PRAXIS.

BDNF augmentation reverses chemotherapy-induced decline in

cognitive function and neurogenesis

Manal T. Usmani1, Robert P. Krattli Jr.1, Sanad M. El-Khatib1, Anh C. D. Le1, Sarah M. Smith2

Janet E. Baulch2, Ding Quan Ng3, Munjal M. Acharya1, 2, §, Alexandre Chan3, 4, §

1Department of Anatomy and Neurobiology, 2Department of Radiation Oncology,

School of Medicine,

3Department of Clinical Pharmacy Practice, 4Department of Pharmaceutical Sciences,

School of Pharmacy and Pharmaceutical Sciences,

University of California, Irvine.

§Co-corresponding authors:

Dr. Munjal M. Acharya

Associate Professor

Department of Anatomy and Neurobiology,

Department of Radiation Oncology,

School of Medicine,

University of California, Irvine

Email: macharya@uci.edu

Dr. Alexandre Chan

Professor of Clinical Pharmacy

Department of Clinical Pharmacy Practice,

Department of Pharmaceutical Sciences,

School of Pharmacy & Pharmaceutical Sciences,

University of California, Irvine

Email: a.chan@uci.edu

The authors declare no relevant conflicts of interest or financial relationships.

Keywords: BDNF, chemotherapy, chemobrain, cognitive function, riluzole, neurogenesis

ABSTRACT

Cancer-related cognitive impairment (CRCI) greatly affects the quality of life of millions of cancer survivors. In our clinical studies, breast cancer patients treated with doxorubicin (Adriamycin®, ADR) experienced a significant reduction in the blood levels of brain-derived neurotrophic factor (BDNF) that was associated with a higher risk of CRCI. BDNF has been shown to promote survival, differentiation, and maintenance of in vivo dentate neurogenesis, and chemotherapy induces a plethora of physiological and cellular alterations, which includes the decline in neurogenesis and that has been linked with cognitive impairments. Our past rodent studies in CRCI have also shown a significant decline in the dentate neurogenesis that was accompanied by cognitive decline. In this study, using a mouse model of ADR-induced cognitive decline, we tested the impact of riluzole (RZ), an orally active BDNF-enhancing medication that is FDA-approved for amyotrophic lateral sclerosis. ADR-treated mice receiving RZ in the drinking water for one month showed significant improvements in the hippocampal-dependent learning and memory (spatial recognition), fear extinction memory consolidation, and reduced anxiety-like behavior. RZ prevented chemotherapy-induced reductions of BDNF levels in the hippocampus. Importantly, RZ mitigated chemotherapy-induced loss of newly born, immature neurons and dentate neurogenesis. In conclusion, this data provides pre-clinical evidence for a translationally feasible approach to enhance the neuroprotective effects of BDNF to prevent CRCI.

INTRODUCTION

Chemotherapy-related cognitive impairment (CRCI), often referred to as "chemobrain" or "chemofog" is prevalent up to 75% of all breast cancer survivors. It encompasses a wide range of symptoms during and after treatment such as memory loss, inability to concentrate, difficulty in thinking, poor response speed and executive functioning [1]. These cognitive changes can be subtle but have a detrimental effect on patients' daily functioning and health-related quality of life.[2] Chemotherapy-induced cognitive decline can pose significant challenges for cancer survivors who wish to resume their day-to-day routine and social roles. Research has suggested that the quality of life and activities of daily living in this group of long-term cancer survivors are immensely affected by subtle neuropsychological changes.

Clinical studies have suggested that brain-derived neurotrophic factor (BDNF) is linked with a positive impact on cognition in patients with cancer [3-7]. BDNF exerts a number of neuro-modulatory effects on the CNS and on the neurosensory, endocrine, and immune systems, and BDNF is also associated with neuronal repair and survival, dendritic and axonal growth, and long-term potentiation. Numerous studies have reported the possible role of BDNF in the pathogenesis of various cognitive disorders, such as Alzheimer's disease (AD), with low serum levels have been correlated with AD and mild cognitive impairment, and high serum BDNF levels have been associated with better cognition in healthy older adults.[8-11] In our previous clinical studies, we have observed that breast cancer patients [7, 12]. receiving doxorubicin and cyclophosphamide were associated with self-perceived concentration deficit during chemotherapy, and patients experiencing the least reduction of plasma BDNF levels between baseline and end of chemotherapy were most protected from cognitive decline. As chemotherapy can be neurotoxic, it is postulated that the effects of chemotherapy on BDNF expression can persist long after the completion of chemotherapy and in cancer survivors, resulting in persistent cognitive decline post chemotherapy [13]. Loss of BDNF may also accelerate aging, which is postulated to be a mechanism that leads to cognitive decline in patients receiving chemotherapy [14].

Supplementation of BDNF has been shown to protect synapses against various toxic insults in neu­rodegenerative diseases including, AD, Huntington's dis­ease, ALS and Parkinson's disease [15] in animal models. These existing data provides strong justification on the need to study the viability on augmenting BDNF levels to reverse chemobrain. To study the viability of this strategy, we have conducted a pre-clinical study to evaluate the efficacy of riluzole (RZ), an orally bioavailable agent that has been shown to increase BDNF [16] and improve cognitive function in the AD models.[17-19] We hypothesize that RZ treatment will reduce the onset and severity of chemotherapy-induced cognitive decline through augmenting BDNF levels.

MATERIALS AND METHODS

Details on materials, experimental methods, behavior, and immunostaining protocols are provided in the Supplemental Information section.

Animals and treatments

All animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC), and according to the NIH guidelines. Our past clinical studies have shown a positive correlation between the low BDNF levels and the severity of CRCI in female breast cancer patients [7, 12]. Thus, for this pre-clinical study, we used adult female mice. Four-month-old female wild type mice (C57BL/6J) were purchased from Jackson Laboratories. Mice were group housed (4 mice per cage), kept on a standard light-dark cycle (12 hour each) 20 °C ± 1 room temperature, and 70% ± 10 humidity; and given a standard rodent chow diet (Envigo Teklad 2020X) by the University Laboratory Animal Recourses (ULAR). Mice received Adriamycin injections (ADR, doxorubicin hydrochloride, Sigma) freshly made in saline (2 mg/kg, once weekly, i.p.) for four weeks as shown in the research design (Fig. 1A). Riluzole (2-amino6-(trifluoromethoxy)benzothiazole, Selleckchem) was dissolved in filter sterilized (0.2 µm, Millipore), warm RO water (reverse osmosis, ULAR) with constant stirring (1 to 2 hours at 45 ⁰C) at a stock concentration of 600 µg per ml. The stock solution was kept frozen at -20 ⁰C until use. The working solution (60 µg per ml) of RZ was prepared twice a week by diluting the stock with sterile RO water. Mice had ad libitum access to either vehicle (sterile RO water) or RZ solution (13 mg/kg per mouse, per day as described [20-22]) throughout the duration of the study. Mice received RZ treatment 24 hours after the last ADR injection (Fig. 1A). To evaluate the impact of ADR or RZ treatment on in vivo neurogenesis, one week after the last ADR injection, mice were administered BrdU (5-bromo-2’-deoxyuridine, 50 mg/kg, i.p., Sigma) made in phosphate-buffered saline (PBS, pH 7.6, Sigma) once daily for 6 days. Mice were divided into four experimental groups (N = 12 to 24 mice per group) including saline-treated control mice receiving the RO water (Con + Veh), ADR-treated mice receiving the RO water (ADR + Veh), saline-treated control mice receiving RZ (Con + RZ), and ADR-treated mice receiving RZ (ADR + RZ).

Cognitive testing

To determine the effect of RZ-mediated BDNF augmentation on cognitive function after chronic chemotherapy, mice were administered cognitive function and anxiety-like behavior tests four weeks after the initiation of RZ treatment. Over two to three weeks, cognitive function tasks included novel place recognition (NPR), elevated plus maze (EPM) for anxiety-like behavior, and, lastly, fear extinction memory consolidation task (FE). The NPR task determines the ability of animals to explore novel placement of the objects in an unrestricted, non-invasive open environment [23, 24] that is calculated as discrimination index [DI = (Novel/Total exploration time) – (Familiar/Total exploration time)] × 100]. A positive DI indicates that animals spent more time exploring novel spatial locations. In contrast, a negative or zero index shows that animals exhibited little or no preference for the novel placements and spent equal time exploring familiar and novel places. EPM measures anxiety levels based on animal exploration in the elevated open arms versus closed (dark) arms under a brightly lit environment. Anxious animals will spend more time in closed arms compared with open arms. After completing NPR and EPM tasks and about two days of break, animals were administered the FE task. This task determines if chronic chemotherapy or RZ treatment affects hippocampal-dependent fear conditioning and memory consolidation process (see Supplemental Information for details). On the first day of the conditioning phase, mice were presented with three pairs (evenly spaced) of auditory stimulus co-terminating with a mild foot shock. Twenty-four hours later, on the subsequent three days (extinction training phase), animals were presented with 20 tones while in the same contextual environment (odor and visual cues). Twenty-four hours later, the fear test was administered on the final day, where animals were presented with only three tones in the same context. Animals' freezing behavior was recorded using a ceiling-mounted camera in the FE test chamber and recorded by an automated freezing measurement module (FreezeFrame, Coulbourn Instruments). The percentage of time each mouse spent freezing during the tone was then calculated for the conditioning extinction training and testing phases.

Immunohistochemistry, confocal microscopy, and in silico volumetric quantification

After completion of cognitive function tests, mice were euthanized (intra-cardiac perfusion) using saline with heparin (10U/ml, Sigma) and 4% PFA made in 100 mM PBS, pH 7.4 (paraformaldehyde, phosphate-buffered saline, Sigma). Brains were fixed overnight at 4 ⁰C in 4% PFA. Tissues were then cryo-protected (10 to 30% sucrose made in 100 mM PBS, pH 7.4 and 0.02% sodium azide, Sigma) and cryo-sectioned using a cryo-stat (HN525 NX, Microm, Epredia) at the thickness of 30 µm (coronal). To determine the impact of chronic chemotherapy and RZ treatments on the function of the neurogenic niche, serial coronal brain sections (2-3 sections per brain, 8-10 brains per group) through the hippocampal formation were stained using free-floating immunofluorescence protocols as described.[25] Doublecortin (DCX) staining was performed to label newly born, immature neurons [26] using a rabbit anti-DCX primary (1:200; Abcam) and a donkey anti-Rabbit Alexa Fluor 568 secondary antibodies. DCX-positive cells were visualized using fluorescence microscopy as red. For the BrdU-NeuN dual-immunofluorescence staining, tissues were permeabilized to recover the BrdU antigen, and stained using rat anti-BrdU (1:150; Abcam) and rabbit anti-NeuN (1:500, Millipore) primary antibodies. The fluorescence color was developed using donkey anti-rat Alexa Fluor 488 (1:150, Invitrogen) and donkey anti-rabbit Alexa Fluor 568 (1:500, Invitrogen) secondary antibodies. BrdU-positive cells were visualized using fluorescence microscopy as green, and NeuN-positive cells as red.

Single- (DCX) or dual-(BrdU-NeuN) immunofluorescent sections were imaged using a laser-scanning confocal microscope (Nikon Eclipse Ti C2) equipped with a 40× oil-immersion objective lens (1.0 NA) and NIS element AR module (v4.3, Nikon). The high-resolution (1024 to 2048p) z stacks (1 µm thick) were acquired through the 25-30 µm thick brain sections. An adaptive, 3D blinded deconvolution method (ClearView, Imaris v9.2, BitPlane, Inc.) was used to deconvolute and facilitate the fluorescence signal resolution with respective fluorescent wavelengths (510nm, green; 594 nm, red; 447 nm, UV ranges). The deconvoluted IMS format images were analyzed using a 3D algorithm-based Imaris module (v9.2). In Imaris, BrdU and NeuN were 3D modeled using the surface-rendering tool to create the volume of individual neurons. Using an unbiased, dedicated co-localization channel, the number of BrdU+ cells co-labeled with NeuN+ neurons was individually enumerated. The number of DCX+ cells was quantified using the spot analysis tool. All Imaris-based in silico analyses were conducted using automated batch processing modules. The criteria were applied uniformly for all experimental groups by an experimenter blinded to the group IDs to avoid bias.

BDNF ELISA

To determine the impact of RZ on the hippocampal BDNF, mice receiving ADR or RZ were euthanized for 4-weeks after the initiation of RZ treatment, and BDNF ELISA was performed as described [27]. Brains were immediately extracted from the skull (N = 8-9 mice per group). The hippocampus was micro-dissected from each cerebral hemisphere, flash-frozen by immersing the cryo-vials in the liquid nitrogen and stored at -80°C until assayed. Each hippocampus was weighed and transferred into 500 μl ice-cold lysis buffer (NPER Neuronal Protein Extraction Reagent, ThermoScientific) containing sodium orthovanadate (0.5 mM, Santa Cruz), phenyl-methylsulfonyl fluoride (PMSF, 1 mM, Santa Cruz), aprotinin (10 μg/ml, Santa Cruz), and leupeptin (1 μg/ml; Santa Cruz). Tissues were then sonicated individually, centrifuged at 4 °C, and the supernatants were collected and diluted at 1:5 or 1:10 with ice-cold Dulbecco's PBS (Gibco). The supernatants were acidified to pH 2.6 and then neutralized to pH 7.6. The BDNF levels were assayed using a commercially available ELISA kit (E-EL-M0203, Elabscience Biotechnology) and uncoated ELISA plates (Nunc MaxiSorp, Biolegend). The colorimetric measurements were performed at 450 μm wavelength using a microplate reader (BioTek SynergyMx).

Statistical analysis

The statistical analyses of the cognitive function, biochemical, and immunohistochemical data were conducted using one-way ANOVA to confirm overall significance (GraphPad Prism, v8.0). All data are presented as the mean ± SEM. For the chemotherapy or RZ treatment analysis, two-way ANOVA and Bonferroni's multiple comparisons test were performed. The Kruskal-Wallis test was used for the analysis of BDNF, DCX and BrdU-NeuN data. Wilcoxon matched-pairs signed-rank test was used to compare exploration of familiar versus novel places by the same animals. All statistical analyses were considered significant for a value of P ≤ 0.05.

RESULTS

Riluzole treatment mitigates chemotherapy-induced cognitive impairments and anxiety-like behavior

Adult WT female mice received chronic ADR treatment (once weekly for four weeks) followed by RZ in drinking water (Fig. 1A). One month after RZ treatment, mice were handled, habituated, and tested on the hippocampal-dependent NPR task (Fig. 1B-C). The open field activity of animals in the open arena was monitored for the day 1 of the habituation phase (no objects). We did not find significant differences between the experimental groups for the total distance travelled and time spent in the central zone (60% of central area) for the open field activity (Suppl. Fig. 1) indicating the absence of neophobic behavior during cognitive testing. During the test phase, the total time exploring both familiar and novel placements of objects did not differ significantly between the experimental groups (Suppl. Fig. S2). Comparison of the animal exploration of the familiar and novel spatial location times (Fig. 1B) revealed significant differences between Control + Vehicle, Control + RZ, and ADR + RZ mice (P's <0.01), but not for the ADR + Vehicle mice, indicating no differences in animal's preference to the novel location. This preference (or not) for the novelty was then calculated as the discrimination index, and we found significant overall group differences between the experimental groups (Fig. 1C, F(3, 52)=10.29, P<0.0001). ADR-treated mice receiving Vehicle (ADR + Vehicle) showed a significantly reduced preference to explore the novel location compared to the Con + Vehicle (P<0.0001), Con + RZ (P<0.001), and ADR + RZ (P<0.002) mice. This behavior is also reflected in the heat map of mice exploration activity for the novel or familiar spatial locations (Fig. 1D). Importantly, ADR-treated mice receiving RZ did not show reductions in the discrimination index, and exploration for the novel location was comparable to the Con + Vehicle and Control + RZ mice.

To determine if chronic ADR and RZ treatments affected anxiety-like behavior, all groups of mice were administered an elevated plus maze (EPM) task. The EPM task provides a preference for exploring either open or the closed (dark) arms of the maze under the bright ceiling lights. Anxious mice would show a higher preference for the closed versus open arms to avert brightly lit space. We found significant overall group differences between the mice groups for the percentage of time spent in the open arms (Fig. 1E, F(3, 44)=7.69, P<0.0003). ADR + vehicle-treated mice spent significantly less time in the open arms than the Con + Vehicle group (P<0.001). In contrast, ADR + RZ-treated mice showed significantly increased time spent in the open arms (P<0.002, Fig. 1E) than ADR + Vehicle groups indicating reduced anxiety-like behavior following RZ treatment.

In the past, we reported chemotherapy-induced impairments in the contextual fear memory and fear extinction memory consolidation.[26, 28] Thus, to evaluate if RZ treatment exerts beneficial effects or not, animals were administered the fear extinction memory consolidation task (FE, Fig. 2). During day 1, the conditioning phase of FE testing, all groups of mice (Con + Vehicle, Con + RZ, ADR + Veh, and ADR + RZ) showed comparable associative learning as indicated by increased time spent freezing during the tone-shock conditioning phase (Fig. 2A; T1-T3; 40 to 48% on T3). Twenty-four hours later, during the extinction training phase (days 1-3), mice were presented with 20 tones per day (5s intervals) in the same context as the conditioning phase while no foot shock was administered. The ADR + Vehicle group continued to show increased freezing compared to the Con + Vehicle and ADR + RZ groups (Fig. 2A; Extinction Training Day 1-3, P<0.02). These data indicated that RZ treatment to the chronically-exposed ADR animals mitigated impairments in the ability to dissociate the learned response (freezing) to a prior aversive event (tone-shock pairing). Twenty-four-hour post-extinction training phase, the mice were administered extinction testing (3 tones, 120s intervals) in the same testing environment was used for extinction training. We found significant group effects in the extinction test (Fig. 2B; F(3, 56)=6.10, P=0.001). ADR + Vehicle mice failed to abolish fear memories during this retrieval testing and exhibited increased freezing that was ameliorated by the RZ treatment (Fig. 2B; P=0.02 vs. ADR + RZ group).

These cognitive function and anxiety-like behavior data demonstrated that chronic chemotherapy-induced impairments were mitigated by RZ treatment.

Riluzole restores in vivo BDNF levels in the ADR-treated mice brains

Our past human studies have shown that reduction of BDNF levels is linked with acute and persistent self-reported cognitive impairments in cancer survivors treated with cytotoxic agents including cyclophosphamide and ADR [7, 12]. In the mouse model of AD, RZ treatment has been shown to improve cognitive function [22]. To link favorable cognitive and behavioral outcomes in the ADR-treated mice receiving RZ, we conducted an ELISA-based estimation of BDNF in the micro-dissected hippocampus. We found a significant overall group effect in the hippocampal BDNF levels (Fig. 3; F(3, 28)=43.64, P<0.0001). Chronic ADR treatment leads to about a 45% decrease in the BDNF levels compared with either Control + Vehicle or Control + RZ groups (P's <0.001). RZ treatment significantly restored BDNF levels in the mice exposed to chronic ADR (P<0.001). This data indicates that RZ augmented BDNF in the chemotherapy-exposed mice and served as one of the contributory factors in the neuro-cognitive improvements.

BDNF augmentation reverses chemotherapy-induced loss of newly born neurons

BDNF has been shown to play critical role in maintaining neurogenesis and neuronal plasticity in the brain [29, 30]. In our past studies, we have shown that exposure to chronic chemotherapy adversely affect neurogenesis [26]. We assessed the impact of chronic ADR and RZ treatments on the expression of doublecortin (DCX), a surrogate marker of the newly born, immature neurons in the hippocampal dentate gyrus. Following differentiation from neural stem cells, newly born neurons express DCX in the hippocampal granule cell layers (GCL) and sub-granular zone (SGZ) within three hours and up to 12 to 14 days in vivo. We conducted immunofluorescence staining and quantification for the DCX+ neurons in the hippocampal GCL and SGZ. We found significant overall group differences in the number of DCX+ cells (Fig. 4; F(3, 44)=10.92, P<0.0001). The chronic ADR treatment (ADR + Vehicle) led to a significant reduction in the number of DCX+ cells in hippocampal GCL and SGZ compared to either Control + Vehicle and Control + RZ groups (Fig 4A-D, P<0.01 and P<0.002 respectively). Treatment with RZ significantly increased the number of DCX+ cells in the ADR + RZ groups compared with ADR + Vehicle groups (P<0.0001) indicating protective effects against chronic chemotherapy.

BDNF enhancement reverses chemotherapy-induced decline in neurogenesis

In the rodent hippocampus, after differentiation from neural stem cells, the maturation of newly born neurons takes approximately 26-27 days to express mature neuronal marker (NeuN). Next, we assessed the impact of riluzole treatment and BDNF enhancement on the process of neurogenesis in vivo. One week after completion of chronic ADR treatment (Fig. 1A), mice were treated with BrdU to label proliferating neural stem cells. The assessment of neurogenesis (BrdU+-NeuN+ dual-labeled cells) was conducted 7-8 weeks after BrdU injection. We found overall significant effects for the percentage of BrdU+ cells expressing NeuN (Fig. 5A-D, a1-d1, BrdU+-NeuN+; F(3, 24)=5.77, P<0.005). ADR + Vehicle groups showed a significantly reduced percentage of BrdU+-NeuN+ cells compared with Control + Vehicle and Control + RZ groups (Fig. 5E, P's <0.01). On the other hand, treatment with RZ (or augmentation of BDNF), significantly increased BrdU+-NeuN+ cells in the ADR + RZ group compared with ADR + RZ group (P<0.003). This data corroborates our findings on the immature neurons (Fig. 4, DCX) and assert that RZ treatment is neuroprotective against chemotherapy-induced loss of neurogenesis.

DISCUSSION

This study provided pre-clinical evidence that RZ reverses chemotherapy-induced cognitive decline in a mouse model. Notably, in chemotherapy-exposed mice, we have observed that RZ increased BDNF levels, which was associated with neurocognitive improvements. Furthermore, it is evident that augmentation of BDNF reverses chemotherapy-induced loss of newly born neurons and the decline in neurogenesis. Our results echoed with other studies reported in the literature where treatment with riluzole arrested cognitive decline in mouse models of AD likely mediated through protection of hippocampal neurons [49, 50]. In cultured mouse astrocytes, RZ is shown to stimulate BDNF and glial cell line-derived neurotrophic factor (GDNF) synthesis [31].

The basis of this study stems from the breadth of our and others' clinical data showing a positive correlation between the low BDNF levels with cognitive dysfunction in cancer patients receiving cytotoxic chemotherapy [3-7]. We postulated that the effects of chronic chemotherapy on BDNF expression could persist long after the completion of chemotherapy and in cancer survivors, resulting in the long-term CRCI [13]. In our past clinical studies [7, 12, 32], we consistently observed a statistically significant decrease in plasma BDNF levels at 6 and 12 weeks after breast cancer patients were treated with ADR and cyclophosphamide (ADR-CYP). This association between reduced BDNF and the higher risk of cognitive decline was evident 24 months post-treatment. These findings suggest that increasing or augmenting BDNF levels in vivo provides a therapeutic avenue to reverse short- and long-term CRCI.

It has been well established that along with other neurotrophins, BDNF plays a major role in cognitive performance, maintenance of neurogenesis and neuronal plasticity in the brain [29, 30]. BDNF is highly expressed in the hippocampus, cortex, and basal forebrain and has an important role in regions that are vital to cognitive function, particularly, hippocampal-dependent learning and memory [33]. BDNF plays a neuromodulatory role in the maintenance of LTP [10, 11, 30] that contribute in the learning and memory consolidation [34]. Several studies have linked reductions in BDNF in the pathogenesis of cognitive disorders, such as AD, with low serum levels correlated with AD and mild cognitive impairment, and high serum levels associated with better cognition in healthy older adults [8-11]. Our past pre-clinical studies have also shown a significant decline in newly born neurons (DCX+), hippocampal neurogenesis (BrdU-NeuN) and loss of neuronal structure and synaptic density that was accompanied by cognitive dysfunction following chronic ADR or CYP treatment [26, 28]. As chemotherapy diminishes dentate neurogenesis, it is plausible that it is partially linked to a decline in the hippocampal BDNF levels in the chemo-treated brains. In the rat models of chemobrain, physical exercise prevented decline in newly born neurons, neurogenesis, apoptosis and reduced cognitive impairments [35, 36]. Physical exercise or running has been shown to increase BDNF levels which is linked with improved synaptic plasticity, spinogenesis, dentate neurogenesis and improved cognitive function in rodents [37]. Several human studies evaluated non-pharmacological interventions (such as cognitive training) or integrative therapies to improve from CRCI by increasing BDNF levels [38-41]. In the past, our clinical study evaluated whether cognitive training may improve CRCI in adolescent and young adult cancer patients [38] . Albeit the sample size and effect size were small, we found an improving trend of BDNF levels in patients undergoing cognitive training compared to the controls, with fewer patients reporting cognitive problems among those receiving cognitive training. Taken together, these pre-clinical and clinical studies supported our pharmacological approach to augment BDNF in vivo to protect from the loss of neurogenesis and thereby cognitive function following exposure to chemotherapy.

A number of mitigation strategies have been investigated to clinically manage CRCI, which include both non-pharmacological and pharmacological strategies [42, 43]. Unfortunately, the poor effectiveness for majority of the proposed intervention was likely due to the poor ability for these to target the underlying mechanisms associated with CRCI. Recent research has also evaluated whether novel modalities such as stem cells [44] or exosomes [45] are able to reverse CRCI. Hence, the utility of translationally feasible pharmacological strategy in the context of non-CNS cancer therapy has yet to be established and remains an area of limited research. Our strategy to augment BDNF in CRCI has also been trialed in a number of neurological complications such as AD [46], Parkinson’s Disease[47] and depression [48]. In the AD, it has been shown that CREB-mediated transcription is down regulated by β-amyloid peptide in the amyloid plaques, which can contribute to a decrease of CREB-regulated BDNF levels. Non-pharmacological strategies such as exercise can improve learning and memory impairment by upregulating the BDNF pathway in animal models of AD [46] and in Parkinson’s disease [47]. Investigators have also evaluated the efficacy of utilizing Mediterranean diet to improvement plasma BDNF levels in patients diagnosed with depression [48]. It is important note that current interventions are mostly focused on non-pharmacological strategies, with a lack of pharmacological agents demonstrating the effectiveness to improve disease as well as associated BDNF levels.

We acknowledge that in addition to protection of neurogenesis, RZ-mediated beneficial neuroprotective effects in the chemobrain model are likely mediated through other mechanisms as well including maintenance of neuronal plasticity or LTP, protection of dendritic structure, spine density, synapses and reduction of chemotherapy-induced gliosis. For example, AD mice treated with RZ had long-term memory comparable to the age-matched controls that was hypothesized to be mediated through reducing glutamate release during early AD as glutaminergic hyperactivity is known to cause a neurotoxic effect on hippocampal neurons [49, 50]. The eventual goal of our approach is to test this pharmacologic strategy in a clinically relevant, cancer bearing model receiving combination chemo- or radiation-therapy to broaden its prospects to ameliorate CRCI.

CONCLUSION

In summary, our data provides pre-clinical evidence for a translationally-feasible pharmacological approach to augment BDNF in vivo to mitigate cancer chemotherapy-induced adverse impact on neurogenesis and cognitive function.

ACKNOWLEDGEMENTS

The project described was supported by the National Center for Research Resources and the National Center for Advancing Translational Sciences (NCATS), NIH, through Grant UL1 TR001414 to A.C. American Cancer Society (ACS) Research Scholar Grant (RSG-17–146–01-CCE), and NIH R01 CA251110 awards to M.M.A. We also thank the support of the UCI Chao Family Comprehensive Cancer Center Optical Biology shared resource supported by the NIH (award P30CA062203). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

FIGURE LEGENDS

Figure 1. Treatment with riluzole reverses chronic chemotherapy-induced cognitive dysfunction. (A) Schematic representation of research design: Eight to ten weeks old wild type (C57BL/6J) female mice were injected with vehicle (Veh, saline) or an anthracycline, Adriamycin (ADR, doxorubicin hydrochloride, 2 mg/kg, i.p.), once weekly for four weeks. Twenty-four hours after the last ADR injection, mice were given riluzole (RZ) treatment in the drinking water (13 mg/kg) and continued on RZ till the end of the study. ADR-treated animals that received drinking water served as the vehicle group. To assess the in vivo neurogenesis, one week after the last ADR injection, mice were administered with BrdU (5-Bromo-2’-deoxyuridine, 50 mg/kg, i.p., once daily for six days). One month after initiation of RZ treatment, mice were administered a spatial memory retention test (novel place recognition, NPR), an anxiety test (elevated plus maze, EPM) followed by a fear extinction memory consolidation task (F. Ext.). After completion of cognitive testing, mice were euthanized and brains were collected for the biochemical and immunohistochemistry analyses. (B) ADR-treated mice spent significantly less time exploring novel spatial location. Time spent exploring the novel placements of objects during the test phase of the NPR task show that Control + Vehicle, Control + RZ and ADR + RZ mice spent significantly more time exploring novel versus familiar location whereas ADR + Vehicle mice spent comparable time exploring both locations indicating an inability to recognize the novel location. (C) The tendency to explore a novel spatial location was derived from the Discrimination Index, calculated as ([Novel location exploration time/Total exploration time] – [Familiar location exploration time/Total exploration time]) × 100. ADR treatment significantly impaired cognitive function as shown by the reduced preference towards the novel place in the ADR + Vehicle mice compared to the Control + Vehicle, Control + RZ, and ADR + RZ mice. Importantly, ADR-exposed mice receiving RZ treatment did not show a decline in spatial exploration. (D) Examples of heat maps depicting animals exploring novel or familiar spatial locations in each experimental group during the NPR task. (E) ADR + Vehicle mice showed increased anxiety levels as indicated by reduced time (percentage) spent in the open arms on the EPM tasks. In contrast, Control + Vehicle and ADR + RZ treated mice showed significantly higher time spent in the open arms. Data are presented as mean ± SEM (N =12-16 mice per group). P values were derived from ANOVA and Bonferroni's multiple comparisons test.

Figure 2. Riluzole treatment alleviates chemotherapy-induced deficits in fear memory consolidation. (A) Treatment with either ADR or RZ did not impair the acquisition of conditioned fear memory as shown by the elevated freezing following a series of three-tone and shock pairings (80 dB, 0.6 mA, T1–T3). Subsequently, 24 h later, fear extinction training was administered every 24 h (20 tones) for 3 days. Each data point for the days 1-3 is presented as average of percentage time freezing for 5 tones. All mice showed a gradual decrease in the freezing behavior (Days 1–3), however, ADR + Vehicle mice spent a significantly higher time freezing compared with Control + Vehicle or ADR + RZ mice. (B) Twenty-four hours after the extinction training, Control + Vehicle and Control + RZ mice showed abolished fear memory (reduced freezing) compared with ADR + Vehicle mice. Importantly, ADR-treated mice receiving RZ (ADR + RZ) were able to successfully abolish fear memory (reduced freezing) compared with the ADR + Vehicle group. Data are presented as mean ± SEM (N = 10-20 mice per group). P values were derived from ANOVA and Bonferroni's multiple comparisons test.

Figure 3. Riluzole treatment restores hippocampal BDNF in the chemotherapy-exposed mice. 8-10 weeks old WT female mice received chronic ADR treatment (2 mg/kg, i. p., once weekly for 4-weeks) followed by riluzole (RZ) treatment (13 mg/kg) in drinking water for four weeks. An ELISA-based estimation of BDNF from micro-dissected mice hippocampus showed chemotherapy-induced reductions in the ADR + Vehicle group. Importantly, RZ treatment to the ADR-exposed mice showed significant restoration of BDNF levels. Data are presented as mean ± SEM (N = 8 mice per group). P values were derived from ANOVA and Bonferroni's multiple comparisons test.

Figure 4. BDNF augmentation reverses chemotherapy-induced decline in newly born neurons. Adult female mice treated with chronic ADR (2 mg/kg, i.p., once weekly for 4-weeks) and 24 h later received riluzole (RZ, 13 mg/kg) in drinking water for six to seven weeks. Newly born, immature neurons (doublecortin, DCX) in the hippocampal sub-granular zone (SGZ) and granule cell layer (GCL) were assessed using immunofluorescence staining, confocal microscopy and 3D algorithm-based unbiased spot analysis of DCX+ cells (red; DAPI, blue). ADR treatment significantly reduced the number of DCX+ neurons in the hippocampus compared with either Control + Vehicle or Control + RZ groups. ADR-treated mice receiving RZ showed a significantly higher number of DCX+ cells compared with the ADR + Vehicle group. Data are presented as mean ± SEM (N = 12 mice per group). P values were derived from ANOVA and Bonferroni's multiple comparisons test. Scale bars, 40 μm, A–D.

Figure 5. BDNF enhancement reverses chemotherapy-induced decline in neurogenesis. WT adult female mice treated with chronic ADR (2 mg/kg, i. p., once weekly for 4-weeks) received riluzole (RZ, 13 mg/kg) 24 h later in drinking water for six to seven weeks. One week after the last ADR injection mice were treated with BrdU and hippocampal neurogenesis was quantified using BrdU-NeuN dual-immunofluorescence staining in the dentate gyrus four to five weeks after BrdU treatment. Compared to the Control + Vehicle (A, a1) and Control + RZ (B, b1), ADR + Vehicle mice (C, c1) showed a significant decline in neurogenesis, as indicated by the reduced numbers of the percentage of BrdU+ cells (red) differentiating into the mature neurons (green, NeuN; C, c1, E). RZ treated mice retained significantly higher numbers of dual-fluorescent cells (BrdU+-NeuN+), similar to the Control + Vehicle group (E). Orthogonal z stacks (a1–d1) for the representative dual-labeled BrdU+-NeuN+ cells (white arrows) are shown for each group (A–D). Data are presented as mean ± SEM (N = 6-8 mice per group). P values were derived from ANOVA and Bonferroni's multiple comparisons test. GCL, granule cell layer; DH, dentate hilus. Scale bars, 50 μm, A–D, and 5 μm, a1–d1.

SOURCES:

1. Cheung YT, Shwe M, Tan YP, Fan G, Ng R, Chan A: Cognitive changes in multiethnic Asian breast cancer patients: a focus group study. Ann Oncol 2012, 23(10):2547-2552.

2. Ng T, Dorajoo SR, Cheung YT, Lam YC, Yeo HL, Shwe M, Gan YX, Foo KM, Loh WK, Koo SL et al: Distinct and heterogeneous trajectories of self-perceived cognitive impairment among Asian breast cancer survivors. Psychooncology 2018, 27(4):1185-1192.

3. Bury-Kaminska M, Szudy-Szczyrek A, Nowaczynska A, Jankowska-Lecka O, Hus M, Kot K: Chemotherapy-Related Differences in Cognitive Functioning and Their Biological Predictors in Patients with Multiple Myeloma. Brain Sci 2021, 11(9).

4. Jehn CF, Becker B, Flath B, Nogai H, Vuong L, Schmid P, Luftner D: Neurocognitive function, brain-derived neurotrophic factor (BDNF) and IL-6 levels in cancer patients with depression. J Neuroimmunol 2015, 287:88-92.

5. Zimmer P, Mierau A, Bloch W, Struder HK, Hulsdunker T, Schenk A, Fiebig L, Baumann FT, Hahn M, Reinart N et al: Post-chemotherapy cognitive impairment in patients with B-cell non-Hodgkin lymphoma: a first comprehensive approach to determine cognitive impairments after treatment with rituximab, cyclophosphamide, doxorubicin, vincristine and prednisone or rituximab and bendamustine. Leuk Lymphoma 2015, 56(2):347-352.

6. Guo JC, Yang YJ, Zheng JF, Guo M, Wang XD, Gao YS, Fu LQ, Jiang XL, Fu LM, Huang T: Functional rs6265 polymorphism in the brain-derived neurotrophic factor gene confers protection against neurocognitive dysfunction in posttraumatic stress disorder among Chinese patients with hepatocellular carcinoma. J Cell Biochem 2019, 120(6):10434-10443.

7. Yap NY, Tan NYT, Tan CJ, Loh KW, Ng RCH, Ho HK, Chan A: Associations of plasma brain-derived neurotrophic factor (BDNF) and Val66Met polymorphism (rs6265) with long-term cancer-related cognitive impairment in survivors of breast cancer. Breast Cancer Res Treat 2020, 183(3):683-696.

8. Gunstad J, Benitez A, Smith J, Glickman E, Spitznagel MB, Alexander T, Juvancic-Heltzel J, Murray L: Serum brain-derived neurotrophic factor is associated with cognitive function in healthy older adults. J Geriatr Psychiatry Neurol 2008, 21(3):166-170.

9. Shimada H, Makizako H, Doi T, Yoshida D, Tsutsumimoto K, Anan Y, Uemura K, Lee S, Park H, Suzuki T: A large, cross-sectional observational study of serum BDNF, cognitive function, and mild cognitive impairment in the elderly. Front Aging Neurosci 2014, 6:69.

10. Teixeira AL, Barbosa IG, Diniz BS, Kummer A: Circulating levels of brain-derived neurotrophic factor: correlation with mood, cognition and motor function. Biomark Med 2010, 4(6):871-887.

11. Zhang XY, Chen DC, Xiu MH, Haile CN, Luo X, Xu K, Zhang HP, Zuo L, Zhang Z, Zhang X et al: Cognitive and serum BDNF correlates of BDNF Val66Met gene polymorphism in patients with schizophrenia and normal controls. Hum Genet 2012, 131(7):1187-1195.

12. Ng T, Lee YY, Chae JW, Yeo AHL, Shwe M, Gan YX, Ng RCH, Chu PPY, Khor CC, Ho HK et al: Evaluation of plasma brain-derived neurotrophic factor levels and self-perceived cognitive impairment post-chemotherapy: a longitudinal study. BMC Cancer 2017, 17(1):867.

13. Alhowail AH, Bloemer J, Majrashi M, Pinky PD, Bhattacharya S, Yongli Z, Bhattacharya D, Eggert M, Woodie L, Buabeid MA et al: Doxorubicin-induced neurotoxicity is associated with acute alterations in synaptic plasticity, apoptosis, and lipid peroxidation. Toxicol Mech Methods 2019, 29(6):457-466.

14. Molinari C, Morsanuto V, Ruga S, Notte F, Farghali M, Galla R, Uberti F: The Role of BDNF on Aging-Modulation Markers. Brain Sci 2020, 10(5).

15. Nagahara AH, Tuszynski MH: Potential therapeutic uses of BDNF in neurological and psychiatric disorders. Nat Rev Drug Discov 2011, 10(3):209-219.

16. Wilkinson ST, Kiselycznyk C, Banasr M, Webler RD, Haile C, Mathew SJ: Serum and plasma brain-derived neurotrophic factor and response in a randomized controlled trial of riluzole for treatment resistant depression. J Affect Disord 2018, 241:514-518.

17. Fessel J: Prevention of Alzheimer's disease by treating mild cognitive impairment with combinations chosen from eight available drugs. Alzheimers Dement (N Y) 2019, 5:780-788.

18. Gulyaeva NV: Hippocampal hyperglutamatergic signaling matters: Early targeting glutamate neurotransmission as a preventive strategy in Alzheimer's disease: An Editorial Highlight for "Riluzole attenuates glutamatergic tone and cognitive decline in AbetaPP/PS1 mice" on https://doi.org/10.1111/jnc.15224. J Neurochem 2020.

19. Whitcomb DJ, Molnar E: Is riluzole a new drug for Alzheimer's disease? J Neurochem 2015, 135(2):207-209.

20. Gourley SL, Espitia JW, Sanacora G, Taylor JR: Antidepressant-like properties of oral riluzole and utility of incentive disengagement models of depression in mice. Psychopharmacology (Berl) 2012, 219(3):805-814.

21. Hunsberger HC, Weitzner DS, Rudy CC, Hickman JE, Libell EM, Speer RR, Gerhardt GA, Reed MN: Riluzole rescues glutamate alterations, cognitive deficits, and tau pathology associated with P301L tau expression. J Neurochem 2015, 135(2):381-394.

22. Okamoto M, Gray JD, Larson CS, Kazim SF, Soya H, McEwen BS, Pereira AC: Riluzole reduces amyloid beta pathology, improves memory, and restores gene expression changes in a transgenic mouse model of early-onset Alzheimer's disease. Transl Psychiatry 2018, 8(1):153.

23. Barker GR, Bird F, Alexander V, Warburton EC: Recognition memory for objects, place, and temporal order: a disconnection analysis of the role of the medial prefrontal cortex and perirhinal cortex. J Neurosci 2007, 27(11):2948-2957.

24. Barker GR, Warburton EC: When is the hippocampus involved in recognition memory? J Neurosci 2011, 31(29):10721-10731.

25. Baulch JE, Acharya MM, Allen BD, Ru N, Chmielewski NN, Martirosian V, Giedzinski E, Syage A, Park AL, Benke SN et al: Cranial grafting of stem cell-derived microvesicles improves cognition and reduces neuropathology in the irradiated brain. Proc Natl Acad Sci U S A 2016, 113(17):4836-4841.

26. Christie LA, Acharya MM, Parihar VK, Nguyen A, Martirosian V, Limoli CL: Impaired cognitive function and hippocampal neurogenesis following cancer chemotherapy. Clin Cancer Res 2012, 18(7):1954-1965.

27. Smith SM, Giedzinski E, Angulo MC, Lui T, Lu C, Park AL, Tang S, Martirosian V, Ru N, Chmielewski NN et al: Functional equivalence of stem cell and stem cell-derived extracellular vesicle transplantation to repair the irradiated brain. Stem Cells Transl Med 2020, 9(1):93-105.

28. Acharya MM, Martirosian V, Chmielewski NN, Hanna N, Tran KK, Liao AC, Christie LA, Parihar VK, Limoli CL: Stem cell transplantation reverses chemotherapy-induced cognitive dysfunction. Cancer Res 2015, 75(4):676-686.

29. Huang EJ, Reichardt LF: Neurotrophins: roles in neuronal development and function. Annu Rev Neurosci 2001, 24:677-736.

30. Leal G, Bramham CR, Duarte CB: BDNF and Hippocampal Synaptic Plasticity. Vitam Horm 2017, 104:153-195.

31. Mizuta I, Ohta M, Ohta K, Nishimura M, Mizuta E, Kuno S: Riluzole stimulates nerve growth factor, brain-derived neurotrophic factor and glial cell line-derived neurotrophic factor synthesis in cultured mouse astrocytes. Neurosci Lett 2001, 310(2-3):117-120.

32. Yap NY, Toh YL, Tan CJ, Acharya MM, Chan A: Relationship between cytokines and brain-derived neurotrophic factor (BDNF) in trajectories of cancer-related cognitive impairment. Cytokine 2021, 144:155556.

33. Lu B, Nagappan G, Lu Y: BDNF and synaptic plasticity, cognitive function, and dysfunction. Handb Exp Pharmacol 2014, 220:223-250.

34. Morris RG, Anderson E, Lynch GS, Baudry M: Selective impairment of learning and blockade of long-term potentiation by an N-methyl-D-aspartate receptor antagonist, AP5. Nature 1986, 319(6056):774-776.

35. Park HS, Kim CJ, Kwak HB, No MH, Heo JW, Kim TW: Physical exercise prevents cognitive impairment by enhancing hippocampal neuroplasticity and mitochondrial function in doxorubicin-induced chemobrain. Neuropharmacology 2018, 133:451-461.

36. Winocur G, Wojtowicz JM, Huang J, Tannock IF: Physical exercise prevents suppression of hippocampal neurogenesis and reduces cognitive impairment in chemotherapy-treated rats. Psychopharmacology (Berl) 2014, 231(11):2311-2320.

37. Vivar C, Potter MC, van Praag H: All about running: synaptic plasticity, growth factors and adult hippocampal neurogenesis. Curr Top Behav Neurosci 2013, 15:189-210.

38. Gooch M, Mehta A, John T, Lomeli N, Naeem E, Mucci G, Toh YL, Chan A, Bota DA, Torno L: Feasibility of Cognitive Training to Promote Recovery in Cancer-Related Cognitive Impairment in Adolescent and Young Adult Patients. J Adolesc Young Adult Oncol 2021.

39. Palmer ACS, Zortea M, Souza A, Santos V, Biazus JV, Torres ILS, Fregni F, Caumo W: Clinical impact of melatonin on breast cancer patients undergoing chemotherapy; effects on cognition, sleep and depressive symptoms: A randomized, double-blind, placebo-controlled trial. PLoS One 2020, 15(4):e0231379.

40. Tong T, Pei C, Chen J, Lv Q, Zhang F, Cheng Z: Efficacy of Acupuncture Therapy for Chemotherapy-Related Cognitive Impairment in Breast Cancer Patients. Med Sci Monit 2018, 24:2919-2927.

41. Hartman SJ, Weiner LS, Nelson SH, Natarajan L, Patterson RE, Palmer BW, Parker BA, Sears DD: Mediators of a Physical Activity Intervention on Cognition in Breast Cancer Survivors: Evidence From a Randomized Controlled Trial. JMIR Cancer 2019, 5(2):e13150.

42. Karschnia P, Parsons MW, Dietrich J: Pharmacologic management of cognitive impairment induced by cancer therapy. Lancet Oncol 2019, 20(2):e92-e102.

43. Mayo SJ, Lustberg M, H MD, Nakamura ZM, Allen DH, Von Ah D, M CJ, Chan A, Olson K, Tan CJ et al: Cancer-related cognitive impairment in patients with non-central nervous system malignancies: an overview for oncology providers from the MASCC Neurological Complications Study Group. Support Care Cancer 2020.

44. Boukelmoune N, Laumet G, Tang Y, Ma J, Mahant I, Singh SK, Nijboer C, Benders M, Kavelaars A, Heijnen CJ: Nasal administration of mesenchymal stem cells reverses chemotherapy-induced peripheral neuropathy in mice. Brain Behav Immun 2021, 93:43-54.

45. Koh YQ, Tan CJ, Toh YL, Sze SK, Ho HK, Limoli CL, Chan A: Role of Exosomes in Cancer-Related Cognitive Impairment. Int J Mol Sci 2020, 21(8).

46. Azimi M, Gharakhanlou R, Naghdi N, Khodadadi D, Heysieattalab S: Moderate treadmill exercise ameliorates amyloid-beta-induced learning and memory impairment, possibly via increasing AMPK activity and up-regulation of the PGC-1alpha/FNDC5/BDNF pathway. Peptides 2018, 102:78-88.

47. Palasz E, Wysocka A, Gasiorowska A, Chalimoniuk M, Niewiadomski W, Niewiadomska G: BDNF as a Promising Therapeutic Agent in Parkinson's Disease. Int J Mol Sci 2020, 21(3).

48. Sanchez-Villegas A, Galbete C, Martinez-Gonzalez MA, Martinez JA, Razquin C, Salas-Salvado J, Estruch R, Buil-Cosiales P, Marti A: The effect of the Mediterranean diet on plasma brain-derived neurotrophic factor (BDNF) levels: the PREDIMED-NAVARRA randomized trial. Nutr Neurosci 2011, 14(5):195-201.

49. Hascup KN, Findley CA, Britz J, Esperant-Hilaire N, Broderick SO, Delfino K, Tischkau S, Bartke A, Hascup ER: Riluzole attenuates glutamatergic tone and cognitive decline in AbetaPP/PS1 mice. J Neurochem 2020.

50. Lesuis SL, Kaplick PM, Lucassen PJ, Krugers HJ: Treatment with the glutamate modulator riluzole prevents early life stress-induced cognitive deficits and impairments in synaptic plasticity in APPswe/PS1dE9 mice. Neuropharmacology 2019, 150:175-183.