Behavioral performance and microglial status in mice after moderate dose of proton irradiation

Cognitive impairment is a remote effect of gamma radiation treatment of malignancies. The major part of the studies on the effect of proton irradiation (a promising alternative in the treatment of radio-resistant tumors and tumors located close to critical organs) on the cognitive abilities of laboratory animals and their relation to morphological changes in the brain is rather contradictory. The aim of this study was to investigate cognitive functions and the dynamics of changes in morphological parameters of hippocampal microglial cells after 7.5 Gy of proton irradiation. Two months after the cranial irradiation, 8- to 9-week-old male SHK mice were tested for total activity, spatial learning, as well as long- and short-term hippocampus-dependent memory. To estimate the morphological parameters of microglia, brain slices of control and irradiated animals each with different time after proton irradiation (24 h, 7 days, 1 month) were stained for microglial marker Iba-1. No changes in behavior or deficits in short-term and long-term hippocampus-dependent memory were found, but an impairment of episodic memory was observed. A change in the morphology of hippocampal microglial cells, which is characteristic of the transition of cells to an activated state, was detected. One day after proton exposure in the brain tissue, a slight decrease in cell density was observed, which was restored to the control level by the 30th day after treatment. The results obtained may be promising with regard to the future use of using high doses of protons per fraction in the irradiation of tumors.


Introduction
Proton therapy (PT) centers have been intensively developing around the world over the last decade, because proton therapy is a promising method of treatment for numerous types of cancer, given the accumulated large amount of data on its benefits when applied to radio-resistant tumors and those located closely to critical organs, compared with conventional X-rays.To date, more than 250,000 patients worldwide have received PT; PT has proven to be particularly in demand in the treatment of brain tumors and in pediatrics (Tommasino and Durante 2015;Thomas and Timmermann 2020).The main advantage of protons is used in PT, namely the specific absorbed dose distribution: relatively low energy of particles at the input to the biological tissue, the presence of the Bragg peak, i.e., maximum energy release at the end of the particle path at a set depth directly in the tumor and a sharp energy drop to zero after the peak, which prevents damage to healthy surrounding tissues (Paganetti et al. 2019).
It is well known today that cognitive impairment is a side effect of conventional radiation treatment of tumors using photon radiation sources.Study of the mechanisms of dysfunction of different parts of the central nervous system (CNS) in response to radiation action is a pressing problem concerning the search for methods to reduce the risk of developing such side effects.It was shown that in irradiation of the head and neck area, when a total focal dose is as high as 50-70 Gy, the highly radiosensitive population of neuronal stem and progenitor cells in the subgranular zone of the dentate gyrus dies, which leads to inhibition of neurogenesis: a decrease in the formation of new neurons and impairment of their functions.Increasing damage to CNS cells is observed in a long-term period after irradiation as a result of the development of neuroinflammation associated with severe oxidative stress induced by the activation of microglial cells and the secretion of proinflammatory cytokines by them (Boyd 2021).The impairment of neurogenesis in the presence of such inflammation eventually leads to reduction in the ability to learn and to memory decline.Microglia are resident immune macrophages, which can monitor the microenvironment for events such as tissue damage, infections, or homeostatic disturbances, and rapidly respond to any pathological changes in the brain parenchyma: excitotoxicity, neurodegenerative lesions, ischemia, and direct damage to tissues, including after exposure to radiation (Fu 2012;Ginhoux 2010).They play a key role in the development, homeostasis, and innate immunity of the CNS.Microglia are exceptionally diverse in their morphological characteristics and actively change the shape of their processes and soma in response to various stimuli.Thus, analysis of the morphological plasticity of microglia provides insight into the inflammatory status of brain tissues and allows assessment of the severity of damage caused by a range of factors, such as oxidative stress, inflammation, or trauma.A series of studies have revealed non-pathological functions of microglia, including regulation of synaptic and structural plasticity during learning and memory consolidation (Sipe 2016;Tremblay 2010).Literature data indicate that microglial cells are the main effectors of synaptic changes, which increases the likelihood of the participation of microglia in radiation-induced damage to neurons and in cognitive dysfunction (Cornell et al. 2022;Liu et al. 2022).
In radiation therapy of tumors, in general, the mode of fractionated irradiation is used with doses of 2 Gy per fraction and a total dose of 40-60 Gy.In previous work, we examined early delayed effects of accelerated carbon ions on mice with whole-body irradiation at a dose of 1.5 Gy and those of protons in cranial irradiation at a dose of 1.8 Gy on cognitive functions.The results demonstrated that the parameters of behavior did not change in irradiated animals 3 months after exposure: the level of anxiety was not increased, the exploratory behavior pattern was pronounced, and hippocampus-dependent memory deficits were not observed.However, in a long-term memory test, the group of animals irradiated with protons made fewer mistakes in finding a hidden target hole, which may be indicative of higher integrity of memory traces (Sorokina et al. 2020).In another study, in examining the effect of a dose of 0.7 Gy of accelerated carbon ions applied as whole-body irradiation in mice, it was found 2 months after exposure that animals exhibited an altered behavioral pattern characterized by anxiety and a hippocampus-dependent memory deficit; however, there was no impairment of episodic memory.In the dorsal hippocampus of irradiated mice, a decrease in the number of cells was observed with the most pronounced decrease in cell density found in the dentate gyrus, and no Fluoro-Jade B staining-positive regions were detected.No pathological cells were found in the irradiated and control groups.Thus, total irradiation with a low dose of carbon ions at this stage may result in a cognitive deficit in mice with no signs of pathological neurodegenerative changes (Sorokina et al. 2021).Literature data on the effect of proton irradiation in a therapeutic dose range on cognitive abilities of laboratory animals and their relation to morphological changes in the brain, particularly to the response of microglia to irradiation, are rather contradictory, as the studies differ in the characteristics of the irradiation used, doses, duration, methods of observation, and animal models (Kiffer et al 2019).
New approaches and methods for the enhancement of a therapeutic effect of protons are being actively developed in recent years to improve PT and extend its scope of use.Hypofractionated irradiation is one of such methods, in which the dose of individual fractions is increased while reducing the number of fractions, which reduces not only the time and cost of treatment, but also the relapse rate (Belyakova et al. 2020).This method has gained momentum because numerous modern medical centers use the technology of scanning the entire volume of a target with a pencil beam of protons with modulated intensity for uniform irradiation, which due to high-precision irradiation of the target enables one to increase the dose in the target and significantly reduces the radiation burden to healthy tissues.
The aim of this study was to investigate the cognitive functions and the dynamics of morphological parameters of the hippocampal microglial cells after cranial irradiation of mice with a pencil scanning beam of protons with energy of 85-100 MeV at a dose of 7.5 Gy.

Proton irradiation
Experiments were performed on 8-to 9-week-old white outbred SHK male mice weighing 26-28 g.Mice were kept under standard conditions in the vivarium and had free access to dry food and water (Institute of Theoretical and Experimental Biophysics).Mice from experimental groups (n = 30) and from controls (n = 20) were kept under the same conditions throughout the experiment.A total of 50 mice were used.
Prior to irradiation, mice were anesthetized with an intraperitoneal injection of xylazine (Interchemie, The Netherlands) at a dose of 0.7 mg/kg and Zoletil 100 (Virbac, France) at a dose of 3.4 mg/kg.Each mouse was irradiated individually in an open polypropylene cuvette with its head fixed on a polyethylene foam platform.
For the development of a plan of irradiation, three-dimensional images of a cuvette with an animal were obtained using a cone beam computed tomography scanner.The irradiated volume of the heads of mice was 1 cm 3 .
Irradiation was performed at the Prometheus proton therapy complex (Physical Technical Center of Lebedev Physical Institute, Protvino).Using the method of proton pencil beam scanning from one direction along the given volume, the modified Bragg peak was formed on the area of the animal's brain.The energy of protons at the output of the accelerator ranged from 85 to 100 MeV, with the beam sigma being 2.8-3.6 mm.The absorbed dose in the brains of mice was 7.5 Gy. Figure 1 presents three projections of a tomogram of a mouse in a cuvette for irradiation in the window of the irradiation planning program.Color gradient shows the absorbed dose from the proton beam (the beam is directed along the YO axis); the redder color demonstrates the larger dose.Irradiation was performed in pulsed mode, with a pulse duration of 200 ms and a cycle of 2 s.In this cyclic mode of operation, protons are injected into the channel each cycle, accelerated to the set energy, and then released into the set target.The dose rate was 1.5 Gy/min.The dose control was performed using a clinical dosimeter, based on a diamond detector (IFTP, Russia), and a dosimetry film (Gafchromic radiotherapy film EBT2, USA).The error in determining the absorbed dose of protons was about 5%.
Control animals were also anesthetized, transported to the radiation source, and placed under conditions of irradiation simulation.For histological analysis, control animals were sacrificed by decapitation after anesthesia with isoflurane at the same time periods as the irradiated animals-24 h, 1 week and 1 month after proton irradiation.Since no differences were found, the data were combined into one control group, presented in the table and figures.
The study was approved by the Commission on biological safety and biological ethics of the ITEB RAS (protocol no.28/2020).Experiments were performed in compliance with the requirements of the Federation of European Laboratory Animal Science Associations (FELASA).

Histological analysis
Brain slices of control (n = 8) and irradiated animals (n = 18) at different times after proton irradiation (24 h, 7 days, 1 month, n = 6 for each period) were stained for microglial marker Iba-1.Free-floating slices (35 µm) from perfused animals were incubated with Triton X (0.3%) in PBS three times for 5 min, followed by blocking solution (BSA 1%, Triton X 0.3% in PBS) for 2 h.Then, the rabbit primary anti-Iba-1 antibody (1:1000; Wako, Japan) was added and slices were incubated overnight at 4 °C.Next day, the goat anti-rabbit secondary antibodies (1:1000; Alexa Fluor 488, ThermoFisher, USA) were added for 2 h.After washout in PBS with 0.3% TritonX-100, slices were mounted on gelatinized covers in Fluoromount media (Sigma-Aldrich, USA).Immunostained samples were analyzed under a Nikon E200 fluorescence microscope.To make a proper comparison, equivalent regions of the hippocampus were chosen for all the groups.Photomicrographs of fluorescent areas taken at 10× were quantified with the ImageJ software (NIH, USA).The following morphological parameters of Iba-1 + cells were assessed using custom macros: cell perimeter, cell area, circularity (4*π*(area/perimeter^2)), maximum and minimum span, and their ratio for convex hull (convex hull span ratio, CHSR) (Young and Morrison 2018).The perimeter and area of cells allow to assess cell hypertrophy, circularity allows to evaluate the transition to the activated amoeboid stage.The increase in CHSR (elongation) indicates another type of activation, the so-called bipolar/rod microglia, which appears in the hippocampus in neurological diseases (Giordano et al. 2021).

Behavioral tests
For behavioral tests, 12 control and 12 irradiated animals were used.To assess general activity and anxiety levels, spatial learning, and short-and long-term hippocampusdependent memory 2 months after exposure, an open field test, a Barnes maze, and a novel object recognition test were used, which were accompanied by automatic video tracking of mice using custom software.The procedure for conducting behavioral tests was previously described in detail in our paper (Sorokina et al. 2021).

Statistical analysis
For behavioral experiments, the analysis of significance of the differences between the groups was performed using the Mann-Whitney U test (significance level p < 0.05).Statistical comparison of learning curves was carried out using the analysis of variance (ANOVA) in the IgorPro 8 software package for statistical analysis.Unpaired Student's t test was used to compare the morphological parameters of cells.

Results
Table 1 demonstrates the results of the analysis of morphological parameters of hippocampal microglial cells at different times after cranial irradiation of mice with protons at a dose of 7.5 Gy.On average, 40 sections were analyzed from each mouse.Cell density analysis showed a slight, but not significant, decrease in the number of cells in the 24-h and 1-week groups.One month after irradiation, the density of microglial cells in the hippocampus was the same as in the control.Significant changes were found in the morphology of microglia, these changes becoming more pronounced over time after irradiation.Irradiation led to hypertrophy of microglial cells; considerable differences were found in the perimeter and area of the analyzed cells (see Fig. 2).Despite an initial tendency toward roundness apparent 1 day after irradiation, a decrease in circularity and an increase in CHSR were observed over longer periods (1 week and 1 month).
In the group of irradiated animals, a large group of microglial cells did not differ morphologically from the control (supplementary Fig. 1); these non-activated cells were characterized by small soma sizes and thin branching processes.However, among cells with larger area parameters, cells from irradiated animals predominated (Fig. 3).The area and perimeter of cells in irradiated animals were greater at the same values of the maximum span compared to controls, which indicates more pronounced branching and thickening of processes (see Fig. 3).
Behavioral tests were used to assess general activity and anxiety levels, spatial learning, and short-and long-term hippocampus-dependent memory 2 months after exposure.
As shown in Table 2, in the open field test in mice irradiated with protons at a dose of 7.5 Gy, the velocity and distance traveled do not differ from those in control animals.In addition to this, no differences were found in the frequency of entries into and time spent in the center of the arena in irradiated and control mice.
To estimate the exploratory activity, vertical activity was additionally analyzed by the number of rears, and to estimate the emotional state, the number of grooming, defecation, and urination acts was considered.These observations were necessary for a more consistent estimate of the presence of stress.A low level of defecation/urination in presence of unaltered locomotor activity indicated the absence of anxiety in experimental groups in contrast to the control (data not shown).
Figure 4 presents learning curves of mice, demonstrating the change in the time of seeking a goal box in the Barnes maze with the increasing number of trials.The figure indicates that both control animals and the group of mice irradiated at a dose of 7.5 Gy demonstrated learning throughout nine sessions, which indicates the absence of impairment of hippocampus-dependent memory in the animals.
In conducting a long-term memory test, on the 3rd day (probe 1) and on the 9th day (probe 2) after learning, the mice irradiated at a dose of 7.5 Gy demonstrated no significant differences compared to the control group, but there was a tendency for the average time of seeking the goal box to increase (Fig. 5).
In the novel object recognition (NOR) test (Fig. 6), we did not reveal intergroup differences in the total time of exploration of objects at the stage of familiarization (cumulative exploration time was 8.6 ± 3.3 s in control and 7.4 ± 2.2 s in the 7.5 Gy group).However, both groups of animals preferred to explore one of the objects (object 1).The least interesting one, which was examined for less time (37% in control, and 36% in the 7.5 Gy group), was replaced with a new one during the testing phase.The presentation of a novel object did not significantly affect the cumulative exploration time in the control (4.5 ± 2.3 s); however, the irradiated animals spent more time exploring the objects in the testing phase (13.6 ± 5 s, p < 0.05).Interestingly, after replacing the object with a new one, the exploration of this zone of the open field remained at the same level in the control (43%), while the animals after irradiation paid less attention to the new object (24%).With the formula of index of stimulus recognition or discrimination index (DI) described in the earlier manuscript (Sorokina et al. 2020), we calculated this preference for novelty and indicated that the DI for control mice and irradiated animals was 0.103 and − 0.537, respectively.The NOR test showed that the recognition memory did not remain intact in mice 2 months following 7.5 Gy of proton irradiation.
Thus, 2 months after cranial irradiation of mice with protons at a dose of 7.5 Gy, no changes in behavior or deficits in short-term and long-term hippocampus-dependent memory were detected; however, an impairment of episodic memory was observed.

Discussion
In the brains of mature mice and rats after 5-10 Gy of gamma irradiation, apoptosis is observed, direct radiationinduced damage to hippocampal neurons occurs, and the density of dendritic spines changes with the following deterioration in cognitive abilities (Chakraborti et al. 2012;Kumar et al. 2013).It has been established that irradiation of the hippocampus leads to an increase in apoptosis in this area, and the arrest of neurogenesis is connected with changes in the microenvironment, including impairment of microvascular angiogenesis as well as an increase in the number and activation level of microglia inside the zone of neurogenesis, which may contribute to the aggravation of side effects of radiation therapy.In addition, there is abundant evidence of the connection of reduced neurogenesis with inflammation because anti-inflammatory drugs can restore and even enhance neurogenesis after cranial irradiation (Monje et al. 2003).
Most studies on the effect of proton irradiation on cognitive abilities of laboratory animals and their connection with morphological brain changes were conducted in the low-dose (< 1 Gy) and medium-dose (< 4 Gy) range.The results of those studies are contradictory.This may be due to many factors, including the use of protons of different energies, various parameters of the irradiation facilities, various exposure conditions (only on the head or on the whole body), different testable cognitive tasks that involve different areas of the brain, various experimenters and laboratories that may affect the work with the subjects.For instance, at doses of 0.1-1 Gy of whole-body exposure (150-250 MeV), an impairment of spatial memory was found in mice in the Barnes maze 3-6 months after irradiation (Rudobeck et al. 2017) as well as a change in the activity of animals in the open field up to the 9th month of observation (Pecaut et al. 2002;Kiffer et al. 2018).Rabin and colleagues reported that whole-body exposure to 0.35-2 Gy of protons (1000 MeV) produced a disruption of performance on a variety of cognitive tasks in rats, including plus maze, operant responding and novel object recognition (Rabin et al. 2008(Rabin et al. , 2015)).Davis et al. (2014) reported that there was no effect of headonly exposure to 0.25-2 Gy of protons (150 MeV) on the psychomotor vigilance test, on a measure of attention or on the acquisition of a visual discrimination task in rats.O'Banion and colleagues (Sweet et al. 2014) have reported that whole-body exposure to 0.1-2 Gy protons (1000 MeV) did not affect hippocampal-dependent contextual freezing, but did decrease hippocampal neurogenesis and ICAM-1 immunoreactivity in mice.Similarly, Raber et al. (2016) discovered that contextual or cued fear memory was not affected in the animals whole body irradiated in the dose range from 0.1 to 2 Gy (150 MeV).
Data on the effect of higher doses of protons (2-4 Gy) on the behavior of laboratory animals are scarce too.It was demonstrated (Pecaut et al. 2002) that whole-body  1 3 irradiation of female C57BL/6 mice with protons at doses of 3-4 Gy (250 MeV) resulted in inhibition of locomotor and exploratory activity as well as in a decrease in the ability of an animal to stay on the rotarod.In model experiments on cranial irradiation of monkeys with protons at a dose of 3 Gy (170 MeV), neither in cognitive functions of an animal nor in the concentrations of monoamines and their metabolites in peripheral blood any significant changes were found (Belyaeva 2017); however, in the work of Shtemberg et al. (2013) at the same cranial dose in rats, inhibition of locomotor activity, orientation, and exploratory behavior was found as well as an increase in situational anxiety.In another study of Shtemberg et al. (2015), it was shown that when rats were whole body irradiated with protons at a dose of 2 Gy (170 MeV), there was a tendency for the functional impairment of long-term working memory, already formed by the moment of irradiation, to progress, whereas irradiation at a dose of 1 Gy did not result in a significant decrease in the learning rate.Exposure to protons at either dose did not affect the formation and execution of the passive avoidance reflex.When rats were cranial exposed to protons at a dose of 1.5 Gy (165 MeV), an improvement in working memory of the animals was detected; however, this effect was inverted at an increase of a dose up to 3 Gy (Shtemberg et al. 2013).Other results (Shukitt-Hale et al. 2004) showed that there was no effect of 1.5-4 Gy protons (250 MeV) whole-body irradiation on neurochemical and behavioral end points, including dopaminergic functioning, amphetamine-induced conditioned taste aversion learning, and spatial learning and memory as measured by the Morris water maze.
We believe that one of the promising directions in the development of proton therapy is the development and application of new schemes of hypofractionated tumor irradiation, in which the dose of single fractions is increased concomitant with a reduction in the number of fractions, which reduces not only the time and cost of treatment, but also the relapse rate (Belyakova et al. 2020).The estimated dose for hypofractionated irradiation is in the dose range from 2.5 to 10-20 Gy depending on the type of cancer, its location, stage, associated complications in the patient, and so on (Fields et al. 2017).The most simple hypofractionated radiotherapy regimen that has been well documented for use is whole-organ treatment with 10 Gy in a single fraction with repeat doses up to three total treatments given every 4 weeks.A novel approach has now been developed to reduce neurocognitive toxicity in therapeutic regimens requiring whole-brain radiation, such as for the treatment of medulloblastoma, one of the most common brain tumors in children.Recent studies show that high dose rates (> 40 Gy/s; FLASH) may confer less toxicity on irradiated healthy tissue and may reduce neurocognitive toxicity compared to conventional dose rates (~ 1 Gy/s).The results of the authors (Williams et al. 2022) suggest that the striatal and prefrontal cortex are sensitive areas in young rats (11 days old) to whole-brain proton irradiation at a dose of 5 or 8 Gy with reduced toxicity from ultrahigh dose rates of proton radiation (> 40 Gy/s; FLASH) at 5 Gy.This gave us reason to choose the 7.5 Gy dose in the current study.
The results obtained by us in the open field test and the Barnes maze showed that 2 months after irradiation of the Fig. 4 The learning curve during trials: the effect of 7.5 Gy of protons on spatial learning and memory performance in mice in the Barnes maze test 2 months after irradiation.No significant differences were detected in latency to get the goal box (n = 12 per group).All data are presented as mean and SEM Fig. 5 The effect of protons at a dose 7.5 Gy on spatial memory in mice in the Barnes maze 2 months after irradiation.Probe 1, testing animals in the maze on day 3 after learning; probe 2, testing animals in the maze on day 9 after learning.Data are presented as mean and SEM brain with protons at a dose of 7.5 Gy, the behavior parameters did not change in mice: the level of anxiety was not increased, the exploratory behavior pattern was pronounced, and the deficit of hippocampus-dependent memory was not observed.In the work of Williams et al. (2020), it was reported that whole-brain proton exposure impaired egocentric (Cincinnati water maze) and allocentric learning and caused reference memory deficits (Morris water maze), but did not affect proximal cue learning or swimming performance in rats that had received a single dose of 11, 14, 17, or 20 Gy irradiation, 6 weeks post-irradiation.
There are no other data in the literature on the effect of protons on the cognitive functions of animals in the investigated doses, but there are works on the low-LET radiation effect in the same dose range.In a study by Raber et al. (2004), it was shown that localized X-irradiation (10 Gy) to the hippocampus/cortex induced the hippocampal-dependent spatial learning and memory impairments in the Barnes maze, but not the Morris water maze 3 months after exposure.No nonspatial learning and memory impairments were detected.The cognitive impairments were associated with reductions in proliferating Ki-67-positive cells and doublecortin-positive immature neurons in the subgranular zone of the dentate gyrus, which indicates a contributory role of reduced neurogenesis in the pathogenesis of radiationinduced cognitive impairment.In contrast, Tome et al. demonstrated that mice exposed to 10 Gy using either X-rays whole-brain radiation or hippocampal sparing radiation did not exhibit anxiety-like behaviors (Tomé et al. 2015).In the paper of Ueno et al. (2019), reduced depressive-like behaviors within 10 days were found in mice submitted to 20 Gy brain X-rays ionizing irradiation.The reason for these discrepancies in anxiety-like behaviors between those studies is not clear; nevertheless, it might be related to the fact that the radiation-induced brain injury appears in a dose-and time-dependent fashion (Jiang et al. 2015).Constanzo et al. (2020), after X-ray irradiation of a rat brain (6 mm 3 ) at a dose of 37 Gy, observed a loss of spatial memory, disinhibition of anxiety-like behavior, hyperactivity, and hypersensitivity to pain, but without significant changes in motor coordination and grip strength.
The results obtained by us in the NOR test showed that in irradiated animals, despite their high exploratory activity, the ability to recognize new objects was impaired 2 months after cranial irradiation with protons at a dose of 7.5 Gy.This in turn can be explained by a disruption in the functional relationship between the hippocampus and medial prefrontal cortex, although these changes may be caused by different factors.Episodic memory in mice also declined 3 months after irradiation with protons at doses of 0.1, 0.25, 0.5, and 2 Gy (Rabin 2015;Parihar et al. 2015).Impey et al. (2016) and Casadesus et al. (2004) found memory impairment in mice in the NOR test 2 weeks after irradiation at doses of 0.1 and 0.4 Gy and 3 months after irradiation at a dose of 1.5 Gy.Williams and colleagues (2020) detected that in rats that had received a single dose of 11, 14, 17, or 20 Gy whole-brain proton irradiation, 6 weeks post-irradiation, there were no effects on NOR.
A number of studies have shown that the altered behavior pattern under irradiation is often accompanied by neuroinflammation, when microglial cells enter an activated state (Raber et al. 2016;Lumniczky et al. 2017).One of the most important functions of glia is organization of the inner immune system of the brain.Resident macrophages of nervous tissue, microglia, become activated in case of lesions, traumas, infection, or disruption of the blood-brain barrier.
Neuroinflammation is adaptive in nature in the beginning, exercising protective functions, but in excessive activation of the inflammatory process, it becomes a pathological factor.When microglia are damaged, their activation state is reflected Fig. 6 The effect of protons at a dose 7.5 Gy on recognition memory in mice in the NOR test 2 months after irradiation.*p < 0.05 All data are presented as mean and SEM by a morphological shift from complexly branched cell bodies to hypertrophied ones initially, and then to a bushy and amoeboid (rounded) shape (Schmitt et al. 1998).These changes can be visualized using Iba-1 immunohistological staining.
There are few studies in the literature on the effect of whole-body proton and ions irradiation on microglia in mice.In the study of Raber et al. (2016), it was shown that there was a negative correlation between a measure of novel object recognition and the number of newly born activated microglia in the dentate gyrus of mice at 1 month after 0.1 Gy of proton (150 MeV) exposure.Altered macrophagederived chemokine (MDC) and eotaxin levels were detected 3 months after exposure.These data demonstrate the sensitivity of novel object recognition for detecting cognitive injury 3 months after exposure to proton radiation, and that newly born activated microglia and inflammation might be involved in this injury.In the work of Rola et al. (2008), a dose-related decrease in hippocampal neurogenesis and a dose-related increase in the numbers of newly born activated microglia were shown from 0.5 to 4.0 Gy of 56 Fe.Those findings were similar to that reported after X-irradiation (Mizumatsu et al. 2003).In a study by Krukowski et al (2018), it was found that helium exposure does not impact mice recognition memory (NOR test) early (18 + days) after exposure, while microglia are depleted.However, both 0.15 Gy or 0.5 Gy of wholebody helium exposure cause deficits in recognition memory measured 90+ days post exposure.
Data on the effect of high doses of radiation on the morphology and status of microglia are much less, and they were obtained mainly by X-ray irradiation.In the review of Boyd et al. (2021), it is discussed that irradiation using a high dosage will elicit a neuroinflammatory response.An increase in microglial density has been found in the rat cerebellum following a whole-brain dose of 6 Gy (Zhou et al. 2017).Osman et al. (2020) observed that a whole-brain X-ray dose of 8 Gy on the juvenile murine brain induced transient microglial activation, as demonstrated through changes in microglial morphology and density.Microglial activation was also associated with a transient increase in apoptotic cell levels, as well as a simultaneous increase in both pro-and anti-inflammatory genes.A number of in vivo or in vitro studies have shown that following low-LET irradiation higher than 7 Gy, microglia produced high levels of ROS, NO, pro-inflammatory cytokines, and intercellular adhesion molecule 1 (ICAM-1) (Liu et al. 2022).In the study of Hong with colleagues (1995), it was found that the initial response of the brain to irradiation involves expression of the same inflammatory gene products which are probably responsible for clinically observed early symptoms of brain radiotherapy.Responses were radiation dose dependent, but were not found below 7 Gy.This also gave us reason to choose the 7.5 Gy dose in the current study.In the study (Menzel et al. 2018), the radio-resistance of microglia was analyzed independently of their physiological brain environment and compared to other mononuclear cells from spleen and brain after X-irradiation with 7 Gy or 30 Gy.A significantly higher survival rate of isolated microglia 4 h after X-irradiation with 30 Gy accompanied by a decreased proliferation rate was observed.Irradiation of organotypic hippocampal slice cultures with 7 and 30 Gy revealed a highly and significantly decreased cell number, morphological changes, and an increase in migration velocity of microglia.Furthermore, cell loss, increased soma siz,e and process length of microglia were also found in BM chimeras irradiated with 9.5 Gy 5 weeks after irradiation.The authors presented new evidence implying that microglia are not a homogeneous population of radio-resistant cells and report on long-term alterations of microglia that survived irradiation.
In our current work, the immunohistological analysis of microglial cell density demonstrated a slight decrease in their number in the mice irradiated with protons at a dose of 7.5 Gy at 24 h and 1 week after irradiation, but 1 month after irradiation the microglial cell density in the hippocampus recovered to the control level.These results are consistent with the known data on the dynamics of density and proliferation of microglia in response to a damaging effect (Kondo et al. 2009;Li et al. 2021;Gao et al. 2022).The initial decrease in microglial density after exposure to stress is regarded as a mechanism for suppression of the inflammatory response (Li et al. 2021), whereas a decrease in density due to apoptotic death of some cells is apparently induced by stronger stressors; in less severe damage, such as ischemia (Gao et al. 2022), rapid activation of microglial cells without a significant decrease in their density is observed.The main factor causing both death and activation of surviving microglial cells is reactive oxygen species (Palwinder et al. 2006).Regardless of the original nature of damage, the accumulation of ROS has been shown to directly stimulate microglial NADPH oxidase, leading to subsequent hydrogen peroxide production, which acts as a mitogenic signal for microglia.Thus, oxidative stress induced in brain tissue, in addition to the damaging effect in the first days after irradiation, leads to the activation of proliferation and the restoration of microglial density in 30 days.We also found significant changes in the morphology of microglia: an increase in cell volume was observed as early as a day after irradiation.Hypertrophy is typical of the transition of microglia to an active state; however, in presence of a decrease in the cell density, such morphological changes may rather be associated with the initiation of apoptosis in some cells.A week after irradiation and, more noticeably, a month later, the size of the cells significantly increased and their processes expanded, while the hypertrophy was not accompanied by deramification (rounding) of the cells, which is characteristic of activated branched microglia (Boche et al. 2013).Such cells are incapable of migration and active phagocytosis, but they actively produce inflammatory factors (Stence et al. 2001;Boche et al. 2013;Fu et al. 2014).Similar results were obtained in gamma-irradiated mice at a dose of 10 Gy: with minor morphological changes, the microglial cells on the 30th day after exposure significantly increased the expression of activation markers (Hinkle et al. 2019).In the work of Rodina (2021), after irradiation of the heads of mice with protons at doses of 4 and 8 Gy, a long-term dose-dependent reduction in the number of resting microglial cells and an increase in the proportion of activated microglial cells were observed, which indicated the development of neuroinflammation 2 months after exposure.In the activated state, microglial cells can support chronic inflammatory reactions (Davis et al. 1994;Boche et al. 2013), which, in their turn, may result in various cognitive deficits.

Conclusion
The results obtained in our work demonstrated a change in the morphology of microglial cells in the hippocampus, indicative of a transition of the cells to an activated state, and a slight decrease in cell density as early as a day after exposure, which returned to the control level 1 month after cranial irradiation of mice with protons at a dose of 7.5 Gy.Two months after exposure, no disturbances in the exploratory behavior and spatial learning were found, but a deficit in long-term episodic memory was observed.Probably, a dose of protons of about 7.5 Gy, on the one hand, impose a significant damaging effect on the brain, inducing oxidative stress in which the cellular changes observed in the hippocampus in the first days after irradiation lead to the activation of proliferation and, in 30 days, to the restoration of microglial density that, in its turn, serves as a deterrent to a further inflammatory cascade, exercising a protective function.On the other hand, a dose of 7.5 Gy used in the work is insufficient for a sustained morphological response of activated microglia 30 days after irradiation.The results obtained may become promising in the prospect of using high doses of protons per fraction for irradiation of tumors of the head, in which the volume of irradiated area would be significantly smaller, which would probably reduce not only the area of inflammatory response, but also the risk of hippocampus-mediated cognitive deficits.Several authors also suggest that reducing early damage to the microvascular network and the brain parenchyma will be able to prevent the spread of remote effects of radiotherapy.The inhibition of microglial activation may also be useful for improving hippocampus-dependent spatial memory after irradiation (Jang et al. 2010;Wadhwa et al. 2017).
The evidence obtained shows that in planning proton treatment of tumors of the head, it is important to consider the sensitivity and adaptiveness of brain tissue, which will help to preserve particular cognitive functions.It is also necessary to develop pharmacological approaches including inhibition of early postradiation inflammatory processes in the brain, which would make it possible in the future to prevent the development of long-term cognitive dysfunction.Overall, our study suggests that the slow and gradual beginning of cell damage in the hippocampus at a dose of 7.5 Gy of protons provides an opportunity for the emergence of adaptive reactions, limiting acute neuroinflammation, and preventing further damage of brain cells.The brain structures that do not have progenitor properties probably adapt better to higher doses and may be exposed to a more intense irradiation mode.Thus, planning radiation treatment with regard to sensitivity and adaptiveness of brain structures in which the tumor is localized may prevent negative consequences and preserve cognitive functions and, moreover, the quality of life in patients.

Fig. 1
Fig.1Preparation of an animal irradiation session at the Prometheus proton therapy complex: projection of the animal obtained using the cone beam computed tomography.Color gradient shows the absorbed

Fig. 2
Fig. 2 Changes in the morphology of microglia at different times after irradiation.Top: photographs of brain sections with immunohistochemical staining for Iba-1, magnification 10 × and 40 × .Bottom: averaged morphological parameters at different times after irra-

Table 1
Morphological parameters of microglia at different times after irradiation

Table 2
Activity parameters in mice in the open field test 2 months after irradiation.All data are presented as mean and SEM