical factors. (S) ROC curve of nomogram.

Characteristics of immune microenvironment of high and low ubiquitination subtypes

Some studies have shown that the high expression of ubiquitin specific proteinase 7 (USP7) can reduce the recognition and attack of immune cells on tumors, reduce the degree of immune infiltration of tumor cells in cervical cancer, and promote the development of tumors [27]. We analyzed the immune microenvironment scores of high and low ubiquitination subtypes and found that the Stromal Score, Immune Score, and ESTIMATE score of low ubiquitination subtypes were higher, and the tumor purity was lower (Fig. 4A-D). The abundance of natural killer cells and activated B cells in the low ubiquitination subtype is higher than that in the high ubiquitination subtype (Fig. 4E). Most immune cells exhibit varying degrees of negative correlation with gene signature (Fig. 4F), which may be due to high ubiquitination can induce apoptosis of immune cells [28]. CASP4 is considered a gene that promotes apoptosis of immune cells, and CASP6 is also considered a pro-apoptotic gene. We found that the expression of CASP4 and CASP6 in the high ubiquitination subtype was significantly increased, which led to more apoptosis of immune cells and decreased immune infiltration in the high ubiquitination subtype (Fig. 4G).

Fig. 4.

Figure 4

Open in a new tab

(A) Difference of Stromal Score between high ubi and low ubi subtype. (B) Difference of Immune Score between high ubi and low ubi subtype. (C) Difference of TumorPurity between high ubi and low ubi subtype. (D) Difference of Immune Score between high ubi and low ubi subtype. (E) The abundance of immune cells in high and low ubiquitination subtypes. (F) Correlation analysis between ubiquitination score, gene signatures and immune cells. (G) Differential expression of apoptosis-related factors in high and low ubiquitination subtypes.

Analysis about gene signatures

Further analysis of 15 gene signatures showed that except for PPP2CA, ARL4D, CTDNEP1 and EHPA2 genes, the remaining 11 genes were differentially expressed between the disease group and the normal group (Fig. 5A). EPHA2 had the highest frequency of somatic mutations (Fig. 5B). There is a complex interaction between these 15 gene markers (Fig. 5C). Most gene signatures present positive correlation, for example, PPIA and MALSU1 present strong positive correlation, and MESD and MALSU1 present weak negative correlation (Fig. 5D). MALSU1 and ARL4A were amplification genes in LUAD samples, while RGS20, SAC3D1, EPHA2, TNIP2, PTTG1IP, PPP2CA and JAGN1 were deletion genes (Fig. 5E). We selected the five gene signatures with the highest expression differences and validated their expression in vitro through RT-QPCR technology. The results showed that all of the five genes had differential expression (Fig. 5F).

Fig. 5.

Figure 5

Open in a new tab

(A) The expression difference of gene signatures between normal and tumor. (B) Waterfall plot of somatic mutation of gene signatures. (C) Interaction between gene signatures. (D) Correlation and significance analysis of 15 genes. (E) Copy number variations of gene signatures. (F) Difference expression of gene signatures between A549 cell and BEAS.2B cell.

Identification and analysis of NMF subtypes based on differential genes

According to the decomposition principle of non-negative matrix, we select the previous point (Cophenetic=3) with the largest coefficient decrease as the number of NMF subtypes (Fig. 6A,B). The survival analysis shows that the survival probability of C3 subtype is significantly lower than that of C1 and C2 (Fig. 6C). The clinical information also shows that the proportion of high ubiquitination subtypes, Stage III and Stage IV in C3 subtype is significantly higher than that of other subtypes, the primary site of tumor is not significantly different from other subtypes (Fig. 6D). There are also differences among the three NMF subtypes in other clinical characteristics such as age, tumor grade (Supplement Fig. 3A-H).

Fig. 6.

Figure 6

Open in a new tab

(A) Coefficient plot of NMF cluster analysis. (B) Consensus matrix of NMF cluster analysis. (C) Survival analysis of NMF subtypes. (D) Classification bar-plot of clinical information about NMF subtypes. (E) Differences in tumor microenvironment scores of NMF subtypes. (F) Survival analysis of ubiquitination subtypes stratified by NMF subtypes. (G) Survival analysis of Radiation treatment stratified by NMF subtypes. (H) Volcanic map of C2 vs C1 differential gene pathway enrichment. (I) Volcanic map of C3 vs C1 differential gene pathway enrichment.

The stemness index is an indicator to evaluate the similarity between tumor cells and stem cells, which is related to the degree of tumor dedifferentiation [29]. There is a significant difference between the tumor stemness of the three subtypes. The tumor stemness index of C3 subtype is the highest, and that of C1 subtype is the lowest (Supplement Fig. 3I,J). This means that tumors of C3 subtype have a high degree of dedifferentiation and a high degree of malignancy.

The immune characteristics of the three subtypes differ separately, The immune score and stromal score of C2 subtype are higher than those of the other two subtypes (Fig. 6E). Analysis of the abundance of immune cells in the three subtypes also showed that the abundance of immune cells in C2 was higher (Supplement Fig. 3K). Cancer Immune Cycle refers to the cycle process of interaction between tumor and immune system. It mainly includes tumor antigen expression, antigen presentation, T cell recognition, activation and expansion, T cell infiltration, tumor cell apoptosis and antigen presentation. The transport of immune cells to the tumor in C2 is more active than in C1 and C3, which may also be one of the reasons for the higher immune microenvironment score in C2 (Supplement Fig. 3L).

To further understand the situation of survival analysis, we conducted survival analysis for NMF subtype under a specific subtype. We found that in the C2 subtype of NMF, the median survival of the low ubiquitination subtype is longer (Fig. 6F), and the median survival after radiotherapy is shorter (Fig. 6G). In the low TMB subtype, the median survival of the C3 subtype of NMF is significantly lower than C1, while there is no difference in the high TMB subtype (Supplement Fig. 4A). In the low CNV subtype, the median survival of the C3 subtype of NMF is significantly lower than C1 and C2, and there is no difference in the high CNV subtype (Supplement Fig. 4B).

We conducted pathway enrichment analysis of differential genes for three NMF subtypes. Compared with C1, many up-regulated genes in C2 are enriched in phagocytosis, recognition, and endopeptidase activity pathway (Fig. 6H). Most up-regulated genes in C3 are enriched in nuclear division, and down-regulated genes are enriched in humoral immune response (Fig. 6I). The pathway enrichment analysis of C2 and C3 differential genes shows that there are many up-regulated genes in C2 and many of them are enriched in immune-related pathways (Supplement Fig. 4C).

Characteristics of tumor microorganisms in three subtypes

More and more studies show that microbes play an important role in the progress of cancer [30]. Tumor microorganisms are closely related to tumor microenvironment [31]. Therefore, it is necessary to analyze the microbial composition of the three NMF subtypes. We use alpha diversity to measure the diversity or richness of different species in each subtype. Beta diversity is used to measure the difference in microbial diversity among the three subtypes. Bray-Curtis distance is used as an indicator to measure beta diversity. The Shannon index, Simpson index and enrichment index of C3 subtype are significantly higher than those of the other two subtypes (Fig. 7A). Beta diversity analysis showed that the three subtypes could be well distinguished (Fig. 7B). There are differences in the distribution of different microbes in different NMF subtypes. Heatmap indicates differences in microbial abundance between high and low ubiquitination subtypes and NMF subtypes (Fig. 7D).

Fig. 7.

Figure 7

Open in a new tab

(A) Alpha diversity of three NMF subtypes. (B) Beta diversity of three NMF subtypes. (C) Boxplot of microbial abundance differences among three subtypes. (D) Microbes distribution of three NMF subtypes. (E) The relationship between intratumor microbes and immune escape and inflammatory factors in C1 subtype. (F) The relationship between intratumor microbes and immune escape and inflammatory factors in C2 subtype. (G) The relationship between intratumor microbes and immune escape and inflammatory factors in C3 subtype.

Overall, the microbial abundance of C2 is lower than that of C1 and C3. For example, the content of microorganisms such as Acidiplasma, Corynebacterium, and Atopodium in C2 is significantly lower than that of C1 and C3. (Fig. 7C). The microbial species diversity in the C3 subtype is higher than that in the C1 subtype.

Intratumor microbes may affect the prognosis of LUAD patients by affecting tumor immune escape and inflammatory factors

In order to further clarify the effect of intratumor microbes on the prognosis, we analyzed the correlation between microbial abundance and immune escape. Immune escape refers to the ability of cancer cells to escape attack and killing by the host immune system through a variety of mechanisms, which is one of the important mechanisms leading to tumor development and progression [32]. The results showed that the microbes can up-regulate CD47, IDO1, IDO2, down-regulate PTEN and other genes closely related to immune escape, and the C3 subtype is more complex and diverse than C1. The intratumor microbes may promote the immune escape of tumors, resulting in worse prognosis of LUAD patients.

Some inflammatory factors play an important role in tumor formation, progression and prognosis. For example, IL-1, IL-6 and TNF-α Such inflammatory factors can promote the proliferation and survival of tumor cells, and also inhibit the apoptosis and immune monitoring of tumor cells, thus promoting the formation and progress of tumor [33]. Compared with C1, the intratumor microbes in C3 subtype has a more significant effect on up-regulating inflammatory factors such as NFKB, STAT3, IL-2, which indicates that the intratumor microbes may lead to worse prognosis by promoting the formation of inflammatory factors (Fig. 7E-G).

Relationship between intratumor microbes and chemotherapy

Microbes will also affect the efficacy of chemotherapy. The presence of such microbes as Sulfolobus and Atopodium will reduce the efficacy of some drugs. Cyanotherce and loktanella may improve the efficacy of chemotherapy drugs. The same microbe may also have different effects on different drugs (Fig. 8).

Fig. 8.

Figure 8

Open in a new tab

Relationship between intratumor microbes and chemotherapeutic drugs.

Summary of clinical classification and prediction of immunotherapy for NMF subtypes and immune subtypes

In order to further clarify the correlation between various clinical parameters and the ubiquitination model, we ranked LUAD patients according to the ubiquitination score from small to large, and summarized the relevant clinical information. For example, the high ubiquitination subtype is closely related to the high EREG.mRNA subtype, but not to the primary site of the tumor (Fig. 9A).

Fig. 9.

Figure 9

Open in a new tab

(A) Summary of clinical information. (B) Differences in expression of immune checkpoints in high and low ubiquitination subtypes. (C) Difference in ubiquitination scores between immune response and non-response groups (D) Dysfunction differences between high and low ubiquitination subtypes. (E) Exclusion differences between high and low ubiquitination subtypes. (F-H) Ubiquitination scoring model for predicting immune response in cohort Imvigor-210 cohort. (I-K) Ubiquitination scoring model for predicting immune response in cohort GSE91061 cohort.

Immune-checkpoint blockers have significant therapeutic effects in anti-tumor treatment [34]. We first examined the expression differences of immune checkpoints in various subtypes. In high and low ubiquitination subtypes, there were significant differences in the expression of BTLA, CTLA4, CD80 and other immune checkpoints (Fig. 9B). The expression of most immune checkpoints in the C2 subtype is relatively high (Supplement Fig. 5A). The Ubiquitination score in the non-responder group is significantly higher than that in the responder group (Fig. 9C). The high ubiquitination subtype has a stronger immune rejection effect (Fig. 9D-E), and its immunotherapeutic effect may be worse, which is also consistent with previous analysis. To make our model more widely applicable, We predict the immunotherapeutic effect of the Ubiquitination Scoring Model in other immunotherapy cohorts. The results also show that the treatment effect of low ubiquitination subtypes is better, and proved that the model has a good ability to predict immunotherapy (Fig. 9F-K). IPS is used to predict the response of patients to the treatment of immune checkpoint inhibitors. Among NMF subtypes, patients with C2 subtype respond more positively to the treatment (Supplement Fig. 5B).

Drug sensitivity analysis

cMap database matches the corresponding small molecule compounds according to the different genes between high and low ubiquitination subtypes and annotates the mechanism of action (MoA) of small molecular compounds. The lower the score of the compound, the better predicted therapeutic effect of the compound. It is predicted that the treatment effect of RHO kinase inhibitor III, PD-169316 is better (Supplement Fig. 6).

Discussion

Lung cancer, as one of the most common malignant tumors, has always been one of the difficult problems in the medical field because of its high incidence rate and mortality. Ubiquitination is considered to be the key factor determining the fate of proteins, marking proteins and degrading them in proteasomes [35]. Previous studies have shown that ubiquitination is closely related to many diseases, including cancer [36]. However, few studies have used subtypes about ubiquitination to predict the prognosis of LUAD patients. Exploring the impact of ubiquitination on the immune microenvironment and the effect of immunotherapy, and the possible influence mechanism of tumor microorganisms will help to provide new ideas for the treatment of lung adenocarcinoma.

In this paper, a total of 1314 samples from four lung adenocarcinoma data sets were analyzed for risk pathways, and ubiquitination was found to be the common risk pathway. We first established three subtypes using unsupervised clustering in the GSE42127 cohort. Preliminary analysis showed that excessive ubiquitination modification may lead to poor prognosis. To further analyze the characteristics of ubiquitination subtypes, we used WGCNA in the TCGA-LUAD cohort to screen the module genes with the highest correlation with ubiquitination modification, and used univariate cox and lasso regression analysis to further obtain the gene signature to construct a ubiquitination scoring model. The ROC curve proves that the model has good predictive ability, and the nomogram shows that the survival rate of the high ubiquitination subtype is lower. There are differences in clinical and immune characteristics between high and low ubiquitination subtypes. Patients with high ubiquitination subtypes have a higher mortality rate. The high ubiquitination subtype has a lower degree of immune infiltration, which is likely because high ubiquitination promotes apoptosis of immune cells.

According to the differential genes of high and low ubiquitination subtypes, patients were also divided into three subtypes using unsupervised clustering, and the clinical characteristics, immune characteristics, and microbial characteristics within the tumor were analyzed for these three subtypes. The clinical stage of C3 subtype is later, mortality is higher, matrix score and immune score are the lowest, tumor stemness is the highest, and tumor microbial diversity and abundance are the highest. The C2 subtype has a lower clinical stage, lower mortality, the highest matrix and immune scores, and a lower overall abundance of tumor microorganisms. The clinical characteristics of C1 and C2 are similar, but the tumor has the lowest degree of stemness, and the remaining characteristics of C1 are between C2 and C3. Further analysis showed that intratumor microbes within the C3 subtype have a more significant role in promoting immune escape and the expression of inflammatory factors, which may also be one of the reasons for the high degree of malignancy of C3 subtype tumors. In the GSE42127 and TCGA-LUAD cohorts, we constructed ubiquitination subtypes using different methods based on ubiquitination related genes. The results showed that there were three ubiquitination subtypes in both lung cancer cohorts, and the three ubiquitination subtypes had different immune phenotypes and tumor microbial characteristics.

Currently, there are various tumor markers in lung adenocarcinoma, such as anthracene-9,10-dione (CA9), Neuron-Specific Enolase [37] and so on, which can be used as an auxiliary diagnostic tool for lung adenocarcinoma, but there are also limitations such as low specificity and poor sensitivity. The gene signature obtained in this study provides a new approach to solving this problem. Among the 15 gene signatures used to construct the ubiquitination scoring model, the enzyme encoded by PPIA gene is a member of the peptide-prolyl cis-trans isomerase (PPIase) family. PPIases enzyme catalyzes the cis-trans isomerization of proline imine peptide bonds in oligopeptides and accelerates the folding of proteins. Using CRISPR-Cas9 technology to knock down the PPIA gene or use small molecular inhibitors to inhibit PPIA makes multiple myeloma (MM) cells sensitive to proteasome inhibitors, further indicating that PPIA is an effective therapeutic target for multiple myeloma [38]. The protein encoded by RGS20 gene belongs to the family of regulator of G protein signaling (RGS) proteins. The manipulation of RGS20 expression significantly affects the cell viability in the PC cell model. Some studies have shown that RGS20 promotes the development of cancer by regulating the PI3K/AKT signal pathway in cancer cells [39]. The expression level of SAC3D1 [40,41], PPP2CA [42,43], PTTG1IP [44] genes is significantly increased in tumor cells and can be used as a cancer marker.

Microorganisms play an important role in the occurrence of diseases. The abundance of Atopobium is higher in the high ubiquitination subtype. Atopobium is a common bacterium in human flora. At present, some studies have shown that it may be involved in the occurrence of some diseases. For example, Atopodium may be associated with vaginitis, periodontitis, gastrointestinal diseases, urinary tract infections and other diseases [45]. The abundance of Lactococcus is relatively high in the low ubiquitination subtype. Lactococcus is a common lactic acid bacterium, which is widely used in the food industry. In addition, some studies have shown that Lactococcus may have certain effects on human immune system and intestinal health, and is closely related to the occurrence of inflammation [46]. In this study, a variety of intratumor microbes have shown correlation with immune regulation, indicating that intratumor microbes are inseparable from the occurrence of cancer. The intratumor microbes will also affect the efficacy of chemotherapy to a certain extent [14]. For example, some bacteria can degrade some chemotherapy drugs through the action of enzymes, making them lose efficacy. In addition, microorganisms can also promote the proliferation and metastasis of tumor cells, thereby further reducing the efficacy of chemotherapy. On the other hand, some microbes can also enhance the efficacy of chemotherapy. For example, some microbes can produce enzymes to enhance the absorption and metabolism of chemotherapy drugs. This may be one of the reasons why Cyanotherce improved the efficacy of Osimertinib in this study.

Immunotherapy aims to activate the immune system of cancer patients and kill cancer cells and tumor tissues by relying on their immune function [47]. Immune Checkpoint Inhibitors (ICIs) is a new and rapidly developing immunotherapy method [48]. At present, research has shown that the abnormal ubiquitination and de ubiquitination of PD-1/PD-L1 affect PD-1/PD-L1 mediated immunosuppression, and ubiquitination modification plays a certain regulatory role in immunotherapy [49]. To explore the impact of ubiquitination on immunotherapy, we applied the ubiquitination scoring model to the Imvigor-210 and GSE91061 cohorts, and the results showed that immunotherapy benefits for low ubiquitination subtypes were higher, which is also consistent with the results in the TCGA cohort.

The innovation of this study is analyzing the characteristics of intratumor microorganisms in lung cancer, which have great potential in cancer prevention and diagnosis. Our research shows that the composition of microorganisms within tumors may be related to the prognosis and survival rate of patients, studying the microorganisms within tumors can provide new biomarkers for predicting the prognosis and survival rate of patients [50]. Tumors are not only composed of malignant cells, but also contain rich microenvironment. By studying the microorganisms within tumors, we can better understand the complexity of tumors [51]. The study of microorganisms in tumors may provide new targets and strategies for microbial therapy, and develop more effective treatment methods.

Due to the exploratory stage of research on microorganisms within tumors, they may face some challenges and problems in their development, such as the complexity of microbial composition, which is very complex and includes multiple microbial species. There may be differences in the microbial composition between different tumor types and individuals, and the microbial composition may also change over time and treatment. This study only provides a preliminary explanation of the possible mechanism of microorganisms within tumors in lung cancer. The specific function of microorganisms in tumor development remains a complex issue, and deep functional research is needed to reveal the interaction mechanism between microorganisms and tumors. To solve these challenges, we need to comprehensively use a variety of strategies. We can develop more accurate and high-throughput technical methods, and use advanced gene sequencing technology, bioinformatics analysis methods, Single-cell sequencing and other technologies to study the composition of microorganisms in tumors. Conduct vitro experiments and animal models to study the specific regulatory mechanisms of microorganisms on tumor growth, immune response, etc.

The research on tumor microbiota will continue to flourish in the next five years, and some important achievements in tumor microbiota research may gradually be translated into clinical applications. At the same time, researchers may develop treatment strategies targeting tumor microbiota. It is possible to inhibit tumor growth or enhance the efficacy of immunotherapy by regulating microbial composition.

Overall, we have constructed a scoring system based on ubiquitination modification, and have proven the accuracy of this scoring system, screening out potential tumor markers. Three ubiquitination modified subtypes were detected in lung adenocarcinoma, and their clinical, immunological characteristics were analyzed. For the first time, we have found that intratumor microbes in lung adenocarcinoma can also have a negative impact on prognosis. At the same time, the immunotherapeutic effects of ubiquitination scoring model and three ubiquitination subtypes were analyzed, this paper provides new ideas and methods for the treatment of lung adenocarcinoma. This study has certain limitations. We did not rule out the impact of contaminants. Some studies have shown that the equivalent species in different samples are mainly contaminants [52]. Although we analyzed microbial differences in different subtypes, some residual contaminants may make the results more complex, In the future, more deep research can be conducted by combining scRNA-seq and 16sRNA sequencing.

Data availability

The datasets generated and/or analysed during the current study are available in the GDC repository (https://portal.gdc.cancer.gov/) and GEO repository (https://www.ncbi.nlm.nih.gov/geo/).

Ethical approval

The data used in this article are from online websites, so ethical approval is not required.

Author Contributions Statement

Siqi Wang conceived and designed the project; Siqi Wang, Pei Liu conducted data collection and analysis; Siqi Wang, Pei Liu, Jie Yu, Tongxiang Liu wrote articles.

Supplementary Materials

Supplement Fig. 1 (A) Selection of soft threshold. (B) Dendrograms corresponding to module genes. (C) Correlation heat map between modular genes and clinical traits. (D) Sample clustering dendrogram with clinical information. (E) genes in the brown module. (F)-(G) Distribution of lasso coefficients of gene signatures. (H) Selection of Cut-off Value.

Supplement Fig. 2 The calibration curves of nomogram in (A) 1 year (B) 3 years (C) 5 years

Supplement Fig. 3 (A)-(H) Bar-plot of classification of relevant clinical features in NMF subtypes. (I)-(J) Difference of stemness index of three NMF subtypes. (K) Immune cells Abundance of Three NMF Types. (L) Differences in Cancer Immunity Cycle among three NMF subtypes.

Supplement Fig. 4 (A) Survival analysis of NMF subtypes stratified by TMB subtypes. (B) Survival analysis of NMF subtypes stratified by CNV subtypes. (C) Volcanic map of C3 vs C2 differential gene pathway enrichment.

Supplement Fig. 5 (A) Differences in expression of immune checkpoints among three subtypes. (B) IPS predicts the efficacy of immunotherapy for three subtypes

Supplement Fig. 6 cMap analysis predicts potential therapeutic drugs

Supplement Table 1 Related pathways and corresponding gene sets

Supplement Table 2 Primer Information

Declaration of Competing Interest

The authors declare that they have no conflict of interest.

Acknowledgments

This research is supported by the National Natural Science Foundation of China (No. 81973977).

Footnotes

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.tranon.2023.101754.

Appendix. Supplementary materials

mmc1.docx (12.8MB, docx)
mmc2.xlsx (172.5KB, xlsx)

References

  • 1.Chen J., Fu Y., Hu J., et al. Hypoxia-related gene signature for predicting LUAD patients' prognosis and immune microenvironment. Cytokine. .2022;152 doi: 10.1016/j.cyto.2022.155820. [DOI] [PubMed] [Google Scholar]
  • 2.Tian Q., Zhou Y., Zhu L., et al. Development and Validation of a Ferroptosis-Related Gene Signature for Overall Survival Prediction in Lung Adenocarcinoma. Front Cell Dev Biol. 2021;9 doi: 10.3389/fcell.2021.684259. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Rizzo A., Cusmai A., Giovannelli F., et al. Impact of Proton Pump Inhibitors and Histamine-2-Receptor Antagonists on Non-Small Cell Lung Cancer Immunotherapy: A Systematic Review and Meta-Analysis. Cancers (Basel) 2022;14:6. doi: 10.3390/cancers14061404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Santoni M., Rizzo A., Mollica V., et al. The impact of gender on The efficacy of immune checkpoint inhibitors in cancer patients: The MOUSEION-01 study. Crit Rev Oncol Hematol. 2022;170 doi: 10.1016/j.critrevonc.2022.103596. [DOI] [PubMed] [Google Scholar]
  • 5.Chen P., Quan Z., Song X., et al. MDFI is a novel biomarker for poor prognosis in LUAD. Front Oncol. 2022;12 doi: 10.3389/fonc.2022.1005962. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Santoni M., Rizzo A., Kucharz J., et al. Complete remissions following immunotherapy or immuno-oncology combinations in cancer patients: the MOUSEION-03 meta-analysis. Cancer Immunol Immunother. 2023;72(6):1365–1379. doi: 10.1007/s00262-022-03349-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Rizzo A. Identifying optimal first-line treatment for advanced non-small cell lung carcinoma with high PD-L1 expression: a matter of debate. Br J Cancer. 2022;127(8):1381–1382. doi: 10.1038/s41416-022-01929-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Fujita Y., Tinoco R., Li Y., et al. Ubiquitin Ligases in Cancer Immunotherapy - Balancing Antitumor and Autoimmunity. Trends Mol Med. 2019;25(5):428–443. doi: 10.1016/j.molmed.2019.02.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Çetin G., Klafack S., Studencka-Turski M., et al. The Ubiquitin-Proteasome System in Immune Cells. Biomolecules. 2021;11:1. doi: 10.3390/biom11010060. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Kanapathipillai M. Treating p53 Mutant Aggregation-Associated Cancer. Cancers (Basel) 2018;10:6. doi: 10.3390/cancers10060154. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Park J., Cho J., Song E.J. Ubiquitin-proteasome system (UPS) as a target for anticancer treatment. Arch Pharm Res. 2020;43(11):1144–1161. doi: 10.1007/s12272-020-01281-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Nejman D., Livyatan I., Fuks G., et al. The human tumor microbiome is composed of tumor type-specific intracellular bacteria. Science. 2020;368(6494):973–980. doi: 10.1126/science.aay9189. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Gong Y., Huang X., Wang M., et al. Intratumor microbiota: a novel tumor component. J Cancer Res Clin Oncol. 2023;149(9):6675–6691. doi: 10.1007/s00432-023-04576-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Geller L.T., Barzily-Rokni M., Danino T., et al. Potential role of intratumor bacteria in mediating tumor resistance to the chemotherapeutic drug gemcitabine. Science. 2017;357(6356):1156–1160. doi: 10.1126/science.aah5043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Colaprico A., Silva T.C., Olsen C., et al. TCGAbiolinks: an R/Bioconductor package for integrative analysis of TCGA data. Nucleic Acids Res. 2016;44(8):e71. doi: 10.1093/nar/gkv1507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Langfelder P., Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008;9:559. doi: 10.1186/1471-2105-9-559. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Coleman S., Kirk P.D.W., Wallace C. Consensus clustering for Bayesian mixture models. BMC Bioinformatics. 2022;23(1):290. doi: 10.1186/s12859-022-04830-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Xu X., He P. Manifold Peaks Nonnegative Matrix Factorization. IEEE Trans Neural Netw Learn Syst. 2022 doi: 10.1109/TNNLS.2022.3212922. [DOI] [PubMed] [Google Scholar]
  • 19.Malta T.M., Sokolov A., Gentles A.J., et al. Machine Learning Identifies Stemness Features Associated with Oncogenic Dedifferentiation. Cell. 2018;173(2):338–354. doi: 10.1016/j.cell.2018.03.034. e15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Li B., Chan H.L., Chen P. Immune Checkpoint Inhibitors: Basics and Challenges. Curr Med Chem. 2019;26(17):3009–3025. doi: 10.2174/0929867324666170804143706. [DOI] [PubMed] [Google Scholar]
  • 21.Auslander N., Zhang G., Lee J.S., et al. Robust prediction of response to immune checkpoint blockade therapy in metastatic melanoma. Nat Med. 2018;24(10):1545–1549. doi: 10.1038/s41591-018-0157-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Xu L., Deng C., Pang B., et al. TIP: A Web Server for Resolving Tumor Immunophenotype Profiling. Cancer Res. 2018;78(23):6575–6580. doi: 10.1158/0008-5472.CAN-18-0689. [DOI] [PubMed] [Google Scholar]
  • 23.Ru B., Wong C.N., Tong Y., et al. TISIDB: an integrated repository portal for tumor-immune system interactions. Bioinformatics. 2019;35(20):4200–4202. doi: 10.1093/bioinformatics/btz210. [DOI] [PubMed] [Google Scholar]
  • 24.Poore G.D., Kopylova E., Zhu Q., et al. Microbiome analyses of blood and tissues suggest cancer diagnostic approach. Nature. 2020;579(7800):567–574. doi: 10.1038/s41586-020-2095-1. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
  • 25.Maeser D., Gruener R.F., Huang R.S. oncoPredict: an R package for predicting in vivo or cancer patient drug response and biomarkers from cell line screening data. Brief Bioinform. 2021;22:6. doi: 10.1093/bib/bbab260. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Lamb J., Crawford E.D., Peck D., et al. The Connectivity Map: using gene-expression signatures to connect small molecules, genes, and disease. Science. 2006;313(5795):1929–1935. doi: 10.1126/science.1132939. [DOI] [PubMed] [Google Scholar]
  • 27.Su D., Ma S., Shan L., et al. Ubiquitin-specific protease 7 sustains DNA damage response and promotes cervical carcinogenesis. J Clin Invest. 2018;128(10):4280–4296. doi: 10.1172/JCI120518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Zou W., Chen X., Shim J.H., et al. The E3 ubiquitin ligase Wwp2 regulates craniofacial development through mono-ubiquitylation of Goosecoid. Nat Cell Biol. 2011;13(1):59–65. doi: 10.1038/ncb2134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Crespo J., Sun H., Welling T.H., et al. T cell anergy, exhaustion, senescence, and stemness in the tumor microenvironment. Curr Opin Immunol. 2013;25(2):214–221. doi: 10.1016/j.coi.2012.12.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Schwabe R.F., Jobin C. The microbiome and cancer. Nat Rev Cancer. 2013;13(11):800–812. doi: 10.1038/nrc3610. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Jiang T., Yang T., Chen Y., et al. Emulating interactions between microorganisms and tumor microenvironment to develop cancer theranostics. Theranostics. 2022;12(6):2833–2859. doi: 10.7150/thno.70719. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Yang Y. Cancer immunotherapy: harnessing the immune system to battle cancer. J Clin Invest. 2015;125(9):3335–3337. doi: 10.1172/JCI83871. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Migliorini P., Italiani P., Pratesi F., et al. The IL-1 family cytokines and receptors in autoimmune diseases. Autoimmun Rev. 2020;19(9) doi: 10.1016/j.autrev.2020.102617. [DOI] [PubMed] [Google Scholar]
  • 34.Wang Z., Wu X. Study and analysis of antitumor resistance mechanism of PD1/PD-L1 immune checkpoint blocker. Cancer Med. 2020;9(21):8086–8121. doi: 10.1002/cam4.3410. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Shaid S., Brandts C.H., Serve H., et al. Ubiquitination and selective autophagy. Cell Death Differ. 2013;20(1):21–30. doi: 10.1038/cdd.2012.72. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Popovic D., Vucic D., Dikic I. Ubiquitination in disease pathogenesis and treatment. Nat Med. 2014;20(11):1242–1253. doi: 10.1038/nm.3739. [DOI] [PubMed] [Google Scholar]
  • 37.Dal Bello M.G., Filiberti R.A., Alama A., et al. The role of CEA, CYFRA21-1 and NSE in monitoring tumor response to Nivolumab in advanced non-small cell lung cancer (NSCLC) patients. J Transl Med. 2019;17(1):74. doi: 10.1186/s12967-019-1828-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Cohen Y.C., Zada M., Wang S.Y., et al. Identification of resistance pathways and therapeutic targets in relapsed multiple myeloma patients through single-cell sequencing. Nat Med. 2021;27(3):491–503. doi: 10.1038/s41591-021-01232-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Shi D., Tong S., Han H., et al. RGS20 Promotes Tumor Progression through Modulating PI3K/AKT Signaling Activation in Penile Cancer. J Oncol. 2022 doi: 10.1155/2022/1293622. 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Han M.E., Kim J.Y., Kim G.H., et al. SAC3D1: a novel prognostic marker in hepatocellular carcinoma. Sci Rep. 2018;8(1):15608. doi: 10.1038/s41598-018-34129-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Liu A.G., Zhong J.C., Chen G., et al. Upregulated expression of SAC3D1 is associated with progression in gastric cancer. Int J Oncol. 2020;57(1):122–138. doi: 10.3892/ijo.2020.5048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Galiger C., Dahlhaus M., Vitek M.P., et al. PPP2CA Is a Novel Therapeutic Target in Neuroblastoma Cells That Can Be Activated by the SET Inhibitor OP449. Front Oncol. 2022;12 doi: 10.3389/fonc.2022.744984. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Liang J., Huang Y., Yang C., et al. The effect of PPP2CA expression on the prognosis of patients with hepatocellular carcinoma and its molecular biological characteristics. J Gastrointest Oncol. 2021;12(6):3008–3021. doi: 10.21037/jgo-21-720. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Repo H., Gurvits N., Löyttyniemi E., et al. PTTG1-interacting protein (PTTG1IP/PBF) predicts breast cancer survival. BMC Cancer. 2017;17(1):705. doi: 10.1186/s12885-017-3694-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Ravel J., Moreno I., Simón C. Bacterial vaginosis and its association with infertility, endometritis, and pelvic inflammatory disease. Am J Obstet Gynecol. 2021;224(3):251–257. doi: 10.1016/j.ajog.2020.10.019. [DOI] [PubMed] [Google Scholar]
  • 46.Bermúdez-Humarán L.G., Aubry C., Motta J.P., et al. Engineering lactococci and lactobacilli for human health. Curr Opin Microbiol. 2013;16(3):278–283. doi: 10.1016/j.mib.2013.06.002. [DOI] [PubMed] [Google Scholar]
  • 47.Zhang Y., Zhang Z. The history and advances in cancer immunotherapy: understanding the characteristics of tumor-infiltrating immune cells and their therapeutic implications. Cell Mol Immunol. 2020;17(8):807–821. doi: 10.1038/s41423-020-0488-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Bagchi S., Yuan R., Engleman E.G. Immune Checkpoint Inhibitors for the Treatment of Cancer: Clinical Impact and Mechanisms of Response and Resistance. Annu Rev Pathol. 2021;16:223–249. doi: 10.1146/annurev-pathol-042020-042741. [DOI] [PubMed] [Google Scholar]
  • 49.Hu X., Wang J., Chu M., et al. Emerging Role of Ubiquitination in the Regulation of PD-1/PD-L1 in Cancer Immunotherapy. Mol Ther. 2021;29(3):908–919. doi: 10.1016/j.ymthe.2020.12.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Liu J., Zhang Y. Intratumor microbiome in cancer progression: current developments, challenges and future trends. Biomark Res. 2022;10(1):37. doi: 10.1186/s40364-022-00381-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Galeano Niño J.L., Wu H., LaCourse K.D., et al. Effect of the intratumoral microbiota on spatial and cellular heterogeneity in cancer. Nature. 2022;611(7937):810–817. doi: 10.1038/s41586-022-05435-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Dohlman A.B., Arguijo Mendoza D., Ding S., et al. The cancer microbiome atlas: a pan-cancer comparative analysis to distinguish tissue-resident microbiota from contaminants. Cell Host Microbe. 2021;29(2):281–298. doi: 10.1016/j.chom.2020.12.001. e5. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

mmc1.docx (12.8MB, docx)
mmc2.xlsx (172.5KB, xlsx)

Data Availability Statement

The datasets generated and/or analysed during the current study are available in the GDC repository (https://portal.gdc.cancer.gov/) and GEO repository (https://www.ncbi.nlm.nih.gov/geo/).

Articles from Translational Oncology are provided here courtesy of Neoplasia Press

RESOURCES