Abstract

Background

   Activating Transcription Factor 4 (ATF4) expression exhibits
   differential patterns across different types of tumors. Besides, the
   pathogenesis of breast cancer is complex, and the exact relationship
   between ATF4 and ATF4 remains uncertain.

Methods

   The analysis of ATF4 expression was conducted by utilizing The Cancer
   Genome Atlas (TCGA) pan-cancer data, while the gene expression profile
   of breast cancer was checked by the comprehensive database-Gene
   Expression Omnibus database. In order to gain a more comprehensive
   understanding of the specific cell types that exhibit ATF4 expression
   within the microenvironment of breast cancer, we conducted a
   single-cell analysis of ATF4 using two distinct datasets of human
   breast cancer ([31]GSE114717 and [32]GSE11088, respectively). The
   spatial distribution of ATF4 within a tissue was demonstrated based on
   datasets obtained from the Human Protein Atlas (HPA) and SpatialDB. The
   clinical prognostic significance of ATF4 was assessed by analyzing
   clinical survival data obtained from TCGA, [33]GSE4830, and
   [34]GSE25055 datasets. We used the R package clusterProfiler to carry
   out an enrichment analysis of ATF4. We assessed how ATF4 impacts the
   growth and movement of breast cancer cell lines. We manipulated ATF4
   levels using plasmid transfection techniques.

Results

   The expression of ATF4 was found to be suboptimal and demonstrated a
   significant correlation with enhanced disease-specific survival
   (p = 0.012) and overall survival (p = 0.032) in breast cancer as well
   as other malignancies. We conducted an analysis to investigate the
   interaction between the infiltration level of immune cells and the
   expression of ATF4, using samples obtained from TCGA with known immune
   cell infiltration scores. Furthermore, a notable positive correlation
   exists between the elevated expression of ATF4 and immune-related
   genomes, specifically those associated with chemokine as well as
   immunity. Subsequent examination revealed a notable augmentation in the
   cytodifferentiation of T cells into regulatory T (Treg) cells within
   tissues exhibiting elevated levels of ATF4 expression. ATF4 exhibits
   notable upregulation in the MDA-MB-231 cell, thereby exerting a
   substantial impact on cell proliferation and migration upon its
   knockdown. Conversely, the overexpression of ATF4 in the MCF7 Luminal A
   breast cancer cell line can also modulate cellular function.

Conclusions

   Our study suggests that ATF4 helps T cells differentiate into Treg
   cells in breast cancer. ATF4 can represent a clinically useful
   biomarker to predict the overall survival rate, especially in patients
   with different subtypes of breast cancer. Provide certain guidance
   value for the development of targeted drugs or inhibitors targeting
   ATF4.

   Keywords: ATF4, Breast cancer, MDA-MB-231, Treg, Biomarker

1. Introduction

   Breast cancer is the most common cancer in women worldwide. It is also
   the leading cause of cancer deaths in women [[35]1]. In recent years,
   the incidence of female breast cancer has risen sharply. The burden of
   the disease is increasing, China leads the world in the number of
   female breast cancer cases and deaths. It accounts for 17.6 % and
   15.6 % of the world's female breast cancer cases and deaths [[36]2].
   Currently, there is a lack of effective markers for early screening and
   diagnosing breast cancer. The advancement of high-throughput sequencing
   technology, has facilitated the identification of an increasing number
   of genes that serve as early tumor markers for breast cancer. However,
   to identify more key driver genes that can influence the immune
   microenvironment of breast cancer and breast cancer progression is
   still necessary. Recently, many researchers have found that ATF4
   expression is upregulated in a variety of tumors and is involved in
   processes related to the regulation of tumor progression [[37]3,[38]4],
   this implies that ATF4 could be a potential target to treat breast
   cancer.

   Activating transcription factor 4 (ATF4) is a member of the ATF/CREB
   family, also known as cyclic adenylic acid (cAMP)-response element
   binding protein 2 (CREB2) [[39]5]. The ATF family is a group of
   transcription factors that contain a basic leucine zip region (bZIP),
   which is associated with protein interactions [[40]5]. It was reported
   that ATF3 could enhance breast cancer-initiating cell features. ATF3
   could also provide feedback on TGFβ [[41]6]. Furthermore, ATF5 has the
   ability to activate PI3K/Akt pathway, which mediates the proliferation
   of tumor cells [[42]7].

   ATF4 plays an essential role in responses induced by stress signals
   such as hypoxia, amino acid deficiency, oxidative stress and
   endoplasmic reticulum stress (ERS) [[43]8]. A study found that ATF4
   increased the multiplication and movement of hepatoma carcinoma cells.
   It did this by increasing the amount of HKDC1 [[44]9], and the
   expression of ATF4 was negative and weakly positive in normal cervical
   tissue cells and strongly positive in cervical cancerous tissue
   [[45]10]. Upregulation of ATF4 expression inhibits the formation of
   extracellular matrix, resulting in reduced adhesion between laryngeal
   cancer cells, themselves and normal tissue cells, which in turn makes
   the tumor cells more invasive [[46]11]. Recently, there are some
   reports on the research of ATF4 regulating breast cancer. For example,
   in an in vivo mouse breast cancer model, ATF4 enhances macrophage
   recruitment through enhanced M-CSF expression and enhances HIF-1α
   expression to promote angiogenesis, thereby promoting breast tumor
   growth [[47]12]. In radiation resistant TNBC, activation of ATF4 can
   increase intracellular levels of reduced glutathione and clearance of
   reactive oxygen species after radiation, leading to radiation resistant
   [[48]13]. In addition, HER2 can facilitate the proliferation and
   migration of breast cancer cells. It upregulates the expression of
   ATF4-ZEB1 [[49]14].

   In this study, we assessed ATF4 expression in different tumor types. We
   analyzed three cohorts: the Cancer Genome Map (TCGA), Genotypic Tissue
   Expression (GTEx), and Gene Expression Omnibus (GEO). Additionally, we
   examined the connection between ATF4 expression and prognosis of breast
   cancer patients. Researchers observed that the expression of ATF4 in
   breast cancer tissues showed a lower level compared to that in adjacent
   and normal tissues. An inverse relationship was observed. The
   expression of ATF4 correlates with lower overall, disease-specific, and
   relapse-free survival rates in individuals with breast cancer.
   Furthermore, it is anticipated that ATF4 plays a role in the chemokine
   signaling pathway. Experts expected it to play a part in both the
   innate and adaptive immune systems. Given the significance of immune
   cell infiltration in relation to the overall survival of individuals
   with cancer, our study sought to investigate the relevance between ATF4
   expression and levels of immune cell infiltration. Our findings
   indicate that tissues exhibiting elevated ATF4 expression are
   associated with heightened infiltration of Treg cells, while
   infiltration of Th cells is diminished. In order to conduct a more
   comprehensive investigation into the expression of ATF4 in various
   subtypes, we employed three negative breast cancer (TNBC) cell lines,
   namely MDA-MB-231, as well as MCF7, a the Luminal A breast cancer cell
   line. Results showed that the expression of ATF4 was elevated in
   MDA-MB-231 cells. Subsequently, the knockdown of ATF4 resulted in a
   notable deceleration of cell proliferation and migration in these
   cells. Conversely, the overexpression of ATF4 in MCF7 cells induced a
   significant acceleration of cell proliferation and migration in this
   cell line. The findings of our research indicate that ATF4 exhibits
   potential roles in various subtypes of breast cancer, thereby
   elucidating the underlying mechanisms by which ATF4 regulates Treg
   cells and the associated cascade of inflammatory factors within the
   microenvironment of breast cancer.

2. Methods

2.1. Data acquisition and analysis

   We obtained clinical data and expression of ATF4 from the GTEx and TCGA
   pan-cancer datasets. This information was obtained from the UCSC Xena
   database, which is accessible in
   ([50]https://xenabrowser.net/datapages/). The complete names of tumor
   abbreviations used in TCGA are provided in [51]Supplementary Table 1.
   To assess ATF4 expression, we obtained tumor tissue samples data from
   the TCGA database. We obtained normal tissue samples by combining
   normal tissues from TCGA and GTEx databases. Analysis was conducted by
   R v4.0.3 software.

2.2. Single-cell analysis of ATF4

   Tumor Immune Single-cell Hub (TISCH) website
   ([52]http://tisch.comp-genomics.org/documentation/) provided
   Single-Cell Analysis of ATF4 expression. The heatmap, scatter diagrams,
   and violin plots were used to quantify the expression of ATF4 in major
   lineage in all cancers.

2.3. Immunohistochemistry

   We collected immunohistochemical staining data from human protein
   profiles ([53]https://www.proteinatlas.org/)Determine the specific
   expression of ATF4 based on different magnification factors. In
   addition, we collected the tissues of two patients and tumor tissues of
   three negative breast cancer mice, and detected the expression of ATF4
   in different breast cancer subtypes.

2.4. Prognosis anaylsis

   Prognosis anaylsis was performed using BEST tool
   ([54]https://rookieutopia.com/app_direct/BEST/) which provided multiple
   gene datasets in TCGA and GEO.

2.5. Correlation and enrichment analysis

   PPI network analysis based on GeneMANIA ([55]http://www.genemania.org)
   [[56]15], which identifies associated genes with ATF4. The genes most
   positively associated with ATF4 were selected for enrichment analysis.
   Gene Ontology (GO) analysis was performed using EnrichGO function in
   the R package “clusterProfiler”. was utilized to carry out Gene
   Ontology (GO) analysis. The Enrich Kyoto Encyclopedia of Genes and
   Genomes (KEGG) function of the R package “clusterProfiler” was utilized
   to conduct KEGG analysis.

2.6. Immune cell infiltration

   The immune cell infiltration scores of TCGA pan-cancer were previously
   obtained using CIBERSORT. We downloaded these scores. To estimate
   immune cell infiltration, we segregated samples from The Cancer Genome
   Atlas (TCGA) into two groups, high-expression ATF4 group and
   low-expression ATF4 group, on the basis of the mid-value expression
   level of ATF4.

2.7. Somatic mutation analysis

   We used RNA-sequencing expression (level 3) profiles, genetic mutation
   for breast cancer patients from the TCGA dataset
   ([57]https://portal.gdc.com). R package “maftools” were used to
   visualize the mutations [[58]16,[59]17].

2.8. Cell culture and plasmid transfection

   The cells utilized in this study were obtained from the Cell Resource
   Center of Shanghai Academy of Biological Sciences, which is the
   affiliation of the Chinese Academy of Sciences. Prior to their use, all
   cell lines were subjected to rigorous testing to confirm the absence of
   mycoplasma contamination. The 4T1, EMT6, MDA-MB-231 and MCF7 cell lines
   were cultivated in Dulbecco's Modified Eagle Medium (DMEM) supplemented
   with 10 % fetal bovine serum (FBS) at a temperature of 37 °C in a
   humidified incubator with 5 % carbon dioxide (CO[2]) concentration. The
   MCF10a cells were cultured in a specialized culture medium. The cell
   culture experiments were conducted in triplicate, with three replicates
   per group for each repetition. The MCF7 cells were transfected with the
   plasmid OE-ATF4, whereas the MDA-MB-231 cells were transfected with the
   plasmid SH-ATF4, resulting in the overexpression and knockdown of ATF4,
   respectively.

2.9. Western immunoblotting

   MCF10a, MCF7 and MDA-MB-231 cells were lysed in RIPA lysis buffer.
   Samples were resolved onto SDS-polyacrylamide gels and imprinted on
   PVDF membranes. Primary antibodies against anti-ATF4 (#11815, cell
   signaling) and GADPH (#2118, cell signaling) were used. Images were
   conducted with a C-Digit chemiluminescent Western blot scanner (LI-COR,
   Lincoln, USA).

2.10. Quantitative real-time RT-PCR

   We used TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA,
   USA) for total RNA extraction, on the basis of the instructions. The
   total RNA sample was subjected to reverse transcription, using a random
   primer and M-MLV Reverse Transcriptase (Takara). The SYBR Green I
   Master Mix reagent by ABI 7500 system (Applied Biosystem, Foster, CA,
   USA) was utilized to carry out quantitative real-time RT-PCR. Primers
   used as shown in [60]Supplementary Table 2. Results were calculated by
   using 2 − (ΔΔCt) method.

2.11. Cell proliferation and migration

   The cell proliferation experiment was detected using the CCK8 reagent
   kit, and the specific operation was carried out according to the
   operating instructions. The proliferation ability was evaluated based
   on the OD values measured at different time points. Cell migration
   experiment, also known as scratch experiment, detects the proportion
   and rate of cell migration over scratches within a certain period of
   time.

2.12. Animal experiments

   The study utilized female nude mice that were 5 weeks old, which were
   acquired from the Shanghai Model Organisms Center located in Shanghai,
   China. The mice were cared for and utilized in accordance with a
   protocol that received approval from the Ethics Committee of the East
   Hospital, which is affiliated with Tongji University in Shanghai,
   China.

   Mice were subcutaneous injected cell suspension (approximately
   5*10^4 cells) to induce breast cancer model. Totally 24 female mice
   (5-week old) were fall into four groups including 4T1-control (4T1),
   ATF4 knockdown in 4T1 (4T1-shATF4), EMT6-control (EMT6), and ATF4
   knockdown in EMT6 (EMT6-shATF4) at random. Tumor tissues were collected
   for further flow cytometry.

2.13. Statistical analysis

   The data are presented as means ± standard error (SE) or median
   (interquartile range, IQR), unless otherwise specified. All figures
   were generated using GraphPad Prism version 6.0 (Graph-Pad Software,
   San Diego, CA). Statistical analysis was conducted using SPSS 23.0
   (SPSS Inc, Chicago, IL, USA). Significant differences were assessed
   using either Student's t-test or one-way ANOVA, followed by the
   Mann-Whitney U test. Statistical significance was attributed to values
   of P < 0.05.

3. Results

3.1. Pan-cancer ATF4 expression analysis

   Initially, we evaluated the expression of ATF4 in a comprehensive
   dataset encompassing multiple cancer types obtained from The Cancer
   Genome Atlas (TCGA) and the Genotype-Tissue Expression (GTEx) project.
   The analysis demonstrated a decrease in ATF4 expression in 16 tumor
   types, namely ACC, BLCA, BRCA, CESC, COAD, ESCA, LAML, LUAD, LUSC, OV,
   PRAD, READ, SKCM, THCA, UCEC, and UCS. In contrast, its expression was
   high in CHOL, DLBC, GBM, HNSC, KIRC, LGG, LIHC, PAAD, TGCT and THYM
   ([61]Fig. 1A). Then we evaluated the RNA expression of ATF4 in various
   organs and found that ATF4 was expressed in all tissues without tissue
   specificity ([62]Fig. 1B). To explore the ATF4 expression in human
   breast cancer, we found that ATF4 was lower expressed in breast cancer
   when compare to normal tissues and para-carcinoma ([63]Fig. 1C).

Fig. 1.

   [64]Fig. 1
   [65]Open in a new tab

   Pan-cancer ATF4 expression analysis. (A) ATF4 expression in tumor and
   normal tissues in TCGA and GTEx pan-cancer data. (B) ATF4 expression in
   tumor cell lines. (C) Correlation between ATF4 expression and RNAss.

3.2. ATF4 expression in human breast cancer

   Using immunohistochemistry, we further validated ATF4 expression in
   breast cancer based on HPA datasets ([66]Fig. 2A left column).
   Furthermore, two patients' breast cancer samples were obtained for
   additional elucidation. The findings verified that, in contrast to
   normal breast and para-carcinoma tissues, ATF4 was not highly expressed
   in breast cancer tissues ([67]Fig. 2A middle column). The mouse triple
   negative breast cancer tissue was chosen for easy comparison due to the
   limited quantity of clinical triple negative breast cancer tissue
   samples. The findings demonstrated that mouse TNBC tumor tissue had
   comparatively high levels of ATF4 expression ([68]Fig. 2A right
   column).

Fig. 2.

   [69]Fig. 2
   [70]Open in a new tab

   Expression of ATF4 in single-cell level.(A) ATF4 expression in tumor
   and normal tissues in breast cancer. (B) ATF4 expression in tumor and
   para-carcinoma tissues in breast cancer. (C–E) ATF4 expression in
   single-cell level in human breast cancer.

3.3. Relationship between the expression of ATF4 expression and the prognosis

   In order to assess the usefulness of ATF4 in predicting the prognosis
   of patients with breast cancer, an analysis was conducted in BEST,
   which included several breast cancer datasets such as TCGA,
   [71]GSE4830, and [72]GSE25055, to examine the interrelationship between
   ATF4 expression and overall survival (OS), disease free survival (DSS),
   and recurrence-free survival (RFS). The results showed that lower
   expression of ATF4 was remarkable related to improved OS, DSS, and RFS
   in TCGA-BRCA (p = 0.032, 0.012 and 0.038 respectively) ([73]Fig. 3A–C).
   When examining various subtypes of BRCA (PAM50), there were notable
   differences in ATF4 expression among Basal, Her2, LumA, LumB, and
   normal breast cancer (p = 0.0042 and p < 0.001, respectively).
   Consistent with the results from TCGA-BRCA, lower ATF4 expression was
   significantly correlated with improved OS and RFS (p = 0.0044 and
   p < 0.001, respectively) ([74]Fig. 3D–G).

Fig. 3.

   [75]Fig. 3
   [76]Open in a new tab

   Expression of ATF4 in breast tissues. (A) Representative images of
   Hematoxylin staining showing different pathological manifestations in
   normal and malignant breast tissues. (B) Representative images of
   immunohistochemistry showing ATF4 expression in benign and malignant
   breast cancer tissues. (C) Expression of ATF4 in spatial level.

3.4. Correlation and enrichment analysis

   In order to ascertain the potential role of ATF4, along with its
   associated pathways, a correlation analysis was conducted between ATF4
   and various biological processes using TCGA dataset ([77]Fig. 4A).
   Additionally, a correlation analysis was performed between ATF4 and
   other genes using Gene MANIA ([78]Fig. 4B). The genes that exhibited
   the strongest positive association with ATF4 were chosen for enrichment
   analysis. We conducted additional analysis on the potential functional
   pathways using the cluster Profiler R package. The findings from the
   enrichment analysis of the “C2: Reactome gene sets” indicate that ATF4
   is predominantly linked to RNA polymerase II transcription, cellular
   responses to external stimuli, transcriptional regulation by TP53, and
   regulation of TP53 activity ([79]Fig. 4C).

Fig. 4.

   [80]Fig. 4
   [81]Open in a new tab

   Association between ATF4 expression and breast cancer prognosis. (A)
   Kaplan–Meier analysis of OS, DSS and RFS in TCGA-BRCA. Results with
   logrank p < 0.05 are shown. (B) ATF4 expression in different subtypes
   of breast cancer and OS in [82]GSE48390. (C) ATF4 expression in
   different subtypes of breast cancer and OS in [83]GSE25055.

   Functional enrichment and GO analysis showed that ATF4 was related to
   macromolecule biosynthetic process. ATF4 was also associated with
   cellular macromolecule biosynthetic process, positive regulation of
   macromolecule metabolic process, positive regulation of nitrogen
   compound metabolic process and positive regulation of macromolecule
   biosynthetic process ([84]Fig. 4D). Furthermore, an examination of the
   KEGG pathway analysis revealed gene sets that were enriched and
   interacted with in pathways related to cancer, infection by the human
   T-cell leukemia virus, Hepatitis B, viral carcinogenesis, and infection
   with the Kaposi sarcoma-associated herpesvirus ([85]Fig. 4E). These
   findings imply that ATF4 is connected to several pathways in breast
   cancer.

3.5. ATF4 expression was associated with T cell infiltration and
differentiation

   We evaluated the immune cell infiltration score of TCGA breast cancer.
   The analysis revealed high levels of infiltration for aDC, B cell, CD8T
   cell, cytotoxic cell, DC, T cell, Th1 cell, NKCD56dim cell, and Treg.
   Conversely, the high-ATF4 expression group exhibited low levels of
   infiltration for Eosinophil, mast cell, T helper cell, Tcm, Tgd, and NK
   cell ([86]Fig. 5A). Other immune cells, including macrophages, did not
   show significant differences in infiltration. Considering the
   functional aspects, indicating that high expression of ATF4 promotes
   the differentiation of T cells to Treg. The tumor's immunosuppressive
   state is related to this. These results suggest that breast cancer's
   immunosuppressive status is related to low-expressed ATF4. We further
   checked our results by the TIMER2 ([87]http://timer.cistrome.org/)
   database [[88][18], [89][19], [90][20]]. We found that ATF4 expression
   was related to the T cells infiltration level, including T cell and
   Treg ([91]Fig. 5B–C). For T cell differentiation, the expression of
   ATF4 was positively correlated with T cells, Th1, Th2 and Treg
   ([92]Supplementary Figs. 1A–S1D) and negatively correlated with Tcm and
   T helper cell ([93]Supplementary Figs. 1E–F). Furthermore, there was a
   significant correlation observed between ATF4 expression and various
   Treg markers in both pan-cancer and breast cancer. These markers
   include CD4, FOXP3, IL2RA, IL7R, TGFB1, and CTLA4, as depicted in
   [94]Fig. 5D–E and [95]Supplementary Figs. S2A–F. The findings of this
   study suggest that decreased ATF4 expression is linked to the
   infiltration of T cells and their differentiation into T helper cells
   or Treg cells within the immunosuppressive microenvironment of breast
   cancer. This regulatory mechanism plays a role in regulating the immune
   response.

Fig. 5.

   [96]Fig. 5
   [97]Open in a new tab

   Function and pathway enrichment analysis of ATF4 in breast cancer. (A)
   correlation analysis between ATF4 and other genes using GeneMANIA.
   (B)Significant Gene Ontology terms of thee hub genes most positively
   associated with ATF4 in “C2:Reactome gene sets”. (C)Significant Gene
   Ontology terms of thee hub genes most positively associated with ATF4,
   including biological processes (BP), molecular function (MF), and cell
   component (CC). (D) Significant KEGG pathways of the hub genes most
   positively associated with ATF4.

3.6. Single-cell analysis of ATF4 in breast cancer

   To further identify the cell types that express ATF4 in breast cancer
   microenvironment, we carried out the single-cell analysis of ATF4 in 2
   single-cell datasets of human breast cancer ([98]GSE114717 and
   [99]GSE11088, respectively). The TISCH web tool was utilized to
   generate a heatmap illustrating the relative expression levels of ATF4
   across various cell types ([100]Fig. 6A). The findings of this study
   revealed that ATF4 exhibited predominant expression in immune cells,
   particularly T proliferative cells, T regulatory cells, and CD8 T
   effector cells ([101]Fig. 6B–C).

Fig. 6.

   [102]Fig. 6
   [103]Open in a new tab

   Correlation between immune cell infiltration and ATF4 in breast cancer.
   (A) Correlation between immune cell infiltration and ATF4 expression,
   right represents positive correlation; the left represents negative
   correlation. (B) Correlation between ATF4 expression and T cell and
   Treg infiltration. (C) Correlation between ATF4 expression and Th1,
   Th2, Treg infiltration. (D) Correlation between ATF4 expression and
   IL7R and IL2A in breast cancer.

3.7. Somatic mutation analysis of ATF4 in triple negative breast cancer

   Considering the high incidence rate and mortality of TNBC, and the high
   expression of ATF4 in TNBC, it is proposed that ATF4 may promote tumor
   progression and metastasis by regulating Treg cells in TNBC. We next
   mainly focused on the ATF4 expression in TNBC. By comparing the
   non-tumorigenic epithelial cell MCF10a, human breast cancer cell MCF7
   and triple negative breast cancer cell MDA-MB-231, it was found that
   the expression of ATF4 in MDA-MB-231 was notably higher than that in
   the other two cell lines, both at mRNA and protein levels ([104]Fig.
   6D–E).

3.8. Knockdown or overexpression of ATF4 can significantly change the
proliferation and migration of breast cancer cells

   To further verify the effect of ATF4 on breast cancer, we selected the
   cell line MCF7, which represents LuminalA, and the cell line
   MDA-MB-231, which represents TNBC. MCF7 overexpressed ATF4 and
   MDA-MB-231 knocked down ATF4 through plasmid transfection, respectively
   ([105]Fig. 7A–B, 7D-7E). Subsequently, the effect of ATF4 expression on
   the proliferation of MCF7 and MDA-MB-231 cells was detected. The
   results showed that after overexpression of ATF4 in MCF7, the
   proliferation rate increased. While after knocking down ATF4, the
   proliferation rate of MDA-MB-231 cells was obviously lower than that of
   the control group ([106]Fig. 7C and F). In order to further explore the
   effect of ATF4 on the migration of TNBC, we used mouse triple negative
   breast cancer cell lines 4T1 and EMT6 to construct a stable viral
   strain that stably knocks down ATF4. The lower the expression of ATF4,
   the slower the migration rate ([107]Fig. 8).

Fig. 7.

   [108]Fig. 7
   [109]Open in a new tab

   Somatic mutation analysis of ATF4. (A) Somatic mutation rate of ATF4 in
   breast cancer. (B) Top ten genes and ATF4 mutation in different groups.
   (C) The mRNA levels of ATF4 in MCF10a, MCF7, MDA-MB-231 cells. (D–E)
   The protein levels of ATF4 (left) and the quantity of the protein level
   (right). * indicates the significant difference compared with MCF10a,
   p < 0.05.

Fig. 8.

   [110]Fig. 8
   [111]Open in a new tab

   Changes in cell migration rate after knocking down ATF4. Comparison of
   cell migration rates of mouse triple negative breast cancer cell lines
   4T1 and EMT6 within 24 h before and after ATF4 is knocked down by
   lentivirus.

3.9. ATF4 regulates the progression of TNBC by regulating of treg cells

   Using the above virus stable transfection strains, mice were
   subcutaneously tumorigenic and divided into four groups: 4T1 group,
   4T1-shATF4 group, EMT6 group, and EMT6-shATF4 group. Retain tumor
   tissue for flow cytometry detection to analyze the composition and
   proportion of immune cells in the tumor tissue. Through comparative
   analysis, it was found that the results were close to the public
   database, that is, in the tumor tissue of TNBC, the expression of ATF4
   was positively related to CD4, CD8, and Treg cells, but there was only
   a statistically significant difference with Treg cells.; There was a
   negative correlation with neutrophils, macrophages, and NK cells, and
   there was no statistical difference ([112]Fig. 9A–H).

Fig. 9.

   [113]Fig. 9
   [114]Open in a new tab

   Correlation between Tregs and ATF4 in TNBC. (A–H) Retain tumor tissue
   for flow cytometry detection to analyze the composition and proportion
   of immune cells in the tumor tissue. Using the above virus stable
   transfection strains, mice were subcutaneously tumorigenic and divided
   into four groups: 4T1 group, 4T1-shATF4 group, EMT6 group, and
   EMT6-shATF4 group. Pearson p < 0.05.

4. Discussion

   The role of ATF4 in the pathogenesis of numerous diseases remains
   incompletely understood, particularly in relation to its involvement in
   different types of tumors, where research efforts have been limited.
   Hence, there is an imperative need to elucidate the precise function of
   ATF4 in the prognosis, advancement, and therapeutic approaches of
   cancer. Prior research has documented divergent functions of ATF4 in
   various types of neoplasms. For instance, in pancreatic ductal
   adenocarcinoma, the perk/eIF2/ATF4 signaling pathway is activated by
   knockout of pum1 gene to promote apoptosis and inhibit the growth,
   invasion and metastasis of tumor cells [[115]21]. Overexpression of
   ATF4 can inhibit tumor proliferation in prostate cancer [[116]22]. On
   the other hand, ATF4 promotes tumor growth of endometrial carcinoma by
   inducing CCL2 expression and macrophage recruitment [[117]23].

   Based on the findings of our study, the expression levels of ATF4 and
   its prognostic significance in pan-cancer were assessed using data from
   TCGA and GTEx obtained from UCSC Xena. Our analysis revealed that ATF4
   exhibited significantly higher expression levels in CHOL, DLBC, GBM,
   HNSC, KIRC, LGG, LIHC, PAAD, TGCT and THYM, compared to normal tissues.
   Conversely, lower expression levels of ATF4 were observed in ACC, BLCA,
   BRCA, CESC, COAD, ESCA, LAML, LUAD, LUSC, OV, PRAD, READ, SKCM, THCA,
   UCEC and UCS. ATF4 expression levels vary across tumor types. This may
   indicate different functions and mechanisms.

   Tumor microenvironment (TME) is composed of host stromal cells and
   tumor cells, including immune system components (such as macrophages
   and lymphocytes), vascular endothelial cells, fibroblasts, muscle
   fibroblasts, mesenchymal stem cells, adipocytes and extracellular
   matrix (ECM) [[118]24]. More and more evidence supports
   clinic-pathological significance of tumor microenvironment immune cells
   to predict the survival status and curative effect of tumor patients
   [[119]25]. Specifically, Regulatory T (Treg) cells expressing Foxp3 are
   essential to prevent autoimmunity and inhibit effective tumor immunity.
   Treg cells infiltrate into tumor tissue, which is usually related to
   the poor prognosis of cancer patients. Removal of Treg cells can
   enhance the anti-tumor immune response, but it may also trigger
   autoimmunity [[120]26]. GO results showed that ATF4 was primarily
   associated with RNA polymerase II transcription, cellular responses to
   external stimuli, transcriptional regulation by TP53 and regulation of
   TP53 activity according to “C2: Reactome gene sets”. In addition, Kyoto
   Encyclopedia of Genes and Genomes (KEGG) pathway analysis indicated an
   enrichment and crosstalk of the hub genes in cancer, human T-cell
   leukemia virus 1 infection and Hepatitis B. Furthermore, the KEGG
   pathway analysis revealed that the hub genes exhibited enrichment and
   crosstalk in various biological processes, including cancer, human
   T-cell leukemia virus 1 infection, and Hepatitis B. Through the
   examination of the association between ATF4 and the infiltration of
   immune cells, it was observed that the group exhibiting high expression
   of ATF4 displayed a notably elevated level of Treg infiltration.
   Conversely, within this group, a decreased level of T helper cell
   infiltration was observed in the context of breast cancer. Furthermore,
   the observed correlation between ATF4 and the expression of
   immune-suppressive genes suggests that ATF4 plays a crucial role in the
   regulation of tumor immunology. In conclusion, it is highly probable
   that ATF4 plays a significant role in the infiltration of immune cells
   and holds potential as a valuable prognostic biomarker for
   triple-negative breast cancer.

   However, in this study, several limitations have been identified. A
   systematic bias in our study was inevitable since we used open
   databases to collect and analyze sequencing data, although we performed
   some wet experiments to verify the ATF4 expression in breast cancer,
   more experiments like the regulation of ATF4 to T-cell differentiation
   need to be conducted. In order to investigate mechanisms and apply
   clinically, well-designed experiments are required.

5. Conclusions

   The findings of our study indicate that ATF4 plays a role in the
   differentiation of T cells into Treg cells in the context of
   triple-negative breast cancer. The ATF4 protein has the potential to
   serve as a valuable biomarker in the clinical setting, as it can be
   used to predict the overall survival outcomes of patients with
   triple-negative breast cancer.

Ethics approval and consent to participate

   Considering that the data are publicly available, no ethics approval is
   required for this study.

Consent for publication

   All authors reviewed and approved the manuscript.

Data availability statement

   Data associated with my study has not been deposited into a publicly
   available repository. The datasets generated and/or analyzed during the
   current study are available from the corresponding author upon
   reasonable request.

Funding

   This work was supported by the National Natural Science Foundation of
   China (82,073,387).

CRediT authorship contribution statement

   Lujing Shao: Writing – original draft, Data curation. Zhounan Zhu:
   Software, Formal analysis. Xinyan Jia: Writing – review & editing,
   Validation. Yabin Ma: Validation, Supervision. Chunyan Dong: Funding
   acquisition, Conceptualization.

Declaration of competing interest

   The authors declare that they have no known competing financial
   interests or personal relationships that could have appeared to
   influence the work reported in this paper.

Acknowledgements