Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 May 14.
Published in final edited form as: Methods Cell Biol. 2023 Mar 25;180:177–197. doi: 10.1016/bs.mcb.2023.02.015

A rigorous behavioral testing platform for the assessment of radiation-induced neurological outcomes

Olivia GG Drayson a, Marie-Catherine Vozenin b,*, Charles L Limoli a
PMCID: PMC11093273  NIHMSID: NIHMS1984528  PMID: 37890929

Abstract

Behavioral testing is a popular and reliable method of neurocognitive assessment of rodents but the lack of standard operating procedures has led to a high variation of protocols in use. Therefore, there exists a strong need to standardize protocols for a combined behavioral platform in order to maintain consistency across institutions and assist newcomers in the field. This paper provides details on the methodology of several behavioral tasks which have been validated in identifying radiation induced cognitive impairment as well as provide guidance on timescales and best practices. The cognitive assessments outlined here are optimized for rodent studies and either target learning and memory (open field task, object in updated location, novel object recognition, object in place, and temporal order) or mood and cognition (social interaction, elevated plus maze, light dark box, forced swim test, and fear extinction). We have utilized this platform successfully in evaluating cognitive injury induced by various radiation types, doses, fractionation schedules and also with ultra-high dose rate FLASH radiotherapy. Recommended materials and software are provided as well as advice on methods of data analysis. In this way a comprehensive behavioral platform is described with broad applicability to assess cognitive endpoints critical to therapeutic outcome.

1. Introduction

Here we describe a comprehensive mouse behavioral testing platform designed to quantify how specific irradiation protocols impact sensitive paradigms of learning, memory, and mood - multifaceted endpoints critical to therapeutic outcome. In this review, we focus on the effects of cranial irradiation (Meyers, 2000), known to be a frontline treatment for the control of CNS malignancies. Nonetheless, our platform has broad applicability, and can be adapted to virtually any irradiation or chemical exposure scenario (whole body, chemotherapy, toxicant) relevant to environmental, occupational, diagnostic, or therapeutic activities where an assessment of follow-up cognitive outcomes is desired. For the purposes of this review, preclinical rodent models subjected to clinically relevant irradiation scenarios are evaluated using selected testing paradigms that have been optimized and validated to uncover potential treatment-induced changes in learning and memory, cognitive flexibility, anxiety, depression, and fear memory extinction spanning multiple regions of the brain (Alaghband et al., 2020; Montay-Gruel et al., 2019, 2021). Importantly, centralizing these activities under rigorous standard operating procedures helps to minimize potential confounders, streamline the approach, and improve animal throughput to provide the necessary sample sizes to ensure statistical rigor and reproducibility.

In the majority of patients with CNS tumors radiation-induced cognitive injury will develop, often including progressive and debilitating impairments in memory, attention and executive function. Preclinical animal studies have employed cognitive assessments including those described in this paper, to evaluate the progression and mechanism of cranial radiation-induced cognitive decline (Acharya, Christie, Lan, et al., 2011; Atwood, Payne, Zhao, et al., 2007; Greene-Schloesser & Robbins, 2012; Meyers & Brown, 2006). Chemotherapy is similarly capable of inducing progressive cognitive injury and so both chemotherapy and combined chemotherapy-radiotherapy treatments have utilized these techniques in tumor-free rodents (Chmielewski-Stivers, Petit, Ollivier, et al., 2021; Dey et al., 2020; John, Kinra, Mudgal, et al., 2021; Loman, Jordan, Haynes, et al., 2019). Tumor-burdened animals could also be tested in this way (Feng, Liu, Chen, Rosi, & Gupta, 2018; Montay-Gruel et al., 2021), but since animals typically succumb to tumors before late radiation effects manifest, and it is a great challenge to separate the impact of the radiotherapy from that of the orthotopic tumor itself, this method is rarely implemented.

In vivo genetically engineered mouse models of primary human brain tumors have been developed and implemented successfully to study pathobiology and efficacy of potential treatments (Akter et al., 2021; Hambardzumyan, Amankulor, Helmy, Becher, & Holland, 2009; Simeonova & Huillard, 2014). While radiotherapy induced cognitive dysfunction is associated with high doses of radiation delivered exclusively to the CNS, the comparatively low doses of mixed-field radiation that astronauts will be exposed during space missions outside Earth’s magnetosphere have been shown to also illicit cognitive deficits observed through behavioral assessments (Acharya, Baulch, Klein, et al., 2019; Parihar et al., 2016; Parihar, Angulo, Allen, et al., 2020). The recently published Phase II NCRP report recommends approaches for conducting behavioral testing in order to study the risk of CNS decrements induced by space radiation exposure (Braby, Raber, Chang, et al., 2019).

Most recently, we have implemented this comprehensive approach in a neurobehavioral platform specifically focused on quantitative assessments of behavioral outcomes in rodents subjected to an innovative irradiation modality (FLASH radiotherapy, FLASH-RT). FLASH-RT involves the delivery of ionizing radiation at ultra-high dose rates (>100 Gy/s) compared to more conventional standard of care dose rates (~0.03 Gy/s) used in the clinic and has shown considerable promise at improving the therapeutic index of cancer treatments and hence an opportunity at revolutionizing clinical radiotherapy (Bourhis, Montay-Gruel, et al., 2019; Bourhis, Sozzi, et al., 2019; Harrington, 2019). At the heart of this promising technology lies the ability to elicit isoefficient tumor growth delay while minimizing significantly normal tissue complications, which has been termed the “FLASH effect” (Montay-Gruel et al., 2021; Vozenin et al., 2019; Vozenin, Hendry, & Limoli, 2019). Our group has a long and extensive history in quantifying behavioral decrements following conventional cranial dose rate irradiation and have used the behavioral tasks described below to assess the impact of various irradiation paradigms on multiple behavioral outcomes (Acharya et al., 2009; Baulch et al., 2016; Leavitt, Acharya, Baulch, & Limoli, 2020). The capability to reproducibly quantify behavioral outcomes has been essential to validate the in vivo FLASH effect, as our more recent reports have been able to demonstrate successfully the protective neurocognitive benefits of FLASH irradiation compared to conventional irradiation using preclinical mouse models (Alaghband et al., 2020; Montay-Gruel et al., 2017, 2019, 2021). For example, comparisons between FLASH-RT and conventional RT from a single fraction whole brain irradiation of 6MeV electrons have shown significant impacts of conventional RT relative to control in memory updating (OUL), object memory (NOR, OiP, TO), and anxiety-like behaviors (LDB, SIT, FST, EPM and FE), which were not found with FLASH-RT (Alaghband et al., 2020; Montay-Gruel et al., 2019). In this regard, we have recognized the importance of providing our multi-institutional collaborators a standardized behavioral platform for intercomparisons of different irradiation modalities (FLASH vs conventional dose rate irradiation), beam parameters, total dose, and fractionation protocols. This critical and necessary prerequisite stands essential to cross-validate different beams across multiple institutions, so that firm conclusions can be drawn from behavioral data in efforts to validate in vivo, the neurocognitive benefits of this promising new technology.

Tests for learning and memory and for disruptions in mood and cognition take advantage of established testing paradigms developed over the years in the field of behavioral neuroscience, selected and/or adapted to provide enhanced sensitivity and cross species relevance. Therefore, datasets derived from carefully conducted tasks can be linked more directly with performance changes observed between rodents and humans. Here we highlight a recent adaptation of a Novel Location Memory task termed the Objects in Updated Locations (OUL) task that provides a cross-species relevant approach for analyzing multiple memory traces in a single behavioral paradigm (Kwapis et al., 2018; Merhav, Riemer, & Wolbers, 2019). This coupled with optimized approaches to Social Interaction Test (SIT), and Fear Extinction (FE) tasks provide highly relevant datasets useful for ascertaining critical benchmarks for such cross-species comparisons of learning and memory (Allen, Morris, Mattfeld, Stark, & Fortin, 2014; Bakker, Kirwan, Miller, & Stark, 2008; Bennett & Stark, 2016; Milad & Quirk, 2002). These tasks can be coupled with more traditional open field exploratory behavioral testing paradigms to provide a comprehensive assessment of the types and magnitude of radiation-induced changes caused by a variety of cranial irradiation paradigms.

2. Materials

Common Consumables/Reagents

  • Bedding

  • 10% Ethanol (for cleaning toys and arena walls in-between rounds)

  • 70% Ethanol (for cleaning at end of day)

Common Equipment

  • 4 square, white arenas (30 cm x 30 cm x 30 cm)

  • Colored tape (for orientation marker)

  • Plastic toys

  • Small magnets (to affix toys to arena floor)

  • Camera (ceiling-mounted, USB 3.0 monochrome CMOS camera (539 fps and 20-megapixel resolution))

  • Lamp (for dim lighting environment of 48 lux)

  • Timer

  • Noldus Ethovision XT software (version 8, Noldus Information Technology)

Task Specific Equipment

  • Elevated Plus Maze (EPM)
    • ELM apparatus including 4 arms in shape of plus sign, elevated from the ground, 2 closed and 2 open arms
  • Light Dark Box (LDB)
    • LDB arena including light compartment (30 × 30 × 27 cm, 915 lux) and dark component (15 × 10 × 27 cm, 4 lux) with a small opening between (7.5 × 7.5 cm)
  • Forced Swim Test (FST)
    • Beaker (20 cm tall with inner diameter of 15 cm)
  • Fear Extinction (FE)
    • Set of eight behavioral conditioning chambers (17.5 × 17.5 × 18 cm, Coulbourn instruments) with steel slat floors (3.2 mm diameter, 8 mm spacing (see Fig. 10)). Chambers are located in sound-attenuating cabinets. The room is lit only by red light lamp for the benefit of the operators but each chamber contains a light as well as a shock source capable of delivering foot shock (0.6 mA) and a speaker for delivery of the tones (16 kHz, 80 dB).
    • FreezeFrame, Harvard Apparatus, Inc
    • Virkon disinfectant
    • Two distinct odor cues (such as 10% vinegar and 10% almond extract)

FIG. 10.

FIG. 10

Open in a new tab

(A) FE arrangement with 3 tone-shock pairings on day 1, 20 tones only on days 2–4 and 3 tones only on day 5. (B) Fear Conditioning (left) and Fear Extinction (right) apparatus with context A (vinegar) and B (almond) scents.

3. Methods

3.1. Animals

The first steps in assessing animals for any behavioral paradigm, whether in house or arriving from off-site locations, is to ensure they are properly acclimated (following any mandatory quarantine periods) to the vivarium for 1–2 weeks. Under certain situations (i.e., if animals are tumor bearing), quarantine periods may be reduced to 2–4 days to facilitate a more rapid testing schedule. Mice are typically maintained in standard housing conditions (20 °C ±1 °C, 70% ±10% humidity, 12 h:12 h light/dark cycle) and provided ab libitum access to standard rodent chow and water (unless otherwise mandated by experimental specifics and/or IACUC regulations). Prior to nearly any behavioral testing paradigm, investigators will also need to implement a defined period of animal handling to familiarize the animals with the personnel conducting the testing, and the period of this handling may need to be adjusted longer for older animals.

The neurobehavioral platform outlined in this review is intended to be run longitudinally on a single cohort of mice. The tasks to be included and the task order are adapted depending on the regions of brain under investigation, wellbeing of the mice (such as tumor burden, travel time or radiation dose), and goals of the strategy. It is strongly recommended to begin with the OUL task due to the length of the task and the interrogation of multiple memory traces. Similarly, the FE task must be administered last due to the electric shock fear memory created that would significantly affect animal behavior in subsequent tests.

The primary disadvantage of conducting longitudinal tests in this way is that experimental animals may show a tendency of fatigue or lethargy after repetitive testing. Therefore, care must be taken to minimize the number of tasks in the testing regimen while still including those relevant to the research goals. Many tests however do provide environmental enrichment which can counteract any onset of fatigue in the experimental animals. Implementation of different testing arenas/contexts and adjustments to the interval between tasks can also help offset testing fatigue. If multiple learning and memory tasks are to be conducted on a single cohort it is paramount to use clearly distinct objects for each task. Tests can also be repeated on a cohort as long as the objects are changed, and a significant interval (>1 week) is kept between the tests. If these constraints are adhered to, the investigator can modify the testing schedule as desired to meet the needs of their project.

3.2. Learning and memory tasks

The testing paradigm described below is designed to quantify radiation-induced deficits in learning and memory through (i) Open Field Testing (OFT), (ii) Object in Updated Location (OUL), (iii) Novel Object Recognition (NOR), (iv) Object in Place (OiP), and lastly (v) Temporal Order (TO). As alluded to above, the longitudinal behavior tests can be administered in any order based on experimental priorities. In our hands, we can recommend the following task order, which is largely based on task rigor, duration of the task, and project driven goals that place a priority on tasks focused on a variety of learning and memory paradigms that generally rely on open field exploration. Other factors need to be considered, such as the need to assess acute vs chronic outcomes, which impacts the overall duration of a given behavioral paradigm as well as the time allotted between longitudinal tests. For parallel testing, where the need to collect data at temporally coincident or restrained timeframes outweighs other variables, then the order of tasks becomes secondary, and the need for higher throughput of larger independent cohorts through a more limited series of tasks becomes rate-limiting. Tasks implementing exploratory behavior are dependent on many variables, such as the specifics of individual objects, an important factor that often necessitates the generation of an “interest index” between different objects (important for the NOR task) that gauges the level of novelty a priori for task optimization. Regardless of the selected strategy, the power calculations for the majority of our behavioral platform routinely call for cohort sizes of at least n = 12 to 16, and clearly depend on age, sex, strain and/or insult under investigation.

3.2.1. Open field task (OFT) and locomotor activity

The open field task (OFT) is a sensorimotor test used to determine general activity levels, gross locomotor activity, and exploration habits in rodents (Kulesskaya & Voikar, 2014). The animal is allowed to freely move about a large, square arena for 10 min where spontaneous activity is recorded (Fig. 1). The Open Field Task is naturally incorporated into the habituation phases of the OUL task (described below) and provides an additional behavioral measure easily derived from nearly any exploration task. The OFT is simple and useful for assessing overt changes in behavior (activity, exploration, anxiety) caused by a variety of cancer related treatments, generalized stress, genetic background and/or pharmacological manipulations, as detailed previously (Dey et al., 2020).

FIG. 1.

FIG. 1

Open in a new tab

Automated tracking of mouse center point during OFT.

3.2.2. Objects in updated location (OUL) task

This task is typically the first test we implement as it requires the longest period (~2 weeks) to complete due to handling and testing requirements. The OUL task is specifically designed as a cross-species task (mouse and human versions exist) that can provide a more direct link between performance changes across species. The task draws upon and extends the standard Novel Location Memory paradigm (NLM) to provide additional information on multiple memory traces over a single behavioral paradigm, where test animals must be able to discriminate between prior overlapping associative memories to recognize the most recent novel locations (Kwapis et al., 2018; Merhav et al., 2019). Objects are presented in pairs of locations at training and later, one object’s location is changed for the update task. On the last day, memory is probed by quantifying exploration of objects presented at all 4 possible locations. In mice, one set of four physical objects is needed for the full task.

The overall scheme of the OUL task is illustrated in Fig. 2. Mice are first habituated to the context for 5 min per day for 4 to 6 consecutive days (during which the OFT described above can be incorporated). For the OUL task it is necessary for the arena to contain a marker of orientation such as masking tape running vertically up one wall. Next, mice undergo object location memory training for 10 min per day for between 3 and 7 consecutive days in which 2 identical objects are positioned in distinct locations (locations A1 and A2). The following day, mice are given a 5-min update session in which one object is moved to a new location and the amount of time spent exploring the moved vs. familiar location is then evaluated (A1 and A3). The next day, mice are tested for 5 min with all four identical objects in distinct locations (A1, A2, A3, and A4) to determine whether the original memory was updated to incorporate the new location information. Time spent exploring the novel object location compared with the updated, original, and initial object locations can then be examined. Memory for the update can be inferred by comparing exploration of the novel object (A4) to that of the updated object (A3). Memory of the original training information can be inferred by comparing exploration of the novel object (A4) to that of the original object (A1). Preference for novel objects is quantified by calculating the Discrimination Index (DI):

DI=(tnovel−tfamiliar)(tnovel+tfamiliar)×100 (1)

where t represents the total time spent exploring the designated object. Exploration is defined as when the animal’s nose is facing towards the object and is within 1 cm. Climbing onto the object, rearing, biting the object or digging nearby are not considered exploration. Animals with nominal brain function will show a tendency to explore novelty, therefore the discrimination indices of both the original and updated memory should be diminished in animals with impaired cognition.

FIG. 2.

FIG. 2

Open in a new tab

OUL arrangement with identical objects in positions A1(original), A2(initial), A3(updated), and A4(novel).

The following tests provide examples of other open field tasks that can be implemented in no particular order but would follow OUL due to their shorter durations of testing, typically requiring from 2 to 5 days. Additional habituation may or may not be necessary and will ultimately depend on the interval between tasks, task repetition and study goals. These tasks can include Novel Object Recognition (NOR), Object in Place (OiP) and Temporal Order (TO) which assess short-term episodic memory dependent on intact hippocampal and/or cortical function (Barker, Bird, Alexander, & Warburton, 2007; Barker & Warburton, 2011).

3.2.3. Novel object recognition (NOR)

The novel object recognition test is one of the most commonly used behavior assessments as it is quick to carry out and can be customized to assess different memory consolidation timeframes. Whereas the OUL task is entirely hippocampal dependent, studies show that the NOR depends predominantly on the post-rhinal and insular cortices as well as hippocampal and medial prefrontal cortex function (Alaghband et al., 2020). Therefore, these tasks are often performed sequentially to assess both spatial and recognition memory impairment.

For the NOR task, animals are habituated to the empty arena (no orientation marker needed) for 4 to 6 consecutive days. If NOR is to quickly proceed OUL, habituation can be reduced to 1 day since the same chambers can be used for both tasks (Fig. 12). On the following day the animals are trained for 5 min with two identical objects affixed at distinct locations. The animals are returned to their cages for 5 min of consolidation while one of the objects is replaced with a novel object (differing in size, shape, and color) in the same location (Fig. 3). The animal is then returned to the arena and allowed to explore for 5 min. Given that this task relies on exploration of a novel object, the object choice is highly influential. The two object types must be as distinctive as possible in order to encourage strong discrimination. Counterbalancing can be conducted as with OiP if there is concern that animals have an innate preference to one object type over the other. Unlike the OUL task, training and testing can take place on the same day when assessments of short-term memory are desired. To investigate long-term memory (memory consolidation) a longer interval must be used between the training and testing phases. In this instance, that interval is typically 24 h, providing ample time for consolidation of prior object memories. The Discrimination Index (DI) is calculated (Eq. 1) by comparing the exploration time of the novel object and that of the original object in the testing round.

FIG. 12.

FIG. 12

Open in a new tab

An example timeline for a full behavioral testing paradigm including a combination of learning and memory, and mood and cognition tasks (created with BioRender). The entire testing procedure should take no more than 4 weeks and can be conducted on a single cohort of mice. This order of tasks is recommended as it has been designed to minimize interference between tasks. Note that if the SIT is to be included, the social mice must arrive and be allowed to adjust to new surroundings a week before the test and should be handled in the 3 days preceding SIT (during LDB and NOR in this case).

FIG. 3.

FIG. 3

Open in a new tab

NOR arrangement with familiar object (green triangle) and novel object (orange circle).

3.2.4. Object in place (OiP)

The OiP task is a test of associative recognition memory involving four objects of varying size, color, and shape. The day after habituation, mice are trained on the initial configuration once for 5 min. The animals are then returned to their cages for 5 min just as in the NOR task. During these 5 min of memory consolidation, 2 of the 4 objects switch locations and the other 2 remain in place. Mice are then returned to the arena for 5 min of testing (Fig. 4). The total exploration time for each object is manually scored and the DI is calculated (Eq. 1) comparing the total time at the switched objects to the time at the fixed objects. Given that four different objects are used in this test, mice may show a preference to certain types than others. To remove this potential bias the objects can be counterbalanced in which the switched objects are changed for half of the cohort.

FIG. 4.

FIG. 4

Open in a new tab

OiP arrangement with 2 possible test configurations if counterbalancing.

3.2.5. Temporal order (TO)

This task assesses temporal order memory, which is also dependent on interactions between the hippocampus, medial prefrontal, and perirhinal cortices. In this task there are two training phases followed by one test trial. Mice are presented with two identical objects separated in location within the arena just as in the training phase of NOR. In the second training phase, the objects are replaced by a second pair of identical objects which differ in size, shape, and color from the original pair. In the test phase, one from each object pair is presented to the animals and the exploration time at each object is measured. Animals with intact brain function show a preference for the object presented earlier in time (blue diamond in Fig. 5). The DI is measured by comparing the exploration of the first object type to that of the second object type.

FIG. 5.

FIG. 5

Open in a new tab

TO arrangement with earlier (blue diamond) and later (yellow star) presented objects.

3.3. Mood and cognition tasks

Disruptions to mood and anxiety-like behaviors can be studied using the following test regimen, (i) Social Interaction Test (SIT), (ii) Elevated Plus Maze (EPM), (iii) Light Dark Box (LDB), and (iv) Forced Swim Test (FST). Based on prior success, it is recommended that the tests be performed in the order as listed but it is not required. The EPM and the LDB tests report similar behaviors and therefore both do not need to be administered. Power calculations typically call for cohort sizes between n = 12 to 16 as with the learning and memory tasks. If embarking on this series of tasks, allow an overall time of 4 weeks for completion.

Cognitive flexibility is a higher form of executive function that can be analyzed through a variety of behavioral tasks. In our hands, we assess this form of cognition through the fear extinction (FE) test, which provides a readout on extinction memory. This task has many variations and has been shown to translate well to analogs of human testing. Because in rodents this task involves a mild aversive electrical shock, it is the terminal behavioral paradigm we administer.

3.1.1. Social interaction test (SIT)

Social interaction is a basic complex behavior that can be quantified in the laboratory setting. The precise analysis of social interaction behavior requires recognition of key social interactions including sniffing in active contact with the “stranger/intruder” animal’s snout, flank, or anogenital area, grooming, or pursuing the “stranger” animal as it actively explores a cage (Gunaydin et al., 2014; Selimbeyoglu et al., 2017; Winslow, 2003; Winslow & Insel, 2004). Importantly, the reciprocal of these interactions can be recorded as avoidance behaviors. Such quantification captures social and environmental situations that occur in natural habitats. The overall protocol is simple, in which behaviors are video-recorded and analyzed to assess active interaction time in a test mouse with a novel mouse (Sato, Mizuguchi, & Ikeda, 2013). This protocol involves direct interaction of two animals without any barrier, thus deficits are reflected by reduction of exploratory behavior toward a novel, same sex, animal.

In the SIT protocol (Sato et al., 2013), the test mouse is habituated alone in the arena for 15 min for 2 days. There is also a handling and habituation step for the strange/intruder mouse prior to testing. On the day of testing, the test mouse is allowed to explore the open field arena for 10 mins after which, a novel mouse (body weight similar or less to the test mouse) is introduced and allowed to remain in the arena for 10 mins. The active social interaction is recorded from a ceiling-mounted, closed-circuit digital camera (Fig. 6). Experimenters blinded to the cohorts then score the social interaction or avoidance behavior from the videos. Aggressive behavior is not treated as a dependent measure and does not contribute to the interaction or avoidance scores. The total time spent interacting and total time spent avoiding are the endpoints recorded from this task.

FIG. 6.

FIG. 6

Open in a new tab

Photo of example interaction and avoidance behaviors.

3.3.2. Elevated plus maze (EPM)

The EPM task investigates anxiety-like behavior using four elevated arms, two of which are exposed and two are enclosed. The animal can explore five areas: the open arms, the closed arms, or the central zone where the arms intersect. Each animal should be placed in the maze in the central zone facing towards a closed arm (Fig. 7). The test lasts 5 min and the total time in each zone as well as the number of transitions are recorded. The exposed arms create a risk of falling and therefore act as a mild stressor. Anxious mice are likely to transition more frequently between the open and closed arms and spend more time in the closed arms than less anxious mice.

FIG. 7.

FIG. 7

Open in a new tab

Schematic of ELM apparatus from above and side view.

3.3.3. Light dark box (LDB)

The LDB test is designed to test anxiety-like behavior in rodents. This test was designed by Crawley and Goodwin (Crawley & Goodwin, 1980) to relate exploratory behavior as an index for the anxiolytic effect of benzodiazepines. The light dark test utilizes a box (45 cm x 30 cm x 27 cm) where one third of the box is a covered dark chamber and the other two thirds is an open light chamber. Both the chambers are connected with an opening of (7.5 cm x 7.5 cm) which allows the mice to move between the light and dark compartments (Fig. 8). Novel environments and light act as mild stressors to the rodents, which impact their spontaneous exploratory behavior in the testing environment. Mice spend 10 min in the light box arena without any prior habituation or training. An overhead camera records the mouse when it is in the light chamber. The number of transitions between the compartments and the time spent in each compartment is recorded to quantify performance on this test.

FIG. 8.

FIG. 8

Open in a new tab

Schematic of LDB apparatus with passage between light region and dark region.

Entry into a chamber is defined as four paws in the chamber. Mice injected with anxiolytics spend more time in the lighted chamber and have increased transitions whereas anxiogenics decrease both of these measures (Imaizumi, Miyazaki, & Onodera, 1994). In general, radiation-studies have been shown to reduce the amount of time spent in the brightly lit area and/or the number of transitions between light and dark regions, interpreted to represent a generalized increase in anxiety.

3.3.4. Forced swim test (FST)

The forced swim test is a relatively straightforward and widely used model for testing depression, as originally adopted by Porsolt, Bertin, & Jalfre (1978). Naïve mice forced to swim in a transparent cylinder (aversive and confined environment) innately fight to escape the apparatus. Following failed attempts to escape, they become immobile (i.e., float), a behavior generally considered as despair or “depressive-like”. Prior treatment with antidepressants decreases overall immobility and increases the latency to reach the first immobility episode.

Each mouse is placed in a glass beaker (with an inner diameter of 15 cm and depth of 20 cm) filled with tap water (20–25 °C) to a depth of 16 cm. The water depth must be sufficient to prevent the mouse from reaching the bottom of the container with their hind limbs or tail and therefore should be adjusted for size of the mouse (Fig. 9). The test lasts 6 min after which the mouse is dried off and returned to their cage. The test is recorded and the immobility time is scored by independent observers. Immobility or floating is defined as the minimum movement needed to keep the mouse’s head above water. While this test is less often employed than the LDB, past reports have found that cranial irradiation increases immobility times (i.e., floating), interpreted to represent increased depression-like behavior.

FIG. 9.

FIG. 9

Open in a new tab

Illustration of FST with glass beaker.

3.3.5. Fear extinction (FE)

The fear extinction memory task is a modified version of the Fear Conditioning task (Parihar et al., 2016, 2018) in which a smaller number of tone-shock pairings are administered such that extinction (dissociation of the tone-shock) can be followed under a reasonable time frame. The latency to forget the association between the tone and shock provide a quantitative measure of extinction. Importantly, extinction memory is an active process of “unlearning” and a critical measure of cognitive flexibility that is highly relevant to many aspects of human behavior (Milad & Quirk, 2002; Sotres-Bayon, Sierra-Mercado, Pardilla-Delgado, & Quirk, 2012).

Testing is conducted in two similar contexts within a chamber with steel slat floors (Fig. 10). Fear conditioning is conducted in context A using a distinctive scent such as 10% acetic acid in water. Fear extinction and testing is conducted in context B, where the slats are covered by a white plastic panel and a contrasting scent such as 10% almond extract in water. After 2 min of habituation in context A, the fear conditioning phase is started which consists of three tone-shock pairings spaced 120 s apart. The tone (16 kHz, 80 dB) lasts 120 s, terminating with a 0.6 mA foot shock given by the metal slats of the chamber. The following 3 days are extinction training days where mice are habituated to context B for 2 min and then 20 tones are played spaced 5 s apart. The tones are identical in frequency, volume, and duration as during the conditioning phase. On the testing day the mice remain in context B and are presented with 3 tones 120s apart, similar to conditioning day. This particular testing paradigm interrogates both learning and memory elements dependent on intact hippocampal and medial prefrontal cortex (mPFC) function. If, however, the entire testing paradigm was conducted in the same context (context A) then the test would be reliant on the mPFC.

A camera records each chamber over the course of the full test paradigm. An automated scoring software “FreezeFrame” detects motion from the videos and can measure total freezing time based on a user defined motion threshold. An investigator blinded to treatment status manually adjusts the motion threshold on each video determining the distinction between immobility or freezing behavior and nominal motion. This threshold is set using the motion index histogram shown on FreezeFrame. In most cases there will be two peaks in the histogram, one sharp peak to show freezing at motion index 0, and another broader peak between 50 and 100 which indicates activity. The threshold should be set at the trough between these peaks which typically lies between 20 and 30 (Fig. 11). The percentage of time spent freezing is calculated for each mouse over the conditioning, extinction, and testing phases. Animals with intact brain function will exhibit a decrease in freezing behavior over the course of the fear extinction days. During the conditioning phase, the freezing behavior after each shock is analyzed and should increase with each shock if the fear memory is properly created. Past work analyzing behavior after various irradiation paradigms has shown that extinction memory is compromised, where animals subjected to conventional dose rate protocols routinely exhibit an inability to extinguish fear, evidenced by an increased amount of time freezing compared to controls.

FIG. 11.

FIG. 11

Open in a new tab

Examples of motion index histograms generated by FreezeFrame with the correct threshold set.

3.4. Behavioral analysis

Each animal behavior room is equipped with ceiling-mounted, USB 3.0 monochrome CMOS camera (539 fps and 20-megapixel resolution) connected with a dedicated computer. In addition, two tripod mounted USB cameras are available to achieve horizontal recording. An automated tracking software program (Ethovision XT, v. 8.0, Noldus Information Technology, Inc.) is capable of measuring position, velocity and distance travelled. It is implemented for OFT, the habituation phase of OUL/NOR/OiP/TO, and for the EPM and LDB tasks. Despite standardized protocols, experience has shown that certain elements of scoring criteria can be subjective, and particular attention to consistency regarding exploration and related/confounding behavior must be discriminated. For example, the proximal distance between the nose of the mouse and a given object that is considered exploration must be kept constant over the duration of the task, and evaluated against confounders such as escape behavior, climbing or digging. Automated tracking and analysis greatly reduce the experimenter time required to complete a particular task, allowing higher throughput for the suite of tasks proposed, but generally have to be validated against blinded manual counts. Exploration time is manually scored for OUL, NOR, OiP, TO, SIT and FST. The FE videos are automatically scored with a motion-tracking software (FreezeFrame, Harvard Apparatus, Inc.) but a manual threshold is needed to be set for each video by a blinded investigator. All requisite equipment, ceiling mounted cameras and computers are readily available from a variety of vendors to conduct the proposed studies as described (Fig. 12).

4. Summary

Here we have provided a brief template for constructing a comprehensive behavioral paradigm proven to be useful for the quantitative assessment of rodent behavioral performance. The list of tasks has been tailored to interrogate multiple behaviors indicative of select brain regions known (or suspected) of being sensitive to prior radiation exposure. The tasks listed are only a subset of available tests developed over the years by behavioral experts in a variety of fields and are not meant to be exclusive to any particular task, nor are they meant to indicate that this is the best or only option available to those already engaged or deciding to embark on behavioral testing. Clearly, there are many testing options available that must be evaluated based on study specific criteria and hypotheses. The driving force behind the development of the present neurobehavioral testing platform was to provide consistent, reliable, and translational behavioral assessments for both male and female mice subjected to distinct irradiation modalities across multiple institutional sites. Toward this end, we have adapted certain testing details to optimize the testing regimen based on study specifics (tumor bearing or tumor free), acute or chronic post irradiation assessments or to substitute certain tests with improved cross-species relevance. While no single test is not without caveats, experience has shown that an overreliance on a single test can be misleading and may miss more subtle decrements that require longer times to manifest. Similarly, radiation exposure has been shown to increase inter-individual variations within cohorts, so careful attention must be paid to power calculations of sample size. Nuances associated with behavioral testing also point to the need to have a standardized platform, where minor differences in experimental set-up and between those conducting the experiments can be minimized to optimize reproducibility and rigor. In our hands, implementation of this approach has yielded robust data sets to date, and points to the benefits of having a centralized core of activities to facilitate validation of neurobehavioral outcomes (our dependent variable) while other independent variables (beam parameters etc.) are changed to determine what does and does not lead to the biological “FLASH Effect” (Vozenin, Montay-Gruel, Limoli, & Germond, 2020). Similar approaches can no doubt be implemented for the behavioral analyses of other neuropathologies, with the addition or elimination of alternative tests tailored to specific experimental objectives.

Acknowledgments

This work was supported by P01CA244091 and R01CA254892 (CLL, M-CV).

References

  1. Acharya MM, Baulch JE, Klein PM, et al. (2019). New concerns for neurocognitive function during deep space exposures to chronic, low dose-rate, neutron radiation [published correction appears in eNeuro. 2019 Oct 18;6(5):]. eNeuro, 6(4). ENEURO. 0094–19.2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Acharya MM, Christie LA, Lan ML, Donovan PJ, Cotman CW, Fike JR, et al. (2009). Rescue of radiation-induced cognitive impairment through cranial transplantation of human embryonic stem cells. Proceedings of the National Academy of Sciences of the United States of America, 106, 19150–19155. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Acharya MM, Christie LA, Lan ML, et al. (2011). Human neural stem cell transplantation ameliorates radiation-induced cognitive dysfunction. Cancer Research, 71(14), 4834–4845. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Akter F, Simon B, de Boer NL, Redjal N, Wakimoto H, & Shah K. (2021). Pre-clinical tumor models of primary brain tumors: Challenges and opportunities. Biochimica Et Biophysica Acta. Reviews on Cancer, 1875(1), 188458. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Alaghband Y, Cheeks SN, Allen BD, Montay-Gruel P, Doan NL, Petit B, et al. (2020). Neuroprotection of radiosensitive juvenile mice by ultra-high dose rate FLASH irradiation. Cancers (Basel), 12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Allen TA, Morris AM, Mattfeld AT, Stark CE, & Fortin NJ (2014). A Sequence of events model of episodic memory shows parallels in rats and humans. Hippocampus, 24, 1178–1188. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Atwood T, Payne VS, Zhao W, et al. (2007). Quantitative magnetic resonance spectroscopy reveals a potential relationship between radiation-induced changes in rat brain metabolites and cognitive impairment. Radiation Research, 168(5), 574–581. [DOI] [PubMed] [Google Scholar]
  8. Bakker A, Kirwan CB, Miller M, & Stark CE (2008). Pattern separation in the human hippocampal CA3 and dentate gyrus. Science, 319, 1640–1642. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Barker GR, Bird F, Alexander V, & Warburton EC (2007). Recognition memory for objects, place, and temporal order: A disconnection analysis of the role of the medial prefrontal cortex and perirhinal cortex. The Journal of Neuroscience, 27, 2948–2957. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Barker GR, & Warburton EC (2011). When is the hippocampus involved in recognition memory? The Journal of Neuroscience, 31, 10721–10731. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Baulch JE, Acharya MM, Allen BD, Ru N, Chmielewski NN, Martirosian V, et al. (2016). Cranial grafting of stem cell-derived microvesicles improves cognition and reduces neuropathology in the irradiated brain. Proceedings of the National Academy of Sciences of the United States of America, 113, 4836–4841. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Bennett IJ, & Stark CE (2016). Mnemonic discrimination relates to perforant path integrity: An ultra-high resolution diffusion tensor imaging study. Neurobiology of Learning and Memory, 129, 107–112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Bourhis J, Montay-Gruel P, Goncalves Jorge P, Bailat C, Petit B, Ollivier J, et al. (2019). Clinical translation of FLASH radiotherapy: Why and how? Radiotherapy and Oncology. [DOI] [PubMed]
  14. Bourhis J, Sozzi WJ, Jorge PG, Gaide O, Bailat C, Duclos F, et al. (2019). Treatment of a first patient with FLASH-radiotherapy. Radiotherapy and Oncology, 139, 18–22. [DOI] [PubMed] [Google Scholar]
  15. Braby L, Raber J, Chang P, et al. (2019). Report No. 183—Radiation exposure in space and the potential for central nervous system effects: Phase II. NCRP. ISBN: 9781944888145. [Google Scholar]
  16. Chmielewski-Stivers N, Petit B, Ollivier J, et al. (2021). Sex-specific differences in toxicity following systemic paclitaxel treatment and localized cardiac radiotherapy. Cancers (Basel), 13(16), 3973. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Crawley J, & Goodwin FK (1980). Preliminary report of a simple animal behavior model for the anxiolytic effects of benzodiazepines. Pharmacology, Biochemistry, and Behavior, 13, 167–170. [DOI] [PubMed] [Google Scholar]
  18. Dey D, Parihar VK, Szabo GG, Klein PM, Tran J, Moayyad J, et al. (2020). Neurological impairments in mice subjected to irradiation and chemotherapy. Radiation Research, 193, 407–424. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Feng X, Liu S, Chen D, Rosi S, & Gupta N. (2018). Rescue of cognitive function following fractionated brain irradiation in a novel preclinical glioma model. eLife, 7, e38865. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Greene-Schloesser D, & Robbins ME (2012). Radiation-induced cognitive impairment—From bench to bedside. Neuro-Oncology, 14(Suppl. 4), iv37–iv44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Gunaydin LA, Grosenick L, Finkelstein JC, Kauvar IV, Fenno LE, Adhikari A, et al. (2014). Natural neural projection dynamics underlying social behavior. Cell, 157, 1535–1551. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hambardzumyan D, Amankulor NM, Helmy KY, Becher OJ, & Holland EC (2009). Modeling adult gliomas using RCAS/t-va technology. Translational Oncology, 2(2), 89–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Harrington KJ (2019). Ultrahigh dose-rate radiotherapy: Next steps for FLASH-RT. Clinical Cancer Research: An Official Journal of the American Association for Cancer Research, 25, 3–5. [DOI] [PubMed] [Google Scholar]
  24. Imaizumi M, Miyazaki S, & Onodera K. (1994). Effects of a non-xanthine adenosine antagonist, CGS 15943, and a phosphodiesterase inhibitor, Ro 20–1724, in a light/dark test in mice. Methods and Findings in Experimental and Clinical Pharmacology, 16, 717–721. [PubMed] [Google Scholar]
  25. John J, Kinra M, Mudgal J, et al. (2021). Animal models of chemotherapy-induced cognitive decline in preclinical drug development. Psychopharmacology, 238, 3025–3053. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kulesskaya N, & Voikar V. (2014). Assessment of mouse anxiety-like behavior in the light-dark box and open-field arena: Role of equipment and procedure. Physiology & Behavior, 133, 30–38. [DOI] [PubMed] [Google Scholar]
  27. Kwapis JL, Alaghband Y, Kramar EA, Lopez AJ, Vogel Ciernia A, White AO, et al. (2018). Epigenetic regulation of the circadian gene Per1 contributes to age-related changes in hippocampal memory. Nature Communications, 9, 3323. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Leavitt R, Acharya M, Baulch J, & Limoli C. (2020). Extracellular vesicle-derived miR-124 resolves radiation-induced brain injury. Cancer Res. 10.1158/0008-5472.CAN-20-1599. [DOI] [PMC free article] [PubMed]
  29. Loman BR, Jordan KR, Haynes B, et al. (2019). Chemotherapy-induced neuroinflammation is associated with disrupted colonic and bacterial homeostasis in female mice. Scientific Reports, 9, 16490. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Merhav M, Riemer M, & Wolbers T. (2019). Spatial updating deficits in human aging are associated with traces of former memory representations. Neurobiology of Aging, 76, 53–61. [DOI] [PubMed] [Google Scholar]
  31. Meyers CA (2000). Neurocognitive dysfunction in cancer patients. Oncology (Williston Park), 14, 75–79. discussion 9, 81–2, 5. [PubMed] [Google Scholar]
  32. Meyers CA, & Brown PD (2006). Role and relevance of neurocognitive assessment in clinical trials of patients with CNS tumors. Journal of Clinical Oncology, 24(8), 1305–1309. [DOI] [PubMed] [Google Scholar]
  33. Milad MR, & Quirk GJ (2002). Neurons in medial prefrontal cortex signal memory for fear extinction. Nature, 420, 70–74. [DOI] [PubMed] [Google Scholar]
  34. Montay-Gruel P, Acharya MM, Goncalves Jorge P, Petit B, Petridis IG, Fuchs P, et al. (2021). Hypofractionated FLASH-RT as an effective treatment against glioblastoma that reduces neurocognitive side effects in mice. Clinical Cancer Research: An Official Journal of the American Association for Cancer Research, 27, 775–784. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Montay-Gruel P, Acharya MM, Petersson K, Alikhani L, Yakkala C, Allen BD, et al. (2019). Long-term neurocognitive benefits of FLASH radiotherapy driven by reduced reactive oxygen species. Proceedings of the National Academy of Sciences of the United States of America, 116, 10943–10951. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Montay-Gruel P, Petersson K, Jaccard M, Boivin G, Germond JF, Petit B, et al. (2017). Irradiation in a flash: Unique sparing of memory in mice after whole brain irradiation with dose rates above 100Gy/s. Radiotherapy and Oncology, 124, 365–369. [DOI] [PubMed] [Google Scholar]
  37. Parihar VK, Allen BD, Caressi C, Kwok S, Chu E, Tran KK, et al. (2016). Cosmic radiation exposure and persistent cognitive dysfunction. Scientific Reports, 6, 34774. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Parihar VK, Angulo MC, Allen BD, et al. (2020). Sex-specific cognitive deficits following space radiation exposure. Frontiers in Behavioral Neuroscience, 14, 535885. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Parihar VK, Maroso M, Syage A, Allen BD, Angulo MC, Soltesz I, et al. (2018). Persistent nature of alterations in cognition and neuronal circuit excitability after exposure to simulated cosmic radiation in mice. Experimental Neurology, 305, 44–55. [DOI] [PubMed] [Google Scholar]
  40. Porsolt RD, Bertin A, & Jalfre M. (1978). “Behavioural despair” in rats and mice: Strain differences and the effects of imipramine. European Journal of Pharmacology, 51, 291–294. [DOI] [PubMed] [Google Scholar]
  41. Sato A, Mizuguchi M, & Ikeda K. (2013). Social interaction test: A sensitive method for examining autism-related behavioral deficits. Protocol Exchange, 1–9.
  42. Selimbeyoglu A, Kim CK, Inoue M, Lee SY, Hong ASO, Kauvar I, et al. (2017). Modulation of prefrontal cortex excitation/inhibition balance rescues social behavior in CNTNAP2-deficient mice. Science Translational Medicine, 9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Simeonova I, & Huillard E. (2014). In vivo models of brain tumors: roles of genetically engineered mouse models in understanding tumor biology and use in preclinical studies. Cellular and Molecular Life Sciences, 71(20), 4007–4026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Sotres-Bayon F, Sierra-Mercado D, Pardilla-Delgado E, & Quirk GJ (2012). Gating of fear in prelimbic cortex by hippocampal and amygdala inputs. Neuron, 76, 804–812. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Vozenin MC, De Fornel P, Petersson K, Favaudon V, Jaccard M, Germond JF, et al. (2019). The advantage of FLASH radiotherapy confirmed in mini-pig and cat-cancer patients. Clinical Cancer Research: An Official Journal of the American Association for Cancer Research, 25, 35–42. [DOI] [PubMed] [Google Scholar]
  46. Vozenin MC, Hendry JH, & Limoli CL (2019). Biological benefits of ultra-high dose rate FLASH radiotherapy: Sleeping beauty awoken. Clinical Oncology (Royal College of Radiologists). [DOI] [PMC free article] [PubMed]
  47. Vozenin MC, Montay-Gruel P, Limoli C, & Germond JF (2020). All irradiations that are ultra-high dose rate may not be FLASH: The critical importance of beam parameter characterization and in vivo validation of the FLASH effect. Radiation Research, 194, 571–572. [DOI] [PubMed] [Google Scholar]
  48. Winslow JT (2003). Mouse social recognition and preference. Current Protocols in Neuroscience, Chapter 8: Unit 8, 16. [DOI] [PubMed] [Google Scholar]
  49. Winslow JT, & Insel TR (2004). Neuroendocrine basis of social recognition. Current Opinion in Neurobiology, 14, 248–253. [DOI] [PubMed] [Google Scholar]

RESOURCES