- Email: [email protected]
Neurogenesis Recovery Induced by Granulocyte-colony Stimulating Factor in Neonatal Rat Brain After Perinatal Hypoxia
Pediatrics and Neonatology (2013) 54, 380e388
Available online at www.sciencedirect.com
journal homepage: http://www.pediatr-neonatol.com
ORIGINAL ARTICLE
Neurogenesis Recovery Induced by Granulocyte-colony Stimulating Factor in Neonatal Rat Brain After Perinatal Hypoxia Yung-Ning Yang a,b, Chien-Seng Lin c, Chun-Hwa Yang b,d, Yu-Hung Lai b,e, Pei-Ling Wu b,d, San-Nan Yang a,b,d,f,* a
Department of Pediatrics, E-Da Hospital, I-Shou University, Kaohsiung, Taiwan Graduate Institute of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan c Department of Emergency and Critical Care Medicine, Cheng Hsin Rehabilitation Medical Center, Taipei, Taiwan d Department of Pediatrics, Kaohsiung Medical University Hospital, Kaohsiung, Taiwan e Department of Ophthalmology, Kaohsiung Medical University Hospital, Kaohsiung Medical University, Kaohsiung, Taiwan f Department of Marine Biotechnology and Resources, National Sun Yat-sen University, Kaohsiung, Taiwan b
Received Nov 11, 2012; received in revised form Mar 14, 2013; accepted May 14, 2013
Key Words brain; cognitive function; granulocyte-colony stimulating factor; hypoxia; neurogenesis
Background: Perinatal hypoxia can lead to a wide range of neurological deficits depending on the differential vulnerability of the involved brain regions to oxygen deprivation. It remains unclear whether the differential vulnerability to oxygen deprivation leads to altered neurogenesis in the neonatal brain after perinatal hypoxia. The primary objective was to investigate whether perinatal hypoxia induces deleterious changes in neurogenesis within three representative brain regions (dentate gyrus of the hippocampus, midbrain, and temporal cortex), with regards to common pathological areas clinically. The secondary objective was to investigate whether granulocyte-colony stimulating factor (G-CSF) therapy exerts beneficial effects in neurogenesis in neonatal rat brains subjected to experimental perinatal hypoxia. Materials and methods: Rat pups were subjected to experimental perinatal hypoxia on the tenth day of life (P10). They were then given G-CSF (30 mg/kg, single injection/day, intraperitoneal injection, P11e16). The neurogenesis efficacy was analyzed on P17 and the radial-arm maze task, a memory task for higher cognitive functions such as problem-solving abilities, was evaluated on P37e58.
* Corresponding author. Department of Pediatrics, Kaohsiung Medical University Hospital, Graduate Institute of Medicine, Kaohsiung Medical University, No. 100, Tzyou 1st Road, Sanmin District, Kaohsiung City 807, Taiwan. E-mail address: [email protected] (S.-N. Yang). 1875-9572/$36 Copyright ª 2013, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. All rights reserved. http://dx.doi.org/10.1016/j.pedneo.2013.04.011
G-CSF and neurogenesis in the neonatal brain
381
Results: Perinatal hypoxia caused a significant decrease in neurogenesis within the three representative brain regions, and this deleterious outcome was alleviated by G-CSF (p < 0.05). In addition, the G-CSF therapy markedly improved the decreased performance of long-term cognitive functions induced by perinatal hypoxia (p < 0.05). Conclusion: This study suggests that G-CSF may be a potentially beneficial therapy, at least in part, through universal recovery of neurogenesis effects in the neonatal brain after perinatal hypoxia insult. Copyright ª 2013, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. All rights reserved.
1. Introduction In immature brains, cerebral hypoxia contributes to devastating and lifelong neurological impairment different from the behavioral impairment in mental retardation and cerebral palsy.1 Previous studies have reported that hypoxic injury in young rats can cause extensive neuronal damage.2e5 Perinatal hypoxia causes brain damage in multiple brain regions including the dentate gyrus, midbrain, and temporal lobe.6e10 The dentate gyrus of the hippocampus (an important area for mammalian memory, stress, and depression) is often discussed due to its sensitivity to oxygen. The midbrain (a functional area for motivation and habituation) is associated with seizure after hypoxia insult.11 The temporal lobe with interrelated structure referred to as the hippocampus is an area involved in auditory perception and implicated in learning and memory.12 Although exposure to perinatal hypoxia in these brain regions has been extensively studied, the mechanisms and effects of clinical treatment are still unclear. Granulocyte-colony stimulating factor (G-CSF) is a glycoprotein of 19.6-kDa that has been used to treat patients with neutropenia.13 It has been shown to stimulate the bone marrow to produce stem cells and induce their mobilization from the bone marrow to the brain, then inducing neurogenesis in adult animals with stroke to improve functional outcomes.14e21 Schneider et al18 proposed that G-CSF leads to recovery of hippocampal neurogenesis post ischemia. Many studies have focused on G-CSF therapy in the brains of adults who have had strokes; however, few studies have focused on the beneficial effects in neonatal brains post perinatal hypoxia. Our previous animal model study reported that G-CSF had clinical potential to treat perinatal hypoxia by recovery of neurogenesis within the hippocampal CA1 region.22 Hence, in this study, we investigated the efficacy of G-CSF in the dentate gyrus of the hippocampus, midbrain, and temporal cortex of neonatal brains, common pathological areas after perinatal hypoxia, using an animal model. We also used an eight-arm radial maze task to examine the performance in long-term cognitive functions after perinatal hypoxia with regards to the cued reference and spatial working memory.
2. Materials and Methods 2.1. Experimental animal protocols The Animal Care and Use Committee at Kaohsiung Medical University (Kaohsiung, Taiwan) and National Science
Council (Taipei, Taiwan) approved all experimental procedures. Sprague-Dawley rats were provided with a 12-hour light/dark cycle and housed in the animal care facility. The animals were divided into four experimental groups as follows: vehicle-control, perinatal hypoxia plus vehicle, G-CSF alone, and perinatal hypoxia plus G-CSF on postnatal jDay 10 (P10). The procedures for perinatal hypoxia have been described previously.11,23,24 The brains of 10e12-dayold rats are generally accepted to be comparable to term human infants.25 We removed rats (P10) from the litter and placed them in an airtight chamber (30 cm 30 cm 30 cm) on a heating pad at a temperature of 34 C. The concentration of oxygen was then reduced to 5e7% and maintained at this level. In addition, nitrogen gas was regulated by infusion into the chamber. The rats were used for this study only when at least one toniceclonic seizure was observed with low blood oxygen saturation (<60%) during hypoxia. Recombinant human G-CSF (Kyowa Hakko Kirin Co. Ltd., Tokyo, Japan; 30 mg/kg, single injection/day, intraperitoneal injection, P11e16) or vehiclesaline (P11e16) was injected subcutaneously into the rat pups 1 day after perinatal hypoxia had occurred. The dosage of G-CSF used (30 mg/kg) was the most effective dose to alleviate hypoxia-induced damage during the perinatal period according to our previous study.22 Brain tissue specimens of the target brain regions were collected on P17, and the memory functions and spatial learning were analyzed at P37eP58.
2.2. Brain tissue slice preparation On P17, slices (400 mm) of the dentate gyrus, midbrain, and temporal lobe were collected and immediately placed into artificial cerebrospinal fluid in a humidified atmosphere of 95% oxygen/5% carbon dioxide at 34.0 0.5 C in an incubation chamber for an equilibrium period of at least 1 hour.23 To obtain the target brain regions, the brain slices were incubated with ice-cold oxygenated artificial cerebrospinal fluid and then cut into a tissue blocks (0.1 cm 0.5 cm) within 15 seconds of slicing the tissue. The brain tissues were then immediately frozen at 80 C until analysis, and all tissue slices were only ever thawed once.
2.3. Evaluation of neurogenesis Bromodeoxyuridine (BrdU), a cell type-specific marker, was used to identify neurogenesis, and double immunofluorescence analysis of laser-scanning confocal microscopy was performed as previously described.18,20,26 The brains were
382 cut coronally at 30 mm with washing for 20 minutes in phosphate-buffered saline containing 0.2% TritonX-100, pH 7.3 at room temperature with gentle rocking, followed by incubation for 1 hour at room temperature in blocking solution with gentle rocking on P17. BrdU injections (P14e16, 50 mg/kg q. 12 hours, subcutaneously) were used to detect the cell proliferation induced by G-CSF. Neuronal nuclei (NeuN)-staining was performed to detect neuron cells and the region between 2.8 mm and 4.0 mm behind the bregma was of interest for neuron counting.27 In each group, brain regions were taken from a total of 10 sections for each of the 10 animals. The expressions of NeuN-positive and BrdUpositive double immunofluorescence were tested following the instructions of the Roche BrdU labeling and detection kit. Each coronal section was first treated with primary BrdU antibody conjugated with FITC (1:250), followed by treatment with neuron-specific antibodies (NeuN, 1:500; Chemicon, Temecula, CA, USA). The tissue sections were analyzed with an Olympus laser-scanning confocal microscope, with green (FITC) and red (Cy3) fluorochromes on the slides excited by a laser beam at 488 nm and 543 nm, respectively.
Y.-N. Yang et al
2.4. Eight-arm radial maze task As in our previous study,28 an eight-arm radial maze was used to train rats on two memory tasks. We placed bait in half of the symmetrical arms, and a black textured insert was put in each of the remaining four arms. All of the rats performed one training session/day for 15 days. They were placed in the center of the maze, facing a different arm in a quasirandom order in each session. The rat was removed from the maze when it had eaten the bait from all four arms or after 15 minutes, whichever came first. After the bait in an arm had been eaten, it was not replenished. The arms containing the textured insert served as reference points in the memory task, and the arms without the insert served as spatial cues in the memory task because the rats could not distinguish the arms that they had visited. In this test, two memory deficits could be recorded: (1) working memory error (WME), a rat went to an arm with bait that it had already visited; and (2) reference memory error (RME), a rat visited an arm with no bait. If the forepaws of the rat crossed the midline of the arm, it was defined as having visited that arm. The rats were defined as having
Figure 1 Effects of granulocyte-colony stimulating factor (G-CSF) therapy in the declining expression of neurogenesis in the dentate gyrus caused by perinatal hypoxia. Colocalization of neuronal nuclei (NeuN, red for neuron identification)- and bromodeoxyuridine (BrdU, green for cell nucleus detection)-positive cells, as assessed on P17, was identified by doubleimmunofluorescence staining under laser-scanning confocal microscopy. The NeuN-positive cells (1), BrdU-positive cells (2), and the combined pictures of the NeuN- and BrdU-positive cells are as follows: (A1e3) vehicle-control; (B1e3) G-CSF alone (30 mg/kg); C1e3 perinatal hypoxia plus vehicle; and (D1e3) perinatal hypoxia plus G-CSF (30 mg/kg). The arrows reveal the representative newly generated neurons. Bar Z 50 mm.
G-CSF and neurogenesis in the neonatal brain
383
Figure 1
successfully completed the cued reference memory and spatial working memory tasks when there were no RMEs or WMEs for at least 2 days within a 3-day period.
2.5. Statistical analysis All data are presented as mean standard error of the mean. Statistical differences were determined by use of one way analysis of variance (ANOVA) followed by a Bonferroni’s t test for posthoc multiple comparisons. A statistical significance level of p < 0.05 was applied to all tests.
3. Results 3.1. Evaluation of neurogenesis We first evaluated whether the G-CSF therapy increased the expression of neurogenesis within the dentate gyrus of the neonatal brains after perinatal hypoxia. A declining trend in the coexpression of NeuN-positive with BrdUpositive cells was revealed in the perinatal hypoxia plus vehicle group (Figure 1C) compared with the vehiclecontrol rats (Figure 1A). By contrast, an increase in the number of BrdU-positive cells colocalizing with NeuNpositive cells was observed in the perinatal hypoxia plus G-CSF group (Figure 1D). In the midbrain, a decrease in the coexpression of BrdUpositive with NeuN-positive cells was observed in the
(continued)
perinatal hypoxia group plus vehicle (Figure 2C) compared with the vehicle-control rats (Figure 2A). However, the perinatal hypoxia plus G-CSF therapy group showed an increase in the number of BrdU-positive cells colocalizing with NeuN-positive cells (Figure 2D). Furthermore, as shown in Figure 3, the perinatal hypoxia plus vehicle group (Figure 3C) had a significantly decreased coexpression of BrdU-positive with NeuN-positive cells compared with the vehicle-control (Figure 3A) and perinatal hypoxia plus G-CSF therapy groups (Figure 3D) in the temporal cortex. The three brain areas in the G-CSF alone group (Figures 1B, and 3B, 2B) were not significantly different, as compared with the vehicle control group. According to our previous study, there was no significant difference of long-term learning and memory between G-CSF alone and vehicle control groups.22 Therefore, in the subsequent experiments we did not perform radial maze for G-CSF alone group. In quantitative analysis (Figure 4), the average number of cells exhibiting double-staining with NeuN-positive and BrdU-positive within the counted areas showed that G-CSF therapy had significantly improved the perinatal hypoxiainduced decrease in neurogenesis, confirming the double staining of immunofluorescence findings.
3.2. Eight-arm radial maze task After repeated daily training with the eight-arm maze task, the long-term cognitive deficits in the working and cued
384
Y.-N. Yang et al
Figure 2 Granulocyte-colony stimulating factor (G-CSF) therapy restored neurogenesis in the midbrain post perinatal hypoxia. The neuronal nuclei-positive cells (1), the bromodeoxyuridine-positive cells (2), and the combined pictures of the neuronal nuclei- and bromodeoxyuridine-positive cells (3) are as follows: (A1e3) vehicle-control; (B1e3) G-CSF alone (30 mg/kg); (C1e3) perinatal hypoxia plus vehicle; and (D1e3) perinatal hypoxia plus G-CSF (30 mg/kg). The arrows reveal the representative newly generated neurons. Bar Z 50 mm.
G-CSF and neurogenesis in the neonatal brain
385
Figure 3 Granulocyte-colony stimulating factor (G-CSF) therapy restored neurogenesis in the temporal cortex post perinatal hypoxia. The neuronal nuclei-positive cells (1), the bromodeoxyuridine-positive cells (2), and the combined pictures of the neuronal nucleiand bromodeoxyuridine-positive cells (3) are as follows: (A1) vehicle-control; (B1) G-CSF alone (30 mg/kg); (C1) perinatal hypoxia plus vehicle; and (D1) perinatal hypoxia plus G-CSF (30 mg/kg). The arrows reveal the representative newly generated neurons. Bar Z 50 mm.
386
Y.-N. Yang et al
Figure 4 Quantified analysis of the effects of granulocyte-colony stimulating factor (G-CSF) in the decreased expressions of neurogenesis in the dentate gyrus, midbrain, and temporal cortex postperinatal hypoxia. The experimental groups are indicated as follows: vehicle-control (C), G-CSF alone (G), perinatal hypoxia plus vehicle (H), perinatal hypoxia plus G-CSF (HþG). (n Z 10 animals for each experimental group.) Colocalization of neuronal nuclei (NeuN)-positive and bromodeoxyuridine (BrdU)-positive cells are indicated new neuronal cells. NeuN-positive and BrdU-negative cells are indicated as old neuronal cells. NeuN-negative and BrdU-positive cells are non-neuronal cells. * p < 0.05 when compared with the vehicle-control group. Data are presented as mean standard error of the mean.
reference memory tasks in the rats post perinatal hypoxia are shown in Figure 5. The perinatal hypoxia plus vehicle group (n Z 11 animals, p < 0.05) had more mean RMEs and WMEs compared with the vehicle-control group in the first few days of life. The experimental perinatal hypoxia plus G-CSF (n Z 10 animals, p < 0.05) group had an alleviated mean RME and a trend of lower mean WME compared with the perinatal hypoxia plus vehicle rats in the first few days of life. Furthermore, the perinatal hypoxia plus vehicle group had a significant increase in the number of total mean RMEs and total mean WMEs compared with the vehicle-control group. The perinatal hypoxia plus G-CSF group showed improved total mean RMEs and total mean WMEs compared with the perinatal hypoxia plus vehicle group. These results suggest that perinatal hypoxia led to a decline in the ability of spatial memory formation, and that the insult caused by the perinatal hypoxia could be rescued after G-CSF therapy.
4. Discussion In this study, we showed that systemic injections of G-CSF had beneficial effects after perinatal hypoxia on neonatal rat brains in several brain regions. G-CSF therapy significantly abated the decline in neurogenesis in the dentate gyrus, midbrain, and temporal lobe of the neonatal brains after experimental perinatal hypoxia. In addition, working and cued reference memory impairments of the perinatal hypoxia rats could be also rescued by this therapy. Using the eight-arm radial maze task, rats in the perinatal hypoxia group were shown to have significantly reduced spatial memory; however, the perinatal hypoxia with G-CSF therapy group showed significant improvements in the damage caused by perinatal hypoxia. In addition, the histological results showed an increase in neurogenesis in the dentate gyrus in the perinatal hypoxia with G-CSF therapy group compared with the perinatal hypoxia group. In previous studies, the hippocampus (including the dentate gyrus) and temporal cortex have been proven to be related to spatial working memory,12,29e31 and G-CSF has been
proven to play an important role in the neural cells with regards to proliferation, differentiation, and integration in the dentate gyrus for the hippocampus.32 However, the mechanism by which G-CSF acts in rat pups with experimental perinatal hypoxia is unclear. The current study provides evidence that G-CSF administration following injury due to perinatal hypoxia seems to improve working and cued reference memory by enhancing neurogenesis in the dentate gyrus of the hippocampus. Interestingly, the difference in RMEs and WMEs between the three groups seemed to decrease after repeat daily training. This spontaneous improvement of spatial memory in the perinatal hypoxia group may be due to repeat training, because an enriched environment is thought to accelerate neuronal development in the dentate gyrus of rats.33 Our radial maze task findings may suggest that early rehabilitation is useful in alleviating cognitive impairment caused by perinatal hypoxia. The benefits and possible mechanisms after G-CSF therapy have been reported to be through several effects.17 In the central nervous system, G-CSF therapy acts in neurons through G-CSF receptors.15,18 G-CSF may affect neuroprotection through neurogenesis, antiapoptotic, antiinflammatory, neurotrophic, and angiogenesis effects in the mammalian adult brain.15e21,34,35 Previous studies have shown that G-CSF has neurogenesis effect in nonischemic or healthy brain of adult rodents.18 However, in this study there was no statistical difference in neurogenesis between the G-CSF alone and vehicle-control group. Thus, it is noteworthy that this discrepancy may be due to different animal species and ages, as well as different hypoxia protocol. cAMP response element-binding protein phosphorylated at serine 133 (pCREBSer-133) has been shown to lead to synaptogenesis, neuronal survival with neurogenesis, and synaptic plasticity in the brain.36e38 In addition, CREB is a DNA-binding transcription factor and plays an important role in the regulation of some immediate-early genes.39 Previously, we reported that G-CSF therapy alleviates the decreased in pCREBSer-133 and postsynaptic density protein
G-CSF and neurogenesis in the neonatal brain
Figure 5 Impairment of the long-term performance of working and cued reference memory was found using an eightarm radial maze task after experimental perinatal hypoxia. (A) One trial/day over 15 trials, average amount of cued reference memory errors, and the inset box reveals the average of total mean reference memory errors. (B) One trial/day over 15 trials, average amount of working memory errors, and the inset box reveals the average total mean working memory errors. Vehicle-control ( n Z 10 animals), perinatal hypoxia plus vehicle (O; n Z 11 animals), perinatal hypoxia plus granulocyte-colony stimulating factor (;; n Z 10 animals), are labeled as C, H, and H þ G, respectively. Statistical analysis of the long-term performance of the working and cued reference memory revealed that the rats with perinatal hypoxia plus vehicle significantly differed from those in the vehiclecontrol and perinatal hypoxia plus granulocyte-colony stimulating factor groups. *p < 0.05, when compared with the vehicle-control group.
95 with N-methyl-D-aspartate receptor subunits within the CA1 region of the hippocampus after perinatal hypoxia, and provides beneficial effects in long-term cognitive deficits.22 However, whether the neuroprotective effects of G-CSF are also displayed in multiple brain regions after experimental perinatal hypoxia was unknown. Hence, we investigated this in the current study and showed that the efficacy of GCSF therapy in neurogenesis after perinatal hypoxia was not only in the CA1 region of the hippocampus but also in the dentate gyrus, midbrain, and temporal region with longterm cognitive effects. The neurogenesis effect of G-CSF may thus work in several brain regions and even in the
387 whole brain. Extensive brain damage is caused by perinatal hypoxia, and G-CSF may provide substantial brain neurogenesis. In addition, G-CSF therapy enhanced neurogenesis but did not rescue all old neural and non-neural cells (Figure 4). The specificity of G-CSF may provide a new therapy strategy in the future. However, it is still unclear what type of neuronal cells, surviving neurons, or neuronal progenitor cells that G-CSF acts on to drive to neurogenesis in this study. Indeed, the G-CSF receptor can be expressed on various cell types, including hematopoietic stem cells (HSCs).40 G-CSF can enhance translocation of HSCs to the ischemic brain and improve neuronal plasticity and vascularization.20 Thus, in our experiments, there is a possibility that neurogenesis effect of G-CSF in perinatal hypoxia pups may also due to mobilizing HSCs into the brain and facilitating neurogenesis through neurotropic factors released or stimulated by HSCs. In conclusion, this study showed that G-CSF therapy was able to recover the perinatal hypoxia-induced decline in neurogenesis in rats, and, at least in part, improve the longterm functional outcomes (in the impaired performance in spatial working and cued reference memory). In addition, environment enrichment may provide a better outcome in perinatal hypoxia subjects, and early rehabilitation may be useful. This study suggests a useful and new potential neuroprotective strategy that may be well tolerated with fewer side effects in the neonatal brain after hypoxic episodes. Although we are far from being able to apply this therapy in human neonates, our results provide an insight into neurological function defects in patients after neonatal hypoxia.
Acknowledgments This study was supported in part by grants from E-Da Hospital, Kaohsiung City, Taiwan (EDAHP102016), National Science Council, Taipei City, Taiwan (NSC 98-2320-B-037022-MY3), and Kaohsiung Medical University Hospital, Kaohsiung City, Taiwan (KMUH96-6R-23, KMUH-97-7R06, KMUH98-8R-12).
References 1. du Plessis AJ, Volpe JJ. Perinatal brain injury in the preterm and term newborn. Curr Opin Neurol 2002;15:151e7. 2. Arvin KL, Han BH, Du Y, Lin SZ, Paul SM, Holtzman DM. Minocycline markedly protects the neonatal brain against hypoxicischemic injury. Ann Neurol 2002;52:54e61. 3. Fan LW, Lin S, Pang Y, Rhodes PG, Cai Z. Minocycline attenuates hypoxia-ischemia-induced neurological dysfunction and brain injury in the juvenile rat. Eur J Neurosci 2006;24: 341e50. 4. Gluckman PD, Pinal CS, Gunn AJ. Hypoxic-ischemic brain injury in the newborn: pathophysiology and potential strategies for intervention. Semin Neonatol 2001;6:109e20. 5. Sizonenko SV, Sirimanne E, Mayall Y, Gluckman PD, Inder T, Williams C. Selective cortical alteration after hypoxic-ischemic injury in the very immature rat brain. Pediatr Res 2003;54: 263e9. 6. Pugh P, Adlaf E, Zhao CS, Markwardt S, Gavin C, Wadiche J, et al. Enhanced integration of newborn neurons after neonatal insults. Front Neurosci 2011;5:45.
388 7. Malberg JE, Eisch AJ, Nestler EJ, Duman RS. Chronic antidepressant treatment increases neurogenesis in adult rat hippocampus. J Neurosci 2000;20:9104e10. 8. Gould E, Tanapat P, McEwen BS, Flu ¨gge G, Fuchs E. Proliferation of granule cell precursors in the dentate gyrus of adult monkeys is diminished by stress. Proc Natl Acad Sci U S A 1998; 95:3168e71. 9. Sato K, Morimoto K, Ujike H, Yamada T, Yamada N, Kuroda S, et al. The effects of perinatal anoxia or hypoxia on hippocampal kindling development in rats. Brain Res Bull 1994;35: 167e70. 10. Reinebrant HE, Wixey JA, Gobe GC, Colditz PB, Buller KM. Differential effects of neonatal hypoxic-ischemic brain injury on brainstem serotonergic raphe nuclei. Brain Res 2010;1322: 124e33. 11. Chen WF, Chang H, Huang LT, Lai MC, Yang CH, Wan TH, et al. Alterations in long-term seizure susceptibility and the complex of PSD-95 with NMDA receptor from animals previously exposed to perinatal hypoxia. Epilepsia 2006;47:288e96. 12. Jarrard LE, Davidson TL, Bowring B. Functional differentiation within the medial temporal lobe in the rat. Hippocampus 2004; 14:434e49. 13. Frampton JE, Lee CR, Faulds D. Filgrastim. A review of its pharmacological properties and therapeutic efficacy in neutropenia. Drugs 1994;48:731e60. 14. Heard SO, Fink MP. Counterregulatory control of the acute inflammatory response: granulocyte colony-stimulating factor has anti-inflammatory properties. Crit Care Med 1999;27: 1019e21. 15. Scha ¨bitz WR, Kollmar R, Schwaninger M, Juettler E, Bardutzky J, Scho ¨lzke MN, et al. Neuroprotective effect of granulocyte colony-stimulating factor after focal cerebral ischemia. Stroke 2003;34:745e51. 16. Yata K, Matchett GA, Tsubokawa T, Tang J, Kanamaru K, Zhang JH. Granulocyte-colony stimulating factor inhibits apoptotic neuron loss after neonatal hypoxia-ischemia in rats. Brain Res 2007;1145:227e38. 17. Solaroglu I, Jadhav V, Zhang JH. Neuroprotective effect of granulocyte-colony stimulating factor. Front Biosci 2007;12: 712e24. 18. Schneider A, Kru ¨ger C, Steigleder T, Weber D, Pitzer C, Laage R, et al. The hematopoietic factor G-CSF is a neuronal ligand that counteracts programmed cell death and drives neurogenesis. J Clin Invest 2005;115:2083e98. 19. Gibson CL, Jones NC, Prior MJ, Bath PM, Murphy SP. G-CSF suppresses edema formation and reduces interleukin-1beta expression after cerebral ischemia in mice. J Neuropathol Exp Neurol 2005;64:763e9. 20. Shyu WC, Lin SZ, Yang HI, Tzeng YS, Pang CY, Yen PS, et al. Functional recovery of stroke rats induced by granulocyte colony-stimulating factor-stimulated stem cells. Circulation 2004;110:1847e54. 21. Sheibani N, Grabowski EF, Schoenfeld DA, Whalen MJ. Effect of granulocyte colony-stimulating factor on functional and histopathologic outcome after traumatic brain injury in mice. Crit Care Med 2004;32:2274e8. 22. Chen WF, Hsu JH, Lin CS, Jong YJ, Yang CH, Huang LT, et al. Granulocyte-colony stimulating factor alleviates perinatal hypoxia-induced decreases in hippocampal synaptic efficacy and neurogenesis in the neonatal rat brain. Pediatr Res 2011; 70:589e95. 23. Chen WF, Chang H, Wong CS, Huang LT, Yang CH, Yang SN. Impaired expression of postsynaptic density proteins in the hippocampal CA1 region of rats following perinatal hypoxia. Exp Neurol 2007;204:400e10.
Y.-N. Yang et al 24. Yang SN, Huang LT, Wang CL, Chen WF, Yang CH, Lin SZ, et al. Prenatal administration of morphine decreases CREBSerine-133 phosphorylation and synaptic plasticity range mediated by glutamatergic transmission in the hippocampal CA1 area of cognitive-deficient rat offspring. Hippocampus 2003;13:915e21. 25. Romijn HJ, Hofman MA, Gramsbergen A. At what age is the developing cerebral cortex of the rat comparable to that of the full-term newborn human baby? Early Hum Dev 1991;26:61e7. 26. Ko ¨hr G. NMDA receptor function: subunit composition versus spatial distribution. Cell Tissue Res 2006;326:439e46. 27. Yang SN, Huang CB, Yang CH, Lai MC, Chen WF, Wang CL, et al. Impaired SynGAP expression and long-term spatial learning and memory in hippocampal CA1 area from rats previously exposed to perinatal hypoxia-induced insults: beneficial effects of A68930. Neurosci Lett 2004;371:73e8. 28. Lin CS, Tao PL, Jong YJ, Chen WF, Yang CH, Huang LT, et al. Prenatal morphine alters the synaptic complex of postsynaptic density 95 with N-methyl-D-aspartate receptor subunit in hippocampal CA1 subregion of rat offspring leading to long-term cognitive deficits. Neuroscience 2009;158:1326e37. 29. Oler JA, Markus EJ. Age-related deficits on the radial maze and in fear conditioning: hippocampal processing and consolidation. Hippocampus 1998;8:402e15. 30. Lee I, Kesner RP. Encoding versus retrieval of spatial memory: double dissociation between the dentate gyrus and the perforant path inputs into CA3 in the dorsal hippocampus. Hippocampus 2004;14:66e76. 31. Diaz A, Limon D, Cha ´vez R, Zenteno E, Guevara J. Ab25-35 injection into the temporal cortex induces chronic inflammation that contributes to neurodegeneration and spatial memory impairment in rats. J Alzheimers Dis 2012;30:505e22. 32. Diederich K, Sevimli S, Do ¨rr H, Ko ¨sters E, Hoppen M, Lewejohann L, et al. The role of granulocyte-colony stimulating factor (G-CSF) in the healthy brain: a characterization of G-CSF-deficient mice. J Neurosci 2009;29:11572e81. 33. Liu N, He S, Yu X. Early natural stimulation through environmental enrichment accelerates neuronal development in the mouse dentate gyrus. PLoS One 2012;7:e30803. 34. Minnerup J, Heidrich J, Wellmann J, Rogalewski A, Schneider A, Scha ¨bitz WR. Meta-analysis of the efficacy of granulocyte-colony stimulating factor in animal models of focal cerebral ischemia. Stroke 2008;39:1855e61. 35. Ohki Y, Heissig B, Sato Y, Akiyama H, Zhu Z, Hicklin DJ, et al. Granulocyte colony-stimulating factor promotes neovascularization by releasing vascular endothelial growth factor from neutrophils. FASEB J 2005;19:2005e7. 36. Bito H, Takemoto-Kimura S. Ca(2þ)/CREB/CBP-dependent gene regulation: a shared mechanism critical in long-term synaptic plasticity and neuronal survival. Cell Calcium 2003; 34:425e30. 37. Tojima T, Ito E. Signal transduction cascades underlying de novo protein synthesis required for neuronal morphogenesis in differentiating neurons. Prog Neurobiol 2004;72:183e93. 38. Garoflos E, Stamatakis A, Mantelas A, Philippidis H, Stylianopoulou F. Cellular mechanisms underlying an effect of "early handling" on pCREB and BDNF in the neonatal rat hippocampus. Brain Res 2005;1052:187e95. 39. Hughes PE, Alexi T, Walton M, Williams CE, Dragunow M, Clark RG, et al. Activity and injury-dependent expression of inducible transcription factors, growth factors and apoptosisrelated genes within the central nervous system. Prog Neurobiol 1999;57:421e50. 40. Liu F, Poursine-Laurent J, Link DC. Expression of the G-CSF receptor on hematopoietic progenitor cells is not required for their mobilization by G-CSF. Blood 2000;95:3025e31.
Available online at www.sciencedirect.com
journal homepage: http://www.pediatr-neonatol.com
ORIGINAL ARTICLE
Neurogenesis Recovery Induced by Granulocyte-colony Stimulating Factor in Neonatal Rat Brain After Perinatal Hypoxia Yung-Ning Yang a,b, Chien-Seng Lin c, Chun-Hwa Yang b,d, Yu-Hung Lai b,e, Pei-Ling Wu b,d, San-Nan Yang a,b,d,f,* a
Department of Pediatrics, E-Da Hospital, I-Shou University, Kaohsiung, Taiwan Graduate Institute of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan c Department of Emergency and Critical Care Medicine, Cheng Hsin Rehabilitation Medical Center, Taipei, Taiwan d Department of Pediatrics, Kaohsiung Medical University Hospital, Kaohsiung, Taiwan e Department of Ophthalmology, Kaohsiung Medical University Hospital, Kaohsiung Medical University, Kaohsiung, Taiwan f Department of Marine Biotechnology and Resources, National Sun Yat-sen University, Kaohsiung, Taiwan b
Received Nov 11, 2012; received in revised form Mar 14, 2013; accepted May 14, 2013
Key Words brain; cognitive function; granulocyte-colony stimulating factor; hypoxia; neurogenesis
Background: Perinatal hypoxia can lead to a wide range of neurological deficits depending on the differential vulnerability of the involved brain regions to oxygen deprivation. It remains unclear whether the differential vulnerability to oxygen deprivation leads to altered neurogenesis in the neonatal brain after perinatal hypoxia. The primary objective was to investigate whether perinatal hypoxia induces deleterious changes in neurogenesis within three representative brain regions (dentate gyrus of the hippocampus, midbrain, and temporal cortex), with regards to common pathological areas clinically. The secondary objective was to investigate whether granulocyte-colony stimulating factor (G-CSF) therapy exerts beneficial effects in neurogenesis in neonatal rat brains subjected to experimental perinatal hypoxia. Materials and methods: Rat pups were subjected to experimental perinatal hypoxia on the tenth day of life (P10). They were then given G-CSF (30 mg/kg, single injection/day, intraperitoneal injection, P11e16). The neurogenesis efficacy was analyzed on P17 and the radial-arm maze task, a memory task for higher cognitive functions such as problem-solving abilities, was evaluated on P37e58.
* Corresponding author. Department of Pediatrics, Kaohsiung Medical University Hospital, Graduate Institute of Medicine, Kaohsiung Medical University, No. 100, Tzyou 1st Road, Sanmin District, Kaohsiung City 807, Taiwan. E-mail address: [email protected] (S.-N. Yang). 1875-9572/$36 Copyright ª 2013, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. All rights reserved. http://dx.doi.org/10.1016/j.pedneo.2013.04.011
G-CSF and neurogenesis in the neonatal brain
381
Results: Perinatal hypoxia caused a significant decrease in neurogenesis within the three representative brain regions, and this deleterious outcome was alleviated by G-CSF (p < 0.05). In addition, the G-CSF therapy markedly improved the decreased performance of long-term cognitive functions induced by perinatal hypoxia (p < 0.05). Conclusion: This study suggests that G-CSF may be a potentially beneficial therapy, at least in part, through universal recovery of neurogenesis effects in the neonatal brain after perinatal hypoxia insult. Copyright ª 2013, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. All rights reserved.
1. Introduction In immature brains, cerebral hypoxia contributes to devastating and lifelong neurological impairment different from the behavioral impairment in mental retardation and cerebral palsy.1 Previous studies have reported that hypoxic injury in young rats can cause extensive neuronal damage.2e5 Perinatal hypoxia causes brain damage in multiple brain regions including the dentate gyrus, midbrain, and temporal lobe.6e10 The dentate gyrus of the hippocampus (an important area for mammalian memory, stress, and depression) is often discussed due to its sensitivity to oxygen. The midbrain (a functional area for motivation and habituation) is associated with seizure after hypoxia insult.11 The temporal lobe with interrelated structure referred to as the hippocampus is an area involved in auditory perception and implicated in learning and memory.12 Although exposure to perinatal hypoxia in these brain regions has been extensively studied, the mechanisms and effects of clinical treatment are still unclear. Granulocyte-colony stimulating factor (G-CSF) is a glycoprotein of 19.6-kDa that has been used to treat patients with neutropenia.13 It has been shown to stimulate the bone marrow to produce stem cells and induce their mobilization from the bone marrow to the brain, then inducing neurogenesis in adult animals with stroke to improve functional outcomes.14e21 Schneider et al18 proposed that G-CSF leads to recovery of hippocampal neurogenesis post ischemia. Many studies have focused on G-CSF therapy in the brains of adults who have had strokes; however, few studies have focused on the beneficial effects in neonatal brains post perinatal hypoxia. Our previous animal model study reported that G-CSF had clinical potential to treat perinatal hypoxia by recovery of neurogenesis within the hippocampal CA1 region.22 Hence, in this study, we investigated the efficacy of G-CSF in the dentate gyrus of the hippocampus, midbrain, and temporal cortex of neonatal brains, common pathological areas after perinatal hypoxia, using an animal model. We also used an eight-arm radial maze task to examine the performance in long-term cognitive functions after perinatal hypoxia with regards to the cued reference and spatial working memory.
2. Materials and Methods 2.1. Experimental animal protocols The Animal Care and Use Committee at Kaohsiung Medical University (Kaohsiung, Taiwan) and National Science
Council (Taipei, Taiwan) approved all experimental procedures. Sprague-Dawley rats were provided with a 12-hour light/dark cycle and housed in the animal care facility. The animals were divided into four experimental groups as follows: vehicle-control, perinatal hypoxia plus vehicle, G-CSF alone, and perinatal hypoxia plus G-CSF on postnatal jDay 10 (P10). The procedures for perinatal hypoxia have been described previously.11,23,24 The brains of 10e12-dayold rats are generally accepted to be comparable to term human infants.25 We removed rats (P10) from the litter and placed them in an airtight chamber (30 cm 30 cm 30 cm) on a heating pad at a temperature of 34 C. The concentration of oxygen was then reduced to 5e7% and maintained at this level. In addition, nitrogen gas was regulated by infusion into the chamber. The rats were used for this study only when at least one toniceclonic seizure was observed with low blood oxygen saturation (<60%) during hypoxia. Recombinant human G-CSF (Kyowa Hakko Kirin Co. Ltd., Tokyo, Japan; 30 mg/kg, single injection/day, intraperitoneal injection, P11e16) or vehiclesaline (P11e16) was injected subcutaneously into the rat pups 1 day after perinatal hypoxia had occurred. The dosage of G-CSF used (30 mg/kg) was the most effective dose to alleviate hypoxia-induced damage during the perinatal period according to our previous study.22 Brain tissue specimens of the target brain regions were collected on P17, and the memory functions and spatial learning were analyzed at P37eP58.
2.2. Brain tissue slice preparation On P17, slices (400 mm) of the dentate gyrus, midbrain, and temporal lobe were collected and immediately placed into artificial cerebrospinal fluid in a humidified atmosphere of 95% oxygen/5% carbon dioxide at 34.0 0.5 C in an incubation chamber for an equilibrium period of at least 1 hour.23 To obtain the target brain regions, the brain slices were incubated with ice-cold oxygenated artificial cerebrospinal fluid and then cut into a tissue blocks (0.1 cm 0.5 cm) within 15 seconds of slicing the tissue. The brain tissues were then immediately frozen at 80 C until analysis, and all tissue slices were only ever thawed once.
2.3. Evaluation of neurogenesis Bromodeoxyuridine (BrdU), a cell type-specific marker, was used to identify neurogenesis, and double immunofluorescence analysis of laser-scanning confocal microscopy was performed as previously described.18,20,26 The brains were
382 cut coronally at 30 mm with washing for 20 minutes in phosphate-buffered saline containing 0.2% TritonX-100, pH 7.3 at room temperature with gentle rocking, followed by incubation for 1 hour at room temperature in blocking solution with gentle rocking on P17. BrdU injections (P14e16, 50 mg/kg q. 12 hours, subcutaneously) were used to detect the cell proliferation induced by G-CSF. Neuronal nuclei (NeuN)-staining was performed to detect neuron cells and the region between 2.8 mm and 4.0 mm behind the bregma was of interest for neuron counting.27 In each group, brain regions were taken from a total of 10 sections for each of the 10 animals. The expressions of NeuN-positive and BrdUpositive double immunofluorescence were tested following the instructions of the Roche BrdU labeling and detection kit. Each coronal section was first treated with primary BrdU antibody conjugated with FITC (1:250), followed by treatment with neuron-specific antibodies (NeuN, 1:500; Chemicon, Temecula, CA, USA). The tissue sections were analyzed with an Olympus laser-scanning confocal microscope, with green (FITC) and red (Cy3) fluorochromes on the slides excited by a laser beam at 488 nm and 543 nm, respectively.
Y.-N. Yang et al
2.4. Eight-arm radial maze task As in our previous study,28 an eight-arm radial maze was used to train rats on two memory tasks. We placed bait in half of the symmetrical arms, and a black textured insert was put in each of the remaining four arms. All of the rats performed one training session/day for 15 days. They were placed in the center of the maze, facing a different arm in a quasirandom order in each session. The rat was removed from the maze when it had eaten the bait from all four arms or after 15 minutes, whichever came first. After the bait in an arm had been eaten, it was not replenished. The arms containing the textured insert served as reference points in the memory task, and the arms without the insert served as spatial cues in the memory task because the rats could not distinguish the arms that they had visited. In this test, two memory deficits could be recorded: (1) working memory error (WME), a rat went to an arm with bait that it had already visited; and (2) reference memory error (RME), a rat visited an arm with no bait. If the forepaws of the rat crossed the midline of the arm, it was defined as having visited that arm. The rats were defined as having
Figure 1 Effects of granulocyte-colony stimulating factor (G-CSF) therapy in the declining expression of neurogenesis in the dentate gyrus caused by perinatal hypoxia. Colocalization of neuronal nuclei (NeuN, red for neuron identification)- and bromodeoxyuridine (BrdU, green for cell nucleus detection)-positive cells, as assessed on P17, was identified by doubleimmunofluorescence staining under laser-scanning confocal microscopy. The NeuN-positive cells (1), BrdU-positive cells (2), and the combined pictures of the NeuN- and BrdU-positive cells are as follows: (A1e3) vehicle-control; (B1e3) G-CSF alone (30 mg/kg); C1e3 perinatal hypoxia plus vehicle; and (D1e3) perinatal hypoxia plus G-CSF (30 mg/kg). The arrows reveal the representative newly generated neurons. Bar Z 50 mm.
G-CSF and neurogenesis in the neonatal brain
383
Figure 1
successfully completed the cued reference memory and spatial working memory tasks when there were no RMEs or WMEs for at least 2 days within a 3-day period.
2.5. Statistical analysis All data are presented as mean standard error of the mean. Statistical differences were determined by use of one way analysis of variance (ANOVA) followed by a Bonferroni’s t test for posthoc multiple comparisons. A statistical significance level of p < 0.05 was applied to all tests.
3. Results 3.1. Evaluation of neurogenesis We first evaluated whether the G-CSF therapy increased the expression of neurogenesis within the dentate gyrus of the neonatal brains after perinatal hypoxia. A declining trend in the coexpression of NeuN-positive with BrdUpositive cells was revealed in the perinatal hypoxia plus vehicle group (Figure 1C) compared with the vehiclecontrol rats (Figure 1A). By contrast, an increase in the number of BrdU-positive cells colocalizing with NeuNpositive cells was observed in the perinatal hypoxia plus G-CSF group (Figure 1D). In the midbrain, a decrease in the coexpression of BrdUpositive with NeuN-positive cells was observed in the
(continued)
perinatal hypoxia group plus vehicle (Figure 2C) compared with the vehicle-control rats (Figure 2A). However, the perinatal hypoxia plus G-CSF therapy group showed an increase in the number of BrdU-positive cells colocalizing with NeuN-positive cells (Figure 2D). Furthermore, as shown in Figure 3, the perinatal hypoxia plus vehicle group (Figure 3C) had a significantly decreased coexpression of BrdU-positive with NeuN-positive cells compared with the vehicle-control (Figure 3A) and perinatal hypoxia plus G-CSF therapy groups (Figure 3D) in the temporal cortex. The three brain areas in the G-CSF alone group (Figures 1B, and 3B, 2B) were not significantly different, as compared with the vehicle control group. According to our previous study, there was no significant difference of long-term learning and memory between G-CSF alone and vehicle control groups.22 Therefore, in the subsequent experiments we did not perform radial maze for G-CSF alone group. In quantitative analysis (Figure 4), the average number of cells exhibiting double-staining with NeuN-positive and BrdU-positive within the counted areas showed that G-CSF therapy had significantly improved the perinatal hypoxiainduced decrease in neurogenesis, confirming the double staining of immunofluorescence findings.
3.2. Eight-arm radial maze task After repeated daily training with the eight-arm maze task, the long-term cognitive deficits in the working and cued
384
Y.-N. Yang et al
Figure 2 Granulocyte-colony stimulating factor (G-CSF) therapy restored neurogenesis in the midbrain post perinatal hypoxia. The neuronal nuclei-positive cells (1), the bromodeoxyuridine-positive cells (2), and the combined pictures of the neuronal nuclei- and bromodeoxyuridine-positive cells (3) are as follows: (A1e3) vehicle-control; (B1e3) G-CSF alone (30 mg/kg); (C1e3) perinatal hypoxia plus vehicle; and (D1e3) perinatal hypoxia plus G-CSF (30 mg/kg). The arrows reveal the representative newly generated neurons. Bar Z 50 mm.
G-CSF and neurogenesis in the neonatal brain
385
Figure 3 Granulocyte-colony stimulating factor (G-CSF) therapy restored neurogenesis in the temporal cortex post perinatal hypoxia. The neuronal nuclei-positive cells (1), the bromodeoxyuridine-positive cells (2), and the combined pictures of the neuronal nucleiand bromodeoxyuridine-positive cells (3) are as follows: (A1) vehicle-control; (B1) G-CSF alone (30 mg/kg); (C1) perinatal hypoxia plus vehicle; and (D1) perinatal hypoxia plus G-CSF (30 mg/kg). The arrows reveal the representative newly generated neurons. Bar Z 50 mm.
386
Y.-N. Yang et al
Figure 4 Quantified analysis of the effects of granulocyte-colony stimulating factor (G-CSF) in the decreased expressions of neurogenesis in the dentate gyrus, midbrain, and temporal cortex postperinatal hypoxia. The experimental groups are indicated as follows: vehicle-control (C), G-CSF alone (G), perinatal hypoxia plus vehicle (H), perinatal hypoxia plus G-CSF (HþG). (n Z 10 animals for each experimental group.) Colocalization of neuronal nuclei (NeuN)-positive and bromodeoxyuridine (BrdU)-positive cells are indicated new neuronal cells. NeuN-positive and BrdU-negative cells are indicated as old neuronal cells. NeuN-negative and BrdU-positive cells are non-neuronal cells. * p < 0.05 when compared with the vehicle-control group. Data are presented as mean standard error of the mean.
reference memory tasks in the rats post perinatal hypoxia are shown in Figure 5. The perinatal hypoxia plus vehicle group (n Z 11 animals, p < 0.05) had more mean RMEs and WMEs compared with the vehicle-control group in the first few days of life. The experimental perinatal hypoxia plus G-CSF (n Z 10 animals, p < 0.05) group had an alleviated mean RME and a trend of lower mean WME compared with the perinatal hypoxia plus vehicle rats in the first few days of life. Furthermore, the perinatal hypoxia plus vehicle group had a significant increase in the number of total mean RMEs and total mean WMEs compared with the vehicle-control group. The perinatal hypoxia plus G-CSF group showed improved total mean RMEs and total mean WMEs compared with the perinatal hypoxia plus vehicle group. These results suggest that perinatal hypoxia led to a decline in the ability of spatial memory formation, and that the insult caused by the perinatal hypoxia could be rescued after G-CSF therapy.
4. Discussion In this study, we showed that systemic injections of G-CSF had beneficial effects after perinatal hypoxia on neonatal rat brains in several brain regions. G-CSF therapy significantly abated the decline in neurogenesis in the dentate gyrus, midbrain, and temporal lobe of the neonatal brains after experimental perinatal hypoxia. In addition, working and cued reference memory impairments of the perinatal hypoxia rats could be also rescued by this therapy. Using the eight-arm radial maze task, rats in the perinatal hypoxia group were shown to have significantly reduced spatial memory; however, the perinatal hypoxia with G-CSF therapy group showed significant improvements in the damage caused by perinatal hypoxia. In addition, the histological results showed an increase in neurogenesis in the dentate gyrus in the perinatal hypoxia with G-CSF therapy group compared with the perinatal hypoxia group. In previous studies, the hippocampus (including the dentate gyrus) and temporal cortex have been proven to be related to spatial working memory,12,29e31 and G-CSF has been
proven to play an important role in the neural cells with regards to proliferation, differentiation, and integration in the dentate gyrus for the hippocampus.32 However, the mechanism by which G-CSF acts in rat pups with experimental perinatal hypoxia is unclear. The current study provides evidence that G-CSF administration following injury due to perinatal hypoxia seems to improve working and cued reference memory by enhancing neurogenesis in the dentate gyrus of the hippocampus. Interestingly, the difference in RMEs and WMEs between the three groups seemed to decrease after repeat daily training. This spontaneous improvement of spatial memory in the perinatal hypoxia group may be due to repeat training, because an enriched environment is thought to accelerate neuronal development in the dentate gyrus of rats.33 Our radial maze task findings may suggest that early rehabilitation is useful in alleviating cognitive impairment caused by perinatal hypoxia. The benefits and possible mechanisms after G-CSF therapy have been reported to be through several effects.17 In the central nervous system, G-CSF therapy acts in neurons through G-CSF receptors.15,18 G-CSF may affect neuroprotection through neurogenesis, antiapoptotic, antiinflammatory, neurotrophic, and angiogenesis effects in the mammalian adult brain.15e21,34,35 Previous studies have shown that G-CSF has neurogenesis effect in nonischemic or healthy brain of adult rodents.18 However, in this study there was no statistical difference in neurogenesis between the G-CSF alone and vehicle-control group. Thus, it is noteworthy that this discrepancy may be due to different animal species and ages, as well as different hypoxia protocol. cAMP response element-binding protein phosphorylated at serine 133 (pCREBSer-133) has been shown to lead to synaptogenesis, neuronal survival with neurogenesis, and synaptic plasticity in the brain.36e38 In addition, CREB is a DNA-binding transcription factor and plays an important role in the regulation of some immediate-early genes.39 Previously, we reported that G-CSF therapy alleviates the decreased in pCREBSer-133 and postsynaptic density protein
G-CSF and neurogenesis in the neonatal brain
Figure 5 Impairment of the long-term performance of working and cued reference memory was found using an eightarm radial maze task after experimental perinatal hypoxia. (A) One trial/day over 15 trials, average amount of cued reference memory errors, and the inset box reveals the average of total mean reference memory errors. (B) One trial/day over 15 trials, average amount of working memory errors, and the inset box reveals the average total mean working memory errors. Vehicle-control ( n Z 10 animals), perinatal hypoxia plus vehicle (O; n Z 11 animals), perinatal hypoxia plus granulocyte-colony stimulating factor (;; n Z 10 animals), are labeled as C, H, and H þ G, respectively. Statistical analysis of the long-term performance of the working and cued reference memory revealed that the rats with perinatal hypoxia plus vehicle significantly differed from those in the vehiclecontrol and perinatal hypoxia plus granulocyte-colony stimulating factor groups. *p < 0.05, when compared with the vehicle-control group.
95 with N-methyl-D-aspartate receptor subunits within the CA1 region of the hippocampus after perinatal hypoxia, and provides beneficial effects in long-term cognitive deficits.22 However, whether the neuroprotective effects of G-CSF are also displayed in multiple brain regions after experimental perinatal hypoxia was unknown. Hence, we investigated this in the current study and showed that the efficacy of GCSF therapy in neurogenesis after perinatal hypoxia was not only in the CA1 region of the hippocampus but also in the dentate gyrus, midbrain, and temporal region with longterm cognitive effects. The neurogenesis effect of G-CSF may thus work in several brain regions and even in the
387 whole brain. Extensive brain damage is caused by perinatal hypoxia, and G-CSF may provide substantial brain neurogenesis. In addition, G-CSF therapy enhanced neurogenesis but did not rescue all old neural and non-neural cells (Figure 4). The specificity of G-CSF may provide a new therapy strategy in the future. However, it is still unclear what type of neuronal cells, surviving neurons, or neuronal progenitor cells that G-CSF acts on to drive to neurogenesis in this study. Indeed, the G-CSF receptor can be expressed on various cell types, including hematopoietic stem cells (HSCs).40 G-CSF can enhance translocation of HSCs to the ischemic brain and improve neuronal plasticity and vascularization.20 Thus, in our experiments, there is a possibility that neurogenesis effect of G-CSF in perinatal hypoxia pups may also due to mobilizing HSCs into the brain and facilitating neurogenesis through neurotropic factors released or stimulated by HSCs. In conclusion, this study showed that G-CSF therapy was able to recover the perinatal hypoxia-induced decline in neurogenesis in rats, and, at least in part, improve the longterm functional outcomes (in the impaired performance in spatial working and cued reference memory). In addition, environment enrichment may provide a better outcome in perinatal hypoxia subjects, and early rehabilitation may be useful. This study suggests a useful and new potential neuroprotective strategy that may be well tolerated with fewer side effects in the neonatal brain after hypoxic episodes. Although we are far from being able to apply this therapy in human neonates, our results provide an insight into neurological function defects in patients after neonatal hypoxia.
Acknowledgments This study was supported in part by grants from E-Da Hospital, Kaohsiung City, Taiwan (EDAHP102016), National Science Council, Taipei City, Taiwan (NSC 98-2320-B-037022-MY3), and Kaohsiung Medical University Hospital, Kaohsiung City, Taiwan (KMUH96-6R-23, KMUH-97-7R06, KMUH98-8R-12).
References 1. du Plessis AJ, Volpe JJ. Perinatal brain injury in the preterm and term newborn. Curr Opin Neurol 2002;15:151e7. 2. Arvin KL, Han BH, Du Y, Lin SZ, Paul SM, Holtzman DM. Minocycline markedly protects the neonatal brain against hypoxicischemic injury. Ann Neurol 2002;52:54e61. 3. Fan LW, Lin S, Pang Y, Rhodes PG, Cai Z. Minocycline attenuates hypoxia-ischemia-induced neurological dysfunction and brain injury in the juvenile rat. Eur J Neurosci 2006;24: 341e50. 4. Gluckman PD, Pinal CS, Gunn AJ. Hypoxic-ischemic brain injury in the newborn: pathophysiology and potential strategies for intervention. Semin Neonatol 2001;6:109e20. 5. Sizonenko SV, Sirimanne E, Mayall Y, Gluckman PD, Inder T, Williams C. Selective cortical alteration after hypoxic-ischemic injury in the very immature rat brain. Pediatr Res 2003;54: 263e9. 6. Pugh P, Adlaf E, Zhao CS, Markwardt S, Gavin C, Wadiche J, et al. Enhanced integration of newborn neurons after neonatal insults. Front Neurosci 2011;5:45.
388 7. Malberg JE, Eisch AJ, Nestler EJ, Duman RS. Chronic antidepressant treatment increases neurogenesis in adult rat hippocampus. J Neurosci 2000;20:9104e10. 8. Gould E, Tanapat P, McEwen BS, Flu ¨gge G, Fuchs E. Proliferation of granule cell precursors in the dentate gyrus of adult monkeys is diminished by stress. Proc Natl Acad Sci U S A 1998; 95:3168e71. 9. Sato K, Morimoto K, Ujike H, Yamada T, Yamada N, Kuroda S, et al. The effects of perinatal anoxia or hypoxia on hippocampal kindling development in rats. Brain Res Bull 1994;35: 167e70. 10. Reinebrant HE, Wixey JA, Gobe GC, Colditz PB, Buller KM. Differential effects of neonatal hypoxic-ischemic brain injury on brainstem serotonergic raphe nuclei. Brain Res 2010;1322: 124e33. 11. Chen WF, Chang H, Huang LT, Lai MC, Yang CH, Wan TH, et al. Alterations in long-term seizure susceptibility and the complex of PSD-95 with NMDA receptor from animals previously exposed to perinatal hypoxia. Epilepsia 2006;47:288e96. 12. Jarrard LE, Davidson TL, Bowring B. Functional differentiation within the medial temporal lobe in the rat. Hippocampus 2004; 14:434e49. 13. Frampton JE, Lee CR, Faulds D. Filgrastim. A review of its pharmacological properties and therapeutic efficacy in neutropenia. Drugs 1994;48:731e60. 14. Heard SO, Fink MP. Counterregulatory control of the acute inflammatory response: granulocyte colony-stimulating factor has anti-inflammatory properties. Crit Care Med 1999;27: 1019e21. 15. Scha ¨bitz WR, Kollmar R, Schwaninger M, Juettler E, Bardutzky J, Scho ¨lzke MN, et al. Neuroprotective effect of granulocyte colony-stimulating factor after focal cerebral ischemia. Stroke 2003;34:745e51. 16. Yata K, Matchett GA, Tsubokawa T, Tang J, Kanamaru K, Zhang JH. Granulocyte-colony stimulating factor inhibits apoptotic neuron loss after neonatal hypoxia-ischemia in rats. Brain Res 2007;1145:227e38. 17. Solaroglu I, Jadhav V, Zhang JH. Neuroprotective effect of granulocyte-colony stimulating factor. Front Biosci 2007;12: 712e24. 18. Schneider A, Kru ¨ger C, Steigleder T, Weber D, Pitzer C, Laage R, et al. The hematopoietic factor G-CSF is a neuronal ligand that counteracts programmed cell death and drives neurogenesis. J Clin Invest 2005;115:2083e98. 19. Gibson CL, Jones NC, Prior MJ, Bath PM, Murphy SP. G-CSF suppresses edema formation and reduces interleukin-1beta expression after cerebral ischemia in mice. J Neuropathol Exp Neurol 2005;64:763e9. 20. Shyu WC, Lin SZ, Yang HI, Tzeng YS, Pang CY, Yen PS, et al. Functional recovery of stroke rats induced by granulocyte colony-stimulating factor-stimulated stem cells. Circulation 2004;110:1847e54. 21. Sheibani N, Grabowski EF, Schoenfeld DA, Whalen MJ. Effect of granulocyte colony-stimulating factor on functional and histopathologic outcome after traumatic brain injury in mice. Crit Care Med 2004;32:2274e8. 22. Chen WF, Hsu JH, Lin CS, Jong YJ, Yang CH, Huang LT, et al. Granulocyte-colony stimulating factor alleviates perinatal hypoxia-induced decreases in hippocampal synaptic efficacy and neurogenesis in the neonatal rat brain. Pediatr Res 2011; 70:589e95. 23. Chen WF, Chang H, Wong CS, Huang LT, Yang CH, Yang SN. Impaired expression of postsynaptic density proteins in the hippocampal CA1 region of rats following perinatal hypoxia. Exp Neurol 2007;204:400e10.
Y.-N. Yang et al 24. Yang SN, Huang LT, Wang CL, Chen WF, Yang CH, Lin SZ, et al. Prenatal administration of morphine decreases CREBSerine-133 phosphorylation and synaptic plasticity range mediated by glutamatergic transmission in the hippocampal CA1 area of cognitive-deficient rat offspring. Hippocampus 2003;13:915e21. 25. Romijn HJ, Hofman MA, Gramsbergen A. At what age is the developing cerebral cortex of the rat comparable to that of the full-term newborn human baby? Early Hum Dev 1991;26:61e7. 26. Ko ¨hr G. NMDA receptor function: subunit composition versus spatial distribution. Cell Tissue Res 2006;326:439e46. 27. Yang SN, Huang CB, Yang CH, Lai MC, Chen WF, Wang CL, et al. Impaired SynGAP expression and long-term spatial learning and memory in hippocampal CA1 area from rats previously exposed to perinatal hypoxia-induced insults: beneficial effects of A68930. Neurosci Lett 2004;371:73e8. 28. Lin CS, Tao PL, Jong YJ, Chen WF, Yang CH, Huang LT, et al. Prenatal morphine alters the synaptic complex of postsynaptic density 95 with N-methyl-D-aspartate receptor subunit in hippocampal CA1 subregion of rat offspring leading to long-term cognitive deficits. Neuroscience 2009;158:1326e37. 29. Oler JA, Markus EJ. Age-related deficits on the radial maze and in fear conditioning: hippocampal processing and consolidation. Hippocampus 1998;8:402e15. 30. Lee I, Kesner RP. Encoding versus retrieval of spatial memory: double dissociation between the dentate gyrus and the perforant path inputs into CA3 in the dorsal hippocampus. Hippocampus 2004;14:66e76. 31. Diaz A, Limon D, Cha ´vez R, Zenteno E, Guevara J. Ab25-35 injection into the temporal cortex induces chronic inflammation that contributes to neurodegeneration and spatial memory impairment in rats. J Alzheimers Dis 2012;30:505e22. 32. Diederich K, Sevimli S, Do ¨rr H, Ko ¨sters E, Hoppen M, Lewejohann L, et al. The role of granulocyte-colony stimulating factor (G-CSF) in the healthy brain: a characterization of G-CSF-deficient mice. J Neurosci 2009;29:11572e81. 33. Liu N, He S, Yu X. Early natural stimulation through environmental enrichment accelerates neuronal development in the mouse dentate gyrus. PLoS One 2012;7:e30803. 34. Minnerup J, Heidrich J, Wellmann J, Rogalewski A, Schneider A, Scha ¨bitz WR. Meta-analysis of the efficacy of granulocyte-colony stimulating factor in animal models of focal cerebral ischemia. Stroke 2008;39:1855e61. 35. Ohki Y, Heissig B, Sato Y, Akiyama H, Zhu Z, Hicklin DJ, et al. Granulocyte colony-stimulating factor promotes neovascularization by releasing vascular endothelial growth factor from neutrophils. FASEB J 2005;19:2005e7. 36. Bito H, Takemoto-Kimura S. Ca(2þ)/CREB/CBP-dependent gene regulation: a shared mechanism critical in long-term synaptic plasticity and neuronal survival. Cell Calcium 2003; 34:425e30. 37. Tojima T, Ito E. Signal transduction cascades underlying de novo protein synthesis required for neuronal morphogenesis in differentiating neurons. Prog Neurobiol 2004;72:183e93. 38. Garoflos E, Stamatakis A, Mantelas A, Philippidis H, Stylianopoulou F. Cellular mechanisms underlying an effect of "early handling" on pCREB and BDNF in the neonatal rat hippocampus. Brain Res 2005;1052:187e95. 39. Hughes PE, Alexi T, Walton M, Williams CE, Dragunow M, Clark RG, et al. Activity and injury-dependent expression of inducible transcription factors, growth factors and apoptosisrelated genes within the central nervous system. Prog Neurobiol 1999;57:421e50. 40. Liu F, Poursine-Laurent J, Link DC. Expression of the G-CSF receptor on hematopoietic progenitor cells is not required for their mobilization by G-CSF. Blood 2000;95:3025e31.