Aldehyde Dehydrogenase, a Therapeutic Target in Chordoma: Analysis in 3D Cellular Models

Abstract

Chordomas are rare, slow-growing tumors of the axial skeleton. These tumors are locally aggressive and refractory to conventional therapies. Radical surgery and radiation remain the first-line treatments. Despite these aggressive treatments, chordomas often recur and second-line treatment options are limited. The mechanisms underlying chordoma radioresistance remain unknown, although several radioresistant cancer cells have been shown to respond favorably to aldehyde dehydrogenase (ALDH) inhibition. The study of chordoma has been delayed by small patient cohorts and few available models due to the scarcity of these tumors. We thus created cellular 3D models of chordoma by using low-adherence culture systems. Then, we evaluated their radiosensitivity using colony-forming and spheroid size assays. Finally, we determined whether pharmacologically inhibiting ALDH increased their radiosensitivity. We found that 3D cellular models of chordoma (derived from primary, relapse, and metastatic tumors) reproduce the histological and gene expression features of the disease. The metastatic, relapse, and primary spheroids displayed high, medium, and low radioresistance, respectively. Moreover, inhibiting ALDH decreased the radioresistance in all three models.

Keywords: 3D models; aldehyde dehydrogenase; chordoma; combination therapy; hypoxia; radioresistance; radiotherapy.

Figure 1

Chordoma spheroids recapitulate the main histological and morphological features of the disease: at 7 days of culture, (a) chordoma markers brachyury, CD24, cytokeratins AE1/AE3, and EMA (40× objective) and their quantification were immunostained and (b) spheroid morphology was imaged by an electron microscopy. The arrows represent vacuoles, the stars represent junctions, the squares represent the extracellular matrix, f is the filopodia, and g is the glycogen granules.

Figure 2

Radioresistant chordoma spheroid environment: at 7 days of culture, (a) extracellular matrix components were immunostained, including proteoglycans with alcian blue and collagen with sirius red (20× objective). (b) The graph represents the median and confidence interval at 95% of the measure of the area of spheroids using live imaging from day 0 to day 21 after seeding for each spheroid. (c) The graph represents the median and confidence interval at 95% of the luminescence fluorescence intensity (LFI) correlated with cell survival from day 0 to day 21 after seeding for each spheroid. (d) The number of proliferative Ki67+ cells was evaluated, and the median and confidence intervals at 95% for each spheroid were quantified. (e) The hypoxic zones within spheroids were mapped by pimonidazole and HIF-1α staining. Each experiment was conducted in triplicate and repeated three times for spheroid proliferation assessment or twice for extracellular matrix and hypoxia-relative staining. For a comparison of the areas and cell survival, statistical analysis included 2-way ANOVA with time comparison, presented at the bottom of each graph, and cell line comparisons, presented on the right-hand side of the graphs. The comparison of the number of Ki67+ cells between spheroids was determined using a one-way ANOVA analysis. Significant p-values are indicated as follows: p < 0.05 *, p < 0.01 **, p < 0.001 ***, and p < 0.0001 ****.

Figure 3

Chordomas exhibit three different levels of radioresistance: (a) DNA double-stranded breaks induced by radiation and quantified using γH2AX foci staining, (b) images and (c) a graph representative of the number of colonies formed in untreated conditions or after 2 Gy of X-rays, and (d) images and (e) a graph representative of the spheroid size in untreated conditions or after 2 Gy of X-rays. Each experiment was repeated three times in triplicate. Comparisons between untreated and radiation-treated groups were analyzed with a two-way ANOVA. Significant p-values are indicated as follows: p < 0.01 **.

Figure 4

Aldehyde dehydrogenase (ALDH) is a promising radiosensitizing target in chordoma: (a) a graph representing ALDH gene expression scores in 13 chordoma patients, (b) representative images of ALDH3A2 staining after 2 Gy of X-rays or in untreated controls and quantification, (c) images representative of an Aldefluor assay quantifying ALDH1 and 3 activity in response to 2 Gy of X-rays in U-CH12 spheroids, (d) graph representative of the percent of ALDH^high cells, and (e) graph representative of the percent of ALDH^high cells relative to untreated conditions. Diethylaminobenzaldehyde (DEAB), an inhibitor of ALDH activity, was used as a negative control. The experiment was performed three times in triplicate for U-CH12 and CH22 spheroids, allowing for a comparison between the UT and 2 Gy groups using 2-way ANOVA. The p-value is indicated for each spheroid.

Figure 5

ALDH inhibition decreases radioresistance in chordoma: (a) a graph representative of three independent experiments of DIMATE IC50 in chordoma spheroids after 48 h of treatment at concentrations ranging from 0.1 to 25 μM, (b) images and (c) a graph representative of the CH22 spheroid size relative to untreated conditions after treatment with X-rays or after a combined treatment with X-rays and DIMATE, (d) images and (e) a graph representative of the number of colonies formed in UT conditions or after treatment with DIMATE as a monotherapy or combined with 2 Gy of X-rays in CH22 spheroids, and (f) a graph representative of spheroid cell death over 48 h in UT conditions or after treatment with DIMATE alone or in combination with 2 Gy of X-rays. Each experiment was conducted three times in triplicate. The statistical comparison between each group was determined using a one-way ANOVA. Significant p-values are indicated as follows: p < 0.05 *, p < 0.01 **, p < 0.001 ***, and p < 0.0001 ****.