D‑galactose induces astrocytic aging and contributes to astrocytoma progression and chemoresistance via cellular senescence

Introduction

Cellular senescence was originally reported as definite proliferative capacity in human fibroblasts culture, which reflects one particular type of senescence produced by the absence of endogenous telomere activities (1). Apart from telomere erosion, various noxious stimuli such as DNA lesions and reactive oxygen species (ROS) have also been shown to induce a senescent growth arrest in vitro (2–5). Cellular senescence was originally proposed to prevent the growth of damaged cells, indicating a safeguard against cancer (6,7). Subsequent studies during past decades convincingly demonstrated that senescent cells are involved in aging, aging-related dysfunction and chronic diseases (8,9). Intriguingly, senescent cells have a complex senescence-associated secretory phenotype (SASP). SASPs were postulated that they contributed to the development and processing of age-associated pathologies including malignant transformation by changing tissue microenvironment (10–12).

In the brain, the aging process is associated with neuronal degeneration and loss, which eventually lead to cognitive impairment. Thus, aging is widely recognized as one the most prominent risk factors for Alzheimer Disease (AD). Brain aging is typically accompanied with the suppression of innate immunity, favoring a pro-inflammatory status (13). Astrocytes in aging brains can trigger SASP, presenting a typical proinflammatory phenotype, suggesting that senescent astrocytes drives a low-level, chronic inflammatory status in aged brains (14). Therefore, models of senescence may permit investigations into potential cellular mechanisms of astrocytic senescence, in which treatments for delaying or preventing brain aging and aging-related neuroinflammation or subsequent degeneration could be developed.

Chronic low dose D-galactose treatment leads to accelerated aging in rodent and Drosophila models. Interestingly, rodents chronically treated with D-galactose showed progressive decline in learning and memory abilities, neurodegeneration and a damaged immune system (15–17). Drosophila models exhibited shortened lifespans and increasing oxidative stress. While D-galactose-induced models have been widely reported for the study of brain aging processes and drug candidate screening, how galactose affects organismal brain aging remains unclear. In the present study, we report evidence indicating D-galactose-induced astrocytic senescence.

Materials and methods

Reagents and cell culture

D-galactose and Bay 11–7082 were purchased from Sigma-Aldrich (Merck KGaA). Astrocytic CRT and U373-MG Uppsala cells (kindly provided by Professor Chul-hee Choi, Korea Advanced Institute of Science and Technology, South Korea) were maintained in RPMI-1640 medium (Thermo Fisher Scientific, Inc.) with 10% heat-inactivated fetal bovine serum (FBS), 100 U/ml of penicillin, and 100 µg/ml of streptomycin (Thermo Fisher Scientific, Inc.) as previously described (18). N2a cells (ATCC® CCL-131™) were grown in Dulbecco's Modified Eagles medium (DMEM; Thermo Fisher Scientific, Inc.) supplemented with 1 or 10% heat-inactivated FBS (Thermo Fisher Scientific, Inc.), 100 U/ml of penicillin and 100 µg/ml of streptomycin. Primary rat astrocytes were maintained in 10% FBS-DMEM containing 1% nonessential amino acids (Gibco; Thermo Fisher Scientific, Inc.). Stable cell line CRT-MG/IL-8p-d2EGFP cells were prepared and maintained as previously described (19). All the cells were incubated at 37°C in a 5% CO2 atmosphere.

Senescence induction and senescence-associated β-galactosidase (SA-β-gal) staining

Human astrocytic CRT cells and rat primary astrocytes were treated with varying doses of D-galactose resolved in culture medium (0–60 g/l) at 37°C and the cell viability after 72 h exposure was determined. The half-maximal inhibitory concentration, 50 g/l, of D-galactose was used for the subsequent experiments. SA-β-gal staining was performed using an SA-β-gal kit (cat. no. 9860, Cell Signaling Technology, Inc.) in accordance with the manufacturer's protocols to confirm cellular senescence. The cells were fixed for 10–15 min at room temperature, then rinsed twice with PBS and stained with staining solution at a final pH of 6.0 overnight. The SA-β-gal-positive cells were seen as blue, and were counted under a phase-contrast microscope (magnification, ×100; 5 fields per view analyzed). The experiment was repeated three times in each group.

Cell viability assay

Cell viability was evaluated by a WST-1 assay, which is based on the enzymatic cleavage of the tetrazolium salt WST-1 to formazan by cellular mitochondrial dehydrogenase present in viable cells. In brief, after 24 h following treatment, 20 µl of WST-1 was added to each well (12-well plate with 1×105 cells per well) and the plates were incubated at 37°C for 2 h. The plates were employed for centrifugation (300 × g) at room temperature for 3 min and 100 µl of the medium was withdrawn and analyzed by measuring the absorbance at a wavelength of 450 nm using microplate reader (Tecan Group, Inc.).

Preparation of conditioned medium (CM) from CRT cells

Astrocytic CRT cells (8×105 cells in 90 mm petri dish) were exposed to D-galactose (50 g/l) at 37°C for 10 days to induce premature secretory senescence. Then, 8 days following senescence induction, cells were cultured with complete culture medium (without 10% FBS) for the collection of CM. The CM was collected from both pre-senescent (untreated cells) and senescent cells after 48 h. CM were centrifuged for 20 min at 800 × g at 4°C, filtered through 0.22 µm bottle-top filters (Sartorius Stedim Biotech) and used for subsequent experiments. CM was used to treat U373-MG and N2a cells with different percentages (25, 50, and 100%) at 37°C for 24 h. Cells were then exposed with or without 0.5 mM Temozolomide (TMZ, Sigma-Aldrich; Merck KGaA) at 37°C for 48 h followed by cell viability tests using the WST-1 method and observations of morphological changes by microscopy (magnification, ×100; 6 fields per view analyzed).

Interleukin (IL)-6 and IL-8 levels in the CM were quantified using the Human IL-6 and IL-8 ELISA kits (cat. nos. D6050 and D8000C, R&D Systems, Inc.) according to the manufacturer's protocols. Briefly, 100 µl of culture media normalized by cell number was added in the wells with five replicates and incubated for 2.5 h at room temperature with gentle agitation. Then, the medium was discarded and wells were washed with washing buffer for four times. After that, 100 µl of human IL-6 or IL-8 conjugate were added to each well and plates was incubated for 2 h at room temperature with gentle agitation. Following washing with washing buffer, 100 µl of substrate solution was added to each well and incubated for 20 min at room temperature. Lastly, 50 µl of stop solution was added and wells were assayed with a microplate reader set to 450 nm within 30 min.

Immunoblotting

Whole cell lysate (20 mg) was prepared from various treatments. The samples (30 µg determined by Bradford Assay) were loaded onto 10% SDS-PAGE and separated by electrophoresis for 2 h at 100 V. Protein were transferred polyvinylidene difluoride membrane for 1 h at 110 V. The following primary antibodies were used: Antibody p16ink4a, p53, p21, PARP, Cleaved caspase 3, c-Myc and Tju1 (rabbit polyclonal, 1:2,000, Abcam, cat. nos. ab51243; ab32389; ab109520; ab74290; ab2302; ab39688 and ab18207, respectively); senescence marker protein 30 (SMP30; mouse monoclonal, 1:1,000, Santa Cruz Biotechnology, Inc., cat. no. sc-390098), GAPDH as a loading control (mouse monoclonal, 1:4,000, Santa Cruz Biotechnology, Inc., cat. no. sc-47724). Membranes were washed and incubated with horseradish peroxidase-conjugated secondary antibodies (1:5,000; Cell Signaling Technology, Inc., cat. nos. sc-2370 and sc-2380). The bands from the western blots were densitometrically visualized by ECL detection and the signals quantified using ImageJ software (version 1.5.1, National Institutes of Health).

For the immunofluorescence assay, the cells were first washed twice with PBS, then fixed with 4% paraformaldehyde for 15 min and permeabilized with buffer (0.15% Triton X-100 in PBS) for 20 min at room temperature. Cells were blocked with 3% bovine serum albumin (Sigma-Aldrich; Merck KGaA) for 30 min and then incubated with primary antibodies against NF-κB p65 (rabbit polyclonal, 1:200, Santa Cruz Biotechnology, Inc., cat.no. sc-8008), and IL-6 (mouse monoclonal, 1:1,000, Abcam) overnight at 4°C. Following washing three times for 5 min each with PBS and 0.05% Triton X-100. Cells were incubated with appropriate Alexa Fluor® secondary antibodies 488 or 594 (Invitrogen; Thermo Fisher Scientific, Inc., cat. nos. A-11008; A21209) to detect the signal at room temperature for 1.5 h. After another set of washing, images were captured with a Nikon i2 U fluorescence microscope (magnification, ×100; Nikon Corporation).

Optical density analysis was conducted to quantify the fluorescence signal and was performed by using CellProfiler Software (2.2.0, http://cellprofiler.org). Each evaluation was conducted on five fields randomly selected for each of the target proteins.

Statistical analysis

All data were expressed as the mean ± standard deviation except unless otherwise indicated. P-values were generate using a Student's t-test or one-way analysis of variance; P<0.05 was considered to indicate a statistically significant difference. A protected Fisher's Least Significant Difference post hoc test was used for multiple comparisons. All calculations were performed using GraphPad Prism software (v7.0, GraphPad Software, Inc.).

Results

D-galactose induces the senescence of astrocytic CRT cells and rat primary astrocytes

To determine whether D-galactose can induce the senescence of astrocytic cells, we first cultured astrocytic CRT cells and rat primary astrocytes in the absence or presence of varying doses of D-galactose (0–60 g/l) and observed the cell viability after 72 h treatment (Fig. 1A). Treatment with D-galactose significantly suppressed cellular viability in a dose-dependent manner from 40 g/l of D-galactose upwards. Importantly, a higher dose of D-galactose (60 g/l) induced a considerable cell death (profound apoptotic bodies) as observed under microscopy. Then, we investigated further the classic markers of the senescence-related phenotype. The astrocytic CRT and rat primary cells exhibited flattened morphology and significantly increased levels of SA-β-gal activity compared with the control (Fig. 1B). Additionally, the expression of hallmark regulatory proteins, such as p16, p53, p21 and SMP30 were markedly elevated after D-galactose exposure (Fig. 1C). Taken together, these observations demonstrated that, in a relative long-term of 3 days with D-galactose, astrocytic cells were susceptible to the stress, thereby inducing a characteristic senescent phenotype.

Figure 1. D-gal treatment suppresses cell viability and induces cellular senescence of astrocytic CRT cells and rat primary astrocytes. (A) Astrocytic CRT cells and rat primary cells were incubated with different concentrations of D-gal (0, 10, 20, 40 and 60 g/l) for 72 h. Then, cells were analyzed for their viability. (B) Astrocytic CRT cells and rat primary astrocytes were treated with 50 g/l D-galactose for 5 days, and the percentage of cells expressing senescence-associated-β-galactosidase was determined by light microscopy (top panel) and counting (bottom panel; both magnification, ×100). (C) Extracts from non-senescent or senescent (D-gal) astrocytic CRT cells and rat primary astrocytes were assayed for the indicated senescence marker proteins (p16, p21, SMP30 and p53) by western blotting. GAPDH served as loading control. *P<0.05 and **P<0.01 vs. control group. Three independent experiments were conducted. D-gal, D-galactose; SMP30, senescence marker protein 30.

Senescent astrocytic CRT cells develop the SASP phenotype

Since the treatment of D-galactose induced typical the senescent phenotype in CRT cells, we next examined at the intracellular level the expression of proinflammatory cytokines IL-6 and IL-8, the major SASP components in rodent and human cells (20). As expected, D-galactose significantly increased IL-6 expression by astrocytic CRT-cells compared with the control (Fig. 2A). In addition, significantly increased IL-8 expression at the transcriptional level was also exhibited by the stable cell line CRT-MG/IL-8p-d2EGFP cells compared with the control (Fig. 2B). To explore the molecular mechanism, we examined the expression of NF-κB by immunofluorescence, which has been revealed to stimulate the transcription of many SASP genes (10), and the results indicated a significant increase in the nuclear translocation after D-galactose exposure compared with the control (Fig. 2C). Our findings suggested that D-galactose increased the expression of SASP-related proteins through activating the NF-κB pathway as early as 6 days after treatment as determined by elevated secretions of IL-6 and IL-8, which was abrogated by a NF-κB pathway inhibitor, Bay 11–7082 (Fig. 2D), suggesting that it may act as a potential target for inhibition on development of SASP in astrocytes.

Figure 2. D-gal treatment induces the SASP of astrocytic CRT cells. (A) Intracellular levels of IL-6 were identified and assessed by immunofluorescence (magnification, ×100). (B) The levels of IL-8 were confirmed and assessed within the stable cell line CRT-MG/IL-8p-d2EGFP cells (magnification, ×100). (C) Quantification of nuclear translocation of NF-κB in astrocytic cells (magnification, ×200). (D) Secreted levels of IL-6 and IL-8 were determined by ELISA. For the SASP assay, astrocytic CRT cells were treated with 50 g/l D-gale for 10 consecutive days and the medium was changed every 3 days. *P<0.05 and **P<0.01 vs. D-gal group. Con, control; D-gal, D-galactose; IL, interleukin; NF-κB, nuclear factor-κB; SASP, senescence-associated secretory phenotype.

CM from senescent astrocytes promotes brain tumor cell viability and survival from chemotherapy agent treatment

To explore the role of CM in the process of cellular senescence, we used two different brain tumor cell lines: U373-MG and N2a. CM was used to treat U373-MG and N2a cells at various concentrations (25–100% of culture medium; Fig. 3A and D). Treatment of U373-MG and N2a cells (with 10% FBS) promoted cell viability, suggesting a pro-tumoral action. Furthermore, U373 cells treated with different concentrations of CM were exposed to 0.5 mM temozolomide, and strong chemotherapy resistance was observed with sustained cell viability compared with the control treated with temozolomide (Fig. 3B). This resistance was further confirmed by the decreased levels of activated apoptosis-associated molecules cleaved caspase-3 and PARP (Fig. 3C). Intriguingly, N2a cell-derived neurons (with 1% FBS) were observed as tumor-like neuroblasts after treatment with CM. The results herein showed that neuron-specific class III β-tubulin expression levels were markedly decreased, and tumor-like alterations were observed with reduced dendritic extensions. In addition, we also observed the increased expression of c-Myc protein (Fig. 3E), which is important for cancer cell proliferation (21). Therefore, these studies suggest a potential key role of CM derived from senescent astrocytic cells in human brain tumor progression.

Figure 3. SASP of astrocytic CRT cells promotes brain tumor viability and chemotherapy resistance. (A) U373-MG cells were treated with the indicated concentrations of CM from senescent astrocytic CRT cells, after which cell viability was analyzed by a WST-1 assay. (B) U373-MG cells were treated with normal medium or indicated concentrations of CM for 24 h, after which cells were plated with fresh medium supplemented with 0.5 mM TMZ for 48 h. Finally, the cell viability was determined by a WST-1 assay. (C) Extracts from non-treated or TMZ treated, CM-treated U373-MG cells and TMZ-treated, were assay for cleaved-caspase3 and PARP by western blotting. (D) N2a cells were treated with the indicated concentrations of CM from senescent astrocytic CRT cells, then the viability was analyzed by a WST-1 assay. (E) After treatment with CM, protein extracts from N2a cells were analyzed for Tuj1 and c-Myc expression by western blotting for the indicated tumor and neuronal markers; microscopic analysis was also conducted (magnification, ×100). GAPDH served as the loading control. *P<0.05 and **P<0.01 vs. control group. Con, control; CM, conditioned medium; PARP, poly (ADP-ribose) polymerase; TMZ, temozolomide; Tuj1, neuron-specific class III β-tubulin.

Discussion

Our findings suggest that treatment with D-galactose might induce astrocytic cellular senescence and promote brain aging pathologies, including the onset of brain cancer by the mechanism that comprises the SASP. Senescent cells are not limited to irreversible cell-cycle arrest. Of note, senescent cells have undergone various changes in protein expression and secretion, ultimately leading to SASP(11). Coppe et al (22), Orjalo et al (23) and Davalos et al (24) first discovered that senescent cells could trigger tumorigenesis in neighboring malignant cells, and disrupt normal tissue structure and function. Additionally, the accumulation of senescent cells after chemotherapy gives rise to chemoresistance and cancer recurrence (25). Our study proposed that senescent astrocytes induced by D-galactose develop SASP via activating the NF-κB pathway, suggesting a potential mechanism of cellular senescence in the normal brain aging process.

Of note, low-dose administration of D-galactose is a widely recognized model for brain aging in rodents. Increasing evidence has been reported that D-galactose treatment induces memory impairment, synaptic dysfunction, oxidative stress and neurodegeneration (26,27); however, little is known about the underlying mechanism. Elevated levels of D-galactose trigger ROS accumulation in the brain, ultimately inducing oxidative stress and the production of advanced glycation end products (AGEs). These changes have been shown to be involved in physiologically aging and in neurodegenerative diseases, such as AD and Parkinson's disease, indicated by the presence in senile plaques and neurofibrillary tangles (28,29). Additionally, the accumulation of ROS activates the mitochondrial apoptotic pathway via cytochrome c release (30) and markedly decreased glutamine synthetase expression in the brain of mice (31). Collectively, the majority of studies have reported on mitochondrial damage by ROS and the interaction of AGE-receptor for AGE in the brain (32). Our study herein addressed a novel possibility in vitro: D-galactose might induce brain inflammation caused by senescent astrocytes and the SASP. The SASP involves potent inflammatory cytokines such as IL-6 and IL-8, which may alter tissue environments. This is in agreement with previous reports in epithelial cell models (33,34). In summary, our findings reported astrocytic SASP as a potential mechanism by which D-galactose treatment induces brain aging and aging-related pathologies, including the development of brain cancer. Understanding how the SASP is regulated and how it may be prevented is useful in clinical practice. The model described in the present study provided a basis for the screening and development of potential therapeutic strategies to reduce deleterious effects of astrocytic senescence during the normal process of aging.

Acknowledgements

Not applicable.

Funding

The present study was supported by the Intelligent Synthetic Biology Center of the Global Frontier Project, funded by the Ministry of Education, Science and Technology (grant no. 2011-0031967).

Availability of data and materials

All data generated or analyzed during this study are included in this published article.

Authors' contributions

JH wrote the paper. JH and YY performed the experiments. JH, JX, MS and SK designed the experiments and improved the manuscript. All authors read and approved the manuscript and agree to be accountable for all aspects of the research in ensuring that the accuracy and integrity of any part of the work are appropriately investigated and resolved.

Ethics approval and consent to participate

Not applicable.

Patient consent to participate

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References