Abstract.
Background and Purpose: Colorectal cancer (CRC) is one of the cancers with the highest incidence in which APC gene mutations occurs in almost 80% of patients. This mutation leads to β-catenin aberrant accumulation and an uncontrolled proliferation. Apoptosis evasion, changes in the immune response and microbiota composition are also events that arise in CRC. Tetracyclines are drugs with proven antibiotic and immunomodulatory properties that have shown cytotoxic activity against different tumor cell lines. Experimental Approach: The effect of tigecycline was evaluated in vitro in HCT116 cells and in vivo in a colitis-associated colorectal cancer (CAC) murine model. 5-fluorouracil was assayed as positive control in both studies.Key Results: Tigecycline showed an antiproliferative activity targeting the Wnt/β-catenin pathway and downregulating STAT3. Moreover, tigecycline induced apoptosis through extrinsic, intrinsic and endoplasmic reticulum pathways converging on an increase of CASP7 levels. Furthermore, tigecycline modulated the immune response in CAC, reducing the cancer-associated inflammation through a downregulation of cytokines expression. Additionally, tigecycline favored the cytotoxic activity of CD8+ T lymphocytes, one of the main immune defenses against tumor cells. Lastly, the antibiotic reestablished the gut dysbiosis in CAC mice increasing the abundance of bacterial genera and species, such as Akkermansia and Parabacteroides distasonis , that act as protectors against tumor development. These findings resulted in a reduction of the numbers of tumors and an amelioration of the tumorigenesis process in CAC. Conclusion and Implications : tigecycline exerts a beneficial effect against CRC supporting the use of this antibiotic for the treatment of this disease.
Keywords: colitis-associated colorectal cancer, tigecycline, β-catenin, CD8+ T lymphocytes, microbiota.
Abbreviations: activating transcription factor 6, ATF6; azoxymethane, AOM; Bcl-2-associated X protein, BAX; C/EBP homologous protein, DDIT3; cancer stem cells, CSCs; colorectal associated-cancer, CAC; Colorectal cancer, CRC; dextran sulfate sodium, DSS; disease activity index, DAI; endoplasmic reticulum, ER; immunoglobulin heavy chain-binding protein, HSPA5; inflammatory bowel disease, IBD; Jun-N-terminal kinase, JNK; membrane permeabilization, MOMP; mesenteric lymph nodes, MLN; metalloproteinases, MMP; Principal Coordinates Analysis, PCoA; Ribosomal Database Project, RDP; Truncation of BID, tBID; tumor protein 53, TP53; β-catenin, CTNNB1.
Bullet point summary: Tetracyclines displays immunomodulatory properties and cytotoxic activity in tumor cell lines.
Tigecycline decreases tumorigenesis in vivo inhibiting cell proliferation and modulating immune response and gut dysbiosis.
Tigecycline effects in experimental colorectal cancer supports its consideration for human colorectal cancer management.
Data availability statement: The data that support the findings of this study are available from the corresponding authors upon reasonable request. Some data may not be made available because of privacy or ethical restrictions.
Funding statement: This work was funded by the Junta de Andalucía (CTS 164); the Instituto de Salud Carlos III (Spain) and Fondo Europeo de Desarrollo Regional (FEDER), from the European Union, through the research grants PI18/00826, P18-RT-4930, PI0206-2016, PIE16/00045 and PI19/01058; Spanish Ministry of Science and Innovation (MCIN/AEI/10.13039/501100011033/FEDER) under grant RTI2018-101309-B-C22; and the Chair “Doctors Galera-Requena in cancer stem cell research” (CMC-CTS963). A.J. R-M L. H-G are predoctoral fellows funded by the Spanish Ministry of Science and Innovation (“Programa de Doctorado: Medicina Clínica y Salud Pública” B12.56.1). M.J. R-S and J.A. M-T are predoctoral fellows from the Instituto de Salud Carlos III (FI17/00176). P. D-E is a postdoctoral fellow of Junta de Andalucía (P18-RT-4930). T. V and J. G-M are postdoctoral fellows from the University of Granada. CIBER-EHD is funded by the Instituto de Salud Carlos III.
Author contribution statement: All authors performed the experiments and contributed to the acquisition. F.G., A.J.R.-M. and A.R.-N. carried out the microbiome studies. A.J.R.-M., M.E.R.-C., A.R.-N. and J.G. designed the experiments. All authors drafted the manuscript, read, and approved the final manuscript.
Conflict of interest disclosure: The authors have declared no conflicting interests.
INTRODUCTION
Colorectal cancer (CRC) is the second most diagnosed cancer in women and the third in men. Moreover, CRC is the third most deadly among cancers, responsible for nearly a million deaths in 2020 (GLOBOCAN, 2020). Whilst recent developments/advances in the screening and treatment have reduced the mortality, the survival rate remains low, especially for metastatic CRC, with a 5-year survival rate of 12% (Xie et al. , 2020). Furthermore, there is an elevated percentage of patients that do not respond to the current treatments or show important adverse effects, so the development of new adjuvant therapies against CRC is still necessary. Different risk factors contribute to the development of CRC, from the patient’s genetic background to aging, intestinal inflammation, diet and other environmental factors (Carethers et al. , 2015). The accumulation of genetic mutations leads to an altered balance between the activation of oncogenes and the reduction of tumor suppressor genes (Fearon et al. , 1990). For instance, the mutations in the APC gene in the intestinal epithelium is an early event that occurs in almost 80% of CRC patients (Korinek et al. , 1997) leading to β-catenin aberrant activation and increasing cell proliferation (Shang et al. , 2017). In addition, a proinflammatory environment contributes to potentiate cell proliferation, angiogenesis and cancer progression (Terzic et al. , 2010). In fact, patients with inflammatory bowel disease (IBD) have a significantly elevated risk of CRC development (Clarke et al. , 2019).
Moreover, CRC tumorigenesis and progression has been clearly associated with an imbalanced gut microbiome, commonly termed as dysbiosis. Thus, changes in gut microbiome abundance have been described in patients with CRC (Garrett, 2019) evidencing a critical role for specific bacteria in colon tumorigenesis, like Fusobacterium nucleatum andBacteroides fragilis (Tjalsma et al. , 2012), as well as, a depletion of some bacterial species like Faecalibacterium prausnitzii (Lopez-Siles et al. , 2016).
Considering the involvement of the microbiome in the initiation and progression of CRC, the modulation of the gut microbiota can be considered as a potential strategy for prevention and treatment of CRC (Fong et al. , 2020). In this regard, tetracyclines are a family of broad-spectrum antibiotics that have shown beneficial effects against intestinal inflammation (Garrido-Mesa et al. , 2018) since they present immunomodulatory properties in addition to their antibiotic activity as it is the case of tigecycline. Besides, these antibiotics have shown cytostatic and cytotoxic activity against different tumor cell lines (Kroon et al. , 1984), as well as against metastasis due to their ability to inhibit metalloproteinases (MMP) (Saikaliet al. , 2003). Therefore, it is interesting to explore their potential application for the treatment of CRC. Consequently, in the present study, we evaluated the antitumor effect of tigecycline using a combination of in vitro and in vivo studies carried out on tumoral cells and azoxymethane (AOM) / dextran sulfate sodium (DSS) murine model, respectively.
MATERIAL AND METHODS.
Chemicals, cell culture and treatment.
The human colon cancer cell lines HCT116 and Caco-2 were obtained from the Cell Culture Unit of the University of Granada (Granada, Spain). NCM356 human colonic epithelial cells were kindly provided by Laura Medrano González and Ezra Aksoy (William Harvey Research Institute, Queen Mary University of London, London, UK).
Cells were cultured and treated with tigecycline (Tygacil®, Pfizer, New York, USA), Wnt3a (Sigma-Aldrich, Madrid, Spain) and 5-fluorouracil (5-FU) (Accord Farma, Polanco, Mexico) as described in Supplementary Material and Methods. RNA and cytoplasmic and nuclear proteins were isolated for Taqman qPCR and Western blot, respectively.
Cell proliferation assay.
HCT116 and NCM356 cells were treated with different doses of tigecycline (1 µM - 75 µM) or 5-FU (5 µM - 100 µM) for 48 h. After that, cell viability was measured using an MTS assay (CellTiter 96® AQueous One Solution Cell Proliferation Assay, Promega, Madison, USA) according to the manufacturer’s instructions.
Colony Formation Assay.
HCT116, Caco-2 and NCM356 cells were plated into 6-well plates at 200 cells/well for HCT116 and 400 cells/well for Caco-2 and NCM356. From the following day, fresh media was added supplemented with different doses of tigecycline and 5-FU, every 48 h for 1 week, when the media was removed, the cells were fixed with absolute ethanol and the colonies were stained with 2.3% crystal violet and counted with ImageJ software (Free Software Foundation Inc) after taking the images.
Annexin V and propidium iodide (PI) assay.
Apoptosis and necrosis were evaluated by flow cytometry using FITC Annexin V apoptosis detection kit with PI (Immunostep, Salamanca, Spain). HCT116 cells were treated with different concentrations of tigecycline (1 µM - 50 µM) for 48 h. After the treatment, attached cells and those present in the supernatant were collected, washed with cold PBS and stained with FITC-Annexin V and PI. Labeled cells were acquired on a BD FACsAria Ill (BD Biosciences, Becton, Dickinson and Company, Franklin Lakes, NY, USA) and data were analyzed using the FlowJo v10.6.2 software (FlowJo LLC, Ashland, OR, USA).
TUNEL Assay
In order to confirm the apoptotic capacity of the treatments, a TUNEL assay was carried out in HCT116 cells treated with tigecycline or 5-FU for 48 h (n=3) with the TUNEL assay kit-HRP DAB ab206386 (Abcam, Cambridge, UK) following the manufacturer’s instruction
Apoptosis was also evaluated using the DeadEnd Fluorometric TUNEL system (Promega, Madison, USA) in colonic tissue sections following the manufacturer’s indications.
Full description of the methods can be found in Supplementary Material and Methods.
CAC murine model, DAI evaluation and macroscopic data analysis .
All mice studies were performed following the ‘Guide for the Care and Use of Laboratory Animals’ as promulgated by the National Institute of Health and the protocols approved by the Ethic Committee of Laboratory Animals of the University of Granada (Spain) (Ref. No.23/10/2019/174). C57Bl/6J female mice (7-9 weeks old) from Charles River (Barcelona, Spain) were housed in groups of five mice in each makrolon cage, with an air-conditioned atmosphere, a 12 h light-dark cycle and provided with free access to tap water and food. They were subjected to a process of CAC induction by administering one intraperitoneal initial dose of AOM (Sigma-Aldrich, Madrid, Spain) at 10 mg/kg followed by three cycles of DSS (36-50 KDa, MP Biomedicals, Illkirch Cedex, France) added in the drinking water. Each DSS cycle consisted of 2% DSS for 1 week followed by a recovery period of 2 weeks with normal drinking water. Mice were divided into five groups: healthy control, colorectal associated-cancer (CAC) control, and three CAC groups treated daily with tigecycline 25 mg/kg or tigecycline 50 mg/kg by oral gavage or every three days with 5-FU 15 mg/kg intraperitoneally. The treatment started 50 days after the beginning of the assay, on the day of initiation of the third cycle of DSS and lasted for 7 weeks. The control groups received PBS following the same protocol.
A detailed description of the procedures can be found in Supplementary Material and Methods. Briefly, the disease activity index (DAI) (Camuesco et al. , 2012) was determined daily in each DSS cycle, and the day before euthanasia, both tumor number and size was assessed via colonoscopy and a tumor score was given as described by Becker C et al., (Becker et al. , 2006) by a blind observer.
When mice were sacrificed spleen, mesenteric lymph nodes (MLN) and colon were aseptically removed. The bowels were open longitudinally, the macroscopic tumors were counted, and images were acquired for measuring the size of each tumor with the ImageJ software. Colon samples were taken for RT-qPCR, Western blot, flow cytometry and fixed for histological studies. A detailed description of the procedures can be found in Supplementary Material and Methods.
Microbiota.
Faeces samples from each mouse were aseptically collected on the end-point day of the assay. Stool DNA was isolated using the QIAamp PowerFecal Pro DNA Kit (Qiagen, Hilden, Germany). Total DNA was amplified and a library for the V4–V5 region of 16S rRNA was constructed in accordance with the 16S Metagenomic Sequencing Library Preparation Illumina protocol. Sequencing was executed using the MiSeq 2 × 300 platform (Illumina Inc., San Diego, CA, USA) in accordance with the manufacturer’s instructions.
The resulting sequences were completed, quality-filtered, clustered, and taxonomically assigned on the basis of 97% similarity level against the RDP (Ribosomal Database Project) (Dhariwal et al. , 2017) by using the QIIME2 software package (2021.11 version) and “R” statistical software package (version 3.6.0; https://www.r-project.org/). Alpha diversity and beta diversity were determined by the q2-diversity plugin in QIIME2. Differential abundance analysis was performed using the Wald test implemented DESeq2 v1.30.1 as previously described (Love et al. , 2014) within the “R” statistical software (version 4.0; https://www.r-project.org/). Healthy and the three treated groups were compared with the control CAC-group and EnhancedVolcano package was used for constructing volcano-plots. Log2-fold-change (Log2FC) normalized values and the adjusted P-value (Padj) were used to construct Venn diagrams (Limma package) of differentially microbial communities up-regulated (log2FC>1.5 andPadj<0.05) and downregulated (log2FC<1.5 andPadj<0.05) among groups.
Statistical and correlation analysis
Statistical analysis was performed using the GraphPad Prism version 7 software (GraphPad Software, Inc, San Diego, CA, USA) with statistical significance set at P<0.05 . All data are represented as mean (SEM) of at least 3 independent experiments/biological replicates unless otherwise stated in the figure legends. The Mann-Whitney U test for nonparametric data was used for the analysis of the DAI. For the rest of the data, multiple comparisons between groups were performed using the one-way ANOVA followed by Tukey’s test.
A heat map depicting the macroscopic parameters, molecular changes, and patterns of microbial abundance of untreated CAC mice and those treated with both doses of tigecycline was constructed within the “R” statistical software using the “Hmisc” and “ggplot2” packages. Spearman’s correlations of each parameter were previously calculated with “R” software.
RESULTS