Chemosensitization of HT29 and HT29-5FU Cell Lines by a Combination of a Multi-Tyrosine Kinase Inhibitor and 5FU Downregulates ABCC1 and Inhibits PIK3CA in Light of Their Importance in Saudi Colorectal Cancer

Abstract

Colorectal cancer (CRC) remains one of the main causes of death worldwide and in Saudi Arabia. The toxicity and the development of resistance against 5 fluorouracil 5FU pose increasing therapeutic difficulties, which necessitates the development of personalized drugs and drug combinations.

Objectives: First, to determine the most important kinases and kinase pathways, and the amount of ABC transporters and KRAS in samples taken from Saudi CRC patients. Second, to investigate the chemosensitizing effect of LY294002 and HAA₂₀₂₀ and their combinations with 5FU on HT29, HT29-5FU, HCT116, and HCT116-5FU CRC cells, their effect on the three ABC transporters, cell cycle, and apoptosis, in light of the important kinase pathways resulting from the first part of this study.

Methods: The PamChip^® peptide micro-array profiling was used to determine the level of kinase and targets in the Saudi CRC samples. Next, RT-PCR, MTT cytotoxicity, Western blotting, perturbation of cell cycle, annexin V, and immunofluorescence assays were used to investigate the effect on CRC, MRC5, and HUVEC cells.

Results: The kinase activity profiling highlighted the importance of the PI3K/AKT, MAPK, and the growth factors pathways in the Saudi CRC samples. PIK3CA was the most overexpressed, and it was associated with increased level of mutated KRAS and the three ABC transporters, especially ABCC1 in the Saudi samples. Next, combining HAA₂₀₂₀ with 5FU exhibited the best synergistic and resistance-reversal effect in the four CRC cells, and the highest selectivity indices compared to MRC5 and HUVEC normal cells. Additionally, HAA₂₀₂₀ with 5FU exerted significant inhibition of ABCC1 in the four CRC cells, and inhibition of PIK3CA/AKT/MAPK7/ERK in HT29 and HT29-5FU cells. The combination also inhibited EGFR, increased the preG₁/S cell cycle phases, apoptosis, and caspase 8 in HT29 cells, while it increased the G₁ phase, p21/p27, and apoptosis in HT29-5FU cells.

Conclusion: We have combined the PamChip kinase profiling of Saudi CRC samples with in vitro drug combination studies in four CRC cells, highlighting the importance of targeting PIK3CA and ABCC1 for Saudi CRC patients, especially given that the overexpression of PIK3CA mutations was previously linked with the lack of activity for the anti-EGFRs as first line treatment for CRC patients. The combination of HAA₂₀₂₀ and 5FU has selectively sensitized the four CRC cells to 5FU and could be further studied.

Keywords: 5FU; ABCC1; PIK3CA; Saudi colorectal cancer; drug resistance; kinase pathway profiling.

Figure 1

Molecular structures of (A) 5FU, (B) LY294002, and (C) HAA_2020.

Figure 2

(A) Pathway map of PIK3CA/AKT. Symbols looking like a thermometer represent the kinase analysis data. Red color represents protein levels of phosphorylation. (B) Peptides with the highest signals in the PI3K/AKT network (PIK3CA red-circled).

Figure 3

(A) The highest-ranking networks in the Saudi CRC samples (Supplementary material S1: process network). (B) Gene ontology (GO) processes of the Saudi CRC samples, sorted by statistically significant processes (Supplementary material S1: GO processes).

Figure 4

Total mRNA was isolated from the 10 Saudi patients (p1–p10) and quantified by RT-PCR with (A) ABCB1, (B) ABCC1, and (C) ABCG2 primers. The results were expressed as relative fold change (average ± SD, n = 3, x2 independent experiments) compared with GAPDH (1-fold change). W ^a: wild type KRAS, M ^b: mutated KRAS. Statistical differences (one-way ANOVA and Tukey’s post-hoc): p < 0.05 (*), p < 0.01 (**), p < 0.001 (***) were considered significant.

Figure 4

Total mRNA was isolated from the 10 Saudi patients (p1–p10) and quantified by RT-PCR with (A) ABCB1, (B) ABCC1, and (C) ABCG2 primers. The results were expressed as relative fold change (average ± SD, n = 3, x2 independent experiments) compared with GAPDH (1-fold change). W ^a: wild type KRAS, M ^b: mutated KRAS. Statistical differences (one-way ANOVA and Tukey’s post-hoc): p < 0.05 (*), p < 0.01 (**), p < 0.001 (***) were considered significant.

Figure 5

Total mRNA was isolated from HT29, HT29-5FU, HCT116, and HCT116-5FU cells and quantified by RT-PCR with ABCB1, ABCC1, and ABCG2 primers. The results were expressed as relative fold change (average ± SD, n = 3, ×2 independent experiments) compared with GAPDH (1-fold change). Statistical differences (one-way ANOVA, Tukey’s post-hoc): p < 0.05 (*), p < 0.01 (**), and p < 0.001 (***) were considered significant.

Figure 6

The inhibitory effect (72 h) of vehicle control, 5FU (0.25 µM), HAA₂₀₂₀ (3 µM), or their combination on the expression of mRNA of ABCB1, ABCC1, and ABCG2 in (A) HT29, (B) HT29-5FU, (C) HCT116, and (D) HCT116-5FU cells was quantified by RT-PCR. The data represent the mean ± SD of the fold change related to vehicle control (fold change = 1 dashed line, n = 2, ×2 independent experiments). p < 0.05 (*), p < 0.01 (**), and p < 0.001 (***) were considered significant.

Figure 7

The expression of (A) PIK3CA, AKT, and MAPK7, and (B–D) AKT, p-AKT, ERK, and p-ERK in HT29 cells, and expression of (E) PIK3CA, AKT, and MAPK7, and (F–H): AKT, p-AKT, ERK, and p-ERK in HT29-5FU cells. All cells were treated with vehicle, 5FU (0.25 µM), HAA₂₀₂₀ (3 µM), and their combination for 72 h. Results in (A) and (E) represent the mean ± SD of the mRNA fold change related to vehicle control (fold change = 1 dashed line). Results in (C,D,G,H) represent the mean ± SD of the fold change percentage (%, Y axis) of the relative protein levels normalized to GAPDH, n = 2, ×2 independent experiments. p < 0.05 (*), p < 0.01 (**), and p < 0.001 (***) were considered significant.

Figure 8

Cell cycle analysis of (A) HT29, and (B) HT29-5FU cells (mean ± SD, n = 3) treated for 72 h with either vehicle, 5FU (0.25 µM), HAA₂₀₂₀ (3 µM), or 5FU (0.25 µM) + HAA₂₀₂₀ (3 µM).

Figure 9

Induction of apoptosis in HT29 and HT29-5FU cells (mean ± SD, n = 3, x3 independent experiments), which were treated (72 h) with either (A,F) vehicle, (B,G) 5FU (0.25 µM), (C,H) HAA₂₀₂₀ (3 µM), or (D,I): their combination. Stacked histograms of the different drug treatments in (E): HT29 and (J): HT29-5FU cells. C1: necrosis, C2: late apoptosis, C3: live cells, C4: early apoptosis.

Figure 9

Induction of apoptosis in HT29 and HT29-5FU cells (mean ± SD, n = 3, x3 independent experiments), which were treated (72 h) with either (A,F) vehicle, (B,G) 5FU (0.25 µM), (C,H) HAA₂₀₂₀ (3 µM), or (D,I): their combination. Stacked histograms of the different drug treatments in (E): HT29 and (J): HT29-5FU cells. C1: necrosis, C2: late apoptosis, C3: live cells, C4: early apoptosis.

Figure 10

Detection of the immunofluorescence in (A,B) (expression %): HT29 and (E,F) (expression %): HT29-5FU cells (EGFR and p27: green, caspase-8 and p21: red, stained with DAPI, 15 μm; 40× objective). (I) vehicle control, (II) cells incubated with 5FU, (III): cells incubated with HAA₂₀₂₀, (IV): cells incubated with 5FU and HAA₂₀₂₀ for 72 h. Quantification of EGFR level amounts in HT29 cells by Western blotting (C,D). Quantification of p21 level amounts in HT29-5FU cells by Western blotting (G,H). Results in (D,H) represented the mean ± SD of the fold change percentage (%, Y axis) of the relative protein levels normalized to GAPDH, n = 2, ×2 independent experiments. p < 0.05 (*), p < 0.01 (**), and p < 0.001 (***) were considered significant.

Figure 10

Detection of the immunofluorescence in (A,B) (expression %): HT29 and (E,F) (expression %): HT29-5FU cells (EGFR and p27: green, caspase-8 and p21: red, stained with DAPI, 15 μm; 40× objective). (I) vehicle control, (II) cells incubated with 5FU, (III): cells incubated with HAA₂₀₂₀, (IV): cells incubated with 5FU and HAA₂₀₂₀ for 72 h. Quantification of EGFR level amounts in HT29 cells by Western blotting (C,D). Quantification of p21 level amounts in HT29-5FU cells by Western blotting (G,H). Results in (D,H) represented the mean ± SD of the fold change percentage (%, Y axis) of the relative protein levels normalized to GAPDH, n = 2, ×2 independent experiments. p < 0.05 (*), p < 0.01 (**), and p < 0.001 (***) were considered significant.