Abstract
Non-small cell lung cancer (NSCLC) is the most common type of lung cancer, and multiple evidence has confirmed that C5a production is elevated in NSCLC microenvironment. Although NSCLC cell proliferation induced by C5a has been reported, the involved mechanism has not been elucidated. In this study, we examined the proliferation-related genes (i.e., KLF5, GCN5, and GDF15) and C5a receptor (C5aR) expression in tumor tissues as well as C5a concentration in plasma of NSCLC patients, and then determined the roles of KLF5, GCN5, and GDF15 in C5a-triggered NSCLC cell proliferation and the related mechanism both in vitro and in vivo. Our results found that the expression of KLF5, GCN5, GDF15, C5aR, and C5a was significantly upregulated in NSCLC patients. Mechanistic exploration in vitro revealed that C5a could facilitate A549 cell proliferation through increasing KLF5, GCN5, and GDF15 expression. Besides, KLF5 and GCN5 could form a complex, binding to GDF15 promoter in a KLF5-dependent manner and leading to GDF15 gene transcription. More importantly, GCN5-mediated KLF5 acetylation contributing to GDF15 gene transcription and cell proliferation upon C5a stimulation, the region (−103 to +58 nt) of GDF15 promoter which KLF5 could bind to, and two new KLF5 lysine sites (K335 and K391) acetylated by GCN5 were identified for the first time. Furthermore, our experiment in vivo demonstrated that the growth of xenograft tumors in BALB/c nude mice was greatly suppressed by the silence of KLF5, GCN5, or GDF15. Collectively, these findings disclose that C5a-driven KLF5–GCN5–GDF15 axis had a critical role in NSCLC proliferation and might serve as targets for NSCLC therapy.
Introduction
Complement participates in the processes of inflammatory diseases and malignant tumors [1–3]. Recent studies have revealed that C5a is associated with tumor growth [4, 5], and C5a in tumor microenvironment not only acts as a leukocyte chemoattractant [4, 5], but also promotes tumor cell proliferation [6–8]. Non-small cell lung cancer (NSCLC) is the most common type of lung cancer [9]. Although many researchers have uncovered that inflammatory cytokines or mediators, e.g., C5a are involved in NSCLC carcinogenesis and proliferation [10–12], the mechanism of C5a governing NSCLC cell proliferation remains largely unclear.
It is well known that cell proliferation is associated with the upregulation of transcription factors, transcriptional co-activators and pro-proliferation molecules in response to extracellular stimuli [13, 14]. Reportedly, kruppel-like factor 5 (KLF5), as a transcription factor, can boost breast cancer cell proliferation [15], and activate sox4 or HIF-1α transcription via binding to GC-rich DNA sequences, resulting in lung cancer proliferation [16]. Moreover, as a transcriptional co-activator, general control non-depressible (GCN5) potentiates NSCLC growth by increasing E2F1 and cyclin D1 expression [17] and accelerates glioma development as well [18]. Besides, growth differentiation factor 15 (GDF15) also tends to be an oncoprotein contributing to cancer cell proliferation [19–21]. Given that the pro-proliferation function of KLF5, GCN5, and GDF15 has been confirmed, and the earlier stage of our study also found that KLF5, GCN5, GDF15, C5a and C5aR expression increased in NSCLC patients, how C5a triggers NSCLC cell proliferation and the expression of KLF5, GCN5, or GDF15, and what impact of KLF5, GCN5, or GDF15 on NSCLC cell proliferation in response to C5a and the corresponding mechanism need to be clarified.
Acetylated transcription factors have a pivotal role in the development of cancer or other diseases [22–25]. GCN5, which has acetyltransferase activity, can acetylate transcription factors such as E2F1 [17] or c-Myc and regulate target gene expression in hepatocellular carcinoma and colon cancer [26, 27]. However, the role of KLF5 acetylation mediated by GCN5 in regulating KLF5 function and C5a-induced NSCLC cell proliferation are still unexplored.
In this study, we not only examined the levels of KLF5, GCN5, GDF15, C5aR, or C5a in NSCLC samples, but also evaluated the roles of KLF5, GCN5 and GDF15 in C5a-triggered cell proliferation in vitro and xenograft tumor growth in vivo. Additionally, we also assessed the relationship between these genes and mechanism of them in promoting NSCLC cell proliferation exposed to C5a.
Results
Examination of proliferation-related genes, C5a and analysis of correlation between KLF5, GCN5, GDF15, or C5aR expression, and clinic-pathological data in NSCLC patients
First, we collected fresh tumor and adjacent tissues from 12 NSCLC patients, and total RNAs of every 4 tumor tissues were pooled. Subsequently, transcriptome sequencing was performed, and RNA-seq discovered that 411 genes in T1–T4, 542 genes in T5–T8, and 403 genes in T9–T12 were upregulated (≥2-fold) compared to paired adjacent tissues. Venn diagram showed that 86 genes were co-elevated in all samples (Fig. 1a). Meantime, gene ontology (GO) analysis found that 17.4% of theses 86 genes was involved in cell proliferation, growth, and maintenance (Fig. 1b). Next, we measured the mRNA levels of 12 proliferation-related genes in 40 NSCLC fresh tumor tissues. Real-time PCR exhibited that the mRNA levels of KLF5, HMGA1, FOXM1, EHF, HMGB3, SOX4, SOX9, GCN5, GDF15, MDK, TDGF1, and cyclin D1 were remarkably elevated, especially KLF5, GCN5, and GDF15 (Supplementary Figure 1a). Additionally, KLF5, GCN5 and GDF15 were positively correlated with each other (Supplementary Figure 1b). Subsequently, we observed marked increase of KLF5, GCN5, and GDF15 expression in 185 NSCLC tumor tissues by immunohistochemical (IHC) staining (Fig. 1c, d). To explore the association of C5a with NSCLC proliferation, we detected the plasma C5a of 40 NSCLC patients, and confirmed that C5a level was significantly upregulated (Fig. 1e). IHC also showed the high expression of C5aR (CD88) in tumor tissues of the above-mentioned 185 patients (Fig. 1f, g). Furthermore, we analyzed the correlation between KLF5, GCN5, GDF15, or C5aR expression and clinic-pathological parameters of the 185 NSCLC patients, and found that the expression of these proteins was tightly linked with the tumor size, lymph node metastasis, and TNM stage, but not related to tumor type, patient age and sex (Supplementary Tables 3 to 6). These findings suggest a certain connection between the expression of these genes and NSCLC tumorigenesis.
C5a induces NSCLC cell proliferation and KLF5, GCN5, or GDF15 expression via binding to C5aR
To investigate the mechanism of C5a-induced cell proliferation in NSCLC tumorigenesis, we detected C5aR in 6 human NSCLC cell lines, and found that C5aR expression was prominently upregulated in A549 and PC9 cell lines (Fig. 2a). Thereafter, we stimulated A549 and PC9 cells with recombinant C5a, and examined the cell proliferation by CCK8, cell counting and colony formation, and found that the cell viability, cell counts, and colony numbers were obviously increased when C5a concentration reached 25 or 250 ng/ml, particularly 25 ng/ml (Fig. 2b, d).
To verify that C5a can upregulate KLF5, GCN5, and GDF15 expression, A549 cells were stimulated with 25 ng/ml C5a, and then the mRNA and protein of the three genes were determined by real-time PCR and immunoblotting (IB). It displayed that mRNA and protein of these genes started to increase at 1 h upon C5a, peaked at 2 h and 3 h (GCN5 at 6 h, Fig. 2e, f).
To determine that the cell proliferation and KLF5, GCN5, or GDF15 expression in response to C5a are associated with C5a binding to C5aR, A549 cells were treated with DMEM, C5a, DMSO + C5a, C5aR inhibitor (C5aR-I, W54011) + C5a (C5aR-I + C5a), IgG + C5a, or Anti-C5a + C5a (the data of C5aR-I and Anti-C5a dose determination were not shown). We observed that C5aR-I and anti-C5a notably decreased cell viability (72 h), cell number (7d), colony number (14d) as well as KLF5, GCN5, and GDF15 expression (Supplementary Figure 2a–e). These indicate that the C5a-induced cell proliferation and KLF5, GCN5, or GDF15 expression require C5a-C5aR binding.
Effect of KLF5, GCN5, or GDF15 expression on C5a-driving cell proliferation including their relationship with each other
Given that the KLF5, GCN5, and GDF15 expression was earlier than cell proliferation in response to C5a, we suspected that these proteins might regulate cell proliferation. To elucidate the role and relationship of KLF5, GCN5, and GDF15, we constructed corresponding overexpression or interference plasmids (Supplementary Figure 3a and b), and confirmed that KLF5, GCN5, or GDF15 overexpression improved A549 cell proliferation, while knockdown of them inhibited the C5a-triggered cell proliferation (Fig. 3a–d), suggesting that the C5a-induced upregulation of these genes has an important role in A549 cell proliferation.
When ascertained the up- and downstream relationship within the three genes, we observed that while overexpressing or silencing KLF5 and GCN5 could increase or decrease GDF15 expression (Fig. 3e, f), the change of GDF15 did not affect KLF5 or GCN5 expression (Fig. 3g), implicating that GDF15 is in the downstream of KLF5 and GCN5.
C5a, KLF5, and GCN5 promote GDF15 gene transcription in A549 cells
Since KLF5 and GCN5 could increase GDF15 expression (Fig. 3e, f), further experiments were done to determine whether KLF5 or GCN5 could directly trigger GDF15 gene transcription. Firstly, we demonstrated that GDF15 full-length (FL) promoter (−2078 to +103 nt) activity was greatly boosted by C5a at 1 h, peaked at 2 h and 3 h (Supplementary Figure 4a), and descended markedly when cells were pretreated with C5aR-I or Anti-C5a (Supplementary Figure 4b). Importantly, overexpression or knockdown of KLF5 and GCN5 upregulated or downregulated GDF15 promoter activity (Supplementary Figure 4c). These data hint that C5a-enhanced KLF5 and GCN5 expression can directly trigger and augment GDF15 gene transcription.
KLF5 and GCN5 improve GDF15 gene transcription via forming a complex on GDF15 promoter in A549 cells upon C5a exposure
Because KLF5 and GCN5 can enhance GDF15 promoter activity, the mechanism in C5a-induced GDF15 gene transcription requires to be explored. First, five KLF5-binding elements (RE1 to RE5) on GDF15 promoter were predicted by JASPAR software (Fig. 4a). Then, the plasmids of four truncated GDF15 promoter fragments (truncate 1: −1682 to +103 nt, truncate 2: −1039 to +103 nt, truncate 3: −401 to +103 nt and truncate 4: −55 to +103 nt) were constructed. Further experiments showed that the activity of truncate 4 promoter was acutely reduced compared with FL or other truncated promoters, and did not change after C5a stimulation, KLF5 overexpression or KLF5 knockdown (with C5a incubation) in A549 cells (Fig. 4b–d). These results implicate the effective KLF5-binding sites on GDF15 promoter might locate within −401 to −55 nt and may be RE4 (−122 to −113 nt) and/or RE5 (−71 to 62 nt).
To further identify the exact site of KLF5 binding to GDF15 promoter, chromatin immunoprecipitation (ChIP)-real-time PCR was performed and found that KLF5 could bind to the region (−103 to +58 nt) of GDF15 promoter, which was eminently augmented by C5a (Fig. 4e), demonstrating that this region of GDF15 promoter contains KLF5-binding element (probably RE5). Expectedly, the mutation of RE5 (AGGGCGGGAC to AAAAAAAAAA) could significantly reduce the C5a-triggered activation of the GDF15-FL promoter (Fig. 4f).
Subsequently, we continued to assess the enrichment of KLF5 on GDF15 promoter (−103 to +58 nt) in KLF5 overexpression or knockdown. It displayed that the binding level was greatly increased after KLF5 overexpression, and markedly decreased when KLF5 was silenced even in the present of C5a (Fig. 4g), indicating that KLF5 can directly bind to GDF15 promoter and improve GDF15 gene transcription.
GCN5 is reported to be a co-activator that can be recruited to promoter and enhanced gene transcription [17, 18]. Hence, to clarify whether C5a can induce GCN5 to bind to GDF15 promoter, ChIP and Re-ChIP assays were performed and showed that GCN5 and KLF5 could bind to GDF15 promoter (−103 to +58 nt) through forming a complex (Fig. 4h). During this process, there was a notable reduction of GCN5 binding on GDF15 promoter when interfering KLF5 expression (Fig. 4i). These data reveal that C5a can lead to KLF5 and GCN5 forming a complex and binding to GDF15 promoter in a KLF5-dependent manner, and finally enhance GDF15 gene transcription.
KLF5 acetylation by GCN5 promotes GDF15 gene transcription through elevating KLF5 binding to GDF15 promoter
As we know, GCN5 is a acetyltransferase which can modify protein lysine [17, 18], but whether GCN5 can acetylate KLF5 and whether acetylated KLF5 can affect GDF15 transcription remain unknown. To solve these problems, IP was done and exhibited an interaction between GCN5 and KLF5 as well as an upregulation of KLF5 acetylation following C5a stimulation (Fig. 5a). Additionally, the level of GCN5–KLF5 binding and KLF5 acetylation in A549 cells with C5aR-I or Anti-C5a treatment was markedly downregulated (Supplementary Figure 5), implying that C5a stimulation can result in GCN5–KLF5 combination and KLF5 acetylation via C5a-C5aR. Furthermore, the GCN5–KLF5 binding and KLF5 acetylation were significantly increased by GCN5 overexpression (Fig. 5b), while C5a-induced GCN5–KLF5 binding and KLF5 acetylation remarkably reduced when GCN5 was silenced, even in the present of deacetylase inhibitor trichostatin A (TSA) and nicotinamide (NAM, Fig. 5c). Meantime, the direct binding of purified recombinant KLF5 and GCN5, or the KLF5 acetylation modified by GCN5 was also confirmed by pulldown assay (Fig. 5d) or in vitro acetylation assay (Fig. 5e). These suggest that GCN5 bears a great responsibility in C5a-induced KLF5 acetylation.
To study the impact of GCN5 and acetylated KLF5 on GDF15 gene transcription, we performed ChIP, Re-ChIP, and ChID assays and found that the binding of KLF5 on GDF15 promoter (−103 to +58 nt) was diminished after GCN5 silence (Fig. 5f), and only acetylated KLF5 could greatly bind to the region of GDF15 promoter (Fig. 5g). Therefore, we conclude that the GDF15 gene transcription in C5a-stimulated A549 cells was attributed to KLF5 acetylation by GCN5.
KLF5 acetylation at lysine 335 and lysine 391 via GCN5 regulates GDF15 gene transcription, expression, and cell proliferation triggered by C5a
To further define the role of KLF5 acetylation by GCN5 in GDF15 gene transcription, expression and cell proliferation mediated by C5a, we transfected A549 cells with a HAT (GNAT-SF) domain (Fig. 6a)-deleted GCN5 mutant (△GCN5) followed with C5a treatment for 3 h and uncovered a significant decrease of KLF5 acetylation compared with wild-type (WT) GCN5 (Fig. 6c). On the other hand, according to the mass spectrometry result that KLF5 was acetylated by GCN5 at two highly conserved acetylation sites, lysine 335 (K335) and lysine 391 (K391, Fig. 6b), we mutated the two sites separately or together by changing lysine (K) to non-acetylatable arginine (R). It showed that KLF5 acetylation was remarkably downregulated when K335 or K391 was mutated, and the downregulation was even more eminent upon the two sites mutation (Fig. 6d). Meanwhile, we observed that the HAT domain-deletion of GCN5 and double mutation of KLF5 at K335 and K391 could markedly decrease GDF15 promoter activity, GDF15 expression (Fig. 6e–g), and A549 cells proliferation (Fig. 6h–j). Besides, the exogenous KLF5 acetylation by GCN5 in A549 cells exposed to C5a was also reduced in the condition of previous-mentioned mutations, and so were the promoter activity and expression of GDF15, as well as A549 cell proliferation (Supplementary Figure 6a–f). Together, these results implicate that the KLF5–K335 and K391 acetylation by GCN5 has a pivotal role in mediating GDF15 gene transcription, expression and A549 cell proliferation.
Effect of silencing KLF5, GCN5, or GDF15 gene on mice A549 xenograft tumor growth and proliferation-related protein expression
Since we already found the pro-proliferative effect of KLF5, GCN5, and GDF15 in vitro, we wondered the function of these genes in vivo. To explore this, A549 cells infected stably with lentiviruses carrying shCTR, shKLF5, shGCN5, or shGDF15 (Supplementary Figure 7a and b) were injected subcutaneously into BALB/c nude mice. Thereafter, the xenograft growth was monitored for up to 28 days. The results showed that the volume, size, and weight of LV-shKLF5, LV-shGCN5, or LV-shGDF15 xenografts were prominently less than that of LV-shCTR tumors (Fig. 7a–c). Similarly, a marked suppression of cyclin D1, PCNA and Ki67 expression was also displayed in above-mentioned xenografts (Fig. 7d, e), although no histomorphological differences were seen in all xenografts (Supplementary Figure 7c). Additionally, GDF15 protein was significantly downregulated in LV-shKLF5 and LV-shGCN5 xenograft, and KLF5 acetylation was notably reduced in LV-shGCN5 xenograft (Supplementary Figure 7d and e). The findings in vivo confirmed again that KLF5, GCN5, and GDF15 expression could accelerate NSCLC growth and cell proliferation.
Discussion
Inflammation is a contributor to NSCLC progression [10–12], and C5a can cause tumor cell proliferation [8, 9, 28, 29]. However, how C5a affects tumor cell proliferation in NSCLC still remains obscure.
In this study, we scrutinized some proliferation-related genes, and found that KLF5, GCN5, and GDF15 including C5a and C5aR levels were elevated in the majority of NSCLC patients. We also discovered that KLF5, GCN5, GDF15, and C5aR expression was closely correlated with the tumor size, lymph node metastasis, and TNM stage of NSCLC, indicating that these proteins may have an important function in NSCLC tumorigenesis.
Reportedly, complement could be activated in tumor tissues [2, 30, 31], and lung cancer cells could also generate C5a [3], creating a microenvironment for NSCLC growth and proliferation [11, 12, 30]. Thus, to explore the effect of C5a on NSCLC cell proliferation, the C5aR expression in multiple NSCLC cell lines was detected and the upregulation of C5aR in A549 and PC9 cells was found. Subsequent experiments exhibited that the proliferation of A549 and PC9 cells increased upon C5a stimulation, particularly that of A549 cells. Further studies discovered that the KLF5, GCN5, and GDF15 expression in C5a-treated A549 cells was upregulated and blocking C5aR or neutralizing C5a could remarkably repress the expression of these three proteins and cell proliferation, suggesting that C5a possesses the capacity of promoting these genes’ expression and cell proliferation via binding to C5aR. Notably, some pro-inflammatory factors such as IL-1 and IL-6 can upregulate C5aR expression during the process of other diseases [32, 33]. These factors, which also exist in NSCLC microenvironment [10], might elevate C5aR expression in NSCLC cells, and finally promoting NSCLC development and progression together with C5a.
Recently, researches have confirmed that KLF5 and GCN5 are overexpressed in many cancers [16, 17], and could regulate tumor biological processes [34, 35] as a transcription factor and a transcriptional co-activator respectively. Many studies also consider that GDF15 is a contributor to the development and progression of some malignant tumors [20, 21, 36–38], although several documents have reported that retinoid-related molecule or anti-inflammatory drug can induce tumor cell apoptosis via GDF15 [39–41]. Bruzzese F20 points out that the pro- and anti-tumor activities of GDF15 are highly dependent on the cellular and microenvironmental context. Our present studies revealed that overexpression or knockdown of KLF5, GCN5, and GDF15 genes in A549 cells could promote or inhibit cell proliferation, implicating that these proteins have a pro-proliferation function. Meanwhile, we found that overexpression or knockdown of KLF5 and GCN5 upregulated or downregulated GDF15 expression, but overexpressing or silencing GDF15 gene did not alter KLF5 and GCN5 expression, hinting that GDF15 is a downstream gene of KLF5 and GCN5 in C5a-governing proliferation.
Emerging evidence has shown that GDF15 overexpression can facilitate tumor cells proliferation, e.g., ovarian cancer [37], breast cancer [21], and esophageal squamous cell carcinomas [38]. Our data proved that GDF15 promoter activity was upregulated by C5a stimulation. Correspondingly, overexpression or knockdown of KLF5 and GCN5 also elevated or reduced GDF15 promoter activity. These findings suggest that C5a-induced proliferation is due to the increase of KLF5, GCN5, and GDF15 expression, and KLF5 and GCN5 overexpression has an improving role in GDF15 gene transcription.
To date, KLF5 binding to GC boxes on gene promoters and regulating gene transcription have been reported [14, 42]. Moreover, it is also demonstrated that GCN5 can amplify transcription factor effect [17, 18, 43]. The luciferase assay revealed that KLF5 expression augmented GDF15 promoter activity, and the KLF5-binding elements were located within −401 to −55 nt of GDF15 promoter. Furthermore, we discovered for the first time that KLF5 could bind to the region −103 to +58 nt of GDF15 promoter, containing the response element AGGGCGGGAC, and that C5a could cause the interaction between KLF5 and GCN5, leading them to form a complex in a KLF5-dependent manner, and finally promote GDF15 gene transcription in A549 cells.
Protein function is often regulated by posttranslational modifications such as acetylation [18, 44]. As GCN5 is an acetyltransferase modifying histone or non-histone protein such as transcription factor [43], KLF5 acetylation in GDF15 gene transcription and NSCLC cell proliferation need to be illustrated. In the study, we first disclosed that KLF5 could be acetylated by GCN5 at lysine 335 and lysine 391 via mass spectrometry. Notably, only acetylated KLF5 could bind to GDF15 promoter, and failure in KLF5 acetylation in A549 cells transfected with ΔGCN5 or KLF5–K335R and K391R mutants not only prevented KLF5–GCN5 complex formation, but also lessened the binding of KLF5 to GDF15 promoter. These imply that KLF5 acetylation is essential for KLF5 recruitment to GDF15 promoter and GDF15 gene transcription, expression as well as A549 cell proliferation.
In vivo, we demonstrated delayed growth and proliferation in nude mice xenograft tumors created by inoculating A549 cells stably infected with LV-shKLF5, LV-shGCN5, or LV-shGDF15, indicating that KLF5, GCN5 and GDF15 are necessary in the maintenance of NSCLC growth.
In conclusion, our studies revealed that KLF5, GCN5, GDF15, C5a, and C5aR were increased in the samples of NSCLC patients, and in vitro C5a could induce KLF5, GCN5, or GDF15 expression, promoting A549 cell proliferation. The C5a-elevated KLF5 could recruit GCN5, forming a complex which bound to GDF15 promoter in a KLF5-dependent manner and enhancing GDF15 gene transcription. In this process, KLF5 could be acetylated by GCN5 at lysine 335 and lysine 391, facilitating the binding of KLF5 with GDF15 promoter and upregulating GDF15 expression as well as A549 cell proliferation (Fig. 7d). Besides, GCN5, KLF5, or GDF15 silence reduced cell proliferation in vitro and xenograft tumor growth in vivo. Together, these findings provide clues to the mechanism of C5a-induced NSCLC proliferation, which may also provide targets for human NSCLC therapy.
Materials and methods
Human specimens and animals
The fresh NSCLC tissues and the paired adjacent tissues of 52 patients were collected during the surgery from the First Affiliated Hospital of Nanjing Medical University and Jiangsu Cancer Hospital. Paraffin-embedded NSCLC tumor tissues (N = 185) were provided by National Engineering Center for BioChips or collected from the First Affiliated Hospital of Nanjing Medical University. All samples were from patients who underwent surgical resection without preoperative chemotherapy or radiotherapy. Patients in any other disease conditions (infections, allergies, or other inflammatory diseases and cancers) were excluded. The study was also approved by the Ethics Committee of Nanjing Medical University and informed consent was obtained from all patients participating in this research prior to the experiment. Female BALB/c nude mice (5-week-old) were from Vital River Laboratory Animal Technology Co. Ltd (Beijing, China). Mice were maintained in animal facilities under pathogen-free conditions, and the animal study was approved by the Animal Ethical and Welfare Committee (AEWC) of Nanjing Medical University.
Cell lines and reagents
The human NSCLC cell lines, A549, H1299, H1975, and H1703 were obtained from the American Type Culture Collection (ATCC). The PC9 and SPC-A1 cells were from the European Collection of Authenticated Cell Cultures (ECACC) and the Institute of Biochemistry and Cell Biology of the Chinese Academy of Science, respectively. The human bronchial epithelial cell line (16HBE) was provided by Dr Gruenert (California Pacific Medical Center, San Francisco, CA, USA). All of the cell lines were authenticated by short tandem repeat profiling and free of mycoplasma contamination. The recombinant human C5a was from R&D systems (Tustin, CA, USA), and C5aR antagonist W54011 was provided by Merck Millipore (Darmstadt, Germany). The antibodies (Abs) against C5a (ab135187), GDF15 (ab180929) and cyclin D1 (ab134175) were supplied by Abcam (Cambridge, UK). The KLF5 (sc-22797×) and GCN5 (sc-365321×) Abs were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). The Abs of acetylated lysine (Ac-K, #9441 or #9681), anti-PCNA (#13110) and chromatin immunoprecipitation (ChIP) kit were from Cell Signaling Technology (Danvers, MA, USA). The Abs against C5aR (P21730) and Ki67 (P46013) were from Bioworld Technology (St. Louis Park, MN, USA). DAB substrate kit, C5a ELISA kit and his-tag isolation & pulldown dynabeads were from Thermo Fisher (Waltham, MA, USA). Dual-luciferase reporter assay system kit was purchased from Promega (Madison, WI, USA). The recombinant his-tagged GCN5 was from Cayman Chemical (Ann Arbor, MI, USA), and the acetyl-CoA was from Solarbio (Beijing, China). The anti-flag immunoaffinity resin was supplied by Sigma-Aldrich (St. Louis, MS, USA) and QuikChange II site-directed mutagenesis kit was from Agilent (Santa Clara, CA, USA).
Transcriptome deep sequencing
Total RNA from 12 pairs of NSCLC tumor and adjacent tissues were isolated. The samples were generated by mixing the RNAs from 4 patients and the sequencing library of each mixing sample was prepared. The RNA library was then sequenced on Illummina HiSeq 2500 [45].
Quantitative real-time PCR
Total RNA of tissues or cells was prepared, and the cDNA was then generated. Real-time PCR experiment was performed. The primers for real-time PCR are shown in Supplementary Table 1. The results were normalized to the β-actin expression and analyzed by using the 2−ΔΔCt method [22, 25].
Immunohistochemical staining
The sections of patient NSCLC tumor and mouse xenograft tumor were incubated with Abs against KLF5, GCN5, GDF15, C5aR, cyclin D1, PCNA, and Ki67. The IHC was scored according to staining intensity. For each sample, 500 cells from five randomly chosen fields were counted. For IHC intensity scoring: negative = 0, weak = 1, moderate = 2, strong = 3. For staining area scoring: 0% = 0, 1–25% = 1, 26–50% = 2, 51–75% = 3 and 76–100% = 4. These scores were multiplied to produce the final score: 0–1, negative expression; 2–4, weak positive expression; 6–12, strong positive expression [46].
ELISA detection
The plasma from 40 patients and 40 healthy individuals were collected from the First Affiliated Hospital of Nanjing Medical University. The C5a concentration was measured using ELISA kit according to manufacturer’s instructions.
Plasmids construction
The plasmids of pIRES2-KLF5, pIRES2-GCN5, pIRES2-GDF15, flag-KLF5 were constructed by inserting complete open reading frames (ORF) of human KLF5 (NM_001730.4), GCN5 (NM_021078.2), and GDF15 (NM_004864.2) gene into pIRES2-EGFP vector. Primer sequences are listed in Supplementary Table 2. Then, shKLF5, shGCN5, and shGDF15 plasmids were constructed using pGpU6/GFP/Neo vector and the most effective shRNA was chosen for further experiment. The shRNA targeted sequences are as follows: shKLF5, TGCTGTTCCGCAGACTGCAGTGAAA; shGCN5, GAGGACGTGGCTACCTACAAGGTCA; shGDF15, CCGGATACTCACGCCAGAAGT. Moreover, the mutant GCN5 (ΔGCN5) which deleted acetyltransferase domain and KLF5 of which lysine 335 (K335), lysine 391 (K391), or both two lysine sites (K335/391) were mutated to non-acetylatable arginine (R) were created.
Lentiviral shRNA packing
The plasmids of lentiviral (LV)-shKLF5, LV-shGCN5, and LV-shGDF15 were prepared by Genechem (Shanghai, China). The shRNA sequences used to silence corresponding genes are the same as previous-described most effective shRNAs against the three genes.
Cell culture and transfection
NSCLC tumor cell lines were cultured in DMEM with 10% FBS. For transient transfection, cells were cultured overnight, and the mixture of plasmid and transfection reagent was added, incubated for 48 h. For stable transfection, 2 × 105 A549 cells were incubated with LV at the titer of 4 × 106 TU/ml for 48 h and then challenged by puromycin selection. The transfection efficiency was estimated by GFP and corresponding protein expression.
Immunoprecipitation (IP) assay
Protein extracted from A549 cells and tissues was incubated with protein G-Sepharose beads and pre-immune IgG. After centrifugation, the supernatant was collected and incubated with corresponding Abs. The antigen–antibody complex was precipitated with protein G agarose beads, and IB assay for detecting protein and acetylated lysine was performed [25].
Immunoblotting (IB) analysis
The whole-cell lysates (WCL) were electrophoresed. The blots were probed with primary Abs. Thereafter, the blots were incubated with corresponding secondary Abs, and exposed using regular X-Ray film or Amersham imager 600 (GE, Boston, MA, USA) [22, 25].
Luciferase reporter experiment
GDF15 full-length (GDF15-FL) promoter plasmids were constructed by inserting GDF15 promoter fragment (−2068 to +103 nt) into pGL3-basic vector. The promoter deletion fragments (truncate 1: −1682 to +103 nt, truncate 2: −1039 to +103 nt, truncate 3: −401 to +103 nt and truncate 4: −55 to +103 nt) were cloned into the same vector. Specific primers for promoter luciferase assays are listed in Supplementary Table 2. The GDF15-FL-RE5-Mut plasmid was constructed by mutating the KLF5 response element 5 on GDF15-FL (RE5, −71 to 62 nt, AGGGCGGGAC) to the AAAAAAAAAA sequence. The pGL3-GDF15-FL and truncated 1 to 4 plasmids were transfected into A549 cells. The promoter activity was measured by dual-luciferase reporter reagent [47].
ChIP, Re-ChIP, and ChID assays
Chromatin immunoprecipitation (ChIP) assay was performed using ChIP-validated Abs and ChIP-grade protein G agarose beads. For Re-ChIP, the primary ChIP material was released from the agaroses and subjected to second round of immunoprecipitation using corresponding Abs. For chromatin immunodepletion (ChID), ChIP material isolated from the agaroses was immunoprecipitated again with an anti-Ac-K Ab, and GDF15 promoter fragment in the supernatant immunodepleted with anti-Ac-K was detected [15, 27]. The primers for GDF15 promoter fragments were as follows: primer for −234 to −76 nt, forward: 5′-GGCTGGAATGGTGTCCT-3′, reverse: 5′-CCCCTGCCTGCAGAGTA-3′; primer for −103 to +58, forward: 5′-TCAAACAATCCACCCACCT-3′, reverse: 5′-TTTCCAGTCTAAGCAGGGTG-3′.
His-pulldown assay
The His-pulldown assay was performed according to the manufacturer’s instruction of his-tag isolation & pulldown dynabeads. Briefly, the recombinant his-tagged GCN5 protein (5 μg) were added to 50 μl cobalt-coated magnetic beads and incubated on a roller at room temperature for 10 min. Then the bead/protein complex was washed and resuspended in the pulldown buffer (3.25 mM sodium-phosphate, 70 mM NaCl and 0.01% Tween-20). Subsequently, the flag-tagged KLF5 (5 μg) was added into the bead/protein complex incubated for 1 h. After washing, the mixture was boiled with loading buffer and analyzed by SDS-PAGE [48].
In vitro acetylation assay
The active full-length recombinant GCN5 or heat-inactivate were incubated with the recombinant flag-KLF5 in 50 μl acetyltransferase assay buffer (50 mM Tris-HCl, pH 8.0, 10% glycerol, 0.1 mM EDTA, 1 mM PMSF, and 1 mM dithiothreitol) with 20 μM acetyl-CoA at 37 °C for 30 min. The acetylation of KLF5 was determined by IB using an acetylated lysine antibody [48, 49].
Mass spectrometry analysis
A549 cells were overexpressed with flag-tagged KLF5 fusion protein and then subject to lysis reagent. Each clarified lysate was bound to anti-flag immunoaffinity resin, and then bounded protein complexes were competitively eluted using synthetic flag peptide. The purified flag-KLF5 protein was analyzed by mass spectrometer (ULTRAFLEX-II, Germany) at the Center of Hygienic Analysis & Detection of Nanjing Medical University [25, 50].
Cell proliferation examination
For CCK8, the cells were allowed to grow in a 96-well plate. The optical density (OD) values were documented. For cell counting, the cells transfected with indicated plasmids were seeded at day 0 and digested at day 7, and the live cells were counted. For colony formation, cells were seeded in 6-well plate. On day 14, the cells were stained with 0.1 % crystal violet, and visible colonies were calculated.
Xenograft tumor experiment
A total of 32 nude mice were randomly and blindly divided into four groups (N = 8 per group). The sample size was decided according to the estimation of the PASS software (www.ncss.com/software/pass/) and previous experience. A549 cells infected stably with LV-shRNAs or LV-shCTR were harvested. Then 5 × 106 cells were inoculated subcutaneously into axilla of nude mice to generate an A549 lung adenocarcinoma-bearing mouse model. The tumor volumes were monitored every 4 days and calculated using following formula: volume = 0.5 × length × width2. Twenty eight days after inoculation, the mice were sacrificed, and the tumors were weighed. Additionally, KLF5, GCN5, GDF15 expression, and KLF5 acetylation in xenografts were detected by IP and IB, and the tumor cell proliferation was assessed by cyclin D1, PCNA, and Ki67 staining.
Statistical analysis
Data are presented as means ± S.E.M. of three independent experiments. GraphPad Prism was used for data analysis. Two-tailed t-test was applied to determine significant differences between two groups while data among several groups were compared by one-way ANOVA test with Dunnett’s correction. The relationships between KLF5, GCN5, and GDF15 expression were analyzed by the Spearman’s correlation. The χ2 test was used for the correlations of KLF5, GCN5, GDF15, and C5aR expression in tumor tissues with clinic-pathological parameters of NSCLC patients. Statistics with p value < 0.05 were considered as statistically significant.
Electronic supplementary material
Acknowledgements
This work was supported by grants from the National Natural Science Foundations of China (Nos. 81272532, 81472626, and 81672896). We thank Fuqiang Wang for technical assistance.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Electronic supplementary material
The online version of this article (10.1038/s41388-018-0298-9) contains supplementary material, which is available to authorized users.
References
- 1.Pandey MK, Burrow TA, Rani R, Martin LJ, Witte D, Setchell KD, et al. Complement drives glucosylceramide accumulation and tissue inflammation in Gaucher disease. Nature. 2017;543:108–12. doi: 10.1038/nature21368. [DOI] [PubMed] [Google Scholar]
- 2.Markiewski MM, DeAngelis RA, Benencia F, Ricklin-Lichtsteiner SK, Koutoulaki A, Gerard C, et al. Modulation of the antitumor immune response by complement. Nat Immunol. 2008;9:1225–35. doi: 10.1038/ni.1655. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Nitta H, Murakami Y, Wada Y, Eto M, Baba H, Imamura T. Cancer cells release anaphylatoxin C5a from C5 by serine protease to enhance invasiveness. Oncol Rep. 2014;32:1715–9. doi: 10.3892/or.2014.3341. [DOI] [PubMed] [Google Scholar]
- 4.Sayegh ET, Bloch O, Parsa AT. Complement anaphylatoxins as immune regulators in cancer. Cancer Med. 2014;3:747–58. doi: 10.1002/cam4.241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Piao C, Cai L, Qiu S, Jia L, Song W, Du J. Complement 5a enhances hepatic metastases of colon cancer via monocyte chemoattractant protein-1-mediated inflammatory cell infiltration. J Biol Chem. 2015;290:10667–76. doi: 10.1074/jbc.M114.612622. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Markiewski MM, Lambris JD. Unwelcome complement. Cancer Res. 2009;69:6367–70. doi: 10.1158/0008-5472.CAN-09-1918. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lu Y, Hu XB. C5a stimulates the proliferation of breast cancer cells via Akt-dependent RGC-32 gene activation. Oncol Rep. 2014;32:2817–23. doi: 10.3892/or.2014.3489. [DOI] [PubMed] [Google Scholar]
- 8.Imamura T, Yamamoto-Ibusuki M, Sueta A, Kubo T, Irie A, Kikuchi K, et al. Influence of the C5a-C5a receptor system on breast cancer progression and patient prognosis. Breast Cancer. 2016;23:876–85. doi: 10.1007/s12282-015-0654-3. [DOI] [PubMed] [Google Scholar]
- 9.Zheng X, Chi J, Zhi J, Zhang H, Yue D, Zhao J, et al. Aurora-A-mediated phosphorylation of LKB1 compromises LKB1/AMPK signaling axis to facilitate NSCLC growth and migration. Oncogene. 2018;37:502–11. doi: 10.1038/onc.2017.354. [DOI] [PubMed] [Google Scholar]
- 10.Wang L, Zhang LF, Wu J, Xu SJ, Xu YY, Li D, et al. IL-1beta-mediated repression of microRNA-101 is crucial for inflammation-promoted lung tumorigenesis. Cancer Res. 2014;74:4720–30. doi: 10.1158/0008-5472.CAN-14-0960. [DOI] [PubMed] [Google Scholar]
- 11.Corrales L, Ajona D, Rafail S, Lasarte JJ, Riezu-Boj JI, Lambris JD, et al. Anaphylatoxin C5a creates a favorable microenvironment for lung cancer progression. J Immunol. 2012;189:4674–83. doi: 10.4049/jimmunol.1201654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Afshar-Kharghan V. The role of the complement system in cancer. J Clin Invest. 2017;127:780–9. doi: 10.1172/JCI90962. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Qiu W, Zhang Y, Liu X, Zhou J, Li Y, Zhou Y, et al. Sublytic C5b-9 complexes induce proliferative changes of glomerular mesangial cells in rat Thy-1 nephritis through TRAF6-mediated PI3K-dependent Akt1 activation. J Pathol. 2012;226:619–32. doi: 10.1002/path.3011. [DOI] [PubMed] [Google Scholar]
- 14.Li X, Liu X, Xu Y, Liu J, Xie M, Ni W, et al. KLF5 promotes hypoxia-induced survival and inhibits apoptosis in non-small cell lung cancer cells via HIF-1alpha. Int J Oncol. 2014;45:1507–14. doi: 10.3892/ijo.2014.2544. [DOI] [PubMed] [Google Scholar]
- 15.Jia L, Zhou Z, Liang H, Wu J, Shi P, Li F, et al. KLF5 promotes breast cancer proliferation, migration and invasion in part by upregulating the transcription of TNFAIP2. Oncogene. 2016;35:2040–51. doi: 10.1038/onc.2015.263. [DOI] [PubMed] [Google Scholar]
- 16.Li Q, Dong Z, Zhou F, Cai X, Gao Y, Wang LW. Kruppel-like factor 5 promotes lung tumorigenesis through upregulation of Sox4. Cell Physiol Biochem. 2014;33:1–10. doi: 10.1159/000356645. [DOI] [PubMed] [Google Scholar]
- 17.Chen L, Wei T, Si X, Wang Q, Li Y, Leng Y, et al. Lysine acetyltransferase GCN5 potentiates the growth of non-small cell lung cancer via promotion of E2F1, cyclin D1, and cyclin E1 expression. J Biol Chem. 2013;288:14510–21. doi: 10.1074/jbc.M113.458737. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Liu K, Zhang Q, Lan H, Wang L, Mou P, Shao W, et al. GCN5 potentiates glioma proliferation and invasion via STAT3 and AKT signaling pathways. Int J Mol Sci. 2015;16:21897–910. doi: 10.3390/ijms160921897. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Tanno T, Lim Y, Wang Q, Chesi M, Bergsagel PL, Matthews G, et al. Growth differentiating factor 15 enhances the tumor-initiating and self-renewal potential of multiple myeloma cells. Blood. 2014;123:725–33. doi: 10.1182/blood-2013-08-524025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Bruzzese F, Hagglof C, Leone A, Sjoberg E, Roca MS, Kiflemariam S, et al. Local and systemic protumorigenic effects of cancer-associated fibroblast-derived GDF15. Cancer Res. 2014;74:3408–17. doi: 10.1158/0008-5472.CAN-13-2259. [DOI] [PubMed] [Google Scholar]
- 21.Tu M, Liu X, Han B, Ge Q, Li Z, Lu Z, et al. Vasohibin2 promotes proliferation in human breast cancer cells via upregulation of fibroblast growth factor2 and growth/differentiation factor 15 expression. Mol Med Rep. 2014;10:663–669. [DOI] [PMC free article] [PubMed]
- 22.Zhang J, Li Y, Shan K, Wang L, Qiu W, Lu Y, et al. Sublytic C5b-9 induces IL-6 and TGF-beta1 production by glomerular mesangial cells in rat Thy-1 nephritis through p300-mediated C/EBPbeta acetylation. FASEB J. 2014;28:1511–25. doi: 10.1096/fj.13-242693. [DOI] [PubMed] [Google Scholar]
- 23.Cai K, Wan Y, Wang Z, Wang Y, Zhao X, Bao X. C5a promotes the proliferation of human nasopharyngeal carcinoma cells through PCAF-mediated STAT3 acetylation. Oncol Rep. 2014;32:2260–6. doi: 10.3892/or.2014.3420. [DOI] [PubMed] [Google Scholar]
- 24.Kaypee S, Sudarshan D, Shanmugam MK, Mukherjee D, Sethi G, Kundu TK. Aberrant lysine acetylation in tumorigenesis: Implications in the development of therapeutics. Pharmacol Ther. 2016;162:98–119. doi: 10.1016/j.pharmthera.2016.01.011. [DOI] [PubMed] [Google Scholar]
- 25.He F, Zhou M, Yu T, Zhao D, Zhang J, Qiu W, et al. Sublytic C5b-9 triggers glomerular mesangial cell apoptosis in rat Thy-1 nephritis via Gadd45 activation mediated by Egr-1 and p300-dependent ATF3 acetylation. J Mol Cell Biol. 2016;8:477–91. doi: 10.1093/jmcb/mjw021. [DOI] [PubMed] [Google Scholar]
- 26.Majaz S, Tong Z, Peng K, Wang W, Ren W, Li M, et al. Histone acetyl transferase GCN5 promotes human hepatocellular carcinoma progression by enhancing AIB1 expression. Cell Biosci. 2016;6:47. doi: 10.1186/s13578-016-0114-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Yin YW, Jin HJ, Zhao W, Gao B, Fang J, Wei J, et al. The Histone acetyltransferase GCN5 expression is elevated and regulated by c-Myc and E2F1 transcription factors in human colon cancer. Gene Expr. 2015;16:187–96. doi: 10.3727/105221615X14399878166230. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Vadrevu SK, Chintala NK, Sharma SK, Sharma P, Cleveland C, Riediger L, et al. Complement c5a receptor facilitates cancer metastasis by altering T-cell responses in the metastatic niche. Cancer Res. 2014;74:3454–65. doi: 10.1158/0008-5472.CAN-14-0157. [DOI] [PubMed] [Google Scholar]
- 29.Maeda Y, Kawano Y, Wada Y, Yatsuda J, Motoshima T, Murakami Y, et al. C5aR is frequently expressed in metastatic renal cell carcinoma and plays a crucial role in cell invasion via the ERK and PI3 kinase pathways. Oncol Rep. 2015;33:1844–50. doi: 10.3892/or.2015.3800. [DOI] [PubMed] [Google Scholar]
- 30.Mamidi S, Hone S, Kirschfink M. The complement system in cancer: ambivalence between tumour destruction and promotion. Immunobiology. 2017;222:45–54. doi: 10.1016/j.imbio.2015.11.008. [DOI] [PubMed] [Google Scholar]
- 31.Towner LD, Wheat RA, Hughes TR, Morgan BP. Complement membrane attack and tumorigenesis: a systems biology approach. J Biol Chem. 2016;291:14927–38. doi: 10.1074/jbc.M115.708446. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Takabayashi T, Shimizu S, Clark BD, Beinborn M, Burke JF, Gelfand JA. Interleukin-1 upregulates anaphylatoxin receptors on mononuclear cells. Surgery. 2004;135:544–54. doi: 10.1016/j.surg.2003.09.010. [DOI] [PubMed] [Google Scholar]
- 33.Riedemann NC, Neff TA, Guo RF, Bernacki KD, Laudes IJ, Sarma JV, et al. Protective effects of IL-6 blockade in sepsis are linked to reduced C5a receptor expression. J Immunol. 2003;170:503–7. doi: 10.4049/jimmunol.170.1.503. [DOI] [PubMed] [Google Scholar]
- 34.Du C, Gao Y, Xu S, Jia J, Huang Z, Fan J, et al. KLF5 promotes cell migration by upregulating FYN in bladder cancer cells. FEBS Lett. 2016;590:408–18. doi: 10.1002/1873-3468.12069. [DOI] [PubMed] [Google Scholar]
- 35.Kikuchi H, Takami Y, Nakayama T. GCN5: a supervisor in all-inclusive control of vertebrate cell cycle progression through transcription regulation of various cell cycle-related genes. Gene. 2005;347:83–97. doi: 10.1016/j.gene.2004.12.007. [DOI] [PubMed] [Google Scholar]
- 36.Carrillo-Garcia C, Prochnow S, Simeonova IK, Strelau J, Holzl-Wenig G, Mandl C, et al. Growth/differentiation factor 15 promotes EGFR signalling, and regulates proliferation and migration in the hippocampus of neonatal and young adult mice. Development. 2014;141:773–83. doi: 10.1242/dev.096131. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Griner SE, Joshi JP, Nahta R. Growth differentiation factor 15 stimulates rapamycin-sensitive ovarian cancer cell growth and invasion. Biochem Pharmacol. 2013;85:46–58. doi: 10.1016/j.bcp.2012.10.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Urakawa N, Utsunomiya S, Nishio M, Shigeoka M, Takase N, Arai N, et al. GDF15 derived from both tumor-associated macrophages and esophageal squamous cell carcinomas contributes to tumor progression via Akt and Erk pathways. Lab Invest. 2015;95:491–503. doi: 10.1038/labinvest.2015.36. [DOI] [PubMed] [Google Scholar]
- 39.Kadara H, Schroeder CP, Lotan D, Pisano C, Lotan R. Induction of GDF-15/NAG-1/MIC-1 in human lung carcinoma cells by retinoid-related molecules and assessment of its role in apoptosis. Cancer Biol Ther. 2006;5:518–22. doi: 10.4161/cbt.5.5.2602. [DOI] [PubMed] [Google Scholar]
- 40.Liu T, Bauskin AR, Zaunders J, Brown DA, Pankhurst S, Russell PJ, Breit SN. Macrophage inhibitory cytokine 1 reduces cell adhesion and induces apoptosis in prostate cancer cells. Cancer Res. 2003;63:5034–40. [PubMed] [Google Scholar]
- 41.Kim JS, Baek SJ, Sali T, Eling TE. The conventional nonsteroidal anti-inflammatory drug sulindac sulfide arrests ovarian cancer cell growth via the expression of NAG-1/MIC-1/GDF-15. Mol Cancer Ther. 2005;4:487–93. doi: 10.1158/1535-7163.MCT-04-0293. [DOI] [PubMed] [Google Scholar]
- 42.Dong JT, Chen C. Essential role of KLF5 transcription factor in cell proliferation and differentiation and its implications for human diseases. Cell Mol Life Sci. 2009;66:2691–706. doi: 10.1007/s00018-009-0045-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Wang J, Dixon SE, Ting LM, Liu TK, Jeffers V, Croken MM, et al. Lysine acetyltransferase GCN5b interacts with AP2 factors and is required for toxoplasma gondii proliferation. PLoS Pathog. 2014;10:e1003830. doi: 10.1371/journal.ppat.1003830. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Noritsugu K, Ito A, Nakao Y, Yoshida M. Identification of zinc finger transcription factor EGR2 as a novel acetylated protein. Biochem Biophys Res Commun. 2017;489:455–9. doi: 10.1016/j.bbrc.2017.05.170. [DOI] [PubMed] [Google Scholar]
- 45.Zhang E, Han L, Yin D, He X, Hong L, Si X, et al. H3K27 acetylation activated-long non-coding RNA CCAT1 affects cell proliferation and migration by regulating SPRY4 and HOXB13 expression in esophageal squamous cell carcinoma. Nucleic Acids Res. 2017;45:3086–101. doi: 10.1093/nar/gkw1247. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Zhang H, Ma RR, Wang XJ, Su ZX, Chen X, Shi DB, et al. KIF26B, a novel oncogene, promotes proliferation and metastasis by activating the VEGF pathway in gastric cancer. Oncogene. 2017;36:5609–19. doi: 10.1038/onc.2017.163. [DOI] [PubMed] [Google Scholar]
- 47.Hou M, Zuo X, Li C, Zhang Y, Teng Y. Mir-29b regulates oxidative stress by targeting SIRT1 in ovarian cancer cells. Cell Physiol Biochem. 2017;43:1767–76. doi: 10.1159/000484063. [DOI] [PubMed] [Google Scholar]
- 48.Wan J, Zhan J, Li S, Ma J, Xu W, Liu C, et al. PCAF-primed EZH2 acetylation regulates its stability and promotes lung adenocarcinoma progression. Nucleic Acids Res. 2015;43:3591–604. doi: 10.1093/nar/gkv238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Wang Y, Guo YR, Liu K, Yin Z, Liu R, Xia Y, et al. KAT2A coupled with the alpha-KGDH complex acts as a histone H3 succinyltransferase. Nature. 2017;552:273–7. doi: 10.1038/nature25003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Guruharsha KG, Rual JF, Zhai B, Mintseris J, Vaidya P, Vaidya N, et al. A protein complex network of Drosophila melanogaster. Cell. 2011;147:690–703. doi: 10.1016/j.cell.2011.08.047. [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.