Higher LSD1 in Non‐Seminomatous Germ Cell Tumors and LSD1 Inhibition by NCL1 Suppresses Embryonal Carcinoma Growth

2.1. Human TGCT Specimens

Orchiectomy specimens were originally collected from 57 patients with TGCT at our institute between 2004 and 2015. After exclusion of four pediatric patients, 53 patients with 74 histologically distinct tumor components were included in the revised analysis. Pathological review of the remaining pure teratoma (n = 1) and pure yolk sac tumor (n = 2) cases confirmed postpubertal‐type germ cell tumors and identified germ cell neoplasia in situ (GCNIS) in adjacent testicular tissue. The quantitative LSD1 and Ki‐67 analyses focused on tumor components; non‐neoplastic seminiferous tubules were excluded from the statistical comparisons. Representative images of non‐neoplastic tubules are retained in Figure 1A,B solely as a histological reference. This study was approved by the Institutional Review Board of Nagoya City University Hospital (Approval no. 60‐00‐1271). Written informed consent was obtained from all enrolled patients. Tumor specimens were reviewed by experienced pathologists.

2.2. Chemicals

The synthesis of NCL1 was as described in a previous paper [21].

2.3. TGCT Cell Lines

The human embryonal carcinoma (EC) cell lines, NTERA‐2 and Tera‐1, were obtained from the American Type Culture Collection. The medium used to culture cells was Dulbecco's Modified Eagle's Medium (DMEM) containing 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (P/S). Cells were grown at 37°C in 5% CO₂ with humidity. Cisplatin‐adapted NTERA‐2 cells (NTERA‐2R) were established by repeated exposure of parental NTERA‐2 cells to increasing concentrations of cisplatin, using a previously reported stepwise‐exposure approach with modifications [24]. Initially, cells were exposed to 0.2 μM cisplatin, corresponding to the IC₂₅ of the parental cells, for 48 h. The drug‐containing medium was then replaced with fresh medium, and the cells were allowed to recover until proliferation resumed. This procedure was repeated while gradually increasing the cisplatin concentration to 3 μM. Surviving cells were subsequently maintained in medium containing 1 μM cisplatin. After approximately 5 months of adaptation, a cell population capable of stable serial passaging was established and designated as NTERA‐2R.

2.4. Cell Proliferation Assay

Cell proliferation was evaluated using a WST‐8 Cell Counting Kit (Wako, Osaka, Japan). Testicular germ cell tumor cells in DMEM medium with 10% FBS and 1% P/S were seeded in 96‐well plates (6 × 10³ cells/well) and then treated with 1–100 μM NCL1 alone or in combination with 10 μM chloroquine (CQ) for 48 h. Growth rate was determined by the uptake of WST tetrazolium dye as measured by the rate of dye absorption. For the additional experiments using cisplatin‐adapted cells and evaluating the combined effects of NCL1 and cisplatin, cells were seeded in 96‐well plates at a density of 1 × 10⁴ cells/well. After 48 h of incubation, cells were treated with the indicated concentrations of cisplatin and/or NCL1 for an additional 48 h. In the combination experiments, NCL1 was administered at a fixed concentration of 5 μM together with increasing concentrations of cisplatin. Cell viability was assessed using the WST‐8 assay, and absorbance was measured at 450 nm. The experiments shown in Figure 2C,E,F were each repeated independently three times, with consistent qualitative trends. The data plotted in these panels are from one representative experiment for each assay.

2.5. Chromatin Immunoprecipitation Assay

NTERA‐2 cells were grown for 3 h 30 min in the presence or absence of 30 μM NCL1. Then, 1% formaldehyde was used to cross‐link cells. In a chromatin immunoprecipitation (ChIP) assay, chromatin was immunoprecipitated with an anti‐H3K9‐me2 antibody (Cell Signaling Technology, Beverly, MA, USA). Control antibody was isotype‐specific IgG. DNA was extracted from cells, dissolved in Tris‐EDTA buffer, and amplified in real‐time PCR with SOX2 and OCT4 specific primers: SOX2 F—CAGGAGTTGTCAAGGCAGAG, SOX2 R—GGAAAATCAGGCGAAGAATA; and OCT4 F—GCCACCACCATTAGGCAAAC, OCT4 R—GCGAAGGGACTACTCAACCC.

2.6. Western Blotting Analysis

Cells treated with the indicated concentrations of NCL1 and/or CQ were lysed in SDS buffer, and 10–30 μg protein samples were separated using 12% polyacrylamide gels and transferred to Hybond ECL membranes (GE Healthcare, Piscataway, NJ, USA). Primary antibodies and working dilutions were as follows: LSD1 (Cell Signaling Technology, 1:200), H3K9me1 (Active Motif, 1:400), H3K9me2 (Abcam, 1:300), SOX2 (Cell Signaling Technology, 1:1,000), OCT4 (Abcam, 1:500), NANOG (Abcam, 1:1,000), cyclin B1 (Abcam, 1:200), cyclin D1 (Santa Cruz Biotechnology, 1:200), CDK2 (Santa Cruz Biotechnology, 1:200), CDK4 (Santa Cruz Biotechnology, 1:300), caspase‐3 (Casp 3) (Cell Signaling Technology, 1:1,000), cleaved caspase‐3 (Cl. Casp 3) (Cell Signaling Technology, 1:750), survivin (Cell Signaling Technology, 1:1,000), LC3‐II (Abcam, 1:1,000), p62 (Abcam, 1:250), and beclin‐1 (Abcam, 1:1,000). β‐actin (Sigma‐Aldrich) was used as a loading control.

2.7. Flow Cytometry

NTERA‐2 cells (1 × 10⁵) were exposed to NCL1 for 72 h. Cell suspensions were subsequently prepared and stained with Guava Cell Cycle Reagent in accordance with the manufacturer's protocol (Guava Technologies, Hayward, CA, USA) for cell‐cycle analysis. Cell‐cycle phase distributions were assessed on a Guava PCA Instrument using CytoSoft software. For apoptosis analysis, cells treated with NCL1 for 24 h were stained using a phycoerythrin Annexin V Apoptosis Detection Kit with 7‐aminoactinomycin D according to the manufacturer's protocol (BioLegend, San Diego, CA, USA). Cells were analyzed using a FACSymphony A1 Cell Analyzer (BD, San Jose, CA, USA).

2.8. RNA Sequencing and Ingenuity Pathway Analysis

NTERA‐2 cells (1.5 × 10⁵) were incubated with or without 5 μM NCL1 for 48 h (n = 3 per group). ISOGEN reagent (Nippon Gene, Tokyo, Japan) was used to extract total RNA according to the manufacturer's protocol. The purity and quantification of the extracted RNA were evaluated using a SimpliNano Spectrophotometer (Biochrom Ltd., Cambridge, UK). Then, RNA library preparation and transcriptomic sequencing were performed by Macrogen Japan Corp. (Tokyo, Japan). Differentially expressed genes obtained from Macrogen Japan were imported into the Ingenuity Pathway Analysis (IPA) system (QIAGEN Inc., Düsseldorf, Germany) for canonical pathway analysis.

2.9. Lysosome Localization Using LysoTracker Red DND‐99

NTERA‐2 cells (4 × 10⁴) were seeded in 6‐well plates with DMEM medium containing 10% FBS. Cells were incubated for 48 h and treated with 5 μM NCL1 or both 5 μM NCL1 and 10 μM CQ. Solvent (dimethylsulfoxide [DMSO] and distilled water) was used on control cells. The medium was removed from each dish after 48 h of treatment. Prewarmed (37°C) probe‐containing medium was then added and cells incubated for 30 min. The loading solution was then replaced with fresh medium and cells observed using a BZ‐9000 multifunctional microscope (Keyence Japan, Osaka, Japan).

2.10. Fluorescent Immunocytochemistry Analysis

NTERA‐2 cells were incubated in 8‐well chamber slides with or without 10 μM CQ, and with or without 7.5 μM NCL1 for 12 h. The chamber slides were fixed by immersion in 4% formaldehyde for 10 min at room temperature, followed by permeabilization with digitonin. The primary anti‐LC3II antibody diluted with PBS was added and incubated for 1 h at room temperature. An anti‐rabbit polyclonal antibody (Abcam) at 1/500 dilution was used as the secondary antibody for 30 min at room temperature. DAPI (blue) was used to counterstain nuclei. Fluorescence images were observed and analyzed using an A1 confocal microscope (Nikon, Tokyo, Japan) and analysis software.

2.11. Transmission Electron Microscopy

NTERA‐2 cells (8 × 10⁴) were seeded in 6‐well plates in DMEM medium containing 10% FBS. After 48‐h incubation, cells were exposed to 7.5 μM NCL1, or both 7.5 μM NCL1 and 7.5 μM CQ, for 3 or 24 h. Cells were prefixed in glutaraldehyde (2.5%) in 0.1 M phosphate buffer (pH 7.4) at 4°C. Osmium tetroxide (1%) in 0.1 M phosphate buffer (pH 7.4) was used to postfix specimens for 45 min. A graded series of ethanol was used to dehydrate specimens, and these then embedded in epoxy resin. Ultra‐thin sections were cut with an Ultracut‐S ultramicrotome (LEICA, Wetzlar, Germany) and a diamond knife. These were then stained with 2% uranyl acetate in distilled water for 15 min and a lead staining solution for 5 min. Sections were observed by transmission electron microscopy (TEM) using a JEM‐1011 J (JEOL, Tokyo, Japan) electron microscope at 80 KV.

2.12. In Vivo Subcutaneous NTERA‐2 Model

We used 6‐week‐old male KSN/nu‐nu nude mice from Nippon SLC Inc. (Hamamatsu, Shizuoka, Japan). T‐75 flasks were seeded with NTERA‐2 cells, which were grown to confluence, trypsinized, and counted. NTERA‐2 cells (0.5 × 10⁶) resuspended in a mixture of 50 μL serum‐free DMEM and 50 μL matrigel were injected subcutaneously into the dorsal lumbar region of mice under isoflurane anesthesia. The administration dose of NCL1 in vivo was based on previous reports. In the initial experiment, mice were treated with vehicle or 1.0 mg/kg NCL1 (n = 15 and 15, respectively). NCL1 was injected intraperitoneally twice weekly, beginning 5 days after tumor inoculation, before tumors became measurable. Tumor size (caliper measurement) was measured twice weekly. Mice were subsequently sacrificed 5 weeks after the injection of cells and subcutaneous tumors, blood, livers, and kidneys were collected.

In a separate experiment evaluating established tumors, treatment began approximately 2 weeks after tumor inoculation, after measurable subcutaneous tumors had formed. Mice were randomized to vehicle, NCL1 0.5 mg/kg, or NCL1 1.0 mg/kg groups (n = 10 per group), with randomization stratified by baseline tumor volume. The mean baseline tumor volumes at randomization were 72.2 ± 30.2, 67.5 ± 37.4, and 68.7 ± 35.3 mm³ in the vehicle, NCL1 0.5 mg/kg, and NCL1 1.0 mg/kg groups, respectively. Intraperitoneal treatment was administered twice weekly for 4 weeks. Tumor volumes were calculated from caliper measurements, and body weights were recorded throughout the treatment period. Mice were subsequently sacrificed 4 weeks after the initiation of treatment of NCL1, and subcutaneous tumors, blood, livers, and kidneys were collected. Tumor volumes were calculated using the formula, V = (A × B ²)/2, where V represents volume in mm³, and A and B represent long and short diameters in mm, respectively. Protocols for animal experiments were approved by the Institutional Animal Care and Use Committee of Nagoya City University School of Medical Sciences (Approval number 21‐026, 26‐025).

2.13. Immunohistochemical Analysis

A 1:200 dilution of anti‐LSD1 (Cell Signaling) and anti‐Ki67 (Abcam) was incubated with deparaffinized human tissues. Deparaffinized animal tissues were exposed to 1:100 diluted anti‐CD31 (Santa Cruz) using a Bond‐Max automated staining system (Leica Biosystems) according to the manufacturer's standard protocol. Antibody binding was visualized by immunostaining using a Bond Intense R Detection system (Leica Biosystems) and 3,3′‐diaminobenzidine. In the evaluation of human specimens, LSD1 and Ki‐67 positivity was measured in individual histologically distinct tumor components using microscopy and image‐analysis software (Keyence Japan, Osaka, Japan). In mixed germ cell tumors, each histologically distinct component was evaluated separately. Accordingly, 74 components from 53 patients were treated as separate observations in the statistical analyses, including components obtained from the same patient. The positivity scores of LSD1 and Ki‐67 expression were evaluated in each histologically distinct TGCT component and in non‐neoplastic seminiferous tubules using BZ9000 multifunctional microscopy and analysis software (Keyence Japan, Osaka, Japan). For each tumor component, five measurements were obtained and averaged to generate a single positivity score. CD31‐positive vessels were evaluated in five randomly selected microscopic fields from the tumor tissue of each mouse.

2.14. TUNEL Assay

A terminal deoxy nucleotidyl transferase‐mediated dUTP nick end labeling (TUNEL) assay was used to detect apoptotic cells in deparaffinized tissues. This was undertaken using an In situ Apoptosis Detection Kit from Takara (Otsu, Japan) according to the manufacturer's instructions. For each mouse, five randomly selected microscopic fields from the tumor tissue were evaluated to determine the proportion of TUNEL‐positive cells.

2.15. Statistical Analysis

Quantitative data plotted as group means are presented as mean ± standard deviation (SD), and error bars indicate SD, unless otherwise specified. Individual data points, boxplots, patient characteristics, and correlation coefficients are reported in the formats indicated in the corresponding figures and tables. Student's t‐test was used for comparisons between two groups, and analysis of variance (ANOVA) or the Kruskal–Wallis test was used for comparisons involving three or more groups, as appropriate. Correlations between LSD1 and Ki‐67 positivity were assessed using Pearson's correlation coefficient. For the LSD1 immunohistochemical subtype comparisons, pairwise t‐tests comparing each non‐seminomatous histology with seminoma were adjusted using the Bonferroni method. For the established‐tumor experiment, the reported group comparisons were performed using ANOVA. For the TUNEL and CD31 analyses, measurements from five microscopic fields were averaged to obtain one value per mouse, and individual mice were used as the independent experimental units for statistical comparisons. Two‐sided p < 0.05 was considered statistically significant. Because multiple histologically distinct components from the same patient were analyzed as separate observations, the component‐level immunohistochemical analyses did not account for possible within‐patient correlation.

3.1. LSD1 Positivity Is Higher in Non‐Seminomatous Than Seminomatous Components and Correlates With Ki‐67 Positivity

The demographic and clinicopathological characteristics of the revised cohort are summarized in Table 1. Following exclusion of four pediatric patients, 53 patients were included (median age, 35 years; range, 20–74 years). The remaining pure teratoma (n = 1) and pure yolk sac tumor (n = 2) cases were confirmed as postpubertal‐type tumors, with germ cell neoplasia in situ identified in adjacent testicular tissue. Thirty‐five patients (66.0%) had pure seminoma and 13 (24.5%) had mixed germ cell tumors. In total, 74 histologically distinct tumor components were evaluated, comprising 42 seminomatous (56.7%) and 32 non‐seminomatous (43.3%) components. Non‐neoplastic seminiferous tubules were excluded from the quantitative comparisons; representative images are shown solely as a morphological reference (Figure 1A,B). Representative images of the tumor components are shown in Figure 1C–Q. LSD1 positivity was higher in pooled non‐seminomatous components than in seminoma (p = 0.0104; Figure 1R). Pairwise comparisons of individual non‐seminomatous histologies with seminoma, adjusted using the Bonferroni method, showed higher LSD1 positivity in teratoma and yolk sac tumor components (adjusted p < 0.05 for each). The embryonal carcinoma–seminoma comparison did not reach statistical significance (adjusted p = 0.08; Figure 1R). Because only two choriocarcinoma components were evaluated, the expression pattern in this subtype should be interpreted cautiously. LSD1 positivity correlated positively with Ki‐67 positivity in seminoma (r = 0.52, 95% CI: 0.24–0.72, p < 0.001), non‐seminomatous components (r = 0.46, 95% CI: 0.056–0.73, p < 0.05), and all tumor components (r = 0.57, 95% CI: 0.37–0.72, p < 0.001; Figure 1S–U).

3.2. NCL1 Suppresses Embryonal Carcinoma Cell Growth and Induces Apoptosis

We first examined LSD1 expression in human embryonal carcinoma cell lines. NTERA‐2 and Tera‐1 cells showed high LSD1 expression by western blotting (Figure 2A). To determine whether LSD1 inhibition influences gene‐specific histone methylation, NTERA‐2 cells treated with NCL1 were subjected to ChIP analysis. NCL1 increased H3K9me2 enrichment at the promoter regions of the pluripotency‐related genes SOX2 and OCT4 (Figure 2B). As shown in Figure 2C, NCL1 suppressed cell viability in both human embryonal carcinoma cell lines in a dose‐dependent manner. To further investigate the growth‐inhibitory effect of NCL1 in cells with reduced cisplatin sensitivity, we established cisplatin‐adapted NTERA‐2 cells (NTERA‐2R). The cisplatin IC₅₀ was 0.53 μM in parental NTERA‐2 cells and 3.18 μM in NTERA‐2R cells, representing a 6.0‐fold increase in the adapted cells (Figure 2D). NCL1 also suppressed cell viability in NTERA‐2R cells in a dose‐dependent manner (Figure 2E). Furthermore, the addition of NCL1 at 5 μM to cisplatin further reduced cell viability compared with cisplatin treatment alone in parental NTERA‐2 cells (Figure 2F).

To elucidate the mechanisms underlying NCL1‐induced growth inhibition, we evaluated proteins involved in pluripotency, apoptosis, and cell‐cycle regulation in NCL1‐treated NTERA‐2 cells. LSD1 expression remained largely unchanged, whereas H3K9me1 and H3K9me2 levels increased following NCL1 treatment. Consistent with the ChIP results, the expression of SOX2 and OCT4 was reduced. In addition, NANOG expression was decreased (Figure 3A). NCL1 treatment also reduced the expression of survivin, an inhibitor of apoptosis protein, and increased Cl. Casp 3 levels, whereas total Casp 3 expression remained largely unchanged (Figure 3A). In contrast, the expression of the examined cell‐cycle‐related proteins showed no marked changes. We therefore performed flow‐cytometric analyses of apoptosis and cell‐cycle distribution. NCL1 treatment increased the proportion of apoptotic NTERA‐2 cells in a dose‐dependent manner (Figure 3B), without markedly altering cell‐cycle distribution (Figure 3C).

3.3. NCL1 Provokes Autophagy‐Related Changes and Chloroquine Enhances Cytotoxicity

RNA sequencing followed by IPA suggested modest changes in autophagy‐related pathways after NCL1 exposure (activation z‐score, 1.61). We therefore examined autophagy–lysosome‐related responses and the effect of CQ on NCL1‐associated growth inhibition. NCL1 treatment increased LC3‐II and p62 levels, and these increases were further enhanced by CQ (Figure 4A). Co‐treatment with CQ was associated with lower WST‐8 absorbance than NCL1 alone at the indicated NCL1 concentrations (Figure 4B). The combination group also showed a higher apoptotic fraction in the displayed experiment (Figure 4C). LysoTracker Red fluorescence increased following NCL1 treatment (Figure 4D–F). Furthermore, LC3 puncta formation increased, consistent with LC3‐II accumulation at the cellular level (Figure 4G–I). TEM revealed an increase in autophagosome‐ and autolysosome‐like vesicular structures after NCL1 treatment, with further accumulation of enlarged vesicular structures after CQ co‐treatment (Figure 4J). Together, these findings indicate that NCL1 perturbs autophagy–lysosome dynamics and that pharmacologic lysosomal inhibition enhances NCL1‐associated cytotoxicity in embryonal carcinoma cells.

3.4. NCL1 Suppresses Early Xenograft Growth Accompanied by Increased TUNEL Positivity In Vivo

To assess the effect of NCL1 on tumor progression in vivo, nude mice were injected subcutaneously with NTERA‐2 cells. In the initial experiment, NCL1 treatment was initiated 5 days after tumor inoculation, before tumors became measurable. Tumor growth and the serum AFP level were significantly inhibited in mice of the 1.0 mg/kg NCL1 treated group compared to those in the vehicle control group (Figure 5A,D). NCL1‐treated tumors showed vacuolar degeneration and other histopathological findings consistent with an antitumor effect (Figure 5E,F). A TUNEL assay was used to study the mechanism of attenuated tumor growth by NCL1 in vivo. Compared with the vehicle control, NCL1 treatment significantly increased the number of TUNEL‐positive cells (Figure 5B,G,H). In addition, CD31‐positive vessels were significantly reduced in NCL1 treated tumors (Figure 5C,I,J). These findings suggest that NCL1 suppressed early xenograft outgrowth, accompanied by increased TUNEL‐positive cells and a reduction in CD31‐positive vessels.

To further evaluate the effect of NCL1 against established tumors, a separate experiment was performed in which treatment was initiated approximately 2 weeks after tumor inoculation, after measurable tumors had formed. Mice were randomized to vehicle, NCL1 0.5 mg/kg, or NCL1 1.0 mg/kg groups. No significant differences in tumor growth were observed among the three groups during the 4‐week treatment period (Figure S1A). No significant differences in body weight or relative liver and kidney weights were observed among the three groups. (Figure S1B, Table S1). Taken together, these findings indicate that NCL1 suppressed early NTERA‐2 xenograft outgrowth, accompanied by increased TUNEL positivity and fewer CD31‐positive vessels. In contrast, NCL1 did not significantly inhibit the growth of established xenografts under the treatment conditions examined.