Volume 1(3); Pages: 235-243, 2021 | DOI: 10.21873/cdp.10032
CRISTINE DE SOUZA DUTRA, CAROLINA LUMERTZ MARTELLO, NATHAN ARAUJO CADORE, HENRIQUE BUNSELMEYER FERREIRA, ARNALDO ZAHA, KARINA MARIANTE MONTEIRO
CRISTINE DE SOUZA DUTRA1, CAROLINA LUMERTZ MARTELLO1, NATHAN ARAUJO CADORE1, HENRIQUE BUNSELMEYER FERREIRA1,2, ARNALDO ZAHA1,2 and KARINA MARIANTE MONTEIRO1,2
1Laboratório de Genômica Estrutural e Funcional, Centro de Biotecnologia, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil
2Departamento de Biologia Molecular e Biotecnologia, Instituto de Biociências, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil
Correspondence to: Karina Mariante Monteiro, Centro de Biotecnologia, Universidade Federal do Rio Grande do Sul, Caixa Postal 15005, 91501-970 Porto Alegre, RS, Brazil. E-mail: firstname.lastname@example.org
Received March 26, 2021 | Revised April 22, 2021 | Accepted April 27, 2021
Background: Drug resistance is the main cause of therapy failure in advanced lung cancer. Although non-genetic mechanisms play important roles in tumor chemoresistance, drug-induced epigenetic reprogramming is still poorly understood. Materials and Methods: The A549 cell line was used to generate cells with non-genetic resistance to cisplatin (CDDP), namely A549/CDDP cells. Bioorthogonal non-canonical amino acid tagging (BONCAT) and mass spectrometry were used to identify proteins modulated by CDDP in A549 and A549/CDDP cells. Results: Proteins related to proteostasis, telomere maintenance, cell adhesion, cytoskeletal remodeling, and cell redox homeostasis were found enriched in both cell lines upon CDDP exposure. On the other hand, proteins involved in drug response, metabolic pathways and mRNA processing and splicing were up-regulated by CDDP only in A549/CDDP cells. Conclusion: Our study revealed proteome dynamics involved in the non-genetic response to CDDP, pointing out potential targets to monitor and overcome epigenetic resistance in lung cancer.
Non-small cell lung cancer (NSCLC) is the leading cause of cancer-related death in the world (1). Cisplatin (CDDP)-based chemotherapy is the standard first-line treatment for inoperable, advanced NSCLC (2). However, drug resistance remains a major challenge for successful treatment. Cell models have been widely used to study cancer drug resistance (3, 4). However, the cell models with high levels of resistance used in most studies have shown limited translation to clinical application (5). In fact, cell lines considered clinically relevant exhibit resistance levels similar to those found in cells isolated from patients after chemotherapy (from 2- to 5-fold increase in resistance) (3). These clinically relevant models often show unstable resistance, which suggests that resistance acquisition is mainly due to gene expression reprogramming instead of genetic alterations.
Accumulating evidence has shown that non-genetic mechanisms play an important role in the chemoresistance of a variety of tumors (6, 7). During non-genetic evolution, gene expression programs that improve cancer cell adaptability and survival are selected and/or induced by drug treatment (7-9). However, drug-induced epigenetic reprogramming is still poorly characterized, mainly because it is a highly dynamic process whose analysis is technically challenging.
Bioorthogonal non-canonical amino acid tagging (BONCAT) is a powerful tool to monitor protein dynamics in response to a wide variety of stimuli. In the BONCAT method, an artificial amino acid (e.g. L-azidohomoalanine, a surrogate for L-methionine) carrying an azide or alkyne group is incorporated into newly synthesized proteins, thus allowing for selective detection or purification of tagged proteins by click chemistry (10, 11).
Herein, we performed a proteomic analysis of non-genetic resistance to CDDP in lung cancer cells. We used BONCAT and liquid chromatography-tandem mass spectrometry (LC-MS/MS) in a time-course analysis to selectively label, purify and identify proteins synthesized by CDDP-sensitive and -resistant A549 cells in response to drug exposure.
Cell culture and chemoresistance induction. An in vitro cellular model for studying non-genetic resistance to CDDP was developed from A549 cells. A549 cells were maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS), 100 units/ml penicillin and 100 μg/ml streptomycin in a humidified atmosphere of 5% CO2 at 37˚C. The CDDP-resistant subline, A549/CDDP, was obtained by a stepwise drug selection protocol, in which A549 parental cells (5×105) were continuously exposed to increasing concentrations of CDDP (0.1, 0.2, 0.3, 0.4 and 0.5 μM in 0.9% NaCl solution) for 72 h each. A549/CDDP cells were independently generated three times to produce the biological replicates used in the experiments. CDDP-resistant cells were maintained in culture medium containing 0.5 μM of CDDP to maintain the resistant phenotype. CDDP cytotoxicity was determined by sulforhodamine B (SRB) assay (12).
Metabolic labeling and click enrichment of newly synthesized proteins. For BONCAT assay, A549 and A549/CDDP cells were seeded at a density of 3×105 cells/well into 6-well plates and cultured for 18 h. Cells were conditioned in methionine-free RPMI medium (Thermo Fisher Scientific, Waltham, MA, USA) for 1 h at 37˚C to deplete methionine reserves, then the proteins were metabolically labeled with azidohomoalanine (AHA, 1 mM) for 2, 4 or 8 h in the absence or in the presence of CDDP. Each cell line was exposed to CDDP concentrations corresponding to their respective IC50 values. Cells were lysed and newly synthesized proteins were enriched using the Click-iT Protein Enrichment Kit (Thermo Fisher Scientific), as described previously (11). Briefly, AHA-containing proteins were captured onto an alkyne-agarose resin and non-specifically bound proteins were removed by washing with increasingly stringent buffers containing SDS, urea, isopropanol and acetonitrile. The resin-bound proteins were digested with trypsin and the generated peptides were desalted using Oasis HLB cartridges (Waters Corporation, Milford, MA, USA) following manufacturer’s instructions.
Mass spectrometry analysis. Peptides were analyzed by LC-MS/MS using a nanoACQUITY Ultra-Performance Liquid Chromatography (UPLC) system coupled to a Xevo G2-XS Q-Tof mass spectrometer (Waters Corporation) with a low-flow probe at the source. Peptides were separated by analytical chromatography (Acquity UPLC BEH C18, 1.7 μm, 2.1×50 mm, Waters Corporation) at a flow rate of 8 μl/min, using a 7-85% water/acetonitrile 0.1% formic acid linear gradient over 90 min. The MS survey scan was set to 0.5 s and recorded from 50 to 2,000 m/z. MS/MS scans were acquired from 50 to 2000 m/z, and scan time was set to 1 s. Data were collected in data-independent MSE mode. The mass spectrometry data were deposited to the ProteomeXchange Consortium via the PRIDE (13) partner repository with the dataset identifier PXD021779.
Data analysis and functional annotation. LC-MSE data were processed and searched using ProteinLynx Global Server (PLGS 3.0.3, Waters Corporation). The searches were conducted against Homo sapiens protein sequences retrieved from UniProtKB/Swiss-Prot database, with trypsin as enzyme, maximum of one missed cleavage, fixed carbamidomethyl modification for cysteine residues, and oxidation of methionine as variable modification. Peptides and protein tolerances were set as automatic, allowing minimum fragment ion per protein as 5, minimum fragment ion per peptide as 2, minimum peptide matches per proteins as 1 and false discovery rate (FDR) as 4%. Only proteins identified in two out of three biological replicates were considered for qualitative and quantitative analysis in order to improve confidence and reproducibility. Data sets were normalized using the “auto-normalization” function of PLGS and label-free quantitative analysis was performed from peak intensity measurements (Hi3 method) (14) using PLGS ExpressionE algorithm. Proteins with regulation-probability (P) values below 0.05 or higher than 0.95 were taken as differentially regulated between samples. Functional annotation and enrichment analysis were performed using PANTHER (Protein Analysis Through Evolutionary Relationships) database (15) matched with the Homo sapiens genome. The Fisher’s exact test was used with FDR correction. The plots of most representative and significant biological processes were constructed using ggplot2 R package.
A549/CDDP cells displayed clinically relevant levels of drug resistance, with an IC50 value 3.5-fold higher than that of the parental A549 cells (Figure 1A). A549/CDDP cells presented unstable resistance, gradually resuming the resistance level of the parental cells after ~30-45 days of cultivation in drug-free medium (data not shown). A549/CDDP cells, therefore, represent a cellular model for studying non-genetic resistance to CDDP.
Figure 1. Development of non-genetic resistance to CDDP in A549 cells. (A) Dose-response curve and calculated IC50 values. Values represent the mean±SD from biological triplicates. (B) Biological processes found enriched in the steady-state proteome of A549 and A549/CDDP cells. The dot color indicates the significance of the enrichment [-log10 (FDR- corrected p-values)] and the dot size indicates the number of proteins associated with each process. The vertical grey dashed line represents a fold enrichment of 1.
Protein dynamics induced by CDDP in A549 and A549/CDDP cells were evaluated by BONCAT and LC-MS/MS. Considering all time-points of the experiment (2, 4 and 8 h), a total of 173 and 151 unique proteins were identified from A549 cells cultured in absence and presence of CDDP, respectively. A total of 148 and 153 unique newly synthesized proteins were identified in A549/CDDP cells during the time-course experiment in absence and presence of CDDP, respectively. The complete lists of proteins identified in each experiment and sample are available at https://zenodo.org/record/4640561#.YIhJhJAzZPY.
Initially, we compared the proteins synthesized by A549 and A549/CDDP cells in drug-free medium to identify differences in their steady-state proteome. The complete list of proteins differentially expressed between A549 and A549/CDDP cells during culture in drug-free medium is available at https://zenodo.org/record/4640561#.YIhJhJAzZPY. A549 and A549/CDDP cells presented different expression profiles, with the enrichment of Gene Ontology (GO) terms related to protein folding, stabilization and turnover, telomere organization and maintenance, cellular adhesion, actin cytoskeleton, and cellular response to drug in A549/CDDP cells (Figure 1B). The complete list of GO annotations is available at https://zenodo.org/record/4640561#.YIhJhJAzZPY. Our results suggest that these biological processes were selected/induced by drug treatment during the acquisition of non-genetic resistance by A549/CDDP cells.
Next, proteins synthesized by A549 and A549/CDDP cells in the presence and absence of CDDP were compared, time-point by time-point, to identify proteome changes induced by drug exposure. The complete lists of proteins induced by CDDP in each time-point of the experiment are available at https://zenodo.org/record/4640561#.YIhJhJAzZPY. Considering all time-points of the experiment, proteins induced by CDDP in A549 and A549/CDDP cells are presented in Figure 2A and Table I. GO terms enriched by CDDP in each cell are shown in Figure 2B. GO terms related to telomere maintenance, protein folding and stabilization, cell adhesion, cytoskeleton, and cell redox homeostasis were found enriched upon CDDP exposure in both A549 and A549/CDDP cells, while proteins involved response to drug, metabolic pathways and regulation of mRNA processing/splicing were induced by CDDP only in A549/CDDP cells.
Figure 2. Overview of proteins and GO terms enriched in A549 and A549/CDDP cells after CDDP exposure. (A) Venn diagram of proteins upregulated by CDDP in each cell. (B) Biological processes found enriched after drug exposure. Detailed legend is the same as in Figure 1B.
Herein, we used BONCAT and LC-MS/MS to monitor gene expression reprogramming induced by CDDP in lung cancer cells with distinct phenotypes of drug sensitivity. Our results revealed that proteins involved in proteostasis, telomere maintenance, cell adhesion, cytoskeleton remodeling and cell redox homeostasis were up-regulated by CDDP in both A549 and A549/CDDP cells. These results highlight the importance of these molecular pathways in the non-genetic adaptive response to therapy. Interestingly, the profile of biological processes enriched in A549 cells after CDDP treatment is very similar to those identified in the steady-state proteome of A549/CDDP cells, which reinforces the participation of these molecular mechanisms in promoting cell survival during drug exposure and suggests their positive selection in CDDP-resistant cells. It is also important to note that the biological processes identified in our cellular model of non-genetic resistance resemble those described in the genetic resistance to CDDP (16, 17), which suggests that comparable pro-survival pathways could be activated by genetic and epigenetic mechanisms. In this sense, it is not surprising that most proteins up-regulated by CDDP are related to stress response pathways, as this signaling is of crucial importance in limiting cellular damage and enhancing cell survival.
Chaperones, foldases and proteases involved in the protein quality control network were found up-regulated by CDDP treatment in A549 and A549/CDDP cells. During stress response, cells use quality-control strategies to maintain protein homeostasis (proteostasis) (18). Thus, cells with up-regulated quality control machinery may be more efficient to cope with protein misfolding stress caused by CDDP exposure. In fact, unfolded protein response (UPR) activation is commonly observed in cancer and correlates with drug resistance (19).
Proteins reported to be involved in telomere maintenance, such as TRiC/CCT complex (20) and hnRNPs (21), had their expression increased in A549 and A549/CDDP cells after CDDP exposure. Human telomeric DNA (tandem repeats of 5’-TTAGGG-3’ sequences) is a potential target for CDDP-induced cross-links (22). Cell treatment with CDDP results in markedly shortened telomeres, which can induce apoptosis (23, 24). Thus, enhanced mechanisms of telomere maintenance can be involved in resistance to CDDP-induced apoptosis.
Proteins related to cell adhesion and cytoskeletal rearrangements were found up-regulated by CDDP exposure in both A549 and A549/CDDP cells. Cell adhesion to extracellular matrix (ECM) elicits activation of different pro-survival signaling pathways which contribute to tumor development and chemoresistance, in a mechanism referred as cell adhesion-mediated drug resistance (CAM-DR) (25, 26). In addition, cell adhesion triggers cytoskeleton reorganization, regulating cellular stiffness (26). Cisplatin has been reported to induce considerable remodeling of actin cytoskeleton, increasing stress fibers and cell stiffness (27). CDDP-resistant cell lines showed a significantly higher cell stiffness when compared to their drug-sensitive counterparts (28). Therefore, proteins involved in cell adhesion and cytoskeletal rearrangement could be relevant targets to counteract cellular resistance to CDDP (26, 29, 30).
The expression of detoxifying enzymes was induced by CDDP in A549 and A549/CDDP cells. CDDP generates a robust oxidative stress and, therefore, cells need to develop antioxidant mechanisms to deal with drug toxicity (31). Cells with an increased detoxification capacity have been reported to be more chemoresistant (32). In accordance, our results indicate that detoxifying enzymes may play a relevant role in non-genetic response to CDDP.
On the other hand, some biological processes, such as drug response, metabolic pathways, and mRNA processing and splicing, were up-regulated by CDDP only in A549/CDDP cells, which suggest their relevance to the development and/or maintenance of a drug resistance phenotype.
Proteins identified in A549/CDDP cells associated to drug response include ALDH3A1, HMOX1 and NQO1 proteins. Aldehyde dehydrogenases are markers of cancer stem cells and associated with cancer chemoresistance (33). HMOX1 and NQO1 have cytoprotective roles and enhance resistance to anticancer therapies (34, 35).
mRNA processing and splicing were also differentially represented in A549 and A549/CDDP cells upon CDDP exposure. Post-transcriptional mechanisms increase proteome diversity, enhancing tumor cell adaptation to chemotherapy (36). Therefore, it is likely that these mechanisms play major roles in the development of non-genetic resistance to CDDP.
Regarding mechanisms of transcriptional regulation, we identified the BTF3 transcription factor up-regulated in A549/CDDP cells after CDDP exposure. BFT3 expression has been associated with cancer stem cells (37), which are known to be involved in cancer growth, metastasis and chemoresistance.
Possible differences in the metabolism of A549 and A549/CDDP cells upon CDDP exposure were also detected by our proteomic approach, with the up-regulation of proteins involved in glycolysis and pentose phosphate pathways in A549/CDDP cells. Metabolic reprogramming is often associated with drug resistance, as the altered metabolism can confer adaptive, proliferative, and survival advantages in adverse conditions (38). Our results pointed that metabolic reprogramming could also be a non-genetic resistance mechanism to CDDP.
The results presented herein shed light on the mechanisms of gene expression regulation involved in the non-genetic resistance to CDDP. Knowing these mechanisms is a fundamental step for the development of novel strategies to monitor and counteract non-genetic resistance in cancer.
The Authors declare that no conflicts of interest exist with regard to the present study.
Conceptualization: C.S.D. and K.M.M.; Methodology: C.S.D. and K.M.M; Investigation: C.S.D., C.L.M. and N.A.C.; Formal analysis: C.S.D. and K.M.M.; Visualization: C.S.D.; Project administration: K.M.M.; Funding acquisition: K.M.M.; Resources: H.B.F., A.Z. and K.M.M.; Writing – original draft: C.S.D.; Writing – review and editing: H.B.F., A.Z. and K.M.M.
This work was funded by Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul (FAPERGS), grant number 16/2551-0000 286-0 (K.M.M.). C.S.D. and C.L.M. were supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) scholarships. N.A.C. was supported by a Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) scholarship. We thank the Uniprote-MS (CBiot/UFRGS) for technical support with LC-MS/MS.
Available at: https://zenodo.org/record/4640561#.YIhJhJAzZPY