Blockade of CaV3 calcium channels and induction of G0/G1 cell cycle arrest in colon cancer cells by gossypol

Abstract

Background and purpose: Gastrointestinal tumours overexpress voltage-gated calcium (Ca_V3) channels (Ca_V3.1, 3.2 and 3.3). Ca_V3 channels regulate cell growth and apoptosis colorectal cancer. Gossypol, a polyphenolic aldehyde found in the cotton plant, has anti-tumour properties and inhibits Ca_V3 currents. A systematic study was performed on gossypol blocking mechanism on Ca_V3 channels and its potential anticancer effects in colon cancer cells, which express Ca_V3 isoforms.

Experimental approach: Transcripts for Ca_V3 proteins were analysed in gastrointestinal cancers using public repositories and in human colorectal cancer cell lines HCT116, SW480 and SW620. The gossypol blocking mechanism on Ca_V3 channels was investigated by combining heterologous expression systems and patch-clamp experiments. The anti-tumoural properties of gossypol were estimated by cell proliferation, viability and cell cycle assays. Ca²⁺ dynamics were evaluated with cytosolic and endoplasmic reticulum (ER) Ca²⁺ indicators.

Key results: High levels of Ca_V3 transcripts correlate with poor prognosis in gastrointestinal cancers. Gossypol blockade of Ca_V3 isoforms is concentration- and use-dependent interacting with the closed, activated and inactivated conformations of Ca_V3 channels. Gossypol and Ca_V3 channels down-regulation inhibit colorectal cancer cell proliferation by arresting cell cycles at the G₀/G₁ and G₂/M phases, respectively. Ca_V3 channels underlie the vectorial Ca²⁺ uptake by endoplasmic reticulum in colorectal cancer cells.

Conclusion and implications: Gossypol differentially blocked Ca_V3 channel and its anticancer activity was correlated with high levels of Ca_V3.1 and Ca_V3.2 in colorectal cancer cells. The Ca_V3 regulates cell proliferation and Ca²⁺ dynamics in colorectal cancer cells. Understanding this blocking mechanism maybe improve cancer therapies.

Keywords: CaV3 calcium channels; TTA‐P2; blocking mechanism; gastrointestinal cancers; gossypol; ion channel blockers.

Figure 1.. The upregulation of transcripts encoding for Ca_V3 channels in gastrointestinal tract malignancies is associated with adverse outcomes.

(a) Kaplan–Meier analysis for disease-free survival according to tumour expression of Ca_V3 channels genes (top, CACNA1G; middle, CACNA1H; bottom, CACNA1I). The two-sided log-rank test compared samples expressing low (blue lines) and high (red lines) levels of transcripts for each gene. Significant results were found for CACNA1G (HR 1.4, p = 0.019), CACNA1H (HR 1.6, p = 0.0009) and CACNA1I (HR 1.6, p = 0.002). (b) Transcriptional expression of Ca_V3 genes using the Affymetrix HGU133 Gene Array platform of the NCBI Gene Expression Omnibus (Bartha & Győrffy, 2021). Violin plots show the expression of CACNA1G, CACNA1H and CACNA1I genes in normal, tumoral and metastatic human tissues. Differences in expression levels among groups are statistically significant for CACNA1H (Kruskal–Wallis, p < 0.0001) and CACNA1I (Kruskal–Wallis, p < 0.0001). Results show that the copy number of transcripts of both genes is about two times more expressed in the metastatic group than in normal and tumour groups. (c) Immunohistochemical analysis of Ca_V3 channels in human colon tissue. Representative images of the expression of Ca_V3 channel proteins in normal and tumour human colon tissue. The positive signal (dark-brown precipitate) for all Ca_V3 subunits was more abundant in tumours than in healthy tissue. Additionally, the Ca_V3 proteins were localized in both the cytoplasm and the plasma membrane of cancer cells. Scale bar 50 μm.

Figure 2.. Relative abundance of Ca_V3 calcium channels in CRC cell lines.

(a) Plots show the relative expression (E^−ΔCt × 10⁵) of gene transcripts encoding for Ca_V3 channels (CACNA1G, CACNA1H and CACNA1I) in HCT116, SW480 and SW620 CRC cells. At the transcriptional level, the CACNA1H encoding for the Ca_V3.2 channel, followed by CACNA1G, encoding for Ca_V3.1, were most abundant in these cell lines; CACNA1I, on the other hand, was the least expressed isoform in CRC cells (n = 6). (b) Analysis of Ca_V3 channel expression by epifluorescence microscopy. The image panel shows the staining for calcium channel proteins (red, Alexa Fluor 647), plasma membrane glycoproteins (green, WGA conjugated with Alexa Fluor 488) and nuclei (cyan, DAPI) in HCT116 (upper panel), SW480 (middle) and SW620 cells (lower panel). Ca_V3 channels are concentrated at the plasma membrane of cancer cells. Scale bar 50 μm. (c) Colocalization analysis of Ca_V3 proteins and plasma membrane labelling. Green and red pseudo-colours fluorescent signals from images shown in (b) were used for Costes’ automatic threshold analysis (left). Cytofluorogram plots were obtained for Ca_V3 proteins and plasma membrane staining from the same images (right). The calculated Pearson’s coefficients were ≥ 0.7 for all conditions. * p < 0.05; Mann–Whitney rank sum test.

Figure 3.. Inhibition and voltage-dependence modification of recombinant human Ca_V3 channels by gossypol.

HEK-293 cells stably expressing the human cDNAs encoding Ca_V3 channels were used for the electrophysiological experiments. (a) Representative recordings of the steady-state blockage of Ca_V3.3 channels by the indicated concentrations of gossypol. Ionic currents were obtained by applying the stated voltage protocol every 10 s. (b) Time course of inhibition of Ca_V3.3 channel currents by gossypol. The amplitudes of the peak currents showed in (a) were normalized to the control amplitude (before gossypol exposure) and defined as normalized currents. (c) Concentration-response relationships of Ca_V3 calcium channels inhibited by gossypol (n = 7 – 15). Experimental data were fitted using Hill’s equation to calculate the IC₅₀ values and Hill coefficients (Table S2). (d) Ca_V3.1 currents were obtained in response to the illustrated voltage protocol before (Control, black traces) and after reaching a stationary inhibition by gossypol (Gossypol, red traces). (e) Current-voltage relationships of Ca_V3.1 channels were obtained under the indicated experimental conditions. Peak current amplitudes were normalized to the membrane capacitance (C_m) value of each cell (n = 7) to present the results as current density (pA/pF). (f) Normalized I-V curves of Ca_V3.1 show a leftward shift in the voltage dependence of activation in the presence of gossypol. (g) Voltage-dependence of Ca_V3.1 channels blocking by gossypol. Note that blockage of calcium currents was stronger at more positive potentials (+20 mV vs. ‒30 mV). This characteristic was not observed for Ca_V3.2 or Ca_V3.3 calcium channels. Biophysical parameters of the voltage-dependence of activation in the absence and presence of gossypol are presented in Table S3.

Figure 4.. Gossypol shifts steady-state inactivation curves of Ca_V3 channels to more negative potentials.

(a) Families of Ca_V3.1 channel currents recorded at ‒30 mV after 10-s prepulses to potentials spanning from ‒110 to ‒40 mV before and after exposure to 3 μM gossypol. Only the last 200 ms of prepulses current traces are presented for clarity. (b), (c) and (d) Steady-state inactivation curves of Ca_V3.1, Ca_V3.2 and Ca_V3.3 channels, respectively, in the absence, presence and after washout of gossypol. As voltage protocol illustrates, the peak of calcium currents obtained at ‒30 mV after pre-pulse application to different potentials was normalized to those obtained when the prepulse potential was ‒110 mV for each condition. The normalized currents were plotted as a function of the potential applied in the prepulses (n = 5 – 7). Smooth lines are data fits with the Boltzmann functions. Gossypol induced a leftward shift of the steady-state inactivation curves of the Ca_V3 channels, which is not entirely reversible after the drug washout. (e) The bars show the percentage of blocking of the Ca_V3 calcium currents at ‒30 mV using two different holding potentials (HP), ‒100 and ‒80 mV. At ‒80 mV, where more channels are inactivated, gossypol can block a more significant proportion of Ca_V3.1 and Ca_V3.2 calcium currents than when using ‒100 mV, potential at which the channels are mostly in a closed state and ready to be opened (n = 5 – 7). * p < 0.05, Mann–Whitney rank sum test.

Figure 5.. Gossypol stabilizes the inactivated state of Ca_V3 channels.

(a) Representative recordings showing recovery from the inactivation at −100 mV for Ca_V3.1 channels in the absence (black traces) and presence of 3 μM gossypol (red traces). Ca_V3 channel currents were inactivated by a 150 ms pulse to −30 mV. The membrane potential was stepped to −100 mV for periods ranging from 1 to 2000 ms, and at that time, a 20-ms activating voltage step to −30 mV was applied. Tail current recordings generated by repolarizing to −100 mV are off-scale. The dotted lines indicate the 100% recovery level. (b) Time course of recovery from inactivation at −100 mV for Ca_V3.1 channels in the absence, presence and after washout of gossypol. The plotted values were obtained by normalizing the peak currents during the 20-ms pulse to the peak current in the 150-ms pulse. Smooth lines are one-phase exponential fits of the experimental data (n = 5). (c) and (d) Identical experiments were performed for Ca_V3.2 and Ca_V3.3 calcium channels (n = 4 – 6). (e) Bars show the mean values of the constants tau of recovery from inactivation (𝜏_h) obtained by data fitting with exponential functions for each Ca_V3 channel and condition; gossypol significantly delays the output of Ca_V3 channels from the inactivated state. * p < 0.05, Mann–Whitney rank sum test.

Figure 6.. Gossypol decreases cell number and viability of CRC cells.

(a) CRC cells were incubated with 0.02% DMSO or 10 μM gossypol for 48 h, harvested and counted in a grid chamber using the trypan blue dye exclusion procedure. Gossypol significantly decreased cell number count in all the cell lines (n = 5 – 6). (b) Cells from (a) were analysed by flow cytometry using a viability dye. Graphs show the percentage of viable cells (left) and the percentage of dead cells (right). No change in viability or death percentage in those populations was observed. (c) Representative histograms for HCT116, SW480 and SW480 cell population incubated with Viobility 488/520 Fixable Dye from cells treated with DMSO (gray histograms) or gossypol (orange histograms). (d) Plots show the mean fluorescence intensity (MFI) values for both conditions and every cell line. Gossypol significantly increased the MFI values in all CRC cells. (e) HCT116, SW480 and SW620 cells were incubated in control (0.02% DMSO) and in several concentrations of gossypol for 24 or 48 h (n = 6). Experimental data were fitted using the Hill equation. IC₅₀ values at 24 h were 16.3, 15.5 and 68.6 μM for HCT116, SW480 and SW620, respectively. IC₅₀ obtained at 48 h were 9.2, 8.9 and 23.3 μM for HCT116, SW480 and SW620, respectively. * p < 0.05, Mann–Whitney rank sum test. NS, not statistically different.

Figure 7.. Gossypol inhibits CRC cell proliferation.

Cell proliferation was assessed by the increase in cell confluence percentage monitored with the IncuCyte instrument. (a) The left graph shows HCT116 cell proliferation for 48 h in the absence (black circles) and presence of gossypol 10 μM (red squares). On the right, representative pictures are shown at the beginning and end of the experiments for control and treated conditions. (b) and (c) Similar experiments were performed for SW480 and SW620 cells. In all cases, gossypol significantly inhibited CRC cell proliferation, as demonstrated by a flattened growth curve and almost no change in cell number of treated conditions at zero and 48 h time. * p < 0.05; Mann–Whitney rank sum test.

Figure 8.. Gossypol induces cell cycle arrest at the G₀/G₁ phase.

Flow cytometric analysis of CRC cells incubated with 0.02% DMSO or 10μM gossypol for 48 h. Cell cycle phases were studied by staining DNA with propidium iodide (PI). (a) Scatter and histogram plots for HCT116 cell populations in control (left) and gossypol (right) conditions. Black boxes enclose cell populations at the cell cycle phases. (b) and (c) Similar experiments were performed for SW480 and SW620 cells. Gossypol produced a strong decrease in cell population in the S phase and an important increase in the G₀/G₁ phase for HCT116 and SW480 cells. SW620 cells showed more modest but statistically significant changes in the same direction. (d) The graph summarizes the effects of gossypol on the cell cycle of CRC cells. In all cases, gossypol significantly increased the percentage of cell population in the G₀/G₁ phase while decreasing the cell populations in the S phase. No significant changes were observed in the cell populations at the G₂/M phase. * p < 0.05; Mann–Whitney rank sum test.

Figure 9.. Modulation of Ca_V3 channel expression influences cell viability in CRC cells.

(a) Percentage cell viability of CRC cells transfected as indicated and assessed with the MTT assay. The absorbances were normalized with respect to the control condition (siRNA-CTL and 0.02% DMSO). Downregulation of Ca_V3 channels and co-treatment with gossypol significantly decreased cell viability in all the cell lines (n = 5 – 7). (b) Changes in cell viability of CRC cells in response to the indicated transfection conditions. Overexpression of Ca_V3.2 channels significantly increased cell viability transiently in all cell lines although gossypol co-treatment masked this increase (n = 5 – 7). (c) Flow cytometry analysis of cells stained with propidium iodide. Scatter plots for SW480 cell populations in control, downregulation of Ca_V3 channels and co-treatment with gossypol experimental conditions. Black boxes enclose cell populations at the cell cycle phases. (d) Summary of the effects of downregulating Ca_V3 channels on SW480 cell cycle. Decreased expression of Ca_V3 channels significantly increased the G₂/M cell population without affecting the G₀/G₁ and S phase populations (n = 5). * p < 0.05; Mann–Whitney rank sum test. NS, not statistically different.

Figure 10.. Pharmacological inhibition of Ca_V3 channels reduces vectorial Ca²⁺ flux from the extracellular space to the ER in CRC cells.

(a) Simultaneous Ca²⁺ dynamics in the ER (lower panels) and cytosol (upper panels) in CRC cells. 2 × 10⁶ cells were pre-incubated for 1 h with DMSO (black trace), with the specific inhibitor of Ca_V3 channels, TTA-P2 (green trace), or with Ca_V1 channels inhibitor, nifedipine (magenta trace). After calcium signals reached a steady state, a depolarizing stimulus was applied by adding a solution containing 40 mM K⁺ and 5mM Ca²⁺ (black box) as a means to assess the contribution of Ca_V3 channels to the Ca²⁺ dynamics. Significant increases in ER Ca²⁺ were observed in response to the depolarizing stimulus, but there were very discrete or no changes in cytosolic calcium levels. The participation of the Ca_V1 channel subfamily in this Ca²⁺ flux was explored using the inhibitor nifedipine (magenta trace). (b) Cell-type specific variations of Ca²⁺ in the ER in response to the depolarizing stimulus. Scatter plots show Ca²⁺ increases in the ER of CRC cells treated with specific inhibitors of Ca_V3 (5 μM TTA-P2, green bars) and Ca_V1 channels (1 μM nifedipine, magenta bars). Only TTA-P2, the specific inhibitor of Ca_V3 channels, significantly decreased Ca²⁺ uptake by the ER in HCT116 and SW480 cells. No significant changes were observed when cells were treated with nifedipine (n = 5). * p < 0.05; Mann–Whitney rank sum test. NS stands for not statistically different.