Long-term cognitive eﬀ ects of human stem cell transplantation in the irradiated brain

Purpose : Radiotherapy remains a primary treatment modality for the majority of central nervous system tumors, but frequently leads to debilitating cognitive dysfunction. Given the absence of satisfactory solutions to this serious problem, we have used human stem cell therapies to ameliorate radiation-induced cognitive impairment. Here, past studies have been extended to determine whether engrafted cells provide even longer-term benefi ts to cognition. Materials and methods : embryonic stem cells (hESC) or human neural stem cells (hNSC) were transplanted, and animals were subjected to cognitive testing on a novel place recognition task 8 months later. Results : Grafting of hNSC was found to provide long lasting cognitive benefi ts over an 8-month post-irradiation interval. At this protracted time, hNSC grafting improved behavioral performance on a novel place recognition task compared to irradiated animals not receiving stem cells. Engrafted hESC previously shown to be benefi cial following a similar task, 1 and 4 months after irradiation, were not found to provide cognitive benefi ts at 8 months. Conclusions : Our fi ndings suggest that hNSC transplantation promotes the long-term recovery of the irradiated brain, where intrahippocampal stem cell grafting helps to preserve cognitive function.


Introduction
Ionizing radiation is a first-line treatment to control primary and metastatic brain tumors and can induce a progressive and long-lasting decline in cognition that can severely impact quality of life (Butler et al. 2006, Meyers andBrown 2006). Given the growing population of long-term survivors of intracranial tumors, quality of life has become an increasing concern with no satisfactory, long-term solutions. We have recently demonstrated that intrahippocampal transplantation of human stem cells prevented the development of radiation-induced cognitive impairment in rodents at 1-and 4-months post-surgery (Acharya et al. 2009(Acharya et al. , 2011(Acharya et al. , 2013. Th ese studies provided the fi rst evidence that engraftment of either pluripotent human embryonic stem cells (hESC) or multipotent human neural stem cells (hNSC) could protect the brain from a serious side eff ect of cranial irradiation (IRR).
Th e mechanisms underlying radiation-induced cognitive impairment are not well understood and are likely multifaceted involving microenvironmental factors such as oxidative stress and infl ammation ), which, in turn, can infl uence neural stem/progenitor cell (NSC) populations associated with cognitive function. Recently, we have shown that cranial IRR alters mature neuronal architecture (dendrites and spines) and modulates proteins involved in synaptic function in the hippocampus Limoli 2013, Parihar et al. 2014). Th e hippocampus is a brain region critical for the acquisition (learning), consolidation and retrieval of declarative memories (for review see: Squire 1994, Eichenbaum 2001. Th ese processes modulate the strength and effi cacy of synaptic signaling (i.e., synaptic plasticity), which impacts neurotransmission and provides a mechanism for translating synaptic plasticity into changes in synaptic strength (memory). How radiation impacts these processes was a lifelong interest of Mike Robbins, and much of his work analyzing the capability of peroxisomal proliferating-activated receptor activation (Greene-Schloesser et al. 2014) or inhibition of the renin-angiotensin system (Lee et al. 2012) demonstrated how pharmacologic strategies could be used to attenuate the adverse eff ects of IRR on cognition. Mike Robbins was always keen on elucidating the molecular pathways contributing to radiation-induced cognitive dysfunction (Greene-Schloesser et al. 2013) and much of this passion persuaded our eff orts to use stem cell transplantation as an alternative strategy to ameliorate Abstract Purpose : Radiotherapy remains a primary treatment modality for the majority of central nervous system tumors, but frequently leads to debilitating cognitive dysfunction. Given the absence of satisfactory solutions to this serious problem, we have used human stem cell therapies to ameliorate radiation-induced cognitive impairment. Here, past studies have been extended to determine whether engrafted cells provide even longer-term benefi ts to cognition. Materials and methods : Athymic nude rats were cranially irradiated (10 Gy) and subjected to intrahippocampal transplantation surgery 2 days later. Human embryonic stem cells (hESC) or human neural stem cells (hNSC) were transplanted, and animals were subjected to cognitive testing on a novel place recognition task 8 months later. Results : Grafting of hNSC was found to provide long lasting cognitive benefi ts over an 8-month post-irradiation interval. At this protracted time, hNSC grafting improved behavioral performance on a novel place recognition task compared to irradiated animals not receiving stem cells. Engrafted hESC previously shown to be benefi cial following a similar task, 1 and 4 months after irradiation, were not found to provide cognitive benefi ts at 8 months. Conclusions : Our fi ndings suggest that hNSC transplantation promotes the long-term recovery of the irradiated brain, where intrahippocampal stem cell grafting helps to preserve cognitive function.

Keywords: Human stem cells , transplantation , radiation , cognition , hippocampus
Stem cells restore cognition after irradiation 817 radiation-induced cognitive defi cits. In this study we sought to determine the longer-term benefi ts of human stem cell grafting in the irradiated brain. Th e present fi ndings suggest that grafted hNSC (but not hESC) provide cognitive benefi ts lasting as long as 8 months after IRR and transplantation surgery.

Materials and methods
Animals and cranial IRR procedure All animal procedures described are in accordance with the National Institutes of Health (NIH) and approved by the Institutional Animal Care and Use Committee (IACUC). Immunodefi cient athymic nude (ATN) rats (strain 0N01, Cr:NIH-rnu, X50 colony, NCI Frederick National Laboratory, Frederick, MD, USA) were maintained in sterile housing conditions (20 ° C Ϯ 1 ° C; 70% Ϯ 10% humidity; 12 h:12 h light and dark cycle), and had free access to sterilized diet and water. A total of 48 young (2 month old) ATN rats were divided in four experimental groups: 0 Gy (no IRR) receiving sham-operation surgery with vehicle (Cont-sham, n ϭ 12), 10 Gy (head-only IRR) receiving vehicle and sham-operation surgery (IRR-sham, n ϭ 12), 10 Gy irradiated receiving hESC grafting (IRR ϩ hESC, n ϭ 12) and 10 Gy irradiated receiving hNSC grafting (IRR ϩ hNSC, n ϭ 12). A subset of animals ( n ϭ 4) from each group was followed for 8 months post-transplantation for the long-term study. For the IRR procedures, animals were anesthetized, eyes and body were lead-shielded and were exposed to cranial γ -IRR (10 Gy) using a 137 Cs irradiator (J.L. Shepard, Mark I, San Fernando, CA, USA) at a dose rate of 2.07 Gy/min, as described in detail previously (Acharya et al. 2009(Acharya et al. , 2011. None of the animals received immune-suppression throughout the study.

Transplantation surgery
Th e use of hESC (H9, WiCell Research Institute, Inc., Madison, WI, USA) and hNSC (ENStem-A cell line, EMD Millipore, Billerica, MA, USA) was approved by the Institutional Human Stem Cell Research Oversight Committee (hSCRO). Th e hESC were cultured on a mitotically inactive mouse embryonic fi broblast (MEF) feeder layer (EMD-Millipore) while hNSC were maintained as a monolayer in T25 fl asks in neural expansion media (EMD-Millipore) as described previously (Acharya et al. 2009(Acharya et al. , 2011. For transplantation studies, hESC were used at passages 42 -49 while hNSC were used at passages 5 -9. Two days post-IRR, rats received bilateral, intrahippocamoal transplantation of human stem cells as described in detail previously (Acharya et al. 2009(Acharya et al. , 2011. A total of 4.0 ϫ 10 5 live hESC or hNSC were transplanted in 4 distinct hippocampal sites (1.0 ϫ 10 5 cells per site) per hemisphere using precise streotaxic coordinates (Acharya et al. 2009(Acharya et al. , 2011. Th erefore, a total of 8.0 ϫ 10 5 live human stem cells were transplanted per brain. Th e sham surgery groups ' Control ' (0 Gy) and ' IRR ' (10 Gy) received sterile vehicle (neural expansion media) at the same stereotaxic coordinates. Th e schematic of the research design is shown in Figure 1.

Novel place recognition task
Groups of rats were tested on a novel place recognition (NPR) task at 1 month or 8 months post-transplantation. All 1-month NPR data was previously published (Acharya et al. 2009(Acharya et al. , 2011 and adapted here for comparative purposes. Th e NPR task assesses spatial recognition memory that has been shown to rely on intact hippocampal function (Save et al. 1992, Mumby et al. 2002. We employed a standard protocol (Acharya et al. 2009 involving video recording and live tracking of animals using Noldus Ethovision XT system (v7.0, Noldus Information Technology, Inc., Leesburg, VA, USA). Briefl y, for this study, animals were habituated for two days in an open fi eld arena with two toy objects for a 20-min session for acclimatization. On the third day, a familiarization phase using identical plastic blocks (8 ϫ 3 ϫ 10 cm high) was administered for 5 min and rats were allowed to explore freely in the arena. Rats were then returned to holding cages for a 5-min retention interval. Following this delay, placement of an identical copy of one block was altered, while an identical copy of the other block was placed in the same spatial location as during the familiarization phase, and the rats were again allowed to explore the novel place stimuli freely for the duration of 3 min. Th e time gap between familiarization and test phase was 5 min, thus, referred to as the 5 min test phase . A positive score was counted when the nose of the rat was within 1 cm and pointed in the direction of the object. Time was not scored for rats that were near but not facing the object. Data collection and analysis for behavior studies were carried out blind to the observer.

Statistical analyses
For the behavioral analyses (NPR task), exploration ratio, or the proportion of total time spent exploring the novel spatial location (t novel / t novel ϩ t familiar ), was used as the main dependent measure for statistical analyses. Exploration ratio data were analyzed using univariate ANOVAs for the 5-min test phases. In all cases, we confi rmed that the data were normally distributed using the Kolmogorov-Smirnov test, and that the error variances did not diff er between groups by using Levene ' s test of equality of error variances. When a statistically signifi cant overall group eff ect was found, multiple comparisons were made using Fisher ' s protected least signifi cant diff erence (FPLSD) post hoc tests to compare the individual groups. Additional analyses of recognition memory were conducted using one-sample t -tests to determine whether the mean proportion of time spent exploring the novel spatial location for each group diff ered signifi cantly from chance (i.e., 0.5) and if diff erences existed between short-and long-term time-points. Th e statistical analyses were carried out by PASW statistics (v17.0, SPSS, IBM Inc., Armonk, NY, USA) and p -values less than 0.05 were considered statistically signifi cant.

Human stem cell transplantation improves cognition at 1-and 8-month post-grafting
We have previously shown that transplantation of hESC or hNSC improved cognition at 1-and 4-months post-IRR (Acharya et al. 2009(Acharya et al. , 2011). In the current study, we have extended these experiments to 8-month post-IRR timepoints and compared data for the restoration of cognition at 1 and 8 months following IRR. During the familiarization phase (Figure 2), IRR animals explored signifi cantly less  Exploration ratios (time novel/time novel ϩ time familiar) for the fi rst minute of the 5 min NPR test session are plotted. Irradiated animals (IRR) at 1 month did not explore the novel spatial location more than expected by chance (indicated by dashed line) and showed a signifi cant decline when compared to the non-irradiated Control group ( p ϭ 0.015).
While the cohort of 8-month IRR animals did exhibit a trend toward improved performance they did not spend signifi cantly more time exploring the novel place than expected by chance. Th e IRR ϩ hNSC group at 1-and 8-month post-surgery showed comparable cognitive performance with controls. At the 1-month time the IRR ϩ hESC group showed signifi cant improvement ( p ϭ 0.013) in novel place exploration compared to the IRR group. Error bars indicate the standard error of the mean Ϯ SEM ( n ϭ 12, 1-month data; n ϭ 4, 8-month data). 1-month data adapted from (Acharya et al. 2009(Acharya et al. , 2011. * indicates signifi cant diff erence between 8-month and 1-month IRR groups (i.e., p Ͻ 0.05 on Post hoc, Fisher LSD group comparisons). ϩ , indicates signifi cant diff erence compared to 0.5 (i.e., more than expected by chance, p ' s Ͻ 0.05, one-sample t -test comparison).
than hNSC transplanted animals (IRR ϩ hNSC) at both timepoints (1 month, p ϭ 0.001, n ϭ 12 and 8 months, p ϭ 0.006, n ϭ 4, Figure 2). For animals transplanted with hESC, cognitive benefi ts were not found, and at the 8-month time, hESC transplantation was not benefi cial (Figure 2). At both 1-and 8-month times, hNSC transplanted animals explored signifi cantly more than non-irradiated control animals (1 month, p ϭ 0.02, n ϭ 12, and 8 months, p ϭ 0.044, n ϭ 4, Figure 2). During the 5 min NPR test phase, the 1-month IRR animals spent a signifi cantly lower proportion of time exploring the novel place compared to all other groups ( p ' s Ͻ 0.001, Post hoc, Fisher ' s least square diff erence (LSD) test, Figure 3). Although not statistically diff erent compared to Control, IRR ϩ hESC and IRR ϩ hNSC cohorts, the 8-month IRR group did not spend signifi cantly more time exploring the novel place than expected by chance (dashed line, Figure 3). Transplantation of irradiated animals with hESC improved cognition signifi cantly 1 month after IRR, such that they explored the novel place signifi cantly more ( p ϭ 0.013, Post hoc, Fisher ' s LSD) compared to IRR animals. Th e cognitive benefi ts of engrafted hESC did not persist when animals were evaluated on the NPR task 8 months after IRR, where they were not found to be signifi cantly diff erent from sham-irradiated controls or the 8-month Stem cells restore cognition after irradiation 819 with memory consolidation (Guzowski et al. 2005, Ram í rez-Amaya et al. 2005. Th e long-term cognitive benefi ts of hNSC grafting may involve a facilitation of Arc expression in neurons of the host brain. Should the long term benefi ts of stem cell grafting in the irradiated brain be found more dependent on neurotrophic support from engrafted cells, rather than functional replacement and/or integration of cells into the damaged tissue bed, then this would be consistent with the preponderance of evidence using stem cell therapies in a wide range of animal-based injury models (Benderitter et al. 2014). While further experimentation will be required to address the underlying mechanisms, stem cell therapy in the irradiated brain may ultimately rely on the protection of existing host neuronal circuitry.
IRR-sham group (Figure 3). However, for animals transplanted with hNSC preference for the novel object was improved signifi cantly at both the 1-month and 8-month times after IRR (1 month IRR ϩ hNSC, p ϭ 0.015, n ϭ 12; and 8 months IRR ϩ hNSC, p ϭ 0.009, n ϭ 4, Figure 3). Th ese data demonstrate that engrafted hNSC provide superior cognitive benefi ts when compared to animals grafted with hESC.

Discussion
Th e present results demonstrate that intrahippocampal transplantation of human stem cells were able to provide certain cognitive benefi ts over protracted times after IRR and transplantation surgery. Clinical management of primary and metastatic brain tumors routinely involves cranial IRR treatment plans designed to forestall the advancement of malignant growth. While such therapies remain benefi cial, resultant cognitive dysfunction has the potential to severely compromise quality of life. Th e debilitating clinical side eff ects of cranial IRR have been problematic for decades, and to date, this remains an unmet medical need, with no satisfactory long-term solutions available to alleviate the persistent and progressive neurocognitive sequelae (Butler et al. 2006, Meyers andBrown 2006).
While it remains uncertain precisely how IRR impacts specifi c subpopulations of cells within the central nervous system (CNS) to disrupt cognition, the present results demonstrate the capability of hNSC transplantation to restore neuronal function at the behavioral level at a protracted treatment interval (i.e., 8 months). Past work from our laboratory has characterized the functional decrements in cognition caused by cranial IRR (Acharya et al. 2009(Acharya et al. , 2011 and demonstrated that intrahippocampal transplantation of human stem cells could ameliorate radiation-induced cognitive dysfunction at earlier times (i.e., Յ 4 months). In the present study we analyzed the functional consequences of grafted cells in the brain at an extended time (8 months) after IRR and transplantation surgery. While irradiated animals showed some recovery in cognition over 8 months, present evidence points to the fact that grafted hNSC were able to provide benefi cial eff ects over extended times following a single transplantation surgery (Figures 2 and 3).
Previous work from our group has defi ned the benefi cial eff ects of multiple stem cell types on cognition following IRR, and defi ned the yields and diff erentiated phenotypes associated with those eff ects at 1 and 4 months following treatment (Acharya et al. 2009(Acharya et al. , 2011. We have recently reported that following hNSC transplantation, grafted cells expressed the behaviorally-induced plasticity-related activity-regulated cytoskeleton-associated protein (Arc) and neuron-specifi c nuclear antigen (NeuN) when analyzed 1 month after transplantation, suggesting that grafted cells possess the capability to functionally integrate into hippocampal circuits (Acharya et al. 2011). Arc is known to play key roles in the neuronal mechanisms underlying long-term synaptic plasticity and memory, and provides a reliable marker for detecting neurons that are actively engaged in spatial and contextual information processing associated