Oncogene (2006) 25, 3638–3648
& 2006 Nature Publishing Group All rights reserved 0950-9232/06 $30.00 www.nature.com/onc
ORIGINAL ARTICLE
Ionizing radiation induces apoptosis and inhibits neuronal differentiation in rat neural stem cells via the c-Jun NH2-terminal kinase (JNK) pathway T Kanzawa1, E Iwado1, H Aoki1, A Iwamaru1, EF Hollingsworth1, R Sawaya1, S Kondo1,2 and Y Kondo1 1 Department of Neurosurgery, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA and 2The University of Texas Graduate School of Biomedical Science at Houston, Houston, TX, USA
A substantial number of neural stem cells (NSCs) continue to proliferate and generate neurons in the central nervous system throughout life. Ionizing radiation, an important adjuvant therapy for glioma patients, may damage NSCs and cause neuronal deficits, such as cognitive dysfunction and memory impairment. However, the precise mechanism of radiation effects on death and differentiation of NSCs remains largely unknown. Here, we found that radiation induced apoptosis in NSCs via the mitochondrial pathway, upregulating the ratio of Bax to Bcl-2 and releasing cytochrome c into the cytoplasm. Radiation also inhibited neuronal differentiation of NSCs by 50%. Of the three stress-associated mitogen-activated protein kinases (MAPKs), only c-Jun NH2-terminal kinase (JNK) was activated in NSCs after radiation. Interestingly, JNK inhibition by the specific inhibitor SP600125 rescued NSCs from apoptosis and improved neuronal differentiation. Furthermore, we examined whether radiation directly inhibits neuronal differentiation or not. Radiation did not affect the promoter activity of NeuroD, a basic helix–loop–helix transcription factor that regulates the expression of neuronal differentiation markers. Radiation induced more apoptosis in NeuroD-positive cells than NeuroD-negative cells. We concluded that radiation activates JNK and induces apoptosis, especially in neural progenitor cells, resulting in the inhibition of neurogenesis. Our findings raise the possibility that JNK inhibition has therapeutic potential in protecting NSCs from the adverse effects of radiation. Oncogene (2006) 25, 3638–3648. doi:10.1038/sj.onc.1209414; published online 20 February 2006 Keywords: neural stem cells; ionizing radiation; JNK; apoptosis; neuronal differentiation
Correspondence: Dr Y Kondo, Department of Neurosurgery, Unit BSRB1004, The University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030, USA. E-mail:
[email protected] Received 5 August 2005; revised 8 December 2005; accepted 23 December 2005; published online 20 February 2006
Introduction Cranial irradiation is an important adjuvant therapy for patients with primary and metastatic brain tumors. Although it improves the prognosis of the patients, many long-term survivors suffer neurobehavioral sequelae including cognitive and memory impairment, which implicates dysfunction of the hippocampus (Crossen et al., 1994; Surma-aho et al., 2001). Recently, the hippocampus has been recognized as one of the regions where neural stem cells (NSCs) continue to proliferate and generate neurons throughout life. In the animal model of cranial irradiation, neural stem/precursor cells in the hippocampus undergo apoptosis, and neurogenesis is suppressed (Peissner et al., 1999; Madsen et al., 2003; Limoli et al., 2004). Monje et al.(2002) studied the association between the radiation effects on NSCs and changes in the microenvironment. They attributed the depletion of NSCs and reduction in neurogenesis to inflammation caused by radiation. However, direct effects of radiation on NSCs and their molecular mechanisms still remain unclear. Apoptosis is an essential process that takes place during normal development, for maintenance of homeostasis, and under pathological conditions. Apoptosis is characterized by condensation and fragmentation of the chromatin, as well as cytoplasmic shrinkage. Central components of the apoptotic machinery include Bcl-2 and caspase family (Green and Reed, 1998; Jacotot et al., 1999). Bcl-2 family consists of both antiapoptotic and proapoptotic members. Antiapoptotic members include Bcl-2 and Bcl-XL, whereas proapoptotic members are Bax, Bak, Bcl-XS, Bim, Bad, and so on (Gross et al., 1999). These two groups have distinct localization in the cell. For example, Bcl-2 is an integral protein of the mitochondrial membrane, whereas Bax is present in the cytoplasm when inactivated. Following death signal, Bax translocates to mitochondria and transforms to be integrated into the membrane. Then, Bax alters mitochondrial transmembrane potential (Dcm), whereas Bcl-2 neutralizes the effect by dimerizing with Bax. When Dcm dissipates, cytochrome c is released from the mitochondrion into the cytoplasm, which activates caspases, resulting in the induction of apoptosis (Harris and Johnson, 2001).
Role of JNK in radiation effects on neural stem cells T Kanzawa et al
3639
Mitogen-activated protein kinases (MAPKs) phosphorylate specific proteins in response to extracellular stimuli and regulate a variety of cellular functions including migration, proliferation, differentiation, and programmed death (Davis, 2000; Chang and Karin, 2001). MAPKs comprise c-Jun NH2terminal kinase (JNK), extracellular signal-regulated kinase (ERK), and p38 MAPK. Accumulating evidence shows that the MAPK pathways play an important regulatory role in radiation-induced apoptosis (Verheij and Bartelink, 2000; Dent et al., 2003). Activation of JNK and p38, and inhibition of ERK are associated with the induction of apoptosis (Xia et al., 1995). Furthermore, several studies have demonstrated that JNK phosphorylates Bim, phosphorylated Bim activates Bax, resulting in apoptosis via the mitochondrial pathway (Tournier et al., 2000; Harris and Johnson, 2001; Lei and Davis, 2003). In the context of neuronal differentiation, multiple studies showed that ERK mediates the signal from various stimuli for neuronal differentiation of neural stem/precursor cells (Rueda et al., 2002; Zhao et al., 2003; Harada et al., 2004), but the role that JNK and p38 may play remains unknown. In the present study, we first investigated radiation effects on death and neuronal differentiation of rat NSCs. Radiation induced apoptosis via the mitochondrial pathway and inhibited neuronal differentiation of NSCs. Radiation activated JNK, upregulated Bax, and downregulated Bcl-2, resulting in apoptosis. SP600125, a specific JNK inhibitor, not only
suppressed apoptosis of NSCs, but also improved neuronal differentiation, suggesting that JNK mediates apoptosis and inhibition of neuronal differentiation after irradiation. Results Isolation of rat NSCs We isolated NSCs from the brain of E16 rat embryos. After culturing cells for several days in serum-free medium with basic fibroblast growth factor (bFGF) and epidermal growth factor (EGF), neurospheres were formed. The vast majority of the cells of the neurospheres were nestin-positive (Figure 1a). By definition, NSCs have the capability for self-renewal and differentiation into neurons and glia (Gage, 2000). When kept cultured in the serum-free medium with bFGF and EGF, NSCs continued proliferating for several weeks. To differentiate NSCs, we added retinoic acid to the medium and withdrew bFGF and EGF from the medium and cultured NSCs for 7 days. A subpopulation of NSCs exhibited immunoreactivity to b-tubulin type III, glial fibrillary acidic protein (GFAP), and 20 -30 cyclic nucleotide 30 -phosphodiesterase (CNPase) (Figure 1b–d, respectively), which are the commonly used markers for neurons, astrocytes, and oligodendrocytes, respectively (Ciccolini and Svendsen, 1998; Shamblott et al., 2001). These results confirm that we successfully isolated NSCs that differentiate into neurons, astrocytes, and oligodendrocytes.
Figure 1 NSCs isolated from the brain of E16 embryos differentiate into neurons, astrocytes, and oligodendrocytes. NSCs were isolated from the cortex of E16 rat embryos. After culturing the cells for several days in serum-free DMEM/F12 medium containing bFGF, EGF, and N2 supplement, neurospheres were formed. NSCs in neurospheres were stained by immunocytochemistry for nestin (a). NSCs were seeded onto chamber slides coated with 0.01% poly-L-ornithine and 1 mg/ml fibronectin and cultured in the serum-free DMEM/F12 medium containing 0.5 mM retinoic acid and N2 supplement for 7 days, and immunostained for b-tubulin type III (b), GFAP (c), or CNPase (d). Scale bar, 15 mm. Oncogene
Role of JNK in radiation effects on neural stem cells T Kanzawa et al
3640
Radiation effects on cell viability and the cell cycle in NSCs and cultured astrocytes We irradiated NSCs and rat-cultured astrocytes RNB, and compared the effects on cell viability. Inhibition of cell growth was dose-dependent and mostly timedependent in both cell types (Figure 2a). Using 10-Gy irradiation, the growth of NSCs was inhibited intensely, whereas that of RNB cells was only moderately inhibited. The viability of NSCs fell to 18, 16, and 2% of the control on days 2, 3, and 4, respectively, after 10-Gy irradiation. A large reduction in cell viability occurred within the first 48 h.
RNB Control
400
80 60
RNB
40
Sub-G1: 0.05%
0
0 2
3
0
4
Days ∗
140 120 100 80 60 40 20 0
NSCs ∗ 0
2
∗ 3
4
∗ ∗ ∗
Control
NSC 10Gy
20.6%
33.2%
0 0
10 Gy
1000
FL2-A
400
0
Days
0
1000
FL2-A
NSC Control
400
2.5 Gy 5 Gy 10 Gy
Counts
% Cell Viability
0.2%
20 0
c
Counts
100
0
NSCs
RNB 10Gy
400
Counts
% Cell Viability
RNB
b
120
Counts
a
Next, we performed flow cytometry to detect changes in the cell cycle in both cell types (Figure 2b; Table 1). In RNB cells, the sub-G1 population, which is indicative of apoptosis, increased from 0.0570.006 to 0.370.1% (Po0.05), but still remained within a small percentage less than 1%. In contrast, the most striking change detected in NSCs was a large increase in the sub-G1 population from 20.570.6 to 31.373.1% (Po0.01) of the cells. It should be noted that even without irradiation, NSCs tend to undergo apoptosis (Bieberich et al., 2003). To confirm the induction of apoptosis morphologically, we performed Hoechst 33258 nuclear staining.
1000
FL2-A
0
1000
FL2-A
d ∗
90 80 70 Apoptotic cells (%)
RNB
control
60
10 Gy
50 40 30 20 10
NSCs
0 RNB
NSC
Figure 2 10-Gy irradiation induces apoptosis in NSCs, but not in RNB cells. (a) Cells were seeded in 96-well plates, incubated overnight, and irradiated at 2.5, 5.0, or 10.0 Gy. Cell viability was measured by using the WST-1 assay 2 to 4 days after irradiation, and the value was calculated relative to that of untreated controls from triplicate data. *Po0.05, compared with the viability of RNB cells treated with the same dose of radiation. (b) Cell-cycle analysis with flow cytometry. Cells were treated with 10-Gy irradiation and subjected to cell-cycle analysis 48 h later. The data shown are representative of three independent experiments. FL2-A indicates propidium iodide intensity. (c) Microphotographs of the cell nuclei stained with Hoechst 33258. Cells were exposed to 10-Gy irradiation and stained 48 h later. Arrows indicate cells with apoptotic characteristics, such as condensed or fragmented nuclei. Scale bar, 12 mm. (d) Quantification of the cells with apoptotic morphology in the nucleus. A total of 300 cells in each group were assessed. Values are the mean7s.d. of triplicate experiments. *Po0.05. Oncogene
Role of JNK in radiation effects on neural stem cells T Kanzawa et al
3641 Table 1 Cell-cycle analysis before and after 10-Gy irradiation Sub-G1 (%)
G1 (%)
S (%)
G2/M (%)
RNB Control 10 Gy
0.0570.006 0.370.1
77.677.7 77.571.4
10.071.3 5.170.2
12.471.4 17.271.4
NSCs Control 10 Gy
20.570.6 31.373.1
61.271.8 54.175.2
9.970.4 5.672.0
6.570.5 7.371.6
Condensed or fragmented nuclei (Figure 2c, arrows), which are characteristic of apoptosis, increased significantly from 39.474.7 to 65.3711.1% in NSCs after irradiation (Po0.05) (Figure 2d). In contrast, no change in the RNB cells was observed. IR induces apoptosis in NSCs through the mitochondrial pathway As mitochondria play a major role in the induction of apoptosis, we examined the involvement of mitochondrial damage in radiation-induced apoptosis. When mitochondria are damaged by apoptotic stimuli, the mitochondrial inner membrane potential (Dcm) dissipates, cytochrome c is released into the cytoplasm, and caspase pathways are activated (Jacotot et al., 1999; Desagher and Martinou, 2000; Goldstein et al., 2000). To investigate radiation effect on mitochondria, we performed tetramethylrhodamine ethyl ester (TMRE) staining and immunoblotting of cytochrome c in the cytoplasm. As expected, the percentage of NSCs with a low Dcm increased from 24.471.6 to 75.074.8% after irradiation (Po0.01) (Figure 3a). In contrast, the percentage of RNB cells with a low Dcm increased a little (control: 2.270.5%; 10 Gy: 5.472.9%; Po0.05). Also, the level of expression of cytochrome c in the cytoplasm increased dramatically in NSCs, 24 h after irradiation, whereas the expression of cytochrome c was not detected in RNB cells (Figure 3b). These results indicate that 10-Gy irradiation activates the apoptotic pathway through mitochondria in NSCs but not in RNB cells. Members of the Bcl-2 family are closely related to mitochondria and modulate apoptosis. The Bcl-2 family consists of proapoptotic proteins, such as Bax and Bcl-XS, and antiapoptotic proteins, such as Bcl-2 and Bcl-XL (Gross et al., 1999). The balance of proapoptotic and antiapoptotic proteins determines whether a cell dies or survives (Oltvai and Korsmeyer, 1994; Gross et al., 1999). To determine the involvement of Bcl-2 family proteins in radiation-induced apoptosis, we performed immunoblotting for Bax and Bcl-2 in NSCs before and after irradiation (Figure 3c). The expression of Bax increased at 24 and 48 h after 10-Gy irradiation. On the other hand, the expression of Bcl-2 decreased gradually from 2 to 48 h after irradiation. In addition, the ratio of Bax to Bcl-2 increased markedly at 24 and 48 h after 10-Gy irradiation. These results suggest that an increase in Bax/Bcl-2 expression induces apoptosis in NSCs after irradiation.
Involvement of JNK in radiation-induced apoptosis of NSCs Recently, several studies demonstrated that the MAPK stress pathways are involved in radiation-induced apoptosis (Verheij and Bartelink, 2000; Dent et al., 2003). To investigate whether MAPKs are involved in radiation-induced apoptosis in NSCs, we performed immunoblotting for activated types of JNK, p38 and ERK up to 24 h after irradiation (Figure 4a). Phosphorylated JNK showed an increase of 30 and 60% at 1 and 2 h after irradiation, respectively. Phosphorylated p38 was barely detected before and after irradiation. Phosphorylated ERK1/2 was detected, but remained constant after irradiation. The expression of total JNK, total p38, and total ERK1/2 did not change during the same period. To confirm the involvement of JNK activation, we examined c-Jun, a downstream signal of JNK, after irradiation (Verheij and Bartelink, 2000). Phosphorylated c-Jun was upregulated 30 min after irradiation with a further increase at 2 h, whereas the expression of total c-Jun was constant during the same period (Figure 4b). We then used a specific inhibitor of JNK, SP600125, to determine whether inhibition of JNK recovers radiation-induced damage in NSCs. First, we examined the cytotoxicity of SP600125 on NSCs, because JNK also contributes to normal cell functions (Waetzig and Herdegen, 2005). SP600125 (1–5 mM) inhibited the viability of NSCs in a dose-dependent manner (Figure 4c). SP600125 (5 mM) inhibited most of phosphorylated JNK, but not phosphorylated ERK, after 10-Gy irradiation (Figure 4d). At 48 h after irradiation, the viability of NSCs reduced to 17% of that of untreated control cells. The viability was significantly recovered to 35% in NSCs treated with radiation and 5 mM SP600125 (Po0.05) (Figure 4e). These results suggest that radiation effects on cell viability of NSCs are via the JNK pathway, at least in part. Radiation inhibits neurogenesis via the JNK pathway We next investigated whether radiation affects the differentiation pattern of NSCs. We irradiated NSCs with increasing doses of radiation (2, 4, and 6 Gy) and cultured them under differentiation-permissive conditions. After 7 days, we performed immunocytochemical staining for b-tubulin type III, CNPase, and GFAP. As the majority of the NSCs treated with 4 or 6 Gy died and detached during the 7-day incubation period, we focused on the NSCs irradiated at 2 Gy (Figure 5a and b). Without irradiation, about 25% of the NSCs were b-tubulin type III-positive, 25% were CNPase-positive, and about 50% were GFAPpositive. After 2-Gy irradiation, the percentage of b-tublin type III-positive cells decreased significantly by 50% (Po0.05), whereas that of CNPase-positive cells remained unchanged, and that of GFAP-positive cells increased by 15% (Po0.05) (Figure 5b). To test whether the JNK pathway was involved in the inhibition of neuronal differentiation after irradiation, we again used SP600125. After 2-Gy irradiation, the percentage of b-tubulin type III-positive NSCs Oncogene
Role of JNK in radiation effects on neural stem cells T Kanzawa et al
3642
a
RNB control
RNB 10 Gy 6.5%
RNB
Counts
Counts
2.2% M1
100
101
102 103 FL2-H
104
M1
100
101
75.2%
Counts
Counts
24.1% M1
100
104
10 Gy
control
NSCs
102 103 FL2-H
101
102 103 FL2-H
b
100
104
RNB 10
0
M1
24
101
0
102 103 FL2-H NSCs 10
104
24 h
cytochrome c
β-actin
c
0
2
10
24
48 (hr)
Bax
Bcl-2
β-actin Figure 3 Radiation induces apoptosis in NSCs via the mitochondrial pathway. (a) Mitochondrial membrane potential measured using TMRE staining. Cells were treated with 10-Gy irradiation and subjected to TMRE staining 48 h later. FL2-H indicates red fluorescence intensity for TMRE staining. The data shown are representative of three independent experiments. (b) Expression of cytochrome c in the cytoplasm. Cells were treated with 10-Gy irradiation and cytosolic protein extracts were subjected to immunoblotting. b-actin was used for loading control. The data shown are representative of two independent experiments. (c) Expression of apoptosis-associated proteins after irradiation. NSCs were treated with 10-Gy irradiation, and total protein was isolated at different time points for immunoblotting. b-actin was used for loading control. The data shown are representative of two independent experiments.
increased by about two-fold in NSCs treated with 5 mM SP600125 (Po0.05) (Figure 5c). We also confirmed the change in the protein level using immunoblotting. The expression of b-tubulin type III reduced after 2-Gy irradiation, but recovered with the addition of SP600125 (Figure 5d). These results suggest that radiation inhibits neuronal differentiation via the JNK pathway. Oncogene
Radiation does not affect neuronal differentiation directly So far, our results demonstrated that radiation induces apoptosis and inhibits neuronal differentiation in NSCs via JNK activation. Though we assessed radiationinduced apoptosis 2 days after irradiation, we had to analyse differentiation pattern 7 days after the treatment, because the expression of b-tubulin type III was not induced until culturing the cells for several days
Role of JNK in radiation effects on neural stem cells T Kanzawa et al
3643
a
b 0
0.5
1
2
5
0
0.5
1
2 (hr)
P-c-Jun
24 (hr)
c-Jun
P-JNK
c 120
JNK
100 Cell Viability (%)
P-p38 p38
P-ERK 1/2 ERK 1/2
80 60 40 20 0
d
10 Gy SP600125
+ −
+ +
e
1
2
3 5 SP600125 (uM)
10
120
JNK P-ERK 1/2
Cell viability (%)
100 P-JNK
80 ∗
60 40 20
10 Gy + SP600125
10 Gy
SP600125
0
0 Gy
ERK 1/2
Figure 4 Radiation induces apoptosis in NSCs via the JNK pathway. (a, b) The expression of MAPK and MAPK-associated proteins. NSCs were treated with 10-Gy irradiation and subjected to immunoblotting at different time points. The data shown are representative of three independent experiments. (c) NSCs were treated with indicated concentrations of SP600125 for 48 h and the cell viability was determined using the WST-1 assay. Values are the mean7s.d. of triplicate experiments. (d) NSCs were treated with or without 5 mM SP600125 for 30 min, treated with 10-Gy irradiation, incubated for 2 h, and subjected to immunoblotting. (e) NSCs were treated with 10-Gy irradiation and/or 5 mM SP600125. Forty-eight hours later, cell viability was determined by using the WST-1 assay. Values are the mean7s.d. of triplicate experiments. *Po0.05.
under differentiation-permissive medium. This gap in time course raises two possibilities: (1) radiation induces apoptosis and inhibits neuronal differentiation directly and independently; (2) radiation induces apoptosis especially in neural progenitor cells, resulting in the inhibition of neuronal differentiation. To address this issue, we directed our attention to NeuroD, a basic helix–loop–helix (bHLH) transcription factor that is expressed earlier than b-tubulin type III. bHLH proteins including NeuroD upregulate neuronal differentiationrelated genes such as b-tubulin type III and NeuN (Diez del Corral and Storey, 2001). First, we examined whether radiation inhibits NeuroD at the transcriptional level using a reporter vector. The promoter activity did
not change significantly after irradiation (Figure 6a). Next, we performed double staining for NeuroD and TUNEL to assess radiation-induced apoptosis in NeuroD-positive cells. Few TUNEL-positive cells were detected in untreated NeuroD-positive cells, whereas many TUNEL-positive cells were also NeuroD-positive after irradiation (arrows, Figure 6b). The percentage of TUNEL-positive cells in NeuroD-positive cells increased significantly from 15.473.1 to 41.773.1% (Po0.05) (Figure 6c). On the other hand, the percentage of TUNEL-positive cells in NeuroD-negative cells did not change significantly (from 20.773.0 to 14.071.7%, P>0.1) (Figure 6d). Thus, NeuroD-positive cells were more vulnerable to radiation than NeuroD-negative Oncogene
Role of JNK in radiation effects on neural stem cells T Kanzawa et al
3644
Figure 5 Radiation inhibits neuronal differentiation via the JNK pathway. (a) NSCs were treated with 2-Gy irradiation and cultured under differentiation-permissive conditions. After 7 days, the cells were stained immunocytochemically for differentiation markers. Scale bar, 15 mm. (b) Quantification of the cells that were positive for differentiation markers. Values are the mean7s.d. of triplicate experiments. *Po0.05. (c, d) NSCs were treated with 2-Gy irradiation and/or 5 mM SP600125, cultured under differentiationpermissive conditions for 7 days and subjected to immunocytochemistry (c) or immunoblotting (d) for b-tubulin type III. (c) The percentage of positive-stained cells was calculated. Values are the mean7s.d. of triplicate experiments. *Po0.05. (d) b-actin was used for loading control.
cells. Additionally, we performed double staining for phosphorylated JNK and terminal deoxynucleotidyltransferase-mediated dUTP nick end-labeling (TUNEL) to confirm that activation of the JNK pathway by radiation induced apoptosis in NSCs. Both phosphorylated JNK-positive cells and TUNEL-positive cells increased prominently after 10-Gy irradiation, and many cells were double positive (Figure 6e). The percentage of phosphorylated JNK-positive cells in TUNEL-positive cells increased significantly from 33.474.0 to 83.273.4% (Po0.01) after radiation. These results collectively indicate that radiation induces apoptosis in NSCs mainly via the JNK pathway, particularly in neural progenitor cells, resulting in the inhibition of neurogenesis.
Discussion We have demonstrated that NSCs were more sensitive to radiation than cultured astrocytes. Radiation induced a massive apoptosis in NSCs through the mitochondrial pathway. Radiation upregulated the ratio of Bax/Bcl-2, released cytochrome c into the cytoplasm, and induced Oncogene
apoptosis. On the other hand, radiation inhibited neural differentiation of NSCs. As an upstream signal pathway, JNK activation was detected, and SP600125, a specific JNK inhibitor, inhibited these radiation effects, at least in part. To the best of our knowledge, this is the first study to show that JNK mediates radiation-induced apoptosis and inhibition of neuronal differentiation in NSCs. An increasing number of publications show that the stress-induced MAPKs play an important role in radiation-induced apoptosis (Verheij et al., 1998; Verheij and Bartelink, 2000; Dent et al., 2003). Of the three stress-induced MAPKs, JNK and p38 induce apoptosis, whereas ERK leads to cell survival (Xia et al., 1995; Verheij et al., 1998). Both JNK and p38 have been reported to mediate apoptosis in neural lineage cells. Nitric oxide induced apoptosis through p38 activation in neural progenitor cells (Cheng et al., 2001), whereas nutrient deprivation induced apoptosis through JNK activation in neurons. (Harris and Johnson, 2001). Radiation-induced apoptosis in the central nervous system has been studied extensively in the animal model. Some recent studies revealed that radiation induced apoptosis in neural stem/progenitor cells (Peissner et al.,
Role of JNK in radiation effects on neural stem cells T Kanzawa et al
3645
a
b
NeuroD
TUNEL
Merge
Relative luciferase activity
1.8 1.6 1.4
control
1.2 1 0.8 0.6 0.4
10 Gy
0.2 0
c
10 Gy
control ∗
50
50
d
45 TUNEL-positive cells/ NeuroD-negative cells (%)
TUNEL-positive cells/ NeuroD-positive cells (%)
45 40 35 30 25 20 15 10 5
40 35 30 25 20 15 10 5
0 control
e
0
10 Gy Phospho-JNK
TUNEL
control
10 Gy
Merge
control
10 Gy
Figure 6 Radiation induces apoptosis in a vast majority of NeuroD-positive cells. (a) NSCs were co-transfected with luciferase reporter plasmids for the NeuroD promoter and pGL3-Control. Twenty-four hours later the promoter activity was measured and standardized using the control vector. Values are the mean7s.d. of triplicate experiments. (b–d) NSCs were treated with or without 10-Gy irradiation and double-stained for NeuroD and TUNEL 48 h later. (b) Arrows indicate double-stained cells. Scale bar, 10 mm. The percentage of TUNEL-positive cells in NeuroD-positive cells (c) and that of TUNEL-positive cells in NeuroD-negative cells (d) were calculated and plotted. Values are the mean7s.d. of triplicate experiments. *Po0.05. (e) NSCs were treated with or without 10Gy irradiation and double-stained for phospho-JNK and TUNEL 48 h later. Arrows indicate double-stained cells. Scale bar, 10 mm.
1999; Monje et al., 2002; Limoli et al., 2004). However, few studies reported the involvement of MAPKs in the process. Two studies from the same group revealed that radiation activated JNK in the hippocampus and induced apoptosis, with the main focus on neuronal cell death (Lonergan et al., 2002; Lynch et al., 2003). In line with these studies, we have clearly demonstrated that JNK mediated radiation-induced apoptosis in NSCs. JNK is activated by cytokines such as tumor necrosis factor (TNF) and interleukin-1 (IL-1), and environmental stress such as osmotic stress, redox stress, and
radiation (Davis, 2000). The signal pathways of JNK have been well established. JNK phosphorylates the NH2-terminus of the transcription factor c-Jun that activates the transcription factor AP-1. JNK also activates other transcription factors, such as ATF2 and Elk-1. JNK modulates cellular function in cell proliferation, differentiation, and death by activating these transcription factors. A study demonstrated that c-Jun activation is necessary in stress-induced apoptosis (Behrens et al., 1999). JNK also interacts with nontranscription factors, such as members of the Bcl-2 family, regulating apoptosis. Activated JNK can Oncogene
Role of JNK in radiation effects on neural stem cells T Kanzawa et al
3646
phosphorylate Bcl-2, BAD, and BH3-only proteins, such as Bim, Bmf and Dp5 (Yamamoto et al., 1999; Harris and Johnson, 2001; Lei and Davis, 2003; Yu et al., 2004). When Bcl-2 and these BH3-only proteins are phosphorylated, they cause Bax translocation to mitochondria, which releases cytochrome c from mitochondria, resulting in the induction of apoptosis. Another recent study showed that JNK2 translocates to mitochondria, inducing apoptosis (Eminel et al., 2004). In accordance with these studies, we showed that radiation induced JNK and c-Jun activation, an increase in the Bax/Bcl-2 ratio, cytochrome c release, and apoptosis. However, JNK does not always induce apoptosis. As shown in IL-3-mediated cell survival, JNK can inhibit apoptosis; when JNK phosphorylated BAD, it inhibited apoptosis (Yu et al., 2004). In the neural development, multiple bHLH transcription factors regulate neuronal differentiation in a cascade manner (Ross et al., 2003; Cho and Tsai, 2004). Proneural bHLH factors, such as Neurogenins and Mash, play a role in initiating neurogenesis. On the other hand, neuronal differentiation bHLH factors, such as NeuroD, mediate terminal differentiation. Neurogenins are expressed in NSCs that generate both neurons and glia, whereas NeuroD is specific to postmitotic neuronal cells. These bHLH factors upregulate neuronal differentiation-related genes including b-tubulin type III and NeuN (Diez del Corral and Storey, 2001). For the last series of experiments in the present study, we used NeuroD expression to assess the relationship between radiation-induced apoptosis and the inhibition of neurogenesis. We found that radiation did not affect NeuroD at the transcriptional level and that a vast majority of apoptotic cells were neural progenitor cells. Therefore, we concluded that radiation induced apoptosis through JNK activation, especially in neural progenitor cells, resulting in the inhibition of neurogenesis (Figure 7a). Our findings suggest that JNK inhibition could decrease apoptosis and result in the recovery of neurogenesis after irradiation. However, sustained and systemic JNK inhibition may cause various adverse side effects, because JNK also have other important functions in the development, differentiation, and insulin resistance (Figure 7b) (Waetzig and Herdegen, 2005). For example, JNK1/ mice
a Radiation effects on NSCs Radiation
Apoptosis
JNK
Neurogenesis
SP600125
b Therapeutic intervention JNK inhibition
Apoptosis
Neurogenesis
Other physiological Functions ? Figure 7 Potential scenario of radiation-induced neuronal damage and therapeutic intervention. Oncogene
showed severe impairment in the architecture of the brain, indicating that JNK phosphorylates microtubuleassociated proteins and contributes to normal brain functions (Chang et al., 2003; Bjorkblom et al., 2005). Therefore, Waetzig and Herdegen (2005) proposed the context-specific inhibition of JNK, that is, specific inhibition of damaging actions of JNK. They introduced that some JNK inhibitors including inhibitory peptides are in development as candidate compounds for this purpose. In summary, we investigated the radiation effects on death and differentiation of rat NSCs. Radiation induced JNK activation, upregulation of the BAX/ Bcl-2 ratio, cytochrome c release, and apoptosis in NSCs. Radiation also inhibited neuronal differentiation via the JNK pathway as a result of apoptosis induction in neural progenitor cells. These results implicate a therapeutic application of JNK inhibition for protection of NSCs from radiation-induced adverse effects.
Materials and methods Cell culture Primary astrocyte cultures (RNB cells) were prepared from the cerebral cortex of 1-day-old neonatal rats as described previously (Galea et al., 1992; Kondo et al., 1996). RNB cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum, 4 mM glutamine, 100 U/ml penicillin, and 100 mg/ml streptomycin (Invitrogen, Carlsbad, CA, USA). To assess whether RNB cells were derived from astroglial cells, immunohistochemical staining using anti-GFAP antibody was performed as described previously (Kondo et al., 1992). Cell culture for NSCs was performed as described previously (Johe et al., 1996; Erlandsson et al., 2001). Briefly, embryonic neural tissue samples were dissected from the frontal cortex of Fisher 344 rats on gestation day 16 (E16). The meninges were removed with a sharp forceps under a microscope, and cells were isolated by mechanical dissociation. Cells were then seeded onto a 12-well plate and allowed to proliferate in the serum-free DMEM/F12 medium containing 20 ng/ml bFGF, 20 ng/ml EGF, and N2 supplement (Invitrogen). Cell cultures were maintained at 371C with 5% CO2, and the medium was changed every other day. For differentiation assays, NSCs were seeded onto chamber slides (Nalge Nunc International, Rochester, NY, USA) coated with 0.01% poly-L-ornithine and 1 mg/ml fibronectin (Sigma, St Louis, MO, USA) and cultured in the serum-free DMEM/F12 medium containing 0.5 mM retinoic acid (Sigma) and N2 supplement for 7 days. Cell viability assay and radiation treatment Cell viability was determined with the Cell Proliferation Reagent WST-1 colorimetric assay (Roche Applied Science, Indianapolis, IN, USA) according to the manufacturer’s instruction. Cells were seeded at a concentration of 20 000 cells/100 ml in 96-well plates and pre-incubated overnight. The cells were exposed to radiation at a dose of 2.5, 5.0, or 10.0 Gy at a rate of 3.4 Gy/min with the use of a 137Cs irradiator (US Nuclear Corp., Burbank, CA, USA). After incubation for 1–3 days, WST-1 (10 ml/well) was added to the wells; the absorbance of the dye at 450 nm was measured 2 h later. SP600125 was purchased from Calbiochem (La Jolla, CA, USA). Cells were pretreated with 5 mM SP600125 for 30 min
Role of JNK in radiation effects on neural stem cells T Kanzawa et al
3647 and irradiated. As SP600125 was lysed in DMSO, we used the same concentration of DMSO (0.01%) in culture medium for other groups when we used SP600125 in experiments and compared its effects on cell viability and differentiation. Cell-cycle analysis Cells fixed with ice-cold 70% ethanol were stained with propidium iodide with the Cellular DNA Flow Cytometric Analysis Reagent Set (Boehringer Mannheim, Indianapolis, IN, USA) and analysed for DNA content by using a FACScan (Becton Dickinson, San Jose, CA, USA). The percentage of cells in the sub-G1, G1, S, and G2/M populations was analysed with the CellQuest software program (Becton Dickinson). Hoechst 33258 staining for detection of apoptosis To detect apoptosis morphologically, NSCs were seeded onto chamber slides coated with 0.01% poly-L-ornithine and 1 mg/ml fibronectin, and RNB cells were seeded onto uncoated chamber slides; the cells were then cultured overnight. Afterward, the cells were irradiated and incubated for 48 h. Then, the cells were fixed with 1% paraformaldehyde, stained with 0.5 mg/ml Hoechst 33258 (Sigma), and observed under a fluorescence microscope. Flow cytometric analysis of mitochondrial membrane potential The mitochondrial membrane potential was measured with the cationic, lipophilic dye TMRE (Molecular Probes, Inc., Eugene, OR, USA) by flow cytometry as described previously (Kanzawa et al., 2004). After irradiation, cells were collected and stained with 150 nM TMRE in the dark at 371C for 30 min. The cells were then washed, resuspended in phosphatebuffered saline containing 15 nM TMRE, and analysed with the use of flow cytometry. Western blotting Treated cells were homogenized and incubated in lysis buffer (10 mM Tris-HCl, pH 7.5, 500 mM NaCl, 1 mM EDTA, 20 mg/ml aprotinin and leupeptin, and 1% Triton X-100) for 20 min on ice and centrifuged (15 000 g, 20 min); the supernatant containing soluble total protein was analysed. The protein concentration was determined by using the BioRad Protein Assay. To detect cytochrome c release, cytosolic protein extracts were prepared as described previously (Goldstein et al., 2000). Briefly, cells were washed in phosphatebuffered saline and incubated for 30 min on ice in 300 ml of lysis buffer (68 mM sucrose, 200 mM mannitol, 50 mM KCl, 1 mM EGTA, 1 mM EDTA, 1 mM DTT, and 1 Complete Protease Inhibitor (Boehringer Mannheim)). Cells were then lysed with 80 strokes of a Dounce homogenizer and centrifuged at 41C (800 g), and the supernatant was centrifuged at 14 000 g for 10 min. The supernatant (cytosol) and pellet (mitochondria) were stored at 701C for Westernblot analysis. Equal amounts of protein were separated on a Tris-HCl sodium dodecyl sulfate gel and transferred to a polyvinylidine difluoride membrane. The membrane was subjected to immunoblot analysis with the enhanced chemiluminescence detection system (Amersham, Piscataway, NJ, USA). The level of protein expression was quantified using densitometer and normalized with the intensity of the band of associated loading control. The antibodies used were against Bax, Bcl-2, cytochrome c, phosphorylated c-Jun (Santa Cruz Biotechnology, Santa Cruz, CA, USA), phosphorylated JNK, phosphorylated p38 MAPK, phosphorylated ERK (Cell Signaling Technology, Beverly, MA, USA), b-actin (Sigma), and c-Jun (BD Transduction Lab, San Jose, CA, USA).
Immunostaining Cells were fixed in 1% paraformaldehyde, blocked with 10% normal goat serum, and stained with antibodies against b-tubulin type III (Promega, Madison, WI, USA), CNPase, and GFAP (Chemicon International, Temecula, CA, USA). For DAB staining, samples were visualized with an avidinbiotinylated horseradish peroxidase procedure (Vector Laboratories, Burlingame, CA, USA) followed by a standard diaminobenzidine reaction. To detect nuclei with fragmented DNA, that are characteristic of apoptosis, TUNEL staining with FITC was performed according to the manufacturer’s instruction (Cheminon International) prior to immunostaining. After TUNEL staining, the cells were blocked with 10% normal goat serum and stained with antibodies against NeuroD1 (Cheminon International). Samples were visualized with Alexa Fluor 594-conjugated goat anti-rabbit IgG (Molecular Probes). Luciferase assay The transcriptional activity of BETA2 (NeuroD) in NSCs was determined by the BETA2 promoter-luciferase reporter plasmid kindly provided by Dr Ming-Jer Tsai (Baylor College of Medicine, Houston, TX, USA) (Huang et al., 2000). Cells were plated at a density of approximately 5 104 cells/ml, transfected with the luciferase reporter plasmids (1 mg each) using Fugen 6 transfection Reagent (Roche Applied Science) according to the manufacturer’s instruction, cultured overnight and irradiated. Twenty-four hours later, cells were washed twice with PBS and lysed in the lysis buffer provided with the luciferase kit (Promega). Transcriptional activity was measured using a Microtiter Plate Luminometer (Dynatech Laboratories, Inc., Chantilly, VA, USA). The luciferase reporter vector pGL3-Control (Promega) was co-transfected as control plasmids to standardize transfection efficiency. All experiments were performed at least three times with each plasmid and the results were average values of relative luciferase activity. Statistical analysis The data were expressed as mean7standard deviation (s.d.). Statistical analysis was performed using Student’s t-test (two-tailed). The criterion for statistical significance was Po0.05.
Abbreviations NSC, neural stem cell; bFGF, basic fibroblast growth factor; EGF, epidermal growth factor; GFAP, glial fibrillary acidic protein; CNPase, 20 -30 -cyclic nucleotide 30 -phosphodiesterase; TMRE, tetramethylrhodamine ethyl ester; MAPK, mitogenactivated protein kinase; JNK, c-Jun NH2 terminal kinase; ERK, extracellular signal-regulated kinase. Acknowledgements We thank Dr Ming-Jer Tsai for providing the BETA2 promoter-luciferase plasmid and Mr Donald Norwood for his help in editing the manuscript. Financial support by USPHS Grants CA088936 and CA108558 (S Kondo), a startup fund from The University of Texas M. D. Anderson Cancer Center (S Kondo), and a generous donation from the Anthony D Bullock III Foundation (S Kondo and Y Kondo) is acknowledged. Oncogene
Role of JNK in radiation effects on neural stem cells T Kanzawa et al
3648 References Behrens A, Sibilia M, Wagner EF. (1999). Nat Genet 21: 326–329. Bieberich E, MacKinnon S, Silva J, Noggle S, Condie BG. (2003). J Cell Biol 162: 469–479. Bjorkblom B, Ostman N, Hongisto V, Komarovski V, Filen JJ, Nyman TA et al. (2005). J Neurosci 25: 6350–6361. Chang L, Jones Y, Ellisman MH, Goldstein LS, Karin M. (2003). Dev Cell 4: 521–533. Chang L, Karin M. (2001). Nature 410: 37–40. Cheng A, Chan SL, Milhavet O, Wang S, Mattson MP. (2001). J Biol Chem 276: 43320–43327. Cho JH, Tsai MJ. (2004). Mol Neurobiol 30: 35–47. Ciccolini F, Svendsen CN. (1998). J Neurosci 18: 7869–7880. Crossen JR, Garwood D, Glatstein E, Neuwelt EA. (1994). J Clin Oncol 12: 627–642. Davis RJ. (2000). Cell 103: 239–252. Dent P, Yacoub A, Fisher PB, Hagan MP, Grant S. (2003). Oncogene 22: 5885–5896. Desagher S, Martinou JC. (2000). Trend Cell Biol 10: 369–377. Diez del Corral R, Storey KG. (2001). Nat Rev Neurosci 2: 835–839. Eminel S, Klettner A, Roemer L, Herdegen T, Waetzig V. (2004). J Biol Chem 279: 55385–55392. Erlandsson A, Enarsson M, Forsberg-Nilsson K. (2001). J Neurosci 21: 3483–3491. Gage FH. (2000). Science 287: 1433–1438. Galea E, Feinstein DL, Reis DJ. (1992). Proc Natl Acad Sci USA 89: 10945–10949. Goldstein JC, Waterhouse NJ, Juin P, Evan GI, Green DR. (2000). Nat Cell Biol 2: 156–162. Green DR, Reed JC. (1998). Science 281: 1309–1312. Gross A, McDonnell JM, Korsmeyer SJ. (1999). Genes Dev 13: 1899–1911. Harada J, Foley M, Moskowitz MA, Waeber C. (2004). J Neurochem 88: 1026–1039. Harris CA, Johnson Jr EM. (2001). J Biol Chem 276: 37754–37760. Huang HP, Liu M, El-Hodiri HM, Chu K, Jamrich M, Tsai MJ. (2000). Mol Cell Biol 20: 3292–3307. Jacotot E, Costantini P, Laboureau E, Zamzami N, Susin SA, Kroemer G. (1999). Ann NY Acad Sci 887: 18–30. Johe KK, Hazel TG, Muller T, Dugich-Djordjevic MM, McKay RD. (1996). Genes Dev 10: 3129–3140. Kanzawa T, Germano IM, Komata T, Ito H, Kondo Y, Kondo S. (2004). Cell Death Differ 11: 448–457.
Oncogene
Kondo S, Miyatake S, Iwasaki K, Oda Y, Kikuchi H, Zu Y et al. (1992). Neurosurgery 30: 506–511. Kondo S, Morimura T, Barnett GH, Kondo Y, Peterson JW, Kaakaji R et al. (1996). Oncogene 13: 1773–1779. Lei K, Davis RJ. (2003). Proc Natl Acad Sci USA 100: 2432–2437. Limoli CL, Giedzinski E, Rola R, Otsuka S, Palmer TD, Fike JR. (2004). Radiat Res 161: 17–27. Lonergan PE, Martin DS, Horrobin DF, Lynch MA. (2002). J Biol Chem 277: 20804–20811. Lynch AM, Moore M, Craig S, Lonergan PE, Martin DS, Lynch MA. (2003). J Biol Chem 278: 51075–51084. Madsen TM, Kristjansen PE, Bolwig TG, Wortwein G. (2003). Neuroscience 119: 635–642. Monje ML, Mizumatsu S, Fike JR, Palmer TD. (2002). Nat Med 8: 955–962. Oltvai ZN, Korsmeyer SJ. (1994). Cell 79: 189–192. Peissner W, Kocher M, Treuer H, Gillardon F. (1999). Brain Res Mol Brain Res 71: 61–68. Ross SE, Greenberg ME, Stiles CD. (2003). Neuron 39: 13–25. Rueda D, Navarro B, Martinez-Serrano A, Guzman M, Galve-Roperh I. (2002). J Biol Chem 277: 46645–46650. Shamblott MJ, Axelman J, Littlefield JW, Blumenthal PD, Huggins GR, Cui Y et al. (2001). Proc Natl Acad Sci USA 98: 113–118. Surma-aho O, Niemela M, Vilkki J, Kouri M, Brander A, Salonen O et al. (2001). Neurology 56: 1285–1290. Tournier C, Hess P, Yang DD, Xu J, Turner TK, Nimnual A et al. (2000). Science 288: 870–874. Verheij M, Bartelink H. (2000). Cell Tissue Res 301: 133–142. Verheij M, Ruiter GA, Zerp SF, van Blitterswijk WJ, Fuks Z, Haimovitz-Friedman A et al. (1998). Radiother Oncol 47: 225–232. Waetzig V, Herdegen T. (2005). Trends Pharmacol Sci 26: 455–461. Xia Z, Dickens M, Raingeaud J, Davis RJ, Greenberg ME. (1995). Science 270: 1326–1331. Yamamoto K, Ichijo H, Korsmeyer SJ. (1999). Mol Cell Biol 19: 8469–8478. Yu C, Minemoto Y, Zhang J, Liu J, Tang F, Bui TN et al. (2004). Mol Cell 13: 329–340. Zhao WQ, Alkon DL, Ma W. (2003). J Neurosci Res 72: 334–342.
