In this study, we report our novel findings on the effects of IR on cortical microglia dynamics (Fig. 7). The importance of microglia to the progression of IR-induced changes in the brain has been reported by several studies [9, 10, 13,14,15,16,17,18,19]. However, these studies largely concentrate on static snapshots of microglia at different times after IR. In contrast, microglia are highly dynamic cells that can rapidly change their morphologies and function, and they carry out their roles in the brain through their dynamic interactions with other cell types. To address the dynamic nature of microglia, we performed chronic in vivo imaging using two-photon microscopy to characterize cortical microglia number, morphology, and movement over one month following IR, allowing us to track the same areas of the brain over time to illuminate microglial dynamics on different time scales from minutes to weeks. We show a single dose of 10 Gy IR disrupts homeostatic cortical microglia dynamic behavior. We observed that IR resulted in microglial loss that persisted through one-month post-irradiation, and that microglia redistributed in the cortex, rearranging themselves to account for cell loss and maintain territorial organization. Furthermore, we found a modest dysregulation of microglial displacement in irradiated mice within 2 weeks following IR, indicating changes to microglial mobility. Lastly, we discovered that IR reduced microglia coverage and surveillance capacity, without overtly changing the morphology of individual microglia. These findings demonstrate that a single dose of IR can induce changes in microglial behavior and function, some of which persist over time and could contribute to the manifestation of cognitive deficits.

Fig. 7

Summary schematic showing loss of homeostatic cortical dynamics at different timepoints following cranial irradiation. Created with BioRender.com

Implications of microglia loss and irregular displacement on neurological health

We observed a ~ 30% loss of cortical microglia cells during the one month time course following IR, in line with previous studies that have shown microglial loss following IR both in the hippocampus [20, 57] and whole brain hemispheres [58]. IR may directly harm microglia, causing DNA damage and oxidative stress, leading to cell death and reactivity in surviving microglia. Alternatively, it could indirectly cause microglial injury by damaging nearby cells, triggering factors that result in microglial death and reactivity. Confirmation through tissue analysis using immunohistochemistry for cell death markers is needed to confirm this possibility. The loss of microglia could have strong negative consequences on neurological health and function. As the resident immune cells of the central nervous system, microglia are primarily responsible for defending the brain against pathogens and responding to injury. In a cortical stab wound injury model, IR can impair microglial proliferation and colony stimulating factor 1 receptor expression [59], demonstrating that IR can reduce microglial responses to injury. This impaired injury response coupled with a reduced overall microglial number could therefore render the brain more vulnerable to outside insults. Future experimentation using other secondary insults, such as a laser ablation time course, could further establish the relationship between radiation and diminished injury responses.

In the context of radiation injury, IR is believed to enhance microglial phagocytosis of synaptic elements, contributing to cognitive dysfunction [14, 16, 19, 24]. Depleting microglia in irradiated mice alters levels of synaptic proteins [13, 17], restores radiation-induced changes in spine morphology [13, 17], and rescues cognitive decline [9, 10, 13, 15, 17]. Given the substantial reduction in microglia population reported following IR, the remaining surviving microglia must exhibit strong functional changes to contribute to cognitive decline, despite their reduced numbers. Our study found that microglial distributions were shifted at weeks 1 through 4 following IR, indicating microglia rearrange themselves to account for cell loss and maintain territorial organization. We also observed a subtle effect of IR on cortical microglial displacement within 2 weeks following IR, where fewer microglia were stable in irradiated mice. Microglia soma movement can become irregular during seizures, altered sensory input, localized laser ablation, and in disease [42,43,44]. It is possible that microglia somas in irradiated mice migrate further away from their original locations at these earlier timepoints to respond to different injury signals in their surroundings or that they need to move to account for microglia loss which starts within a week in our experiments.

Implications of reduced coverage and surveillance capacity on neurological health

Changes in microglia process dynamics can have serious implications for brain health and are observed in neurodegeneration, aging and neuroinflammatory models [35,36,37,38,39,40]. For example, microglia have reduced surveying capacity and response to injury with age [35, 36, 38]. In contrast, microglia can exhibit hypermotility that is also implicated in pathogenesis, as seen in mouse models of Alzheimer’s disease [40] and lipopolysaccharide-induced injury [39]. Although the increase in microglial motility that we observed following IR did not reach statistical significance, we did find a difference in process dynamics with reduced microglia coverage and surveillance observed following IR. A reduction in surveying capacity suggests less microglial contacts with their surroundings and an impaired ability to detect pathogens and damage signals. This coupled with the observed cell loss following radiation injury may leave the brain more vulnerable to secondary insults or injuries. Microglia also displayed a heightened process motility relative to coverage and relative to surveillance following irradiation. This could be a compensatory response, whereby microglia increase their process extension and retraction to counterbalance their reduced sampling area. However, relationships between motility and coverage were observed in both control and irradiated mice, suggesting that compensation in motility following IR may be part of a normal microglial compensation mechanism that exists during physiological conditions. Regardless of the mechanism, as microglia contact surrounding cells and synaptic components to engage in synaptic and structural plasticity [31, 32], their altered process dynamics in response to IR could therefore be reflective of their ability to engage in synaptic remodeling, which could lead to dysregulated synaptic phenotypes after IR [23, 24]. However, the implications of these disrupted dynamics on microglial interactions with their environment is speculative without further experimentation.

Cranial irradiation effects on microglial morphologies

Microglia are highly heterogeneous, exhibiting an array of morphologies between and within brain regions that change depending on the function they are performing [56]. Generally, microglia with smaller somas and highly ramified processes are present in healthy adult mice. However, the functional meaning of different microglial morphologies is an area of active investigation. In models of neurodegeneration, aging, and injury, microglia exhibit changes in soma size, as well as process length and thickness [37, 38, 50, 60, 61]. Increases in microglia soma size and retraction of microglial processes could indicate several metabolic changes associated with classical microglia activation, oxidative stress, or increased lysosomal activity [60, 61]. In our study, we observed no changes in cortical microglial soma or process size, ramification, or shape between irradiated and control mice over time. Radiation induced-morphological effects differ depending on the sex, brain region, age at irradiation, timing of examination following irradiation, and type of irradiation used. Others have shown morphological changes in hippocampal microglia following IR [20, 22], which could mean that cortical microglia may be less susceptible to radiation effects than hippocampal microglia. However, the effects of IR on hippocampal microglial morphology at similar doses to our model (8–10 Gy) are reported to be transient, with most microglia resuming a ramified morphology within one day following irradiation [20, 22] and no differences observed between irradiated and control microglia by one month post-irradiation [22, 24]. Persistent changes in hippocampal microglial morphology have been observed at higher doses, which may be worth exploring in the cortex [62]. Although microglial morphology was unaffected in our study, there were functional differences in cortical microglial process dynamics, demonstrating cortical microglia are sensitive to radiation (Fig. 7). Our lack of detection of morphological differences could also be a consequence of small number of cells analyzed due to the limited imaging field of view required to capture fine microglial processes. It is important to note that while we imaged the same area in each animal over time, the microglia analyzed in this exact field of view may not be the same due to their mobility which may be particularly increased by irradiation (Fig. 3). Future studies could provide a more thorough analysis of regional differences in microglial morphological responses to radiation, as most studies have focused on hippocampal effects.

Cortical radiation effects

Most of the radiation literature has focused on hippocampal effects as cognitive deficits are believed to be a result of loss of neuronal structure and impaired neurogenesis in this brain area [23, 63]. We examined the somatosensory cortex (S1) in our study as this area is ideal for chronic in vivo imaging [64] and because microglia play important roles in development, plasticity, and injury response of this area [65,66,67]. However, less is known about radiation effects in this brain region, although it is likely that sensory deficits contribute to cognitive decline post-radiation [68]. S1 receives peripheral sensory input from the thalamus and innervates the secondary somatosensory region that has links to the amygdala and hippocampus. S1 is responsible for sensory perception and identifies tactile characteristics, such as size, shape, texture and pain. This information is used for higher-order processing and problem-solving carried out by other brain areas. Though less is known regarding cortical radiation effects, there are reports of decreased cortical thickness and volume [69,70,71] and defects in sensory processing in patients following radiation treatment [68]. In rodent models, a number of radiation effects in cortical regions have been reported, including tissue necrosis [21], cellular senescence [72], changes in vascularization [21, 73], impaired neurovascular coupling [72, 74], astrocyte activation [21, 75], increased neuronal excitation and injury [26], and deficits in synaptic plasticity [27]. RNA sequencing studies on irradiated cortical tissue show differential expression of genes involved in circadian regulation, cell differentiation, and protein kinase activity [28]. Increased expression of excitatory neurotransmitters and receptors, coupled with increased glutamine/glutamate ratio has been observed, indicating a chemical imbalance in the cortex [29]. Altogether, these discoveries highlight the susceptibility of cortical areas to cranial radiation, which is further supported by the radiation effects on cortical microglial dynamics we report here. However, microglia are a heterogeneous population whose phenotypes and functions are tied to the brain area in which they reside [76], and therefore their contributions to radiation injury are likely brain region-dependent. Future studies examining multiple brain areas could help to uncover molecular mechanisms behind regional differences in radiation responses and how microglia contribute to these differences.

Study limitations

It is important to note limitations to our study. First, our mice lack a functional copy of the fractalkine receptor, which could impact microglia radiation responses. Indeed, there is some evidence that fractalkine can regulate microglia radiation responses [77]. While these mice remain a gold standard for in vivo imaging, newly generated reporter lines which target different loci could be used to replicate these findings [78, 79]. Additionally, our Cx3Cr1 reporter transgenic line also labels peripheral macrophages. Others have sought to determine the extent of peripheral immune cell infiltration following cranial irradiation and found no evidence of infiltrating peripheral cells with doses of 5 or 8 Gy at these timepoints [20, 62, 80]. Furthermore, others report a single dose of 10 Gy does not affect the proportion [58] or result in the recruitment of peripheral macrophages [81], therefore we did not -attempt to distinguish resident microglia from infiltrating cells, although this should be examined in the future. We chose to examine only male mice because a large body of literature has shown adult male mice are more sensitive to the negative cognitive effects of IR compared to females [9, 10, 12, 24, 45,46,47,48]. However, it is possible that for the parameters we measured, female mice may also be affected by, or possibly even more sensitive to IR compared to males. Microglia structural and functional responses are indeed sex-dependent, with studies demonstrating microglia in male and female mice can have differential responses to insults [82, 83]. Though studies indicate males are more sensitive to IR in adulthood compared to females [9, 10, 12, 24, 45,46,47,48], female microglia can be more responsive to cranial irradiation during earlier developmental stages [84]. Sex differences should be examined in the future to determine whether IR impacts male and female microglial dynamics differentially. For our radiation model, we chose a single dose of 10 Gy, based on literature showing a single dose of 8–10 Gy results in cognitive deficits in mice starting at four-week post irradiation [9, 11,12,13,14]. It is possible that higher doses of radiation or a fractionated radiation scheme could have different effects on S1 microglia and this possibility should be explored. We also imaged and assessed microglial parameters over one month following irradiation, as we proposed that microglial behavioral changes would precede cognitive deficits. However, it is possible that microglia exhibit changes in dynamics after one month, as cognitive deficits [57,