- Email: [email protected]
Cyclophosphamide impairs hippocampus-dependent learning and memory in adult mice: Possible involvement of hippocampal neurogenesis in chemotherapy-induced memory deficits
Neurobiology of Learning and Memory 93 (2010) 487–494
Contents lists available at ScienceDirect
Neurobiology of Learning and Memory journal homepage: www.elsevier.com/locate/ynlme
Cyclophosphamide impairs hippocampus-dependent learning and memory in adult mice: Possible involvement of hippocampal neurogenesis in chemotherapy-induced memory deficits Miyoung Yang a,1, Joong-Sun Kim a,1, Myoung-Sub Song a, Sung-Ho Kim a, Seong Soo Kang a, Chun-Sik Bae a, Jong-Choon Kim a, Hongbing Wang b, Taekyun Shin c,*, Changjong Moon a,* a
College of Veterinary Medicine, Chonnam National University and Animal Medical Institute, Gwangju 500-757, South Korea Department of Physiology and Neuroscience Program, Michigan State University, East Lansing, MI 48823, USA c College of Veterinary Medicine and Veterinary Medical Research Institute, Jeju National University, Jeju 690-756, South Korea b
a r t i c l e
i n f o
Article history: Received 14 October 2009 Revised 15 December 2009 Accepted 20 January 2010 Available online 28 January 2010 Keywords: Cyclophosphamide Cognitive impairment Hippocampus Neurogenesis
a b s t r a c t Cyclophosphamide (CYP) is an anti-neoplastic agent as well as an immunosuppressive agent. In order to elucidate the alteration in adult hippocampal function following acute CYP treatment, hippocampus-related behavioral dysfunction and changes in adult hippocampal neurogenesis in CYP-treated (intraperitoneally, 40 mg/kg) mice (8–10-week-old ICR) were analyzed using hippocampus-dependent learning and memory tasks (passive avoidance and object recognition memory test) and immunohistochemical markers of neurogenesis (Ki-67 and doublecortin (DCX)). Compared to the vehicle-treated controls, mice trained at 12 h after CYP injection showed significant memory deficits in passive avoidance and the object recognition memory test. The number of Ki-67- and DCX-positive cells began to decrease significantly at 12 h postinjection, reaching the lowest level at 24 h after CYP injection; however, this reverted gradually to the vehicle-treated control level between 2 and 10 days. We suggest that the administration of a chemotherapeutic agent in adult mice interrupts hippocampal functions, including learning and memory, possibly through the suppression of hippocampal neurogenesis. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction Cyclophosphamide (CYP), a cytotoxic alkylating agent, is used commonly as an anti-neoplastic agent for the treatment of various cancers, as well as an immunosuppressive agent for organ transplantation, systemic lupus erythematosus, and other benign diseases (Dollery, 1999). Previously, the chemotherapeutic effect has been thought not to cross the blood–brain barrier (BBB) when given in standard, clinically relevant doses. However, despite its wide spectrum of clinical uses, CYP is known to cause several adverse effects, including cognitive impairment in humans (van Dam et al., 1998; Vardy & Tannock, 2007). Cognitive impairment occurs in a subset of cancer survivors and is generally subtle. Most evidence suggests an association with chemotherapy, although other factors related to the diagnosis and treatment of cancer may contribute (Vardy & Tannock, 2007). For example, some breast cancer survivors experience cognitive defects following chemotherapy (Ahles & Saykin, 2002; Bar-
* Corresponding authors. Fax: +82 62 530 2841 (C. Moon), +82 64 756 3354 (T. Shin). E-mail addresses: [email protected] (T. Shin), [email protected] (C. Moon). 1 The first two authors equally contributed to this work. 1074-7427/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.nlm.2010.01.006
ton & Loprinzi, 2002; Schagen, Muller, Boogerd, Mellenbergh, & van Dam, 2006). More patients treated with high-dose chemotherapy than patients treated with standard-dose chemotherapy show a defect in cognitive performance compared to healthy control subjects (Ahles et al., 2002; van Dam et al., 1998). Although several mechanisms have been suggested to explain the cognitive impairment associated with chemotherapy (Barton & Loprinzi, 2002; Jansen, Miaskowski, Dodd, Dowling, & Kramer, 2005; Mirkes, 1985), the precise mechanism(s) are poorly understood. To understand the pathogenesis of chemotherapy-induced cognitive impairment and to develop new therapies and prophylaxis for the brain dysfunction, it may be very important to examine the cognitive dysfunction in detail from a behavioral model. Two previous studies have found that CYP treatment induces neurobehavioral changes in rodents in the form of acute impairment (Reiriz et al., 2006) and chronic improvement (Lee et al., 2006) of learning and memory ability. Here, we suggest a possible mechanism for the memory impairment following chemotherapy, in which CYP may inhibit adult hippocampal neurogenesis, which is involved in learning and memory processing (Kim et al., 2008; Raber et al., 2004; Winocur, Wojtowicz, Sekeres, Snyder, & Wang, 2006). A broad consensus holds that the presence of progenitor cells gives rise to new neural cells in the adult brain in various species.
488
M. Yang et al. / Neurobiology of Learning and Memory 93 (2010) 487–494
The adult brain retains at least two active germinal zones: the subgranular zone (SGZ) in the dentate gyrus (DG), which generates new granular cells in the adult hippocampus, and the forebrain subventricular zone (SVZ), which gives rise to granular cells in the olfactory bulb (Cameron, Woolley, McEwen, & Gould, 1993; Temple & Alvarez-Buylla, 1999). Progenitor neural cells in the DG of the adult hippocampus are particularly vulnerable to ionizing radiation (Peissner, Kocher, Treuer, & Gillardon, 1999), which can cause hippocampus-dependent learning and memory impairment (Kim et al., 2008; Winocur et al., 2006). However, little is known about the effect of chemotherapeutic agents on neurogenesis in the DG of the adult hippocampus. In this study, we examined the behavioral alteration of adult mice after CYP injection using hippocampus-dependent learning paradigms including passive avoidance and object recognition memory test. In addition, the effect of cancer chemotherapy on hippocampal neurogenesis was investigated by examining the changes in the expression levels of the proliferating cell marker Ki-67 and the immature neuronal cell marker doublecortin (DCX) in the DG of the hippocampi of adult mice after CYP treatment. 2. Materials and methods 2.1. Animals Male ICR mice (8–10 weeks old) were obtained from a specific pathogen-free colony at Oriental, Inc. (Seoul, Korea). The animals were housed in a room maintained at 23 ± 2 °C, 50 ± 5% relative humidity, artificial lighting from 08:00 to 20:00 h, and 13–18 air changes per hour. The animals were provided with tap water and commercial rodent chow (Samyang Feed, Seoul, Korea) ad libitum. The Institutional Animal Care and Use Committee at Chonnam National University approved the protocols used in this study. 2.2. Drug treatment CYP (Sigma–Aldrich, St. Louis, MO) was dissolved in sterile 0.9% saline. Mice were injected intraperitoneally (i.p.) with 40 mg/kg of CYP. The vehicle group was injected i.p. with 0.9% saline. 2.3. Behavioral analysis Behavioral dysfunction in mice who received CYP (i.p., 40 mg/ kg) was measured by open-field analysis (n = 7 mice/group), analyzing their sensitivity to electric foot-shock (n = 4 mice/group), passive avoidance (n = 9 mice/group), and object recognition memory test (n = 9 mice/group) at 12 h and 10 days after injection. 2.3.1. Open-field test Open-field analysis was used to measure the activity of vehicle (saline)-treated and CYP-treated mice in a novel environment. Parameters including total moving distance, ambulatory movement time, ambulatory movement count, and resting time were determined using an Activity Monitor (MED Associates, St. Albans, VT). 2.3.2. Passive avoidance The passive-avoidance paradigm was used to examine hippocampus-dependent associative memory (Kim et al., 2008; Zhang et al., 2008), in which the animals learned to associate the aversive unconditioned stimulus (mild electric foot-shock) with the conditioned stimulus (the dark chamber). During training, a mouse was introduced to the lighted half of the training chamber (Ugo Basile, Comerio, Italy), and allowed 1 min to explore the area before the trap door was opened. Immediately after the mouse entered
the darkened half, the trap door was closed, and a mild foot-shock (0.5 mA for 1 s) was delivered. The trained mouse remained in the dark chamber for 20 s after the shock, and it was then returned to the home cage. When tested 24 h after training, the trained mouse was re-introduced to the lighted chamber. The time spent in the lighted half before entering the darkened half was scored as the crossover latency and used as an index for memory formation. A cut-off time of 500 s was chosen for the crossover latency. The mice were removed manually from the lighted chamber when the cut-off value was reached. The sensitivity to the electric foot-shock was further tested in the mice. The electric shock threshold (mA) to elicit stereotypic responses (flinch, vocalization, and jump/vocalization) was measured. 2.3.3. Object recognition memory test Another hippocampus-dependent learning paradigm was used (Kim et al., 2008; Zhang et al., 2008). The mice were first habituated in the training/testing acryl chamber (42 cm L, 28 cm W, 20 cm H) for 1 week (15 min/day) before training. The 3.5-cm tall plastic objects for discriminative recognition were cubes, pyramids and cylinders, and they could not be displaced by the mice. The chamber arena and objects were cleaned with 75% ethanol between trials to prevent the build-up of olfactory cues. During training, two differently shaped objects were selected randomly and presented to each mouse for 15 min. At 24 h after training, another set of objects (one old object and one novel object) was presented to the trained mice. If, for example, the cube- and pyramid-shaped objects were presented during training, the cylinder-shaped object would be used as a novel object during testing. The interactions of the mouse with each object, including approaches and sniffing, were scored. If the mouse had memory retention for an old object, it should show preference for the novel object during testing. The percentage of preference was defined as the number of interactions with a specific object divided by the total number of interactions with both objects. 2.4. Tissue sampling The time-dependent effects of CYP on neurogenesis in the adult mouse hippocampus were observed after mice received an intraperitoneal injection of CYP (40 mg/kg). The mice were sacrificed, and the hippocampi were then dissected from each group at 12 h, 24 h, 4 days, and 10 days (n = 6 mice/group) after the injection of CYP. The samples were processed for embedding in paraffin wax after fixation in 4% paraformaldehyde in phosphate-buffered saline (PBS, pH 7.4) and stored at 70 °C for biochemical analysis. 2.5. Terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end-labeling (TUNEL) The level of DNA fragmentation was detected by in situ nick end-labeling (terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end-labeling, TUNEL), which was performed according to the manufacturer’s instructions (ApopTagÒ in situ apoptosis detection kit; Millipore, Billerica, MA). 2.6. Immunohistochemistry Coronal sections (thickness, 5 lm) were cut by microtome and deparaffinized using routine protocols before being exposed to citrate buffer (0.01 M, pH 6.0) and heated in an autoclave for 10 min. All subsequent steps were performed at room temperature. The sections were treated with 0.3% hydrogen peroxide for 20 min to block endogenous peroxidase activity. After three washes in phosphate-buffered saline (PBS), the sections were
M. Yang et al. / Neurobiology of Learning and Memory 93 (2010) 487–494
blocked with 10% normal goat serum (Vector ABC Elite kit, Burlingame, CA), diluted in PBS for 1 h, and then allowed to react with the immunohistochemical markers for proliferating cells and immature progenitor cells, including monoclonal rabbit anti-Ki-67 (DRM004, 1:200 dilution; Acris Antibodies GmbH, Hiddenhausen, Germany) and polyclonal rabbit anti-DCX (1:400 dilution; Cell Signaling Technology, Beverly, MA) antibody, for 2 h. After three washes in PBS, the sections were exposed to biotinylated goat anti-rabbit IgG (1:100 dilution; Vector) for 45 min. After three washes in PBS, the sections were incubated with the avidin–biotin peroxidase complex (Vector ABC Elite kit) for 45 min. After three washes in PBS, the peroxidase reaction was developed for 3 min using a diaminobenzidine (DAB) substrate (SK-4100; Vector) prepared according to manufacturer’s instructions. As a control, the primary antibodies were omitted for a few test sections in each experiment. After the completion of color development, the sections were counterstained with Harris’s hematoxylin for 5 s, washed in running tap water for 20 min, dehydrated through a graded ethanol series, cleared with xylene, and mounted with Canada balsam (Sigma–Aldrich).
489
tions representing the rostral/mid-hippocampus was used. For each mouse, three non-overlapping sections were analyzed, one from each of the three regions of the hippocampus (approximately 50 lm apart). All positively-immunolabeled cells within the SGZ of the supra- and infra-pyrimidal blades of the DG were quantified. The number of immunopositive cells was determined from the values obtained from each DG in the three brain sections. The mean number of immunopositive cells in the three sections of each mouse was taken as n = 1. The number of immunopositive cells was expressed as the mean ± SEM for each group (n = 6). 2.8. Statistical analysis The data are reported as the mean ± SEM and were analyzed using one-way analysis of variance (ANOVA) followed by a Student–Newman–Keuls post hoc test for multiple comparisons. In all cases, a p-value < 0.05 was considered to be significant. 3. Results
2.7. Cell counting
3.1. CYP-treated mice exhibit normal locomotor activity
The number of cells displaying the specific characteristics of proliferating cells (immunopositive for Ki-67) and immature progenitor cells (immunopositive for DCX) in the hippocampus was scored by an observer blinded to the identity of the sample using a histomorphometric approach (Kim et al., 2009). The brain from each mouse was sampled at approximately 2.12 mm behind the bregma. A standardized counting area containing 5-lm-thick coronal sections in a one-in-ten series of sec-
First, we examined the basic locomotor activity of mice at 12 h and 10 days after CYP treatment (40 mg/kg) in a novel environment by open-field analysis. The analysis quantified the overall locomotor activity that could reflect motivation and performance of the mice. Vehicle-treated (control) and CYP-treated mice showed comparable moving distance, ambulatory movement time, ambulatory movement count, and resting time (Fig. 1), indicating normal movement of CYP-treated mice.
Fig. 1. CYP-treated mice show normal locomotor activity in open-field analysis. The data for vehicle-treated controls (12 h after saline injection, n = 7), and CYP-treated (40 mg/kg) mice (12 h, n = 7 and 10 days, n = 7 after treatment) were monitored. (A) Vehicle- and CYP-treated mice showed similar movement distance (p = 0.962 [12 h] and p = 0.941 [10 days] vs. vehicle-treated controls). (B) Vehicle- and CYP-treated mice showed similar ambulatory movement time (p = 0.796 [12 h] and p = 0.871 [10 days] vs. vehicle-treated controls). (C) Vehicle- and CYP-treated mice showed similar ambulatory movement count (p = 0.991 [12 h] and p = 0.930 [10 days] vs. vehicle-treated controls). (D) Vehicle- and CYP-treated mice showed similar resting time (p = 0.535 [12 h] and p = 0.568 [10 days] vs. vehicle-treated controls).
490
M. Yang et al. / Neurobiology of Learning and Memory 93 (2010) 487–494
3.2. CYP-treated mice are impaired for passive avoidance Compared to vehicle-treated control mice, the mice tested 12 h after CYP treatment showed a significant decrease in crossover latency when trained for passive avoidance. All of the mice subjected to this task stepped through the door into the dark compartment in a short period in the acquisition trial (33.31 ± 3.44 s for vehicle controls, 43.98 ± 8.23 s for mice at 12 h after CYP treatment, and 44.47 ± 6.76 s for mice after 10 days CYP treatment), when they received the electric foot-shock. In the memory retention trials carried out 24 h after the acquisition trial, the memory retention time for the vehicle control mice was significantly longer than that of the mice 12 h after CYP treatment (265.7 ± 14.86 s for vehicle controls and 174.09 ± 32.9 s for mice 12 h after CYP treatment, p = 0.022). However, in the trials performed 10 days after CYP treatment, no significant difference in the memory retention times was found in CYP-treated mice compared to vehicle controls (247.64 ± 21.83 s for mice 10 days after CYP treatment, p = 0.328; Fig. 2A). Their sensitivity to the electric foot-shock was tested further (n = 4 for each group), and no significant difference was ob-
served in the threshold current to elicit stereotypic responses including flinch, vocalization and jump/vocalization, between the vehicle controls and CYP-treated mice (Fig. 2B). This suggests that they had comparable sensitivity to the electric foot-shock. 3.3. CYP-treated mice are impaired for object recognition memory We further examined CYP-treated (40 mg/kg) mice by another sensitive hippocampus-dependent paradigm, object recognition memory (Kim et al., 2008). During the 15 min training session, vehicle- and CYP-treated mice showed no significant difference in their interaction with the objects, indicating normal motivation and exploratory activity (Fig. 3A). Vehicle- and CYP-treated mice also displayed an equal preference for the two objects during training (Fig. 3B). When tested 24 h after training, the results indicated that mice 12 h after CYP treatment exhibited impairment of memory, whereas mice 10 days after CYP treatment did not show a deficit in memory (Fig. 3C). Statistical analysis comparing the recognition indexes obtained in training and retention sessions within groups indicated that the mice tested 12 h after CYP treat-
Fig. 2. CYP transiently inhibits memory retention in a passive-avoidance task. (A) Mice (n = 9) trained 12 h after CYP treatment (40 mg/kg) showed lower cross-over latency at 24 h testing after training (p < 0.05, compared to vehicle-treated controls [12 h after saline injection]; n = 9). However, mice (n = 9) trained 10 days after CYP treatment (40 mg/kg) showed normal memory retention for passive avoidance. (B) The electric shock thresholds (mA) to elicit stereotypic responses (flinch, vocalization, and jump/ vocalization) were measured for controls (1 day after saline injection) and mice 12 h and 10 days after CYP injection (n = 4 for each group). The threshold current to elicit stereotypic responses was similar to that of the vehicle controls at both time points following CYP injection. Data are reported as the mean ± SEM.
Fig. 3. CYP transiently decreases object recognition memory. (A) Vehicle- and CYP-treated (40 mg/kg) mice showed comparable interactions with the two objects during training (p = 0.558, 12 h after CYP treatment, n = 9; p = 0.283, 10 days after CYP treatment, n = 9 vs. vehicle-treated controls [12 h after saline injection], n = 9). (B) Vehicle- and CYP-treated mice showed equal preference for the two objects during training. (C) Vehicle-treated controls and mice trained 10 days after CYP treatment showed significant preference for the novel object during testing (24 h after training), but no significant preference for the novel object was observed in mice trained 12 h after CYP treatment (p = 0.146). A significant difference in novel object preference was found between vehicle-treated controls and mice trained 12 h after CYP treatment (p < 0.05).
M. Yang et al. / Neurobiology of Learning and Memory 93 (2010) 487–494
491
Fig. 4. TUNEL reaction in the DG of adult hippocampi. TUNEL-positive apoptotic nuclei were detected rarely in the DG region of the hippocampus with (A) vehicle (12 h after saline injection) and (B) CYP treatment (12 h after CYP injection). The cells were counterstained with hematoxylin. Scale bars, 50 lm.
ment had a complete memory blockade, as the animals showed no significant preference toward the novel object in the testing session (p = 0.146). No significant difference among groups was found in the total time spent exploring both objects during the training trial; the mean ± SE preference to the novel object was 67.68 ± 2.69% in the vehicle control, 55.46 ± 5.05% in the mice 12 h after CYP treatment, and 64.13 ± 3.28% in the mice 10 days after CYP treatment. 3.4. CYP does not induce apoptosis in the hippocampal DG of adult mice Hematoxylin and eosin (H&E) staining did not reveal any unusual hippocampal structure in adult mice 12 h–10 days after CYP (40 mg/kg) treatment (data not shown). Also, in both the vehicleand CYP-treated mice, TUNEL-positive apoptotic cells were quite rare in the DGs of hippocampi (Fig. 4). This suggests that the acute administration of 40 mg/kg CYP does not induce neural apoptosis in the hippocampus of adult mice. 3.5. CYP transiently decreases Ki-67 and DCX expression in the hippocampal DG of adult mice The levels of Ki-67 and DCX in the hippocampus were evaluated semiquantitatively after the administration of CYP using immunohistochemical markers of neurogenesis. Ki-67- (Fig. 5A and B) and DCX-positive cells (Fig. 6A and B) were observed consistently in the DGs of hippocampi in adult control mice. Between 0 and 24 h after CYP administration, the number of Ki-67-positive cells in the DG declined sharply (2.5 ± 0.62 cells/DG, n = 6, p < 0.0001 vs. vehicle control; Fig. 5C and D); however, it increased gradually to the vehicle control level (8.16 ± 0.4 cells/DG, n = 6) between 2 and 10 days (3.8 ± 0.31 cells/DG, p < 0.0001 vs. vehicle control [2 days]; 6.2 ± 0.60 cells/DG, p < 0.05 vs. vehicle control [4 days]; 9.7 ± 0.99 cells/DG, p = 0.190 vs. vehicle control [10 days]; n = 6 per group) after CYP treatment (Fig. 5E). The number of DCX-positive cells in the DG declined markedly (20.33 ± 1.61 cells/DG, n = 6, p < 0.0001 vs. vehicle control; Fig. 6C and D) between 0 and 24 h after CYP treatment, and then returned gradually to the vehicle control level (43.5 ± 2.32 cells/DG, n = 6) between 2 and 10 days (25.33 ± 3.55 cells/DG, p < 0.01 vs. vehicle control [2 days]; 30.67 ± 3.59 cells/DG, p < 0.05 vs. vehicle control [4 days]; 36.83 ± 3.71 cells/DG, p = 0.159 vs. vehicle control [10 days]; n = 6 per group) after CYP treatment (Fig. 6E). These data suggest that, in the short-term, the decline of neurogenesis in the DGs of adult mice with CYP treatment (40 mg/kg i.p.) is reversible. 4. Discussion Several clinical studies have reported cognitive impairment to be a side effect of cancer chemotherapy in cancer patients (Ahles
& Saykin, 2002; Parth, Dunlap, Kennedy, Ordy, & Lane, 1989; van Dam et al., 1998; Waber, Tarbell, Kahn, Gelber, & Sallan, 1992). A number of factors may be protective against cognitive impairments or place individuals at a higher risk for impairments in cognitive function. These factors include concomitant effects of cancer and its treatment (e.g., medications, fatigue, depression, or anxiety), indirect and direct effects of chemotherapy (e.g., chemotherapy-induced anemia or menopause), and individual patient factors (e.g., age, intelligence level, educational level or menopausal status) (Jansen et al., 2005). The existence of CYP-induced cognitive impairment has become generally accepted (Shilling, Jenkins, & Trapala, 2006), although many details of the concept remain controversial or obscure. The reported rates of cognitively impaired patients vary widely, reflecting different definitions of impairment (Shilling et al., 2006). Data regarding the affected cognitive domains and duration of cognitive impairment are far from congruous (Ahles et al., 2002; Fan et al., 2005; Schagen et al., 2002). A previous study in a mouse model reported that the single administration of 40–200 mg/kg CYP induces cognitive deficit in step-down inhibitory avoidance conditioning (Reiriz et al., 2006). This suggests that an injected CYP dose of 40 mg/kg is enough to interrupt brain cognition in mice. Likewise, in the current study, this dose of CYP induced hippocampusdependent learning and memory impairments without alterations in locomotor activity. Furthermore, mice trained at 12 h after CYP treatment showed memory deficit for two hippocampus-dependent learning paradigms, passive avoidance and the object recognition memory test. Conversely, mice trained at 10 days after the treatment did not show any memory defect. These results are similar to those of a previous study reporting that CYP-induced cognitive deficits at 24 h, but the impairment had recovered 1 week after CYP treatment in another hippocampus-dependent memory task, step-down inhibitory avoidance conditioning (Reiriz et al., 2006). This suggests that a single administration of 40 mg/kg CYP transiently induces cognitive impairments in this behavioral model. A consensus holds that the hippocampus is essential to the process of forming and recovering certain types of memory. In the adult brain, the progenitor/stem cells generate new neural cells in the DG of the hippocampus (Cameron et al., 1993; Temple & Alvarez-Buylla, 1999). Hence, adult-generated hippocampal cells are involved in the learning and memory functions of this structure (Squire, Stark, & Clark, 2004; Winocur et al., 2006). The inhibition of the hippocampal neurogenesis with ionizing radiation causes cognitive impairment in adult mice (Raber et al., 2004). Furthermore, the exposure to relatively low-dose radiation induces transient hippocampus-dependent learning and memory deficits in mice (Kim et al., 2008). Therefore, the pattern of memory ability may be associated with the alteration of hippocampal neurogenesis. However, further studies are needed to establish an unequivocal link between the change of hippocampal neurogenesis and the
492
M. Yang et al. / Neurobiology of Learning and Memory 93 (2010) 487–494
Fig. 5. CYP transiently decreases proliferating cells (Ki-67-positive cells) in the DG of adult hippocampi. (A–D) Representative images showing the Ki-67-positive cells in the DGs of adult hippocampi taken from the vehicle controls (24 h after saline injection) and mice 24 h after CYP injection. (E) The number of Ki-67-positive cells in the DG decreased markedly 0–24 h after CYP treatment (40 mg/kg i.p.). The cells were counterstained with hematoxylin (A–D). GCL, granular cell layer; SGZ, subgranular zone. In A and C, scale bars represent 150 lm; in B and D, scale bars represent 20 lm. Data are reported as the mean ± SEM (for E). p < 0.05 vs. vehicle controls (n = 6 per group).
hippocampus-dependent memory impairment caused by chemotherapy. Barton and Loprinzi (2002) have suggested that cognitive impairment due to chemotherapy might result from indirect chemical toxicity and oxidative damage, direct injury to the neurons, inflammation, or a type of autoimmune response. CYP is a cytotoxic drug that can alkylate DNA and prevent the duplication of the genome in dividing cells. As shown in a previous study, CYP arrests the S-phase of the cell cycle and induces apoptosis in embryonic neural progenitor cells of the telecephalon 6–12 h after administration (Ueno, Katayama, Yamauchi, Nakayama, & Doi, 2006). This study demonstrated that CYP did not induce apoptosis in the adult DG, in contrast to the induction of apoptosis in adult neurogenesis by irradiation reported in a previous study (Kim et al., 2008). However, during the period from 12 h to 4 days after injection, CYP transiently decreased the cells positive for Ki-67 (a proliferating cell marker) and DCX (an immature neuronal cell marker) in the DG of adult hippocampi, indicating that CYP transiently inhibits adult hippocampal neurogenesis. This suggests that CYP only suppresses the generation of new neural cells in the SGZ, possibly by transient arrest of the cell cycle, in the DG of the adult hippocampus. In addition, we suggest another possible mechanism for the inhibition of hippocampal neurogenesis caused by CYP treatment.
The toxic effects resulting from metabolites of CYP, such as acrolein or phosphoramide mustard (Mirkes, 1985), reduce cellular resistance to oxidative stress, which can damage the BBB, thus allowing the entry of possible neurotoxic molecules (i.e., cytokines) to the brain (Subramaniam, Subramaniam, & Shyamala, 1994). Chemotherapy-induced pro-inflammatory cytokines release may incur cognitive changes (Ahles et al., 2002; Jansen et al., 2005). Therefore, the inhibition of hippocampal neurogenesis following chemotherapy may be caused by BBB damage and the entry of neurotoxic molecules including pro-inflammatory cytokines. However, regarding the dose of CYP that induced memory facilitation in the present study, it is unclear what plasma levels of CYP had been reached with those doses. Further research should clarify the molecular and cellular mechanisms for the in vivo inhibitory effect of CYP on the generation of progenitor/stem cells in the adult DG. In conclusion, the pattern of hippocampus-dependent memory dysfunction is consistent with the change in neurogenesis after CYP administration, suggesting that the impairment of memory retention might be associated with the change in neural progenitor/stem cell marker proteins in the hippocampus in adult mice. Moreover, this study shows that a single injection of CYP in adult ICR mice can interrupt the functioning of the hippocampus, including learning and memory, possibly through the inhibition of neurogenesis.
M. Yang et al. / Neurobiology of Learning and Memory 93 (2010) 487–494
493
Fig. 6. CYP transiently decreases immature neurons (DCX-positive cells) in the DG of adult hippocampi. (A–D) Representative images showing the DCX-positive cells in the DGs of adult hippocampi taken from the vehicle controls (24 h after saline injection) and mice 24 h after CYP injection. (E) The number of DCX-positive cells in the DG decreased markedly 0–24 h after CYP treatment (40 mg/kg i.p.). The cells were counterstained with hematoxylin (A–D). GCL, granular cell layer; SGZ, subgranular zone. In A and C, scale bars represent 150 lm; in B and D, scale bars represent 20 lm. Data are reported as the mean ± SEM (for E). p < 0.05 vs. vehicle controls (n = 6 per group).
Acknowledgment This work was supported by National Research Foundation of Korea Grant funded by the Korean Government (2009-0068372). References Ahles, T., & Saykin, A. (2002). Breast cancer chemotherapy-related cognitive dysfunction. European Journal of Cancer Care, 3, 84–90. Ahles, T. A., Saykin, A. J., Furstenberg, C. T., Cole, B., Mott, L. A., Skalla, K., et al. (2002). Neuropsychologic impact of standard-dose systemic chemotherapy in longterm survivors of breast cancer and lymphoma. Journal of Clinical Oncology, 20, 485–493. Barton, D., & Loprinzi, C. (2002). Novel approaches to preventing chemotherapyinduced cognitive dysfunction in breast cancer: The art of the possible. Clinical Breast Cancer, 3, 121–127. Cameron, H. A., Woolley, C. S., McEwen, B. S., & Gould, E. (1993). Differentiation of newly born neurons and glia in the dentate gyrus of the adult rat. Neuroscience, 56, 337–344. Dollery, C. (1999). Cyclophosphamide. In Therapeutic drugs (pp. C349–C354). Edinburgh: Churchill Livingstone. Fan, H. G., Houédé-Tchen, N., Yi, Q. L., Chemerynsky, I., Downie, F. P., Sabate, K., et al. (2005). Fatigue, menopausal symptoms, and cognitive function in women after adjuvant chemotherapy for breast cancer: 1- and 2-year follow-up of a prospective controlled study. Journal of Clinical Oncology, 23, 8025–8032. Jansen, C., Miaskowski, C., Dodd, M., Dowling, G., & Kramer, J. (2005). Potential mechanisms for chemotherapy-induced impairments in cognitive function. Oncology Nursing Forum, 32, 1151–1163. Kim, J. S., Jung, J., Lee, H. J., Kim, J. C., Wang, H., Kim, S. H., et al. (2009). Differences in immunoreactivities of Ki-67 and doublecortin in the adult hippocampus in three strains of mice. Acta Histochemica, 111, 150–156.
Kim, J. S., Lee, H. J., Kim, J. C., Kang, S. S., Bae, C. S., Shin, T., et al. (2008). Transient impairment of hippocampus-dependent learning and memory in relatively low-dose of acute radiation syndrome is associated with inhibition of hippocampal neurogenesis. Journal of Radiation Research, 49, 517–526. Lee, G. D., Longo, D. L., Wang, Y., Rifkind, J. M., Abdul-Raman, L., Mamczarz, J. A., et al. (2006). Transient improvement in cognitive function and synaptic plasticity in rats following cancer chemotherapy. Clinical Cancer Research, 12, 198–205. Mirkes, P. E. (1985). Cyclophosphamide teratogenesis: A review. Teratogenesis, Carcinogenesis and Mutagenesis, 5, 75–88. Parth, P., Dunlap, W. P., Kennedy, R. S., Ordy, J. M., & Lane, N. E. (1989). Motor and cognitive testing of bone marrow transplant patients after chemoradiotherapy. Perceptual and Motor Skills, 68, 1227–1241. Peissner, W., Kocher, M., Treuer, H., & Gillardon, F. (1999). Ionizing radiationinduced apoptosis of proliferating stem cells in the dentate gyrus of the adult rat hippocampus. Brain Research Molecular Brain Research, 71, 61–68. Raber, J., Rola, R., LeFevour, A., Morhardt, D., Curley, J., Mizumatsu, S., et al. (2004). Radiation-induced cognitive impairments are associated with changes in indicators of hippocampal neurogenesis. Radiation Research, 162, 39–47. Reiriz, A. B., Reolon, G. K., Preissler, T., Rosado, J. O., Henriques, J. A., Roesler, R., et al. (2006). Cancer chemotherapy and cognitive function in rodent models: Memory impairment induced by cyclophosphamide in mice. Clinical Cancer Research, 12, 5000. Schagen, S. B., Muller, M. J., Boogerd, W., Mellenbergh, G. J., & van Dam, F. S. (2006). Change in cognitive function after chemotherapy: A prospective longitudinal study in breast cancer patients. Journal of the National Cancer Institute, 98, 1742–1745. Schagen, S. B., Muller, M. J., Boogerd, W., Rosenbrand, R. M., van Rhijn, D., Rodenhuis, S., et al. (2002). Late effects of adjuvant chemotherapy on cognitive function: A follow-up study in breast cancer patients. Annals of Oncology, 13, 1387–1397. Shilling, V., Jenkins, V., & Trapala, I. S. (2006). The (mis)classification of chemo-fog– Methodological inconsistencies in the investigation of cognitive impairment after chemotherapy. Breast Cancer Research Treatment, 95, 125–129.
494
M. Yang et al. / Neurobiology of Learning and Memory 93 (2010) 487–494
Squire, L. R., Stark, C. E., & Clark, R. E. (2004). The medial temporal lobe. Annual Reviews of Neuroscience, 27, 279–306. Subramaniam, S., Subramaniam, S., & Shyamala, D. C. (1994). Erythrocyte antioxidant enzyme activity in CMF treated breast cancer patients. Cancer Biochemistry Biophysics, 14, 177–182. Temple, S., & Alvarez-Buylla, A. (1999). Stem cells in the adult mammalian central nervous system. Current Opinion in Neurobiology, 9, 135–141. Ueno, M., Katayama, K., Yamauchi, H., Nakayama, H., & Doi, K. (2006). Cell cycle progression is required for nuclear migration of neural progenitor cells. Brain Research, 1088, 57–67. van Dam, F. S., Schagen, S. B., Muller, M. J., Boogerd, W., vd Wall, E., Droogleever Fortuyn, M. E., et al. (1998). Impairment of cognitive function in women receiving adjuvant treatment for high-risk breast cancer: High-dose versus standard-dose chemotherapy. Journal of the National Cancer Institute, 90, 210–218.
Vardy, J., & Tannock, I. (2007). Cognitive function after chemotherapy in adults with solid tumours. Critical Reviews in Oncology/Hematology, 63, 183–202. Waber, D. P., Tarbell, N. J., Kahn, C. M., Gelber, R. D., & Sallan, S. E. (1992). The relationship of sex and treatment modality to neuropsychologic outcome in childhood acute lymphoblastic leukemia. Journal of Clinical Oncology, 10, 810–817. Winocur, G., Wojtowicz, J. M., Sekeres, M., Snyder, J. S., & Wang, S. (2006). Inhibition of neurogenesis interferes with hippocampus-dependent memory function. Hippocampus, 16, 296–304. Zhang, M., Moon, C., Chan, G. C., Yang, L., Zheng, F., Conti, A. C., et al. (2008). Castimulated type 8 adenylyl cyclase is required for rapid acquisition of novel spatial information and for working/episodic-like memory. The Journal of Neuroscience, 28, 4736–4744.
Contents lists available at ScienceDirect
Neurobiology of Learning and Memory journal homepage: www.elsevier.com/locate/ynlme
Cyclophosphamide impairs hippocampus-dependent learning and memory in adult mice: Possible involvement of hippocampal neurogenesis in chemotherapy-induced memory deficits Miyoung Yang a,1, Joong-Sun Kim a,1, Myoung-Sub Song a, Sung-Ho Kim a, Seong Soo Kang a, Chun-Sik Bae a, Jong-Choon Kim a, Hongbing Wang b, Taekyun Shin c,*, Changjong Moon a,* a
College of Veterinary Medicine, Chonnam National University and Animal Medical Institute, Gwangju 500-757, South Korea Department of Physiology and Neuroscience Program, Michigan State University, East Lansing, MI 48823, USA c College of Veterinary Medicine and Veterinary Medical Research Institute, Jeju National University, Jeju 690-756, South Korea b
a r t i c l e
i n f o
Article history: Received 14 October 2009 Revised 15 December 2009 Accepted 20 January 2010 Available online 28 January 2010 Keywords: Cyclophosphamide Cognitive impairment Hippocampus Neurogenesis
a b s t r a c t Cyclophosphamide (CYP) is an anti-neoplastic agent as well as an immunosuppressive agent. In order to elucidate the alteration in adult hippocampal function following acute CYP treatment, hippocampus-related behavioral dysfunction and changes in adult hippocampal neurogenesis in CYP-treated (intraperitoneally, 40 mg/kg) mice (8–10-week-old ICR) were analyzed using hippocampus-dependent learning and memory tasks (passive avoidance and object recognition memory test) and immunohistochemical markers of neurogenesis (Ki-67 and doublecortin (DCX)). Compared to the vehicle-treated controls, mice trained at 12 h after CYP injection showed significant memory deficits in passive avoidance and the object recognition memory test. The number of Ki-67- and DCX-positive cells began to decrease significantly at 12 h postinjection, reaching the lowest level at 24 h after CYP injection; however, this reverted gradually to the vehicle-treated control level between 2 and 10 days. We suggest that the administration of a chemotherapeutic agent in adult mice interrupts hippocampal functions, including learning and memory, possibly through the suppression of hippocampal neurogenesis. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction Cyclophosphamide (CYP), a cytotoxic alkylating agent, is used commonly as an anti-neoplastic agent for the treatment of various cancers, as well as an immunosuppressive agent for organ transplantation, systemic lupus erythematosus, and other benign diseases (Dollery, 1999). Previously, the chemotherapeutic effect has been thought not to cross the blood–brain barrier (BBB) when given in standard, clinically relevant doses. However, despite its wide spectrum of clinical uses, CYP is known to cause several adverse effects, including cognitive impairment in humans (van Dam et al., 1998; Vardy & Tannock, 2007). Cognitive impairment occurs in a subset of cancer survivors and is generally subtle. Most evidence suggests an association with chemotherapy, although other factors related to the diagnosis and treatment of cancer may contribute (Vardy & Tannock, 2007). For example, some breast cancer survivors experience cognitive defects following chemotherapy (Ahles & Saykin, 2002; Bar-
* Corresponding authors. Fax: +82 62 530 2841 (C. Moon), +82 64 756 3354 (T. Shin). E-mail addresses: [email protected] (T. Shin), [email protected] (C. Moon). 1 The first two authors equally contributed to this work. 1074-7427/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.nlm.2010.01.006
ton & Loprinzi, 2002; Schagen, Muller, Boogerd, Mellenbergh, & van Dam, 2006). More patients treated with high-dose chemotherapy than patients treated with standard-dose chemotherapy show a defect in cognitive performance compared to healthy control subjects (Ahles et al., 2002; van Dam et al., 1998). Although several mechanisms have been suggested to explain the cognitive impairment associated with chemotherapy (Barton & Loprinzi, 2002; Jansen, Miaskowski, Dodd, Dowling, & Kramer, 2005; Mirkes, 1985), the precise mechanism(s) are poorly understood. To understand the pathogenesis of chemotherapy-induced cognitive impairment and to develop new therapies and prophylaxis for the brain dysfunction, it may be very important to examine the cognitive dysfunction in detail from a behavioral model. Two previous studies have found that CYP treatment induces neurobehavioral changes in rodents in the form of acute impairment (Reiriz et al., 2006) and chronic improvement (Lee et al., 2006) of learning and memory ability. Here, we suggest a possible mechanism for the memory impairment following chemotherapy, in which CYP may inhibit adult hippocampal neurogenesis, which is involved in learning and memory processing (Kim et al., 2008; Raber et al., 2004; Winocur, Wojtowicz, Sekeres, Snyder, & Wang, 2006). A broad consensus holds that the presence of progenitor cells gives rise to new neural cells in the adult brain in various species.
488
M. Yang et al. / Neurobiology of Learning and Memory 93 (2010) 487–494
The adult brain retains at least two active germinal zones: the subgranular zone (SGZ) in the dentate gyrus (DG), which generates new granular cells in the adult hippocampus, and the forebrain subventricular zone (SVZ), which gives rise to granular cells in the olfactory bulb (Cameron, Woolley, McEwen, & Gould, 1993; Temple & Alvarez-Buylla, 1999). Progenitor neural cells in the DG of the adult hippocampus are particularly vulnerable to ionizing radiation (Peissner, Kocher, Treuer, & Gillardon, 1999), which can cause hippocampus-dependent learning and memory impairment (Kim et al., 2008; Winocur et al., 2006). However, little is known about the effect of chemotherapeutic agents on neurogenesis in the DG of the adult hippocampus. In this study, we examined the behavioral alteration of adult mice after CYP injection using hippocampus-dependent learning paradigms including passive avoidance and object recognition memory test. In addition, the effect of cancer chemotherapy on hippocampal neurogenesis was investigated by examining the changes in the expression levels of the proliferating cell marker Ki-67 and the immature neuronal cell marker doublecortin (DCX) in the DG of the hippocampi of adult mice after CYP treatment. 2. Materials and methods 2.1. Animals Male ICR mice (8–10 weeks old) were obtained from a specific pathogen-free colony at Oriental, Inc. (Seoul, Korea). The animals were housed in a room maintained at 23 ± 2 °C, 50 ± 5% relative humidity, artificial lighting from 08:00 to 20:00 h, and 13–18 air changes per hour. The animals were provided with tap water and commercial rodent chow (Samyang Feed, Seoul, Korea) ad libitum. The Institutional Animal Care and Use Committee at Chonnam National University approved the protocols used in this study. 2.2. Drug treatment CYP (Sigma–Aldrich, St. Louis, MO) was dissolved in sterile 0.9% saline. Mice were injected intraperitoneally (i.p.) with 40 mg/kg of CYP. The vehicle group was injected i.p. with 0.9% saline. 2.3. Behavioral analysis Behavioral dysfunction in mice who received CYP (i.p., 40 mg/ kg) was measured by open-field analysis (n = 7 mice/group), analyzing their sensitivity to electric foot-shock (n = 4 mice/group), passive avoidance (n = 9 mice/group), and object recognition memory test (n = 9 mice/group) at 12 h and 10 days after injection. 2.3.1. Open-field test Open-field analysis was used to measure the activity of vehicle (saline)-treated and CYP-treated mice in a novel environment. Parameters including total moving distance, ambulatory movement time, ambulatory movement count, and resting time were determined using an Activity Monitor (MED Associates, St. Albans, VT). 2.3.2. Passive avoidance The passive-avoidance paradigm was used to examine hippocampus-dependent associative memory (Kim et al., 2008; Zhang et al., 2008), in which the animals learned to associate the aversive unconditioned stimulus (mild electric foot-shock) with the conditioned stimulus (the dark chamber). During training, a mouse was introduced to the lighted half of the training chamber (Ugo Basile, Comerio, Italy), and allowed 1 min to explore the area before the trap door was opened. Immediately after the mouse entered
the darkened half, the trap door was closed, and a mild foot-shock (0.5 mA for 1 s) was delivered. The trained mouse remained in the dark chamber for 20 s after the shock, and it was then returned to the home cage. When tested 24 h after training, the trained mouse was re-introduced to the lighted chamber. The time spent in the lighted half before entering the darkened half was scored as the crossover latency and used as an index for memory formation. A cut-off time of 500 s was chosen for the crossover latency. The mice were removed manually from the lighted chamber when the cut-off value was reached. The sensitivity to the electric foot-shock was further tested in the mice. The electric shock threshold (mA) to elicit stereotypic responses (flinch, vocalization, and jump/vocalization) was measured. 2.3.3. Object recognition memory test Another hippocampus-dependent learning paradigm was used (Kim et al., 2008; Zhang et al., 2008). The mice were first habituated in the training/testing acryl chamber (42 cm L, 28 cm W, 20 cm H) for 1 week (15 min/day) before training. The 3.5-cm tall plastic objects for discriminative recognition were cubes, pyramids and cylinders, and they could not be displaced by the mice. The chamber arena and objects were cleaned with 75% ethanol between trials to prevent the build-up of olfactory cues. During training, two differently shaped objects were selected randomly and presented to each mouse for 15 min. At 24 h after training, another set of objects (one old object and one novel object) was presented to the trained mice. If, for example, the cube- and pyramid-shaped objects were presented during training, the cylinder-shaped object would be used as a novel object during testing. The interactions of the mouse with each object, including approaches and sniffing, were scored. If the mouse had memory retention for an old object, it should show preference for the novel object during testing. The percentage of preference was defined as the number of interactions with a specific object divided by the total number of interactions with both objects. 2.4. Tissue sampling The time-dependent effects of CYP on neurogenesis in the adult mouse hippocampus were observed after mice received an intraperitoneal injection of CYP (40 mg/kg). The mice were sacrificed, and the hippocampi were then dissected from each group at 12 h, 24 h, 4 days, and 10 days (n = 6 mice/group) after the injection of CYP. The samples were processed for embedding in paraffin wax after fixation in 4% paraformaldehyde in phosphate-buffered saline (PBS, pH 7.4) and stored at 70 °C for biochemical analysis. 2.5. Terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end-labeling (TUNEL) The level of DNA fragmentation was detected by in situ nick end-labeling (terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end-labeling, TUNEL), which was performed according to the manufacturer’s instructions (ApopTagÒ in situ apoptosis detection kit; Millipore, Billerica, MA). 2.6. Immunohistochemistry Coronal sections (thickness, 5 lm) were cut by microtome and deparaffinized using routine protocols before being exposed to citrate buffer (0.01 M, pH 6.0) and heated in an autoclave for 10 min. All subsequent steps were performed at room temperature. The sections were treated with 0.3% hydrogen peroxide for 20 min to block endogenous peroxidase activity. After three washes in phosphate-buffered saline (PBS), the sections were
M. Yang et al. / Neurobiology of Learning and Memory 93 (2010) 487–494
blocked with 10% normal goat serum (Vector ABC Elite kit, Burlingame, CA), diluted in PBS for 1 h, and then allowed to react with the immunohistochemical markers for proliferating cells and immature progenitor cells, including monoclonal rabbit anti-Ki-67 (DRM004, 1:200 dilution; Acris Antibodies GmbH, Hiddenhausen, Germany) and polyclonal rabbit anti-DCX (1:400 dilution; Cell Signaling Technology, Beverly, MA) antibody, for 2 h. After three washes in PBS, the sections were exposed to biotinylated goat anti-rabbit IgG (1:100 dilution; Vector) for 45 min. After three washes in PBS, the sections were incubated with the avidin–biotin peroxidase complex (Vector ABC Elite kit) for 45 min. After three washes in PBS, the peroxidase reaction was developed for 3 min using a diaminobenzidine (DAB) substrate (SK-4100; Vector) prepared according to manufacturer’s instructions. As a control, the primary antibodies were omitted for a few test sections in each experiment. After the completion of color development, the sections were counterstained with Harris’s hematoxylin for 5 s, washed in running tap water for 20 min, dehydrated through a graded ethanol series, cleared with xylene, and mounted with Canada balsam (Sigma–Aldrich).
489
tions representing the rostral/mid-hippocampus was used. For each mouse, three non-overlapping sections were analyzed, one from each of the three regions of the hippocampus (approximately 50 lm apart). All positively-immunolabeled cells within the SGZ of the supra- and infra-pyrimidal blades of the DG were quantified. The number of immunopositive cells was determined from the values obtained from each DG in the three brain sections. The mean number of immunopositive cells in the three sections of each mouse was taken as n = 1. The number of immunopositive cells was expressed as the mean ± SEM for each group (n = 6). 2.8. Statistical analysis The data are reported as the mean ± SEM and were analyzed using one-way analysis of variance (ANOVA) followed by a Student–Newman–Keuls post hoc test for multiple comparisons. In all cases, a p-value < 0.05 was considered to be significant. 3. Results
2.7. Cell counting
3.1. CYP-treated mice exhibit normal locomotor activity
The number of cells displaying the specific characteristics of proliferating cells (immunopositive for Ki-67) and immature progenitor cells (immunopositive for DCX) in the hippocampus was scored by an observer blinded to the identity of the sample using a histomorphometric approach (Kim et al., 2009). The brain from each mouse was sampled at approximately 2.12 mm behind the bregma. A standardized counting area containing 5-lm-thick coronal sections in a one-in-ten series of sec-
First, we examined the basic locomotor activity of mice at 12 h and 10 days after CYP treatment (40 mg/kg) in a novel environment by open-field analysis. The analysis quantified the overall locomotor activity that could reflect motivation and performance of the mice. Vehicle-treated (control) and CYP-treated mice showed comparable moving distance, ambulatory movement time, ambulatory movement count, and resting time (Fig. 1), indicating normal movement of CYP-treated mice.
Fig. 1. CYP-treated mice show normal locomotor activity in open-field analysis. The data for vehicle-treated controls (12 h after saline injection, n = 7), and CYP-treated (40 mg/kg) mice (12 h, n = 7 and 10 days, n = 7 after treatment) were monitored. (A) Vehicle- and CYP-treated mice showed similar movement distance (p = 0.962 [12 h] and p = 0.941 [10 days] vs. vehicle-treated controls). (B) Vehicle- and CYP-treated mice showed similar ambulatory movement time (p = 0.796 [12 h] and p = 0.871 [10 days] vs. vehicle-treated controls). (C) Vehicle- and CYP-treated mice showed similar ambulatory movement count (p = 0.991 [12 h] and p = 0.930 [10 days] vs. vehicle-treated controls). (D) Vehicle- and CYP-treated mice showed similar resting time (p = 0.535 [12 h] and p = 0.568 [10 days] vs. vehicle-treated controls).
490
M. Yang et al. / Neurobiology of Learning and Memory 93 (2010) 487–494
3.2. CYP-treated mice are impaired for passive avoidance Compared to vehicle-treated control mice, the mice tested 12 h after CYP treatment showed a significant decrease in crossover latency when trained for passive avoidance. All of the mice subjected to this task stepped through the door into the dark compartment in a short period in the acquisition trial (33.31 ± 3.44 s for vehicle controls, 43.98 ± 8.23 s for mice at 12 h after CYP treatment, and 44.47 ± 6.76 s for mice after 10 days CYP treatment), when they received the electric foot-shock. In the memory retention trials carried out 24 h after the acquisition trial, the memory retention time for the vehicle control mice was significantly longer than that of the mice 12 h after CYP treatment (265.7 ± 14.86 s for vehicle controls and 174.09 ± 32.9 s for mice 12 h after CYP treatment, p = 0.022). However, in the trials performed 10 days after CYP treatment, no significant difference in the memory retention times was found in CYP-treated mice compared to vehicle controls (247.64 ± 21.83 s for mice 10 days after CYP treatment, p = 0.328; Fig. 2A). Their sensitivity to the electric foot-shock was tested further (n = 4 for each group), and no significant difference was ob-
served in the threshold current to elicit stereotypic responses including flinch, vocalization and jump/vocalization, between the vehicle controls and CYP-treated mice (Fig. 2B). This suggests that they had comparable sensitivity to the electric foot-shock. 3.3. CYP-treated mice are impaired for object recognition memory We further examined CYP-treated (40 mg/kg) mice by another sensitive hippocampus-dependent paradigm, object recognition memory (Kim et al., 2008). During the 15 min training session, vehicle- and CYP-treated mice showed no significant difference in their interaction with the objects, indicating normal motivation and exploratory activity (Fig. 3A). Vehicle- and CYP-treated mice also displayed an equal preference for the two objects during training (Fig. 3B). When tested 24 h after training, the results indicated that mice 12 h after CYP treatment exhibited impairment of memory, whereas mice 10 days after CYP treatment did not show a deficit in memory (Fig. 3C). Statistical analysis comparing the recognition indexes obtained in training and retention sessions within groups indicated that the mice tested 12 h after CYP treat-
Fig. 2. CYP transiently inhibits memory retention in a passive-avoidance task. (A) Mice (n = 9) trained 12 h after CYP treatment (40 mg/kg) showed lower cross-over latency at 24 h testing after training (p < 0.05, compared to vehicle-treated controls [12 h after saline injection]; n = 9). However, mice (n = 9) trained 10 days after CYP treatment (40 mg/kg) showed normal memory retention for passive avoidance. (B) The electric shock thresholds (mA) to elicit stereotypic responses (flinch, vocalization, and jump/ vocalization) were measured for controls (1 day after saline injection) and mice 12 h and 10 days after CYP injection (n = 4 for each group). The threshold current to elicit stereotypic responses was similar to that of the vehicle controls at both time points following CYP injection. Data are reported as the mean ± SEM.
Fig. 3. CYP transiently decreases object recognition memory. (A) Vehicle- and CYP-treated (40 mg/kg) mice showed comparable interactions with the two objects during training (p = 0.558, 12 h after CYP treatment, n = 9; p = 0.283, 10 days after CYP treatment, n = 9 vs. vehicle-treated controls [12 h after saline injection], n = 9). (B) Vehicle- and CYP-treated mice showed equal preference for the two objects during training. (C) Vehicle-treated controls and mice trained 10 days after CYP treatment showed significant preference for the novel object during testing (24 h after training), but no significant preference for the novel object was observed in mice trained 12 h after CYP treatment (p = 0.146). A significant difference in novel object preference was found between vehicle-treated controls and mice trained 12 h after CYP treatment (p < 0.05).
M. Yang et al. / Neurobiology of Learning and Memory 93 (2010) 487–494
491
Fig. 4. TUNEL reaction in the DG of adult hippocampi. TUNEL-positive apoptotic nuclei were detected rarely in the DG region of the hippocampus with (A) vehicle (12 h after saline injection) and (B) CYP treatment (12 h after CYP injection). The cells were counterstained with hematoxylin. Scale bars, 50 lm.
ment had a complete memory blockade, as the animals showed no significant preference toward the novel object in the testing session (p = 0.146). No significant difference among groups was found in the total time spent exploring both objects during the training trial; the mean ± SE preference to the novel object was 67.68 ± 2.69% in the vehicle control, 55.46 ± 5.05% in the mice 12 h after CYP treatment, and 64.13 ± 3.28% in the mice 10 days after CYP treatment. 3.4. CYP does not induce apoptosis in the hippocampal DG of adult mice Hematoxylin and eosin (H&E) staining did not reveal any unusual hippocampal structure in adult mice 12 h–10 days after CYP (40 mg/kg) treatment (data not shown). Also, in both the vehicleand CYP-treated mice, TUNEL-positive apoptotic cells were quite rare in the DGs of hippocampi (Fig. 4). This suggests that the acute administration of 40 mg/kg CYP does not induce neural apoptosis in the hippocampus of adult mice. 3.5. CYP transiently decreases Ki-67 and DCX expression in the hippocampal DG of adult mice The levels of Ki-67 and DCX in the hippocampus were evaluated semiquantitatively after the administration of CYP using immunohistochemical markers of neurogenesis. Ki-67- (Fig. 5A and B) and DCX-positive cells (Fig. 6A and B) were observed consistently in the DGs of hippocampi in adult control mice. Between 0 and 24 h after CYP administration, the number of Ki-67-positive cells in the DG declined sharply (2.5 ± 0.62 cells/DG, n = 6, p < 0.0001 vs. vehicle control; Fig. 5C and D); however, it increased gradually to the vehicle control level (8.16 ± 0.4 cells/DG, n = 6) between 2 and 10 days (3.8 ± 0.31 cells/DG, p < 0.0001 vs. vehicle control [2 days]; 6.2 ± 0.60 cells/DG, p < 0.05 vs. vehicle control [4 days]; 9.7 ± 0.99 cells/DG, p = 0.190 vs. vehicle control [10 days]; n = 6 per group) after CYP treatment (Fig. 5E). The number of DCX-positive cells in the DG declined markedly (20.33 ± 1.61 cells/DG, n = 6, p < 0.0001 vs. vehicle control; Fig. 6C and D) between 0 and 24 h after CYP treatment, and then returned gradually to the vehicle control level (43.5 ± 2.32 cells/DG, n = 6) between 2 and 10 days (25.33 ± 3.55 cells/DG, p < 0.01 vs. vehicle control [2 days]; 30.67 ± 3.59 cells/DG, p < 0.05 vs. vehicle control [4 days]; 36.83 ± 3.71 cells/DG, p = 0.159 vs. vehicle control [10 days]; n = 6 per group) after CYP treatment (Fig. 6E). These data suggest that, in the short-term, the decline of neurogenesis in the DGs of adult mice with CYP treatment (40 mg/kg i.p.) is reversible. 4. Discussion Several clinical studies have reported cognitive impairment to be a side effect of cancer chemotherapy in cancer patients (Ahles
& Saykin, 2002; Parth, Dunlap, Kennedy, Ordy, & Lane, 1989; van Dam et al., 1998; Waber, Tarbell, Kahn, Gelber, & Sallan, 1992). A number of factors may be protective against cognitive impairments or place individuals at a higher risk for impairments in cognitive function. These factors include concomitant effects of cancer and its treatment (e.g., medications, fatigue, depression, or anxiety), indirect and direct effects of chemotherapy (e.g., chemotherapy-induced anemia or menopause), and individual patient factors (e.g., age, intelligence level, educational level or menopausal status) (Jansen et al., 2005). The existence of CYP-induced cognitive impairment has become generally accepted (Shilling, Jenkins, & Trapala, 2006), although many details of the concept remain controversial or obscure. The reported rates of cognitively impaired patients vary widely, reflecting different definitions of impairment (Shilling et al., 2006). Data regarding the affected cognitive domains and duration of cognitive impairment are far from congruous (Ahles et al., 2002; Fan et al., 2005; Schagen et al., 2002). A previous study in a mouse model reported that the single administration of 40–200 mg/kg CYP induces cognitive deficit in step-down inhibitory avoidance conditioning (Reiriz et al., 2006). This suggests that an injected CYP dose of 40 mg/kg is enough to interrupt brain cognition in mice. Likewise, in the current study, this dose of CYP induced hippocampusdependent learning and memory impairments without alterations in locomotor activity. Furthermore, mice trained at 12 h after CYP treatment showed memory deficit for two hippocampus-dependent learning paradigms, passive avoidance and the object recognition memory test. Conversely, mice trained at 10 days after the treatment did not show any memory defect. These results are similar to those of a previous study reporting that CYP-induced cognitive deficits at 24 h, but the impairment had recovered 1 week after CYP treatment in another hippocampus-dependent memory task, step-down inhibitory avoidance conditioning (Reiriz et al., 2006). This suggests that a single administration of 40 mg/kg CYP transiently induces cognitive impairments in this behavioral model. A consensus holds that the hippocampus is essential to the process of forming and recovering certain types of memory. In the adult brain, the progenitor/stem cells generate new neural cells in the DG of the hippocampus (Cameron et al., 1993; Temple & Alvarez-Buylla, 1999). Hence, adult-generated hippocampal cells are involved in the learning and memory functions of this structure (Squire, Stark, & Clark, 2004; Winocur et al., 2006). The inhibition of the hippocampal neurogenesis with ionizing radiation causes cognitive impairment in adult mice (Raber et al., 2004). Furthermore, the exposure to relatively low-dose radiation induces transient hippocampus-dependent learning and memory deficits in mice (Kim et al., 2008). Therefore, the pattern of memory ability may be associated with the alteration of hippocampal neurogenesis. However, further studies are needed to establish an unequivocal link between the change of hippocampal neurogenesis and the
492
M. Yang et al. / Neurobiology of Learning and Memory 93 (2010) 487–494
Fig. 5. CYP transiently decreases proliferating cells (Ki-67-positive cells) in the DG of adult hippocampi. (A–D) Representative images showing the Ki-67-positive cells in the DGs of adult hippocampi taken from the vehicle controls (24 h after saline injection) and mice 24 h after CYP injection. (E) The number of Ki-67-positive cells in the DG decreased markedly 0–24 h after CYP treatment (40 mg/kg i.p.). The cells were counterstained with hematoxylin (A–D). GCL, granular cell layer; SGZ, subgranular zone. In A and C, scale bars represent 150 lm; in B and D, scale bars represent 20 lm. Data are reported as the mean ± SEM (for E). p < 0.05 vs. vehicle controls (n = 6 per group).
hippocampus-dependent memory impairment caused by chemotherapy. Barton and Loprinzi (2002) have suggested that cognitive impairment due to chemotherapy might result from indirect chemical toxicity and oxidative damage, direct injury to the neurons, inflammation, or a type of autoimmune response. CYP is a cytotoxic drug that can alkylate DNA and prevent the duplication of the genome in dividing cells. As shown in a previous study, CYP arrests the S-phase of the cell cycle and induces apoptosis in embryonic neural progenitor cells of the telecephalon 6–12 h after administration (Ueno, Katayama, Yamauchi, Nakayama, & Doi, 2006). This study demonstrated that CYP did not induce apoptosis in the adult DG, in contrast to the induction of apoptosis in adult neurogenesis by irradiation reported in a previous study (Kim et al., 2008). However, during the period from 12 h to 4 days after injection, CYP transiently decreased the cells positive for Ki-67 (a proliferating cell marker) and DCX (an immature neuronal cell marker) in the DG of adult hippocampi, indicating that CYP transiently inhibits adult hippocampal neurogenesis. This suggests that CYP only suppresses the generation of new neural cells in the SGZ, possibly by transient arrest of the cell cycle, in the DG of the adult hippocampus. In addition, we suggest another possible mechanism for the inhibition of hippocampal neurogenesis caused by CYP treatment.
The toxic effects resulting from metabolites of CYP, such as acrolein or phosphoramide mustard (Mirkes, 1985), reduce cellular resistance to oxidative stress, which can damage the BBB, thus allowing the entry of possible neurotoxic molecules (i.e., cytokines) to the brain (Subramaniam, Subramaniam, & Shyamala, 1994). Chemotherapy-induced pro-inflammatory cytokines release may incur cognitive changes (Ahles et al., 2002; Jansen et al., 2005). Therefore, the inhibition of hippocampal neurogenesis following chemotherapy may be caused by BBB damage and the entry of neurotoxic molecules including pro-inflammatory cytokines. However, regarding the dose of CYP that induced memory facilitation in the present study, it is unclear what plasma levels of CYP had been reached with those doses. Further research should clarify the molecular and cellular mechanisms for the in vivo inhibitory effect of CYP on the generation of progenitor/stem cells in the adult DG. In conclusion, the pattern of hippocampus-dependent memory dysfunction is consistent with the change in neurogenesis after CYP administration, suggesting that the impairment of memory retention might be associated with the change in neural progenitor/stem cell marker proteins in the hippocampus in adult mice. Moreover, this study shows that a single injection of CYP in adult ICR mice can interrupt the functioning of the hippocampus, including learning and memory, possibly through the inhibition of neurogenesis.
M. Yang et al. / Neurobiology of Learning and Memory 93 (2010) 487–494
493
Fig. 6. CYP transiently decreases immature neurons (DCX-positive cells) in the DG of adult hippocampi. (A–D) Representative images showing the DCX-positive cells in the DGs of adult hippocampi taken from the vehicle controls (24 h after saline injection) and mice 24 h after CYP injection. (E) The number of DCX-positive cells in the DG decreased markedly 0–24 h after CYP treatment (40 mg/kg i.p.). The cells were counterstained with hematoxylin (A–D). GCL, granular cell layer; SGZ, subgranular zone. In A and C, scale bars represent 150 lm; in B and D, scale bars represent 20 lm. Data are reported as the mean ± SEM (for E). p < 0.05 vs. vehicle controls (n = 6 per group).
Acknowledgment This work was supported by National Research Foundation of Korea Grant funded by the Korean Government (2009-0068372). References Ahles, T., & Saykin, A. (2002). Breast cancer chemotherapy-related cognitive dysfunction. European Journal of Cancer Care, 3, 84–90. Ahles, T. A., Saykin, A. J., Furstenberg, C. T., Cole, B., Mott, L. A., Skalla, K., et al. (2002). Neuropsychologic impact of standard-dose systemic chemotherapy in longterm survivors of breast cancer and lymphoma. Journal of Clinical Oncology, 20, 485–493. Barton, D., & Loprinzi, C. (2002). Novel approaches to preventing chemotherapyinduced cognitive dysfunction in breast cancer: The art of the possible. Clinical Breast Cancer, 3, 121–127. Cameron, H. A., Woolley, C. S., McEwen, B. S., & Gould, E. (1993). Differentiation of newly born neurons and glia in the dentate gyrus of the adult rat. Neuroscience, 56, 337–344. Dollery, C. (1999). Cyclophosphamide. In Therapeutic drugs (pp. C349–C354). Edinburgh: Churchill Livingstone. Fan, H. G., Houédé-Tchen, N., Yi, Q. L., Chemerynsky, I., Downie, F. P., Sabate, K., et al. (2005). Fatigue, menopausal symptoms, and cognitive function in women after adjuvant chemotherapy for breast cancer: 1- and 2-year follow-up of a prospective controlled study. Journal of Clinical Oncology, 23, 8025–8032. Jansen, C., Miaskowski, C., Dodd, M., Dowling, G., & Kramer, J. (2005). Potential mechanisms for chemotherapy-induced impairments in cognitive function. Oncology Nursing Forum, 32, 1151–1163. Kim, J. S., Jung, J., Lee, H. J., Kim, J. C., Wang, H., Kim, S. H., et al. (2009). Differences in immunoreactivities of Ki-67 and doublecortin in the adult hippocampus in three strains of mice. Acta Histochemica, 111, 150–156.
Kim, J. S., Lee, H. J., Kim, J. C., Kang, S. S., Bae, C. S., Shin, T., et al. (2008). Transient impairment of hippocampus-dependent learning and memory in relatively low-dose of acute radiation syndrome is associated with inhibition of hippocampal neurogenesis. Journal of Radiation Research, 49, 517–526. Lee, G. D., Longo, D. L., Wang, Y., Rifkind, J. M., Abdul-Raman, L., Mamczarz, J. A., et al. (2006). Transient improvement in cognitive function and synaptic plasticity in rats following cancer chemotherapy. Clinical Cancer Research, 12, 198–205. Mirkes, P. E. (1985). Cyclophosphamide teratogenesis: A review. Teratogenesis, Carcinogenesis and Mutagenesis, 5, 75–88. Parth, P., Dunlap, W. P., Kennedy, R. S., Ordy, J. M., & Lane, N. E. (1989). Motor and cognitive testing of bone marrow transplant patients after chemoradiotherapy. Perceptual and Motor Skills, 68, 1227–1241. Peissner, W., Kocher, M., Treuer, H., & Gillardon, F. (1999). Ionizing radiationinduced apoptosis of proliferating stem cells in the dentate gyrus of the adult rat hippocampus. Brain Research Molecular Brain Research, 71, 61–68. Raber, J., Rola, R., LeFevour, A., Morhardt, D., Curley, J., Mizumatsu, S., et al. (2004). Radiation-induced cognitive impairments are associated with changes in indicators of hippocampal neurogenesis. Radiation Research, 162, 39–47. Reiriz, A. B., Reolon, G. K., Preissler, T., Rosado, J. O., Henriques, J. A., Roesler, R., et al. (2006). Cancer chemotherapy and cognitive function in rodent models: Memory impairment induced by cyclophosphamide in mice. Clinical Cancer Research, 12, 5000. Schagen, S. B., Muller, M. J., Boogerd, W., Mellenbergh, G. J., & van Dam, F. S. (2006). Change in cognitive function after chemotherapy: A prospective longitudinal study in breast cancer patients. Journal of the National Cancer Institute, 98, 1742–1745. Schagen, S. B., Muller, M. J., Boogerd, W., Rosenbrand, R. M., van Rhijn, D., Rodenhuis, S., et al. (2002). Late effects of adjuvant chemotherapy on cognitive function: A follow-up study in breast cancer patients. Annals of Oncology, 13, 1387–1397. Shilling, V., Jenkins, V., & Trapala, I. S. (2006). The (mis)classification of chemo-fog– Methodological inconsistencies in the investigation of cognitive impairment after chemotherapy. Breast Cancer Research Treatment, 95, 125–129.
494
M. Yang et al. / Neurobiology of Learning and Memory 93 (2010) 487–494
Squire, L. R., Stark, C. E., & Clark, R. E. (2004). The medial temporal lobe. Annual Reviews of Neuroscience, 27, 279–306. Subramaniam, S., Subramaniam, S., & Shyamala, D. C. (1994). Erythrocyte antioxidant enzyme activity in CMF treated breast cancer patients. Cancer Biochemistry Biophysics, 14, 177–182. Temple, S., & Alvarez-Buylla, A. (1999). Stem cells in the adult mammalian central nervous system. Current Opinion in Neurobiology, 9, 135–141. Ueno, M., Katayama, K., Yamauchi, H., Nakayama, H., & Doi, K. (2006). Cell cycle progression is required for nuclear migration of neural progenitor cells. Brain Research, 1088, 57–67. van Dam, F. S., Schagen, S. B., Muller, M. J., Boogerd, W., vd Wall, E., Droogleever Fortuyn, M. E., et al. (1998). Impairment of cognitive function in women receiving adjuvant treatment for high-risk breast cancer: High-dose versus standard-dose chemotherapy. Journal of the National Cancer Institute, 90, 210–218.
Vardy, J., & Tannock, I. (2007). Cognitive function after chemotherapy in adults with solid tumours. Critical Reviews in Oncology/Hematology, 63, 183–202. Waber, D. P., Tarbell, N. J., Kahn, C. M., Gelber, R. D., & Sallan, S. E. (1992). The relationship of sex and treatment modality to neuropsychologic outcome in childhood acute lymphoblastic leukemia. Journal of Clinical Oncology, 10, 810–817. Winocur, G., Wojtowicz, J. M., Sekeres, M., Snyder, J. S., & Wang, S. (2006). Inhibition of neurogenesis interferes with hippocampus-dependent memory function. Hippocampus, 16, 296–304. Zhang, M., Moon, C., Chan, G. C., Yang, L., Zheng, F., Conti, A. C., et al. (2008). Castimulated type 8 adenylyl cyclase is required for rapid acquisition of novel spatial information and for working/episodic-like memory. The Journal of Neuroscience, 28, 4736–4744.