Paclitaxel-Induced Epidermal Alterations: An In Vitro Preclinical Assessment in Primary Keratinocytes and in a 3D Epidermis Model

Abstract

Paclitaxel is a microtubule-stabilizing chemotherapeutic agent approved for the treatment of ovarian, non-small cell lung, head, neck, and breast cancers. Despite its beneficial effects on cancer and widespread use, paclitaxel also damages healthy tissues, including the skin. However, the mechanisms that drive these skin adverse events are not clearly understood. In the present study, we demonstrated, by using both primary epidermal keratinocytes (NHEK) and a 3D epidermis model, that paclitaxel impairs different cellular processes: paclitaxel increased the release of IL-1α, IL-6, and IL-8 inflammatory cytokines, produced reactive oxygen species (ROS) release and apoptosis, and reduced the endothelial tube formation in the dermal microvascular endothelial cells (HDMEC). Some of the mechanisms driving these adverse skin events in vitro are mediated by the activation of toll-like receptor 4 (TLR-4), which phosphorylate transcription of nuclear factor kappa B (NF-κb). This is the first study analyzing paclitaxel effects on healthy human epidermal cells with an epidermis 3D model, and will help in understanding paclitaxel's effects on the skin.

Keywords: 3D epidermis model; NHEK; epidermis; paclitaxel.

Figure 1

Paclitaxel does not modify cell viability and cytotoxicity. NHEK cells were incubated for 24 h with increasing paclitaxel concentrations. (A) Paclitaxel, at the concentrations assayed did not show alterations on cell viability measured by the MTT assay (B) nor in the cytotoxicity measured by LDH assay. Results are expressed as mean ± standard deviation of three independent experiments (n = 3). Multiple comparisons analysis of variance (ANOVA) was followed by the post hoc Bonferroni test. * p < 0.05 vs. control. PTX: paclitaxel. SLS: sodium lauryl sulfate.

Figure 2

Paclitaxel induces a dose-dependent inflammatory cytokine release in a 3D epidermal model. (A) Paraffin section from the 3D epidermis model stained with hematoxylin and eosin. Scale bar 100 µM. The 3D epidermal model was incubated for 24 h with increasing paclitaxel concentrations. (B) IL-1α, (C) IL-6, and (D) IL-8 levels were measured by ELISA. Results are expressed as mean ± standard deviation of three independent experiments (n = 3). Multiple comparisons analysis of variance (ANOVA) was followed by the post hoc Bonferroni test. * p < 0.05 vs. control. PTX: paclitaxel.

Figure 3

Paclitaxel induces a dose-dependent oxidative stress response in normal human epidermal keratinocytes (NHEK) cells and in a 3D epidermal model. (A) NHEK cells were incubated for 4 h with increasing paclitaxel concentrations. Quantification of reactive oxygen species (ROS) levels measured by the H₂DCF-DA assay. Data are expressed as reactive oxygen species (ROS) DCF relative fluorescence units. (B) 3D epidermal model tissues were incubated for 24 h with increasing paclitaxel concentrations. SOD1, Nrf2 and NOX4 mRNA levels were measured by real-time PCR. Data are expressed as 2^−ΔCt. (C) 3D epidermal model tissues were incubated for 24 h with increasing paclitaxel concentrations. SOD1, Nrf2 and NOX4 protein levels were analyzed by western blotting. Quantification was performed by densitometry and normalized to β-actin. Results are expressed as mean ± standard deviation of three independent experiments (n = 3). Multiple comparisons analysis of variance (ANOVA) was followed by the post hoc Bonferroni test. * p < 0.05 vs. control. PTX: paclitaxel.

Figure 4

Paclitaxel induces apoptosis in normal human epidermal keratinocytes (NHEK) cells. (A) NHEK cells were incubated for 24 h with increasing paclitaxel concentrations. Apoptosis was measured by flow cytometric analysis. (A) Representative plots for each paclitaxel concentration are displayed. (B) Apoptosis plots were analyzed by FlowJo software (TreeStar Inc., Ashland, OR, USA). Results are expressed as the mean apoptosis percentage of annexin-positive and propidium iodide-negative cells, which represent early apoptotic cells. (C–E) 3D epidermal model tissues were incubated for 24 h with increasing paclitaxel concentrations. P53, p21, and BCL2 mRNA levels were measured by real-time PCR. Data are expressed as 2^−ΔCt. (F) 3D epidermal model tissues were incubated for 24 h with increasing paclitaxel concentrations. P53 and BCL2 protein levels were analyzed by Western blotting. Quantification was performed by densitometry and normalized to β-actin. Results are expressed as mean ± standard deviation of three independent experiments (n = 3). Multiple comparisons analysis of variance (ANOVA) was followed by the post hoc Bonferroni test. * p < 0.05 vs. control. PTX: paclitaxel.

Figure 5

Paclitaxel inhibits endothelial tube formation in human dermal microvascular endothelial cells (HDMEC) and reduces eNOS and VEGF expression in the 3D epidermis. (A) HDMEC were incubated with increasing paclitaxel concentrations for 16 h, and angiogenesis was analyzed by the endothelial tube formation assay. Representative images of the tubular structures formed are displayed. Top images show the green, fluorescent calcein staining. Bottom images show the overlay generated by WimTube^TM software (Onimagin Technologies SCA, Córdoba, Spain), in which each color represents a structure: blue represents the covered area, red the tubes, white the branching points and yellow the number of loops. (B) Quantitative evaluation of morphological features of the capillary-like network structure. Tube length, total branching points and total loops after treating HDMEC with increasing paclitaxel concentrations. The analysis was performed using WimTube^TM software (Onimagin Technologies SCA, Córdoba, Spain). (C) 3D epidermal model tissues were incubated for 24 h with increasing paclitaxel concentrations. eNOS and VEGF mRNA levels were measured by real-time PCR. Data are expressed as 2^−ΔCt. (D) In vitro 3D epidermal model was incubated for 24 h with increasing paclitaxel concentrations. VEGF protein levels were analyzed by Western blotting. Quantification was performed by densitometry and normalized to β-actin. Results are expressed as mean ± standard deviation of three independent experiments (n = 3). Multiple comparisons analysis of variance (ANOVA) was followed by the post hoc Bonferroni test. * p < 0.05 vs. control. PTX: paclitaxel.

Figure 6

Paclitaxel activates the transcription factor NF-κB in a 3D epidermal model. The 3D epidermis was incubated for 1 h with increasing paclitaxel concentrations. (A) NF-κB and (B) p-NF-κB protein levels were analyzed by Western blotting. Quantification was performed by densitometry and normalized to β-actin. Results are expressed as mean ± standard deviation of three independent experiments (n = 3). Multiple comparisons analysis of variance (ANOVA) was followed by the post hoc Bonferroni test. * p < 0.05 vs. control. PTX: paclitaxel.

Figure 7

The effect of paclitaxel modulating inflammation, oxidative stress, apoptosis angiogenesis, and p-NF-κB is reduced in siRNA-TLR4 transiently transfected keratinocytes. Normal human epidermal keratinocytes (NHEK) were transiently transfected with control siRNA (-) or siRNA-TLR-4 and incubated for 24 h with paclitaxel 3 µM. (A–C) IL-1α, IL-6 and IL-8 supernatant levels were measured by ELISA. (D) Reactive oxygen species (ROS) were measured using H₂DCF-DA assay in NHEK stimulated with paclitaxel for 4 h. (E–J) The expression of SOD1, NOX4, Nrf2, BCL2, eNOS, and VEGF was measured by real-time PCR. Data are expressed as 2^−ΔCt. (K) NHEK cells were incubated for 1 h with paclitaxel concentrations. NF-κB and p-NF-κB protein levels were analyzed by Western blotting. Quantification was performed by densitometry and normalized to NF-κB/β-actin. Results are expressed as mean ± standard deviation of three independent experiments (n = 3). Multiple comparisons analysis of variance (ANOVA) was followed by the post hoc Bonferroni test. * p < 0.05 vs. siRNA (-) Control. # p < 0.05 vs. siRNA (-) PTX 3 µM. PTX: paclitaxel.