Exposure to low (10 cGy) doses of 4 He ions leads to an apparent increase in risk taking propensity in female rats

The planned missions to the Moon and Mars will present more significant health challenges to astronauts compared to low earth orbit missions. During deep space missions, astronauts will be constantly exposed to Space radiation (SR). Multiple rodent studies suggest that < 25 cGy of SR impairs performance in executive functions, which play a key role in advanced cognitive processes, but also regulate response inhibition and impulse control. There is the possibility that SR exposure may exacerbate aberrant behaviors evoked by psychological stress related to exposure to isolated and confined (ICE) hostile environment or independently induce additional aberrant behaviors. This study has determined that female Wistar rats exposed to 10 cGy of 250 MeV/n He had an increased risk taking propensity (RTP)\compared to shams. The increased RTP of the He-exposed rats was associated with significantly increased reaction times during the trials, suggesting a SR-induced loss of processing speed. The response times of the He-exposed rats were even further reduced in trials that immediately followed a loss, raising the possibility that conflict and interference avoidance may be impaired after SR exposure. Whether these findings occur following other types of SR exposure, and/or in male rats remains to be determined.


Introduction
The upcoming long-duration missions to the Moon and Mars will be much more challenging to the physical and mental health of astronauts than ISS missions.The extended period that astronauts will have to endure an isolated and confined (ICE) hostile environment is of considerable concern.Psychological problems during missions in extreme environments are common, even with modern screening techniques, e.g., the incidence of mood or adjustment disorders is >25 % in individuals returning from missions to ICE environments [1].Occasionally, individuals require evacuation for psychological treatment [2], and in some cases have damaged installations or attacked colleagues.However, astronauts will be exposed to other spaceflights stressors that impact CNS functionality.During deep space missions, astronauts will be exposed to Space Radiation (SR), which is a complex field of highly energetic charged particles [3].While the exact SR dose that astronauts will receive while on mission to Mars may vary greatly, doses could reach as high as 1-1.2 Sv (~200 cGy) [3].There is an ever-growing body of evidence from ground-based rodent studies that low (1-25 cGy) of SR doses impairs performance in many cognitive processes (reviewed in [4][5][6][7][8]).Given that both ICE and SR independently impact CNS functionality, there is a distant possibility that the combined effects of these space flight stressors, may increase the incidence and severity of CNS deficits over than seen in isolation.
The deleterious effects of SR on the CNS are not limited to cognition with an increasing number of studies reporting SR-induced changes in rodent psycho-social behavior [9][10][11][12][13][14].While executive functions are best recognized for the role they play in advanced cognitive processes, they also regulate response inhibition, impulse control, affect, motivation and arousal.Should SR-exposure alter such executive functions that regulate impulsivity and risk decision making as it does those regulating cognitive flexibility, there is the possibility that SR exposure may exacerbate aberrant behaviors evoked by psychological stress related to ICE or induce additional aberrant behaviors.
Decision-making requires that individuals perceive the probabilities and risks associated with the presented options and estimate the consequences of each option [15,16].Executive control functions (ECF) are important cognitive components in decision-making, where they are evaluate and learn the outcomes of choices as positive or negative, and apply this information to select and execute future decisions [15].Deficits in ECF are linked to maladaptive risk behaviors [16][17][18][19].Given the well documented effects of SR on executive functions, decision making and consequently risk taking propensity may be markedly changes following SR exposure.
Decision-making in animals share many similar preferences and biases that are seen in human choice behavior [20].The Balloon Analogue Risk Task (BART)) is a computerized measure of risk taking behavior that models real-world risk behavior through the conceptual frame of balancing the potential for reward versus loss.In humans, risk behaviors demonstrated from decision-making tasks like the BART correlate with baseline risk propensities [21,22] and overall risk behaviors in real life [21,23,24].The BART is part of battery of cognitive tasks that astronauts are required to routinely perform during space flight [25,26].A rodent risk decision making task that mimics the BART task have been developed [27].This touchscreen based task consists of four response lights, each of which has a defined win/loss probability, reward size and loss penalty.Rats utilize their superior numerosity [28] and metacognition [29] skills to ascertain the most "profitable" option after taking these multiple variables into consideration.
This study aims to determine whether the SR-induced loss of executive functions is limited to cognitive flexibility performance or also extends to risk-tasking behavior.We have thus established the impact that exposure to 10 cGy of 250 MeV/n He particles has on the risk taking propensity (RTP) of female Wistar rats.

Animals and materials
This study was conducted in accordance with the National Research Council's "Guide for the Care and Use of Laboratory Animals (8th Edition)" at the animal care facilities of Eastern Virginia Health Sciences Center (EVHSC) and Brookhaven National Laboratory (BNL); both of which are accredited by the Association for Assessment and Accreditation of Laboratory Animal Care, International.All procedures were approved by the Institutional Animal Care and Use Committees of EVHSC and BNL, and the rats were under the surveillance of a licensed veterinarian at both locations.
Rat demographics, housing conditions, experimental timeline, and exercise regimen.The female Wistar rats (Hla®(WI)CVF®; Hilltop Lab animals, Inc., Scottsdale, PA, USA) used to determine risk taking propensity are a subgroup from a previous study on SR-induced loss of ATSET performance [30].
The average age upon arrival at EVHSC was 3 months with an average weight of 250 g.The timeline and age of the rats during various experimental procedures relative to the SR exposure are outlined in Table 1.
The rats were paired-housed in ventilated cages (Green Line Techniplast, Italy), maintained on a reversed 12:12 light/dark cycle (light on: 2030; off: 0830) and given ad libitum access to Teklad 2014 chow with municipal water.Rats were implanted with ID-100us RFID transponders (Trovan Ltd, United Kingdom) to facilitate identification of individual animals.Rats were maintained on a treadmill exercise regimen used previously [31] (exercised for 30 min at 25 m/min, twice a week (at ~1200-1400)) for the entire duration of the study except when the rats were housed at BNL.This protocol corresponds to a mild aerobic exercise regimen [32], and can also be considered to be an enrichment procedure that familiarizes the rats with the investigators and being placed in different environments from the home cage.

Irradiation procedure
Twenty-four rats were shipped to BNL, where they were pairedhoused, maintained on a reversed 12:12 light/dark cycle, and given ad libitum access to Teklad 2014 chow with municipal water by bottle.After at least one week of acclimatization, 11 rats were exposed to 10 cGy 250 MeV/n 4 He (LET= 1.6 KeV/μm) particles at the NASA Space Radiation Laboratory (NSRL) (https://www.bnl.gov/nsrl/userguide/SimGCRSim.php).At the time of irradiation, the rats were ~6 months old.
The rats were placed in well-ventilated custom-made "rat hotel" irradiation jigs, constructed of red plastic and exposed to the GCRSim beam sequence at an overall dose rate of 0.5 cGy/min (~20 min exposure).Dose calibration was performed as previously described [33].The 13 sham rats were placed in identical irradiation jigs that remained in the preparation room, while their counterparts were taken into the radiation vault at NSRL.One week after irradiation, the rats were transported back to EVHSC and rehoused under the same conditions described above.
At 16 ± 2 weeks post-irradiation, at ~10 months of age, the rats were placed on a restricted diet for the duration of the STR and RTP testing.The rats received a daily allowance of ~6 g of Teklad 2014 chow, but the exact amount varied daily to maintain an individual rat's weight at ~85 % of its pre-food restriction weight in line with recommended practices [34].

Touchscreen habituation and stimulus response training
At 16 ± 2 weeks post-irradiation, at ~10 months of age, the rats were assessed for performance in a slightly modified version of the rodent gambling task [24].However, Inter-trial interval was maintained at 5 s.All RTP procedures were conducted in Bussey-Saksida rat touch screen chambers ((Model 80604), Lafayette Instruments, Lafayette, IN).All testing was conducted during the dark cycle with the first rat being tested ~2 hours into the 12 h dark cycle.The rats are first habituated to the touchscreen chambers and learn the availability and location of the food reward in the food tray.Thereafter, the rats underwent Stimulus Response (STR) training in accordance with our previously published protocol [31].Briefly, the STR training consists of four stages (Habituation, STR15, STR4 and STR1) that refines the stimulus response skills until in STR1 stage the rats rapidly (< 10 s) respond to a solitary (randomly-located) light stimulus on the touchscreen to gain a food reward.The training procedure was progressive, such that the rats that reach criterion in a stage (Table 2) proceeded onto the next stage the following day.Rats that failed a stage were rested overnight and retested the following day on the same stage until they reached criterion or exceeded the maximum number of sessions allowed (14 trials) In the STR15 stage, the rat learns that touching any one of 15 illuminated holes (within a 3×5 grid) yields a food reward.Next, in the STR4 stage, the rats are required to learn that the food reward is only dispensed when a lit hole is touched.The rats are presented with a block of four (2×2) lit holes; pressing any of the lit holes yields a reward.The illuminated block is randomly located at one of eight locations within the 3×5 grid.The position of the block is changed after any response (i.e., correct selection of a lit hole or incorrect selection of an unlit hole).Rats that reached criterion in the STR4 stage progressed onto the first STR1 (a single hole illuminated) stage, designated as STR1-timed stage.
During the STR1-timed stage, the rats further refine their stimulus response skills to learn that the food reward is only dispensed by selecting a single illuminated hole.In STR1-timed stage, a response must be made within 30 s of the hole illuminating.The location of the illuminated hole was altered after any response (correct or incorrect) to a random location within the 3×5 grid.Once a rat reached criterion, it moved onto the STR1-fast timed stage.
In the STR1-fast timed stage, the rats had to learn that an illuminated light must be touched within 10 s of its appearance to gain a food reward.The location of the illuminated hole is altered after any response (correct or incorrect) to a random location within the 3×5 grid.Once a rat reached criterion it moved onto the RTP task.

Risk taking propensity (RTP) task
At the start of the RTP testing, all rats were at the maximally permissible weight loss, i.e., at ~85 % of their initial weight.Typically rats had been on food restriction for at least a week at this time.A modified version of a previously described rodent gambling task [27] was used to assess risk taking propensity, hereafter refer to as the RTP task.An activation step, similar to that used in our task switching studies [35,36], was added to the procedure.The rat must press a green activation light (situated in the middle hole of the 3 ×5 grid) to activate the RTP task after each trial.This activation step served to increase the accuracy of the response time by removing potential behavioral time confounds, such as self-grooming, persistent searching for the food reward, or being unaware of stimulus presentation.
This RTP task is designed to replicate the impulsivity and risk decision making aspect of BART, consisting of four response lights with cues defining a win/loss probability, reward size and loss penalty (Table 3).Rats had to select a hole within 10 s, failed to do so resulted in a new trial being initiated.
The rats were given a maximum of five daily sessions (50 trials per session) to reach criterion (selection of Holes 6 or 9 in more than 70 % trials, with a minimum number of 30 trials, in two consecutive sessions).If rats reached criterion, the average performance from both sessions was recorded.If a rat failed to reach criterion in two consecutive sessions, the average response in the first three sessions was recorded.
Rats were screened for RTP performance by investigators with no knowledge of the rats' radiation exposure history.

Statistical methods
Data were statistically evaluated using either Mann-Whitney U test (2-sided).All statistical calculations were performed using the appropriate software program within Prism 10.2 (Graphpad Software, San Diego, CA).
Kernel Density Estimation probability profiles were calculated using the Kernel Density Estimation program from Free Statistics Software (v1.2.1) [37], with the Gaussian distribution as Kernel.

Results
None of the rats used in the study demonstrated any physical impairments requiring veterinarian intervention over the course of the study, nor were there any obvious motor deficits during the present study.

Habituation and stimulus response training
All 13 sham and 11 He-exposed started successfully completed the Habituation, STR15, STR4 and STR1 stages.There were no statistical differences in the number of daily sessions the sham and He-exposed rats took to compete any stage.

Risk taking predisposition
Metrics that provide an overall assessment of the performance of the rats in the RTP task are presented in Table 4.There were no significant inter-group differences in the number of sessions required to reach criterion, the trials completed per session, nor the time to complete the session were observed.Similarly, there were no obvious difference in how the sham and He-exposed rats responded during the session (Table 5).
As a group, the sham rats (N=13) selected one of the two "safe" options (holes 6 and 9) in ~74 % of trials during the 2-5 day test period.In contrast, the He exposed rats (N=11) chose the safe option at a significantly (P=0.019,Mann-Whitney; δ= 1.07; CLES= 77.62 %) lower rate (median 54 %) than did the shams (Fig. 1).KDE-generated performance profiles were generated and analyzed to quantify the percentage of rats that could be classified as "risk-prone", i. e., those chose the riskier option at a frequency of ≥50 %.Almost 2-fold more He -exposed (24.3 %) rats were classified as risk prone than shams (14.1 %) (Fig. 2).
A more detailed analysis of the frequency that the rats chose each individual option or high vs low likliehood of reward in depicted in Fig. 3.There were no significant differences in how frequently shams and He-exposed rats selected a specific hole/preference.However, there were significant differences in how frequently the rats selected the two most (or least) profitable options.
An analysis of the response time, i.e., the time elapsed from when the response lights were illuminated until the rat made its selection revealed that He-exposed rats took significantly (P=0.0126,Mann-Whitney δ = 1.274;CLES 81.6 %) longer to make a selection that did the shams (Fig. 4A).On average, the shams made a response within 1.58 ±0.17 s, whereas the mean response time for He-exposed rats was 2.88 ±0.17 s.
To restart the test (after a win or a loss), the rat must first press an activation light.Thus, the increased response time is not a reflection of the willingness of the rats to participate in the task, but the actual response time (processing speed) to make a choice.These data are thus consistent with a He-induced loss of processing speed.Moreover, this disparity in response times between He-exposed and sham rats was even more pronounced in the trials that immediately followed a loss (Fig. 4B).
The response times of the shams during such trials was comparable to that in all the trials (1.57±0.20 s compared to 1.58 ±0.17 s).However, the He-exposed rats took significantly (P=0.0013,Mann-Whitney; δ = 1.77;CLES 89.5 %) longer (4.06±0.90s) to make a response in the postloss trials than did the shams.Moreover, He-exposed rats took more than a second longer to respond in post-loss trials than their overall response time in all trials (4.06±0.90s and 2.88 ±0.17 s respectively).

Discussion
This study has established that female rats exposed to a low dose (10 cGy) of 250 MeV/n He particles have a greater RTP than their sham counterparts.Specifically, He-exposed rats chose the high reward/low success probability option 20 % more frequently that the sham rats, indicative of a SR-induced change in risk preference or tolerance.Thus, exposure to SR leads to multiple CNS sequelae manifested as loss of cognitive function [4], loss of fine motor control [38,39], altered sleep and associated EEG spectra [40], increased social withdrawal [14] and risk taking predisposition.Collectively, these data suggest that the functionality of SR-exposed brains markedly deviates from the norm.Translating ground-based rodent studies into tangible risk estimates for astronauts remains an enormous challenge, but should similar neurocognitive impairments occur in astronauts exposed to such SR doses, each of these sequelae could have significant operational consequences.
While the He-induced increase in RTP raises the possibility that SRexposed individuals may be more likely to engage is risk-taking behavior, it is important to note that this may reflect an increased/ heightened reward-sensitivity of the He-exposed rats.Currently, the only study to look at SR-induced alterations in reward sensitivity/ motivation found no evidence that GCR exposure appreciably modified motivation for reward via economic demand assessments [41], unfortunately there is no data on how He exposure impacts these parameters.Given the well-documented impact that SR has on cognitive processes that utilize executive functions, it seems likely that the increased RTP of the He-exposed rats, may be the result of He-exposure impacting executive cognitive functions that are used to evaluate and learn the outcomes of choices as positive or negative.SR exposure has been shown to impact the ability of male rats to learn the outcome (associative recognition memory) of responding to a specific cue in the SD stage of the ATSET task [4].SR exposure in female rats results in a reduced ability to avoided interference in that process when alternate cues are presented in the CD stage ( [30]).
The increased RTP of the He-exposed rats was associated with a significant increase in the reaction time during the trials.In simple response tasks like the psychomotor vigilance tasks, reaction times are used as a measure of vigilant attention.Proton and GCRexposure does reduce PVT response times [41,42], thus the He-induced increase in response time may be a reflection of loss of vigilant attention.However, the RTP task is a much more complex task, where the rat must make a choice between four possible actions, and thus reaction time reflects the processing speed of the decision-making process as well as the psychomotor vigilance response.The major deficit in female rats exposed to 10 cGy GCRsim was in the mean correct latency (processing speed) in the CD stage of the ATSET task [30].
The underlying cause for the SR-induced loss of processing speed is presently unclear.One construct contributing to processing speed is working memory [43][44][45][46], and it is well documented that low dose SR exposure impairs working memory [5][6][7]47].Other studies raise the possibility that SR exposure may reduce processing speed by altering myelination.Processing speed is sensitive to changes in neuronal myelination [48,49], and He exposure (and other SR ions) results in a significant decrease in the percentage of myelinated axons relative to controls in the hilus of mice [50].In addition, SR exposure alters the g-ratios (the ratio of axon diameter to fiber diameter) of myelinated axons [50].Furthermore, space flight and SR exposure have been associated enhanced autophagy of oligodendrocyte precursors and their loss [51,52].
Further, insight into the impact of He radiation on processing speed was gained from the observation, that the difference in response time between He-exposed and sham rats was most evident in trials after a loss.While sham rats exhibited similar response times in all trials, Heexposed rats took significantly longer to respond in the trials following a loss (Fig. 4).
The RTP task is a free-choice assay, where the rat can make a choice between four possible actions.An inevitable consequence of choosing between different action plans is the potential for conflict between them.During winning trials, the rats get a reward, the size of which is constant to a given location.An inherent feature of the RTP is that these is a predetermined rate of losses in all options.Thus, after a loss the rat may be in conflict, as choosing a previously highly rewarded option (which used to be preferable) is suddenly no difference from when a poorly rewarded option is selected.The ultimate outcome of the presence of conflict in free-choice tasks is frequently manifested as slower response times [53][54][55].A rodent analogue of the Stroop test has recently been developed [56] that could be used in future studies to explore this possibility.When a loss occurs (at a frequency unique to each option), an aversive stimulus (bright light is turned on) during a time out.Thus, an alternate explanation for the longer response times in trials following a loss is that the aversive stimuli and associated time out represents an interruption (where an interfering stimulus requires attention) in the focus of attention.Interruptions distort task-relevant information of the primary task even in simple visual working memory tasks [35].One explanation is that stored information within the focus of attention is eliminated by interruptions, another is that reactivation of information relevant to the primary task requires a reallocation of cognitive resources and demand processes [57,58].Working memory also plays an important role in the cognitive processing of interruptions [59,60], and in humans working memory capacity is an important (positive) predictor of how quickly and accurately individuals can resume tasks following interruptions [61].
In summary, this study has shown that low dose SR (He particles) exposure leads to an increased RTP in female rats.Whether male rats are impacted in the same manner has yet to be determined.This is the first study to demonstrate that SR-induced impairment of executive functions is not confined to just cognitive processes.These data raise the possibility that astronauts on deep space missions where SR exposure reaches the levels used in this study may need to be closely monitored for an increased risk taking behavior.This study further supports the notion that SR exposure impacts processing speed during cognitive loading, and raises the possibility that conflict and interference avoidance may be reduced after SR exposure.Whether these findings occur following other types of SR exposure, and/or in male rats remains to be determined.

Limitations
This study's findings are constrained by the utilization of only female Wistar rats, which limits the generalizability of results as sex-specific differences may exist.However, it is crucial to note that NASA astronaut classes now have close to an equal gender split, necessitating further sex specific research.Future studies will investigate the RTP of male rats.Whether similar changes in RTP performance would be observed after more realistic SR exposure remains to be determined.

Funding
This work was supported by the NASA grant NNX14AE73G.

Declaration of Competing Interest
The authors have no conflicts of interest to declare in relation to the studies reported in this manuscript.

Fig. 1 .
Fig. 1. : Violin plot of the frequency that Sham (open) and 10 cGy He-exposed (stripped) rats chose the two optimally rewarded ("safe") options in the RTP task.Solid horizontal line represents the median, while the dashed lines represents the quartiles.*Represents significance at the P<0.05 level (Mann Whitney).

Fig. 2 .
Fig. 2. :Kernel Density Estimation of probability density of the frequency that sham (Solid curve) and 10 cGy He-exposed (dashed curve) rats selected the "safe" choices in the RTP task.Solid vertical line represents the 50th percentile, which has been used as a classifier for risk averse or risk-prone disposition.

Fig. 3 .
Fig. 3. : Frequency that rats selected each individual option (hole), or combination of options that were the most or least profitable choices throughout the session.Graphs show means ± SEM, Open bars: Sham; hatched bars: 10 cGy He.GCRsim.*Represents significance at the P<0.05 level (Mann Whitney) between the sham and He-exposed selections at each hole/combination of holes.

Fig. 4 .
Fig. 4. : Violin plot of the reaction times for Sham (open) and 10 cGy Heexposed (stripped) rats in the RTP task; A: all trials, B: post-loss trials.Solid horizontal line represents the median, while the dashed line represents the quartiles.**Represents significance at the P<0.005 level (Mann Whitney).

Table 1
Experimental timeline (relative to SR exposure).

Table 2
Criterion during Stimulus Response (STR) training.
* -All sessions in all stages had a maximum duration of 30 min.

Table 3
Reward characteristics for the four options in the RTP task.

Table 4
General RTP performance metrics.

Table 5
Rat engagement during RTP sessions.