Telomerase Reverse Transcriptase Upregulation Attenuates Astrocyte Proliferation and Promotes Neuronal Survival in the Hypoxic–Ischemic Rat Brain
Background and Purpose—Telomerase reverse transcriptase (TERT) is tightly related to the resistance of cells to stress and injury. However, little is known about the roles of TERT in the nervous system. We try to investigate the effects of TERT on the function of astrocytes in developing rat brains subjected to hypoxia–ischemia.
Methods—TERT expression was detected in rat brains with hypoxia–ischemia. In in vitro study, the function of astrocytes with TERT overexpression was measured, and the effects of astrocyte on neuronal apoptosis were examined. Moreover, overexpression or inhibition of TERT was conducted in vivo by gene transduction. Astrocyte proliferation was examined through Ki67 staining. Terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling staining and brain infarct volume calculation were used to detect neuronal injury.
Results—Both TERT mRNA and protein were upregulated in neurons within 2 days but shifted to astrocytes at Day 3 after hypoxia–ischemia. Astrocyte proliferation was inhibited with TERT overexpression due to the upregulation of cell-cycle regulatory protein p15. Meanwhile, the apoptosis of neurons increased, whereas neurons were cocultured with conditioned media from astrocytes with TERT inhibition compared with TERT overexpression due to the decrease of neurotrophin-3 expression in astrocytes. Furthermore, Ki67-positive astrocytes and neuronal injury were found to be inhibited in TERT-overexpressing rat brains with hypoxia–ischemia.
Conclusions—TERT attenuates astrocyte proliferation and promotes neuronal survival in the developing rat brain after hypoxia–ischemia, partly through its enhancement of p15 and neurotrophin-3 expression in astrocytes.
The original function of telomerase reverse transcriptase (TERT) is to construct telomerase activity.1 Telomerase is a reverse transcriptase that adds TTAGGG repeats to telomeres and thus maintains telomeres length and prevents cellular senescence. Telomerase consists of an RNA template, a protein component called TERT, and other associated proteins (eg, telomerase repeating-binding factor-1, TRF-1, and TRF-2).2 TERT possesses reverse transcriptase activity and is the most important regulator and rate-limiting determinant of telomerase activity.3 During the development of the central nervous system, TERT expression and telomerase activity are high in neural progenitor cells but then decrease as differentiation ensues.4 Five days after birth, TERT expression and telomerase activity are barely detectable in rat brains under physiological conditions.5 However, TERT expression is found to be induced in adult rat neurons in response to injuries such as amyloid β-peptide-induced damage and ischemia induced neurotoxicity, although the signaling pathways and mechanisms of TERT induction are not known.6,7
Astrocytes play an important role in regulating the brain microenvironment through excess proliferation,8 extracellular neurotransmitter regulation,9 and neural growth factor secretion.10 Recently, we have demonstrated that TERT was induced in a neonatal rat hypoxic–ischemic (HI) model and plays a protective role in this setting.11 We also found that TERT is induced in cultured astrocytes when subjected to ischemia.12 However, how TERT is activated and the exact mechanisms that TERT plays in neuroprotection after HI are unclear. Therefore, we hypothesized that TERT plays a protective role in regulating neural repair through modulating the function of astrocytes. To validate this hypothesis, we established both in vivo and in vitro ischemia models using postnatal Day 10 rats and cultured astrocytes and investigated the function of TERT in neonatal HI injury.
Materials and Methods
Detection of TERT Expression in the HI Rat
Sprague-Dawley rats were obtained from Medical Animal Center of Sichuan Province (Chengdu, China). All animal research was approved by the Sichuan University Committee on Animal Research. Postnatal Day 10 rats without gender selection were subjected to right common carotid occlusion plus 8% hypoxia for 2.5 hours for ischemia and hypoxia treatment13 (Supplement I; http://stroke.ahajournals.org). The expression of TERT was detected through reverse transcriptase–polymerase chain reaction, Western blot, and immunohistochemistry (Supplement I).
Cell Culture and Gene Transduction
Astrocytes and neurons were prepared from primary cell cultures of cortical tissues from postnatal Day 1 and embryonic Day 16 pups, respectively (Supplement II).
Plasmids, pcDNA-rTERT bearing the full-length cDNA of rat telomerase reverse transcriptase (rTERT), pcDNA-AS bearing the antisense sequence against rTERT, and pcDNA-SE bearing the sense sequence against rTERT were synthesized and constructed by Jinsite Biotechnology (Supplement III).
Plasmid pcDNA-rTERT (pT) was transduced into astrocytes at passage 3 using the Lipofectamine 2000 reagent (Invitrogen) and the control astrocytes were made by transducing mock plasmid (Mo). To inhibit TERT expression, pcDNA-AS (pAs) was transduced into astrocytes, and the control astrocytes were made by transducing with pcDNA-SE (pSe). Gene transduction was performed using the method as described previously.14 The positive clone cells were selected by 400 μg/mL G418 and maintained in Dulbecco's modified Eagle's medium supplemented with 100 μg/mL G418 and 10% fetal bovine serum. Astrocytes at passage 3 after G418 selection were used to perform subsequent experiments.
Detection of Astrocytes Proliferation In Vitro
Astrocytes were plated into a 96-well plate and cultured in serum-free Dulbecco's modified Eagle's medium for 24 hours, then serum (10% fetal bovine serum) and 600 ng/mL 5-bromodeoxyuridine (Sigma) were added into the media for incubation for 24 hours and cells were collected for 5-bromodeoxyuridine assay (Supplement IV). In parallel, cells with only serum induction were collected for cell-cycle distribution analysis using flow cytometry (Supplement V) and for cell-cycle regulator detection through Western blot (Supplement VI).
Determination of the Effect of p15 and p21 on Astrocyte Proliferation
Phosphorothioate antisense and sense oligodeoxynucleotides against cell-cycle regulatory proteins p15 or p21 were synthesized by Jinsite Biotechnology. These oligonucleotides were transduced in pT group astrocytes mediated by Lipofectin TM (Invitrogen; Supplement VII). Cell-cycle distribution analysis was conducted using flow cytometry as described previously. Inhibition of p21 or p15 expression was confirmed by Western blot.
Evaluation of the Functional State of Astrocytes
The expression of glutamate aspartate transporter GLAST, glutamate transporter-1, glutamine synthetase, and calcium regulatory protein S-100B were assessed using Western blot. Neural growth factors produced by astrocytes such as transforming growth factor β1, glial cell line-derived neurotrophic factor, brain-derived neurotrophic factor, and neurotrophin-3 (NT-3) were detected using Western blot and enzyme-linked immunosorbent assays (Supplement VIII).
In Vitro Combined Hypoxia and Glucose Deprivation Model of Astrocytes
All cell culture dishes were cultured in glucose-free Dulbecco's modified Eagle's medium placed in a humidified incubation chamber at 37°C and flushed with a gas mixture of 93% N2/5% CO2/2% O2 for combined hypoxia and glucose deprivation (CHGD). For 5-bromodeoxyuridine analysis, astrocytes were incubated with 600 ng/mL 5-bromodeoxyuridine (Sigma) for 6 hours before CHGD, then at 6, 12, and 24 hours of treatment, the cells were fixed for 5-bromodeoxyuridine analysis (Supplement IV). NT-3 expression in CHGD astrocytes was also detected through Western blot and enzyme-linked immunosorbent assays as described previously.
Determination of the Effect of Astrocyte-Conditioned Media on Neuronal Apoptosis After Hypoxia
Astrocytes were cultured in glucose-free Dulbecco's modified Eagle's medium placed in a humidified incubation chamber at 37°C and flushed with a gas mixture of 93% N2/5% CO2/2% O2 for CHGD treatment. The conditioned media (10 mL) from the astrocytes (at a density of 6.4×106 cells/dish) were collected at 0, 12, and 24 hours of the CHGD and centrifuged at 500 g to remove cellular debris. The supernatants were used for the subsequent tests using the method previously described.15
Astrocyte-conditioned media were transferred in equal volume into 3 wells (1 mL/well) with each well containing 4×105 neurons and incubated for 6 hours in a humidified incubation chamber at 37°C and flushed with a gas mixture of 93% N2/5% CO2/2% O2 for hypoxia treatment.
At 6 hours of hypoxia, neuronal apoptosis was detected using the In Situ Cell Death Detection Kit (Roche) for terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling detection and apoptotic index calculation (Supplement IV).
Overexpression or Inhibition of TERT In Vivo
Plasmids of pcDNA-GFAP-EGFP (Mock), pcDNA-GFAP-TERT-EGFP (T), pcDNA-GFAP-SE-EGFP (SE), and pcDNA-GFAP-AS-EGFP (AS) were constructed by Jinsite Biotechnology (Supplement X) and solved in lipofectamine (7 μg of plasmid in 5 μL of lipofectamine). For TERT overexpression analysis, T and Mock were injected into the right lateral ventricle at 12 hours before HI. For TERT inhibition analysis, AS and SE were injected into the right lateral ventricle at 12 hours before HI. As a vehicle control, lipofectamine mixed with Hanks buffer was injected into the right lateral ventricle (Supplement XX). At 1, 3, 5, and 7 days after HI, rats were euthanized and the brains were removed for detection of the plasmid distribution (Supplement XI) and subsequent tests.
Evaluation of Astrocyte Proliferation and Neuropathological Injury
To evaluate astrocyte proliferation induced by HI injury, rat brains were immunolabeled by glial fibrillary acidic protein and Ki67. The numbers of Ki67(+) astrocytes were counted (Supplement XII).
Brain injury was evaluated by calculating infarct volume using cresyl violet staining (Supplement XIII). Apoptosis of cerebral cortex neuron was detected by terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling and neuronal nuclei double-labeling staining (Supplement XIV).
Data are presented as mean±SEM from at least 3 independent experiments. Student t test was used when comparison was made between 2 groups. Analysis of variance with Fisher post hoc test was used when comparing >2 groups. The probability value <0.05 was defined as the threshold for significance.
TERT Expression Upregulated in the Developing Rat Brain With HI
TERT expression was detected by reverse transcriptase–polymerase chain reaction, Western blot, and immunohistochemistry in the developing rat brain subjected to HI treatment. We found that both TERT mRNA and protein were upregulated in rat brain with the mRNA level peaking at 24 hours and protein level peaking at 72 hours after HI (Figure 1A–B). TERT/NeuN or TERT/glial fibrillary acidic protein double-labeling assays showed that TERT was expressed mainly in neurons within 2 days but shifted mainly to astrocytes from 3 to 4 days after HI. In sham controls, TERT was not detectable (Figure 1C).
TERT Inhibits Astrocyte Proliferation and Reduces Astrocytes in the S Phase
To determine the role of TERT in regulating astrocyte proliferation, plasmid containing TERT, Mock, sense, or antisense fragment were transduced into astrocytes. The positive clones were, respectively, termed as pT, Mo, pAs, and pSe. Western blot analyses demonstrated TERT overexpression in pT-transduced astrocytes but not in other groups (Figure 2A). To test cellular proliferation, 5-bromodeoxyuridine assays and flow cytometry detection were performed using pT and control transduced astrocytes. We found that TERT inhibits astrocyte proliferation and reduces the number of astrocytes in the S phase (Figure 2B).
TERT Upregulates p15 and p21 Expression in Astrocytes
Because TERT was found to affect cell-cycle distribution, we studied whether cell-cycle regulators such as p15, p21, p27, and p53 were involved in this distribution. Western blot showed that p15 and p21 proteins were upregulated in pT but not in control transduced astrocytes (Figure 2C). To further determine the effects of p15 and p21 on astrocyte proliferation, antisense oligonucleotides were used to inhibit p15 and p21 expression. We found that p15 but not p21 inhibition enhanced the number of astrocytes in the S phase (Figure 2D).
TERT Upregulates NT-3 Expression in Astrocytes
To investigate the functional state of astrocytes with TERT overexpression, we analyzed the expression of glutamate aspartate transporter (GLAST), glutamate transporter-1, glutamine synthetase, and calcium regulatory protein S-100B using Western blots. Moreover, we detected neural growth factors produced by astrocytes such as transforming growth factor β1, glial cell line-derived neurotrophic factor, brain-derived neurotrophic factor, and NT-3 using Western blot and enzyme-linked immunosorbent assays. We found that NT-3, but not GLAST, glutamate transporter-1, glutamine synthetase, S-100B, transforming growth factor β1, brain-derived neurotrophic factor, or glial cell line-derived neurotrophic factor, was significantly increased in pT transduced astrocytes (Figure 3A–B). Our findings suggest that the important physiological functions of astrocytes such as the capacity to clear glutamate, synthesize glutamine, and buffer calcium were not changed by TERT overexpression. However, NT-3 upregulation is an important functional change by TERT overexpression in astrocytes.
Proliferation of Astrocytes Subjected to CHGD In Vitro
Because TERT overexpression inhibits astrocyte proliferation, we further determined the proliferation of astrocytes with or without TERT inhibition after CHGD. Astrocytes were subjected to glucose-free media with 93% N2/5% CO2/2% O2 treatment for 6 hours, 12 hours, or 24 hours to mimic HI in vitro, and 5-bromodeoxyuridine analysis was used to detect cell proliferation. We found that cell proliferation in the pT group was much less than that in Mock, pAs, and pSe groups at 6 hours of CHGD treatment. There were no obvious differences of cell proliferation among Mock, pAs, and pSe groups. Accordingly, Western blot showed that the upregulation of TERT, p15, and p21 was found in the pT group but not in other 3 groups at 6 hours of CHGD. Nevertheless, at 12 hours of this treatment, there was no obvious difference of cell proliferation, TERT, p15, or p21 expression in the pT group compared with Mock and pSe groups. However, more proliferating cells and a lower amount of TERT, p15, and p21 were observed in the pAs group compared with the other 3 groups (Figure 4A–B).
TERT Prolongs NT-3 Expression in CHGD Astrocytes
Because TERT overexpression enhanced NT-3 expression in astrocytes, we further determined the roles of TERT in regulating NT-3 expression in astrocytes with CHGD. We found that TERT and NT-3 were induced by CHGD treatment in vitro. TERT expression appeared at 6 hours and gradually increased until 24 hours after CHGD. NT-3 expression appeared earlier at 2 hours and also gradually increased until 24 hours after CHGD. However, when TERT was inhibited using antisense oligonucleotides (pAs), the increased NT-3 expression was inhibited. Enzyme-linked immunosorbent assays showed the NT-3 level in pAs group was only approximately 40% of pSe group at 24 hours of HI, which is similar to the findings detected by Western blots (Figure 4C).
Astrocyte-Conditioned Media Protects Neurons From Hypoxia Damage In Vitro
Because TERT can regulate the expression of neural growth factors of astrocytes such as NT-3, the supernatant from astrocytes with different TERT expression levels might have different neuronal protection potential. To detect the role of astrocytic TERT in neuronal apoptosis after hypoxia, terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling staining was performed to test neuronal apoptosis with TERT sense (pSe) or antisense (pAs) treated astrocyte-conditioned media. We found that neurons cultured with astrocyte-conditioned media without CHGD treatment had significant apoptosis after hypoxia for 6 hours. However, when neurons were cultured in astrocyte-conditioned media with 12 hours or 24 hours of CHGD treatment, apoptosis decreased. Moreover, there were more apoptotic neurons cultured in media from the pAs group compared with the pSe group (Figure 5A). After measuring the apoptotic index, we found that the apoptosis of neurons cultured in media from the pAs group was much higher than that in pSe group (P<0.05; Figure 5B). To further determine the involvement of NT-3 in neuronal protection, we detected the NT-3 amount in astrocytes supernatant using enzyme-linked immunosorbent assays. We found that the NT-3 amount was lower in pAs than that in the pSe group with CHGD treatment for 12 hours and 24 hours (P<0.05; Figure 5C).
Effect of TERT on Astrocyte Proliferation and Brain Injury Induced by HI In Vivo
Plasmids producing TERT protein (T) or TERT antisense oligonucleotides (AS) were used to overexpress or inhibit TERT expression in vivo. Western blot analyses revealed that TERT was overexpressed in the T group and inhibited in the As group, whereas neither Mock (M) nor sense oligonucleotides (SE) affected TERT expression (Figure 6B). Accordingly, the expression of p15 and NT-3 was enhanced or attenuated with TERT overexpression or inhibition (Figure 6A). Double immunolabeling showed that plasmids were expressed mainly in astrocytes but not in neurons after transduction (Figure 6B).
To further study the TERT effect on astrocyte proliferation induced by HI, we studied Ki67 (cellular proliferation marker) expression in astrocytes after HI treatment using immunohistochemistry. In sham controls, Ki67(+) astrocytes were rarely detected. However, Ki67(+) astrocytes were increased at Days 3, 5, and 7 after HI. Moreover, compared with Mock and SE groups, the Ki67(+) astrocytes were significantly decreased in the T group and increased in the AS group at the previously mentioned time points (Figure 6C).
To determine the effect of TERT on brain injury, we measured infarct volume of brains at Days 3, 5, and 7 after HI using cresyl violet staining. We found a significant decrease of brain infarct volume in the T group and an increase of brain infarct volume in the AS group compared with that in Mock or SE groups (Figure 6D). To analyze the effects of TERT on neuronal apoptosis after HI, terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling staining with neuronal marker, neuronal nuclei expression was detected. Terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling-positive neurons were increased at Days 3, 5 and 7 after HI compared with that in sham rats. Moreover, terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling-positive neurons were significantly decreased in the T group and increased in the AS group compared with those in Mock and SE groups (Figure 6D).
In this study, we first demonstrated that TERT upregulation attenuates astrocyte proliferation and promotes neuronal survival in developing rat brains after HI. This newly found function of TERT may be related to the enhancement of p15 and NT-3 expression in astrocytes.
In in vivo experiments, we found that TERT expressed mainly in neurons within 2 days but then shifted to astrocytes the third day after HI. The early regulation of TERT in neurons may suggest an instant reaction of TERT on neurons with HI. This is in agreement with previous reports that TERT upregulation in neurons ameliorates neuronal death induced by HI.7 The later upregulation of TERT in astrocytes suggests a chronic role of TERT, which is unknown at present. Therefore, in this study, we focused on investigating the functions of TERT in astrocytes and the possible mechanisms behind it.
TERT is induced in the developing rat brain and cultured astrocytes after HI. However, the telomerase RNA component and telomerase activity are undetectable. Because telomerase RNA serves as a template for telomere synthesis and is essential for telomerase activity,16 the absence of telomerase RNA may contribute to the lack of telomerase activity in the developing rat brain. Our findings are in line with a previous report that telomerase activity and telomerase RNA are undetectable in rat adult brains.7 Therefore, TERT might exert its functions through a telomerase activity-independent way in neonatal rat brains.
As we know, the original function of TERT is to construct telomerase activity and thus to maintain telomere length during cell division.1 However, recent advances have clearly demonstrated that TERT exerts many other functions beyond maintaining telomere length. These functions include the regulation of Ca2+ distribution,17 metabolism,18 growth factor secretion,19 mitochondria function,20 energy balance,21 and apoptosis.22 However, whether these functions of TERT exist in the central nervous system, especially in developing brains with HI, us not clear. In the present study, we found that TERT did not change the expression of GLAST, glutamate transporter-1, glutamine synthetase, and S-100B, suggesting that the important physiological function of astrocytes such as the capacity to clear glutamate, synthesize glutamine, and buffer calcium are not changed by TERT. However, TERT was found to enhance NT-3 level and prolong the time windrow of NT-3 expression after HI in developing rat brains. Because NT-3 exerts a neuroprotective function in cerebral ischemia,23 the prolongation of NT-3 expression might contribute to the neuroprotective effect of TERT against HI damage in developing rat brains.
Interestingly, we found that TERT upregulation in astrocytes attenuated cell proliferation rather than promotes cell propagation that has been found in other cell types such as vascular endothelium and hepatocyte cells.24,25 To investigate the possible mechanisms behind this, we detected cell-cycle regulators such as p15, p21, p27, and p53 in astrocytes with TERT overexpression. We found that TERT upregulated p15 and p21 proteins. Furthermore, attenuating p15 but not p21 inhibited the effect of TERT on astrocyte, suggesting that TERT inhibited astrocyte proliferation through upregulating p15 expression but not p21. The protein of p15 is an important cyclin-dependent kinase inhibitor, which can bind to either Cdk4 or Cdk6 and inhibit the action of cyclin D. During the G1 phase in cell cycle, retinoblastoma protein is phosphorylated by cyclin D/cdk4 and cyclin D/cdk6 complexes. Phosphorylation of Rb inactivates the complexes and allows the release of E2F, thereby allowing entry of cells into the S phase.26 Therefore, we speculated that p15 upregulation might block the cell-cycle progression of astrocytes through inhibiting the cyclin D-E2F pathway. Unlike p15, p21 upregulation in TERT overexpression astrocytes did not affect cell proliferation and cell-cycle distribution, suggesting that cell-cycle inhibitors might be cell-specific or tissue-specific-dependent.
In in vivo study, we found that TERT attenuated astrocyte proliferation in the developing rat brain subjected to HI. As we know, excessive astrocyte proliferation might be detrimental and contribute to neuronal damage.27 Reactive astrocytes can form a local biochemical and physical barrier, which will attenuate neuronal survival, axonal regeneration, and circuitry re-establishment after brain damage.28 Therefore, early inhibition of proliferating reactive astrocytes would inhibit the accumulation of molecules involved in neuronal damage and thus achieve an environment more suitable for neural repair.29 In this study, we found that with the attenuation of astrocyte proliferation induced by HI, the neuronal apoptosis was also inhibited, suggesting that TERT might exert its neuroprotective function through deleting the detrimental effect of reactive astrocytes. However, whether TERT would enhance axonal regeneration and circuitry re-establishment through inhibiting glial scar formation after HI injury needs further investigation.
In conclusion, TERT upregulation attenuates astrocyte proliferation and neuronal apoptosis partly by enhancing p15 and NT-3 expression in astrocytes. The later activation of TERT in astrocytes at 3 days after HI suggests an extended therapeutic window for HI-induced neonatal brain damage.
Sources of Funding
This work was supported by the National Natural Science Foundation of China (No. 30825039, 30973236, and 31171020 to D.M.; and No. 81172174 to Y.Q.), the Program of Changjiang Scholars and Innovative Research Team in University (IRT0935), and Outstanding Young Scientist' Foundation of Sichuan Province, China (08ZQ026-069 to Y.Q.).
We thank Dr Donna Ferriero and Yvonne Wu from the University of California for careful review of the article.
The online-only Data Supplement is available at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.111.626325/-/DC1.
- Received May 18, 2011.
- Revision received June 24, 2011.
- Accepted June 29, 2011.
- © 2011 American Heart Association, Inc.
- Bodnar AG,
- Ouellette M,
- Frolkis M,
- Holt SE,
- Chiu CP,
- Morin GB,
- et al
- Kang HJ,
- Choi YS,
- Hong SB,
- Kim KW,
- Woo RS,
- Won SJ,
- et al
- Barnabe-Heider F,
- Miller FD
- Blasco MA,
- Funk W,
- Villeponteau B,
- Greider CW
- Passos JF,
- Saretzki G,
- von Zglinicki T
- Bagheri S,
- Nosrati M,
- Li S,
- Fong S,
- Torabian S,
- Rangel J,
- et al
- Di Giovanni S,
- Movsesyan V,
- Ahmed F,
- Cernak I,
- Schinelli S,
- Stoica B,
- et al
- Zhu Z,
- Zhang Q,
- Yu Z,
- Zhang L,
- Tian D,
- Zhu S,
- et al