Label-Free Quantitative Proteomics Analysis of Adriamycin Selected Multidrug Resistant Human Lung Cancer Cells

Abstract

The development of drug resistance in lung cancer is a major clinical challenge, leading to a 5-year survival rate of only 18%. Therefore, unravelling the mechanisms of drug resistance and developing novel therapeutic strategies is of crucial importance. This study systematically explores the novel biomarkers of drug resistance using a lung cancer model (DLKP) with a series of drug-resistant variants. In-depth label-free quantitative mass spectrometry-based proteomics and gene ontology analysis shows that parental DLKP cells significantly differ from drug-resistant variants, and the cellular proteome changes even among the drug-resistant subpopulations. Overall, ABC transporter proteins and lipid metabolism were determined to play a significant role in the formation of drug resistance in DKLP cells. A series of membrane-related proteins such as HMOX1, TMB1, EPHX2 and NEU1 were identified to be correlated with levels of drug resistance in the DLKP subpopulations. The study also showed enrichment in biological processes and molecular functions such as drug metabolism, cellular response to the drug and drug binding. In gene ontology analysis, 18 proteins were determined to be positively or negatively correlated with resistance levels. Overall, 34 proteins which potentially have a therapeutic and diagnostic value were identified.

Keywords: ABC transporters; DLKP; drug resistance; label-free quantitative proteomics; lipid metabolism; response to drug.

Figure 1

Establishment of drug-resistant cell lines. Resistance levels; pDLKP < DLKP-A2B < DLKP-A < DLKP-A5F < DLKP-A10.

Figure 2

Drug resistance and profile of DLKP and DLKP subpopulations. Fold resistance is calculated as Fold Resistance = IC50_{(subpopulation cell line)}/IC50_{(DLKP-parental cells)}. ±SD is obtained from at least 3 independent experiments. The figure is adapted from Keenan et al. [9].

Figure 3

Expression levels of ABCB1 in pDLKP cell line and drug-resistant subpopulations. (A) Normalised abundance of ABCB1 obtained from DE analysis of pDLKP cell line and drug-resistant subpopulations. For each cell line, data is obtained from 3 independent samples and one-way ANOVA is applied to show statistical significance. * and *** represents p < 0.05 and p < 0.001, respectively. (B) Fold-change of ABCB1 expression in pDLKP cell line and drug-resistant subpopulations. The pDLKP cell line is set to 1 for comparison of drug-resistant subpopulations. Each subpopulation is indicated with colors as follows; DLKPA2B (red), DLKP-A (blue), DLKP-A5F (green) and DLKP-A10 (purple). Data are shown as fold-change ± standard deviation from three independent experiments.

Figure 4

Network of proteins involved in fatty acid metabolism and their interaction with ABC-transporters. Proteins related to various biological processes and protein groups are colour coded as follows: Peroxisome related proteins and ABC-transporters are indicated with purple and pink, respectively. Proteins related to biological processes such as Fatty acid degradation, fatty acid metabolism, fatty acid biosynthesis and fatty acid elongation are indicated with red, green, yellow and dark green, respectively. Protein–protein interactions are shown via grey lines with thickness of the band indicating different strengths of the data support (experiments, data mining, neighbourhood co-occurrence are included to contribute confidence) identified by Gene Ontology analysis. The confidence strength was medium (0.400). (Full DE list and expression details, Supplementary File S1).

Figure 5

Expression profiles of upregulated membrane proteins in pDLKP and its drug-resistant subpopulations. For DE analysis 3 independent samples per cell line were analysed and normalised abundances of proteins of membrane proteins plotted against cell variant types. LIMA1 (LIM domain and actin-binding protein 1), ANXA1 (Annexin A1), SRI (Sorcin), CAST (Calpastatin), DIAPH1 (Protein Diaphanous Homolog 1), CLIP1 (CAP-Gly domain-containing Linker Protein 1).

Figure 6

Expression profiles of downregulated membrane proteins in pDLKP and its drug-resistant subpopulations. For DE analysis 3 independent samples per cell line were analysed and normalised abundances of proteins of membrane proteins plotted against cell variant types. TMUB1 (Ubiquitin Like Domain Containing 1), NEU1 (Sialidase-1), FSCN1 (Fascin Actin-Bundling Protein 1).

Figure 7

Western blotting of some of the target proteins (MDR1, FSCN1 and LIMA1) and comparison of abundance levels between MS and western-blotting. Full Western blot images of antibodies are provided in Supplementary Figure S5.