If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Centre for Experimental Neuroscience (Cense), Department of Neurosurgery, Aarhus University Hospital, Aarhus, DenmarkDepartment of Clinical Medicine, Aarhus University, Aarhus, Denmark
Centre for Experimental Neuroscience (Cense), Department of Neurosurgery, Aarhus University Hospital, Aarhus, DenmarkDepartment of Clinical Medicine, Aarhus University, Aarhus, Denmark
Centre for Experimental Neuroscience (Cense), Department of Neurosurgery, Aarhus University Hospital, Aarhus, DenmarkDepartment of Clinical Medicine, Aarhus University, Aarhus, Denmark
Department of Clinical Medicine, Aarhus University, Aarhus, DenmarkDepartment of Oncology & Radiation Therapy, Aarhus University Hospital, Aarhus, Denmark
Department of Clinical Medicine, Aarhus University, Aarhus, DenmarkDepartment of Oncology & Radiation Therapy, Aarhus University Hospital, Aarhus, Denmark
Department of Clinical Medicine, Aarhus University, Aarhus, DenmarkDepartment of Oncology & Radiation Therapy, Aarhus University Hospital, Aarhus, Denmark
Centre for Experimental Neuroscience (Cense), Department of Neurosurgery, Aarhus University Hospital, Aarhus, DenmarkDepartment of Clinical Medicine, Aarhus University, Aarhus, Denmark
Centre for Experimental Neuroscience (Cense), Department of Neurosurgery, Aarhus University Hospital, Aarhus, DenmarkDepartment of Clinical Medicine, Aarhus University, Aarhus, Denmark
Centre for Experimental Neuroscience (Cense), Department of Neurosurgery, Aarhus University Hospital, Aarhus, DenmarkDepartment of Clinical Medicine, Aarhus University, Aarhus, Denmark
Zap Surgical Systems, Inc, San Carlos, CA, USADepartment of Neurosurgery, Stanford University School of Medicine, Stanford, CA, USADepartment of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Stanford, CA, USA
Centre for Experimental Neuroscience (Cense), Department of Neurosurgery, Aarhus University Hospital, Aarhus, DenmarkDepartment of Clinical Medicine, Aarhus University, Aarhus, Denmark
Leibniz Institute for Resilience Research, Mainz, GermanyInstitute for Pathophysiology, University Medical Center of the Johannes Gutenberg-University Mainz, Mainz, Germany
Single dose of 40 Gy radiation did not cause apparent necrosis in minipig's brain.
•
Single dose of 40 Gy radiation increased spontaneous firing rate in visual cortex.
•
Single dose of 40 Gy radiation shortened the peak time of visual evoked potential.
•
Inhibitory neurons were more sensitive to irradiation than excitatory neurons.
•
A circuit-wide, long-lasting modulation effect of low sub-ablative doses of SRS.
Abstract
Background
Modulation of pathological neural circuit activity in the brain with a minimum of complications is an area of intense interest.
Objective
The goal of the study was to alter neurons' physiological states without apparent damage of cellular integrity using stereotactic radiosurgery (SRS).
Methods
We treated a 7.5 mm-diameter target on the visual cortex of Göttingen minipigs with doses of 40, 60, 80, and 100 Gy. Six months post-irradiation, the pigs were implanted with a 9 mm-wide, eight-shank multi-electrode probe, which spanned the radiation focus as well as the low-exposure neighboring areas.
Results
Doses of 40 Gy led to an increase of spontaneous firing rate, six months post-irradiation, while doses of 60 Gy and greater were associated with a decrease. Subjecting the animals to visual stimuli resulted in typical visual evoked potentials (VEP). At 40 Gy, a significant reduction of the P1 peak time, indicative of higher network excitability was observed. At 80 Gy, P1 peak time was not affected, while a minor reduction at 60 Gy was seen. No distance-dependent effects on spontaneous firing rate, or on VEP were observed. Post-mortem histology revealed no evidence of necrosis at doses below 60 Gy. In an in vitro assay comprising of iPS-derived human neuron-astrocyte co-cultures, we found a higher vulnerability of inhibitory neurons than excitatory neurons with respect to radiation, which might provide the cellular mechanism of the disinhibitory effect observed in vivo.
Conclusion
We provide initial evidence for a rather circuit-wide, long-lasting disinhibitory effect of low sub-ablative doses of SRS.
There is an emergent view on the representation of neuropsychiatric disorders as disorders of the bound neuronal network, also termed circuit disorders [
]. Diverse pathologies such as depression or addiction, but also neurodegenerative disorders such as Parkinson's disease, are characterized by hypo- or hyperactivity of network nodes. The modulation of these distinct hubs can result in either well describable movement dysregulation or complex psychiatric phenotypes [
For many decades, widely distributed but highly convergent trajectories of tightly collimated beams of radiation have been used clinically to treat brain pathologies with a cellularly destructive dose of ionizing radiation. This non-invasive ablative procedure, termed stereotactic radiosurgery, enables a large cumulative dose of radiation to be deposited within the target, while minimally irradiating adjacent healthy brain [
]. By severing brain connections, this approach to ablation has also been used to treat functional disorders such as tremor and obsessive-compulsive disease [
]. More recently, a new hypothesized method called radiosurgical neuromodulation or “radiomodulation”, has been proposed to non-destructively alter neural circuit activity, using precisely targeted lower doses below the threshold of cellular lethality [
]. Radiomodulation has garnered increasing attention due to its non-invasive nature, low cost when compared with contemporary implant-based neuromodulation approaches such as deep brain stimulation (DBS) [
Cost-effectiveness of stereotactic radiosurgery with and without whole-brain radiotherapy for the treatment of newly diagnosed brain metastases: clinical article.
The irradiation effect on cellular destruction, necrosis, and lesion formation depends on the dose, volume, tissue type, cell type, animal age, and the time since radiation exposure [
Melatonin exerts a neuroprotective effect against γ-radiation-induced brain injury in the rat through the modulation of neurotransmitters, inflammatory cytokines, oxidative stress, and apoptosis.
]. For example, in a 6-month-post-radiation model, when ionizing radiation was delivered to a 7.5 mm target on the pig motor cortex, grey matter receiving 100 Gy to the 80% isodose line, and subcortical white matter receiving 60 Gy and above, showed evidence of cellular destruction [
Most studies of subnecrotic radiation doses to morphologically normal tissue have involved large field sizes, short post-radiation follow-up, or both. For example, a maximum dose (Dmax) of 20–40 Gy delivered with a 4 mm collimator decreased the frequency of lesion-induced seizure spiking [
]. In another study, 30 to 60 Cobalt Gy equivalents of proton beam radiation to rat hippocampi (2.5 mm single aperture) resulted in upregulation of heat shock protein 72 (HSP-72) but no cell loss or impairment to elicit post-synaptic potentials [
]. Few studies have investigated the chronic effects of subnecrotic doses on small and histologically normal brain targets. In one such large animal (pig) model, [18F]-Fluoro-d-Glucose Positron Emission Tomography (FDG-PET) analysis showed decreased 18F-FDG-uptake within the motor cortex target at a Dmax of 60–120 Gy, and increased FDG uptake at 10–40 Gy doses, when observed 9 months post-SRS [
Effects of focal radiation on [18F]-Fluoro-D-Glucose Positron emission Tomography in the brains of miniature pigs: preliminary findings on local metabolism.
In this study, we irradiated Göttingen minipigs using SRS with varying dose levels (0, 40, 60, 80 and 100 Gy). We ask whether focal irradiation results in a dose-dependent neuromodulatory effect at subnecrotic doses in vivo, six months post-irradiation. We furthermore analyzed whether there is a dose-related distance-dependency of putative network changes regarding the center of the irradiated field. For that, we employed a multi-shank multi-electrode probe inserted into the primary visual cortex, spanning the center of the irradiated zone, where the dose is highest, and the neighboring areas that a gradient of lower doses. We complement the in vivo recordings with a cellular assay, allowing for resolving cell-specific effects of irradiation.
2. Material and methods
2.1 Study design
The aim of this study is to provide evidence for the prospect of using adjusted radiation doses for long-term neuromodulation in a minipig model, using clinical hardware, as a reference for a future clinical study. To this end, six Göttingen minipigs were treated with various doses of ionizing radiation. Six months later, with two additional control animals, an in vivo readout of local circuitry by implanted microelectrode array followed by histology was conducted, using methods as described before and in the Supplementary material [
]. To learn about the cellular basis of the findings in vivo, iPS-derived neuronal co-cultures were treated with various doses of X-ray and imaged 2-, 5-, 7-, 9- and 12-days post-treatment, followed by a cell number assessment.
2.2 Data analysis
In vivo electrophysiology raw data were analyzed using MATLAB. Statistical analyses were performed using GraphPad Prism 9.0. Mann-Whitney test and Kruskal-Wallis test were used. Data are presented as means ± SEM unless otherwise stated. Imaging data on SynFire® neural cultures were processed and analyzed using Harmony High-Content Imaging and Analysis Software V4.9 (PerkinElmer). Values of four technical replicates were averaged for each time point and normalized to D0 pre-treatment values. Student's t-test was used for statistical analyses.
3. Results
3.1 A cortex-penetrating multi-electrode probe spans the irradiated and neighboring region of the primary visual cortex (V1)
Six Göttingen minipigs received various doses of ionizing radiation (40 Gy, 60 Gy, 80 Gy, and 100 Gy prescribed to the 80% isodose level) targeted at the right V1 (Fig. 1A), and two animals served as controls (0 Gy, Table 1). With a collimator size of 7.5 × 7.5 mm, the volumes receiving 80%, 50%, 25%, and 15% doses were ∼0.15, 0.50, 1.5, and 4.0 cm3 (Table 1). All eight animals were closely monitored throughout the six months post-irradiation period for signs of pain and discomfort. We did not observe failure to thrive, or aberrant behavior. No apparent signs for an impaired function of the visual system were observed. Six months post-irradiation, a custom-made multi-electrode probe consisting of 8 shanks and 64 electrodes [
An intracortical implantable brain-computer interface for telemetric real-time recording and manipulation of neuronal circuits for closed-loop intervention.
] with a width of 9 mm was implanted in the V1 target area (Fig. 1B). The probe covered the irradiated and neighboring region (Fig. 1C), and spanned the entire depth of the visual cortex, which is approximately 2 mm in the minipig [
]. With the custom-made 64-channel probe implanted in the right V1 (Fig. 1D&E), we studied the effect of different dose radiation on spontaneous neuronal activity by comparing spike firing rates on sedated animals. We recorded spontaneous activity from 48 (out of 64) electrodes, which were dedicated for recordings. Out of these 48 electrodes, 24 were putatively located in cortical grey matter (the 1st, 3rd, and 4th electrodes of each shank, Fig. 1E&F). Peaks higher than 7.5-times of the standard deviation of the signal were counted as action potentials (Fig. 2A). In control animals, we observed a constant rate of spontaneous activity across electrodes of the same D-V coordinates across shanks (Fig. S1). To assess the functionality of the local V1 circuitry in the representation of visual afferents, we subjected the sedated animals to visual stimuli comprising trains of light flashes [
]. We could reliably detect the typical VEP waveform in the cortical electrodes, including the two components N1 and P1 (Fig. 2B).
Fig. 1Irradiation on the V1 of minipigs and implantation of a multi-electrode probe spanning the irradiated and neighboring region. (A) SRS procedure on the anesthetized intubated minipigs. (B) The exposed V1 cortex (red arrow) and plastic fiducial marker (white dot left of the red arrow) framed by the titanium screw fiducial markers. (C) Schematic view of the probe position in the radiosurgery field with radiation dose color wash guide, illustrated on radiosurgery planning CT. (D) A coronal section of the Göttingen minipig brain. (E) Enlarged V1 area showing approximately the location of implanted probe. Note, that the tissues sections shrink in the process of histological fixation, resulting in an underestimation of cortical thickness regarding the probe. (F) The 8-shank 64-channel multi-electrode probe design. Probe thickness: 50 μm. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
Fig. 2Spontaneous neuronal activity and VEP recorded from the control animal 0 Gy - 1. (A) Representative spontaneous activities. Red triangles mark action potentials detected using a threshold of 7.5-times of the standard deviation. (B) Representative flash VEPs. The red trace is the average of 50 raw traces (shown in grey). Arrow indicates artifacts of LED on/off. The peak time of the two components of VEP (N1 and P1) were measured. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
Optogenetic modulation of a minor fraction of parvalbumin-positive interneurons specifically affects spatiotemporal dynamics of spontaneous and sensory-evoked activity in mouse somatosensory cortex in vivo.
]. First, we assessed the spontaneous activity rate across shanks within each irradiated animal. We found that the firing rate was not dependent to the distance to the radiation center in the probed area for animals irradiated with a dose of 40 Gy (p = 0.4279, Spearman correlation) or 60 Gy (p = 0.7143 for the animal 60 Gy - 1 and p = 0.1150 for the animal 60 Gy - 2) or 80 Gy (p = 0.7750 for the animal 80 Gy - 1 and p = 0.5667 for the animal 80 Gy - 2), similar to the control animals (p = 0.5364 for the animal 0 Gy - 1 and p = 0.9768 for the animal 0 Gy - 2) (Fig. 3A). At 80 Gy, several sites were devoid of any identifiable spikes. At 100 Gy, the activity vanished, and no putative spike could be detected. We next probed, whether the distance to the irradiation center affects visual evoked potential wave form. Within each animal, the N1 and P1 peak times were not correlated to the distance to the radiation center along the probe width for irradiation doses of 40 Gy (p = 0.7930 and 0.3599), 60 Gy (p = 0.4618 and 0.3599, animal 60 Gy - 2), and 80 Gy (p = 0.2992 and 0.5821, the animal 80 Gy - 1) (Fig. 3B&C). Notably, while spontaneous activity at 80 Gy varied drastically between shanks, rather stable VEPs could be recorded. At 100 Gy, no meaningful activity could be recorded due to the probe damage during implantation.
Fig. 3Irradiation on the V1 of minipigs and implantation of a multi-electrode probe spanning the irradiated and neighboring region. Raw data of (A) spontaneous firing rate, (B) VEP N1 and (C) P1 peak times. Recording sites numbers: 24, 24, 24, 21, 24, 24, 24 and 18 for 0 Gy - 1, 0 Gy - 2, 40 Gy, 60 Gy - 1, 60 Gy - 2, 80 Gy - 1, 80 Gy - 2 and 100 Gy animals. Summary of (D) spontaneous firing rate, (E) VEP N1 and (F) P1 peak times. Kruskal-Wallis test. The red line shows the median. (D) The recorded spontaneous firing rates (Hz) are 0.237 ± 0.013 (0 Gy, 2 animals, 48 recording sites), 0.296 ± 0.016 (40 Gy, 1 animal, 24 recording sites), 0.060 ± 0.0094 (60 Gy, 2 animals, 45 recording sites), 0.041 ± 0.010 (80 Gy, 2 animals, 48 recording sites), and 0.0 ± 0.0 (100 Gy, 1 animal, 18 recording sites). (E) The recorded VEP N1 peak times (ms) are 40.04 ± 0.46 (0 Gy, 2 animals, 48 recording sites), 40.14 ± 0.81 (40 Gy, 1 animal, 24 recording sites), 35.00 ± 0.25 (60 Gy, 1 animal, 24 recording sites), and 31.27 ± 0.65 (80 Gy, 1 animal, 24 recording sites). (F) The recorded VEP P1 peak times (ms) are 76.79 ± 1.36 (0 Gy, 2 animals, 48 recording sites), 49.29 ± 0.76 (40 Gy, 1 animal, 24 recording sites), 69.30 ± 1.07 (60 Gy, 1 animal, 24 recording sites), and 71.98 ± 1.07 (80 Gy, 1 animal, 24 recording sites). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
3.3 At the dose of 40 Gy, spontaneous firing rates increased as opposed to 60 Gy and higher doses
As we could not observe changes in spontaneous activity in relation to distance within the individual probe, both in non-irradiated animals as well as in irradiated animals up to 80 Gy, we compared the mean firing rates of all 24 cortical grey matter recording sites between animals (Fig. 3D). Notably, irradiation with 40 Gy significantly increased the mean firing rates (p = 0.0133) compared to non-irradiated control animals. At 60 Gy, the spontaneous activity rate dropped sharply (p < 0.0001), and at 80 Gy the rate dropped further to almost nil mean activity rates.
3.4 40 Gy radiation dose level decreased the VEP P1 peak time
With regard to the peak times upon visual stimulation, 40 Gy irradiation resulted in a decrease of VEP P1 peak time by 35.8% (p < 0.0001, Fig. 3F), while N1 peak time was not affected (Fig. 3E). Note that a decrease in peak time is generally associated with higher excitability [
]. At 60 Gy and 80 Gy, N1 peak time decreased by 12.6% (60 Gy) respectively 22.6% (80 Gy) (p < 0.0001 for both conditions), while P1 peak times remained similar to non-irradiated levels, albeit the 60 Gy animals exhibit a slight, yet significant decrease by 9.8% (p = 0.0074) (Fig. 3E&F).
3.5 40 Gy radiation did not cause apparent cellular disruption or necrosis in minipigs
Six months after irradiation, MRI scans were conducted before the implantation of the probe to obtain stereotaxic coordinates for probe implantation. Tissue changes were visible on the scans as a hypointense area surrounded by a ring of bright gadolinium enhancement (as shown in Fig. 4B, right). Moreover, since the radiation was delivered to the right hemisphere it was possible to visually compare it to the unirradiated left hemisphere and search for any visible tissue changes. MRI showed notable tissue changes only on animals irradiated with doses of 80 Gy and 100 Gy. Based on MRI measures, the approximate volume of the tissue change (mm3) was 1200 ± 146 (80 Gy - 1), 1140 ± 93 (80 Gy - 2), and 984 ± 87 (100 Gy). After the electrophysiology experiment, animals (except the 100 Gy animal, in which the probe was left in the brain for about one month) were sacrificed, and brains were studied with histology (Fig. 4A). The 40 Gy irradiated brain did not show any apparent signs of destructive changes (Fig. 4B&C). In minipigs irradiated at the dose of 60/80/100 Gy at the shown volumes (Table 1), destructive changes including necrosis caused by irradiation were noted. The span and intensity of tissue changes (necrosis, vascular changes, and astroglial reaction) caused by irradiation were correlated with radiation dose (Fig. 4B&C, Table 2). Based on Iba1 staining, the approximate volume of tissue changes (mm3) was 12.6 (60 Gy - 1), 10.1 (60 Gy - 2), 1004.3 (80 Gy - 1), and 589.1 (80 Gy - 2).
Fig. 4The spectrum of the tissue changes observed in MRI and histology following focused ionizing radiation. (A) Demonstration of the analyzed brain slabs. The visual cortex is highlighted in purple, the radiation target in red, and the probe emplacement area in blue. (B) Left: MRI shows the plastic fiducial marker (red arrow) above the visual cortex in the control animal. Right: MRI shows ring enhancement and necrosis (red arrow) in 100 Gy animal. (C) Spectrum of the changes visible after various doses of radiation, N&E − Nissl and Eosin, Iba1 - microglia marker, GFAP - astroglia marker, MBP (myelin basic protein) - myelin staining. Arrows point to small bleeding caused by electrode implantation, ∗ - Necrosis, V - vascular changes (in 0 Gy caused by electrode implantation, in 60 and 80 Gy the effect of the radiation), triangles - glial scar, o - white matter injury in corpus callosum. Scale bars: 2 mm. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
3.6 Inhibitory neurons are more sensitive to X-ray irradiation in comparison to excitatory neurons
For assessing the cellular underpinnings of the observed differential changes in function at 40 Gy versus 60 Gy and beyond, we irradiated an iPS-derived neuronal culture, comprising of excitatory and inhibitory neurons as well as astrocytes [
]. In each well, approximately 600 inhibitory neurons and 1000 excitatory neurons were counted pre-irradiation. Cell cultures were irradiated with doses of 15, 40, 70, 100, and 150 Gy. The number of surviving neurons was quantified at six time points up to 12 days post-irradiation. Depending on the doses, neuron numbers varied post-irradiation. At doses of 15 and 40 Gy, the cell viability of both excitatory and inhibitory cells was not significantly affected throughout the time period of observation (Fig. 5, Fig. S2). At 70 Gy, commencing at day 2 after irradiation, the number of surviving inhibitory neurons significantly dropped (p < 0.001), upon reaching a plateau until 12 days post-irradiation (D12), while the number of surviving excitatory neurons was not affected (Fig. 5C-E). Only at higher doses of 100 Gy and beyond, also the number of excitatory neurons displayed a significant early reduction. Yet, at 100 and 150 Gy, the number of surviving interneurons dropped sharply and were dose dependent (p < 0.0001). At 150 Gy, on D12, while 61.4% of excitatory neurons still survived, only 10.6% of inhibitory neurons remained viable. This differential impact of increasing irradiation doses on cellular survival is reflected in a dose-dependent drop of the ratio of surviving inhibitory versus excitatory cells. The biological effects of radiation are known to depend upon not only dose, but also the volume of tissue irradiated [
]. For these reasons, specific doses and their impact on cellular physiology in cell culture cannot be quantitatively compared to the 6-months post-radiation in vivo results. Regardless, these findings might offer insights into the cellular mechanisms responsible for the disinhibition at low irradiation doses observed in vivo.
Fig. 5The impact of X-ray treatment on human in vitro neural networks. (A) Schematic representation of the timeline from seeding of induced neuronal (iN)/glia co-cultures to the completion of the experiment. The “lightning bolt” symbol on the timeline indicates X-ray treatment on the 23rd day post-plating the stem cells, “DPP23”. (B) Images on day 12 post-irradiation (D12). Representative images of inhibitory (eGFP, green) and excitatory (tdTomato, red) iN cells. Scale bar: 50 μm. (C-E) Quantification of cell numbers post-irradiation. Relative numbers of inhibitory iN cells (C) and excitatory iN cells (D) as compared to D0 (DPP23) pre-treatment. (E) The ratio of inhibitory and excitatory cells. Line graphs show mean and standard deviation (error bars) of data normalized to timepoint D0 (DPP23) pre-treatment. X-ray treatment doses are color coded. N = 4 for all doses (except for the dose of 15 Gy, when N = 3, because one well with too few cells at timepoint D0 pre-irradiation was excluded as an outlier). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
We report a circuit-wide, long-lasting modulation effect directly associated with ionizing radiation intensity in a minipig model. Régis (2013) provided a series of observations regarding biological effects reported with different doses and tissue volumes that supported the use of the general range of doses [
]. In our recent study on the dose effect on minipigs, histology results showed necrotic lesions in the grey matter at 100 Gy and in white matter at 60 Gy [
]. In another study on minipigs, FDG-PET analysis showed decreased 18F-FDG-uptake within the motor cortex target at 60–120 Gy, and increased FDG uptake at 10–40 Gy doses [
Effects of focal radiation on [18F]-Fluoro-D-Glucose Positron emission Tomography in the brains of miniature pigs: preliminary findings on local metabolism.
]. Therefore, we irradiated the animals with the doses of 0, 40, 60, 80 and 100 Gy in this study. The neuromodulatory effects both on spontaneous activity and on visually evoked activity are apparent at irradiation dose levels below the threshold of significant neural cell loss. In our paradigm, the threshold for sub-ablative neuromodulation in the absence of tissue destruction ranges at or below 40 Gy. For assessing neuronal network function, we employed a multi-shank, 64-electrode probe, which allows us to sample not only the region of tissue that received the highest radiation dose, but also neighboring areas that received very little dose. While recordings of spontaneous activity give important insights into the functional architecture and robustness of a local circuit, the representation of sensory afferents represents the second and maybe even most important marker for circuit integrity. A well-established paradigm providing important diagnostic information regarding the functional integrity of the visual system represents the assessment of VEP [
]. Visual evoked responses in V1 do not only reflect local cortical circuit function, but also require the functionality of the ascending pathway of the visual pathway, e.g., the retina, and the lateral geniculate nucleus (LGN).
After the in vivo electrophysiology recordings on minipigs, brains slices were studied with histology. We did not include markers of cell death, e.g., caspase in this study. It has been shown that apoptosis can be detected in a relatively short time after the irradiation (usually hours, in some studies days or weeks range) [
Experimental evidence to support the hypothesis that damage to vascular endothelium plays the primary role in the development of late radiation-induced CNS injury.
]. In our study, animals were investigated around 6 months post-radiation. In our recent study on the dose effect on minipigs, histology results showed necrotic lesions in the grey matter at 100 Gy and in white matter at 60 Gy [
]. But we did not observe an increase in the number of apoptotic cells.
4.1 Low dose irradiation leads to local circuit-wide functional reorganization beyond the irradiation focus
Within each irradiated animal, we found that the spontaneous firing rate or the VEP peak time was not correlated to the distance to the radiation center, which suggests a relatively widespread effect of radiation while the overall function of the neural network is preserved. It is important to stress that, we assessed network function six months after irradiation. It seems likely that, initial changes and neuromodulatory events occurred at the area of irradiation. However, as a response to these focal events, the entire regional circuit is modulated. We reported a similar plasticity-mediated mechanism in the context of multiple sclerosis, in which areas of the cortex not directly affected by demyelination and T-cell integration changed their functional status [
]. Because the electrode probe spanned only 9 mm between shank 1 and shank 8, we were unable to observe the full extent of network changes. Regardless, we have demonstrated that there is network modulation by virtue of the lateral spread between the area maximally dosed (Shank 1), and the area at Shank 8 that received only approximately 15–25% of that dose (Table 1). Additionally, the VEP analysis suggested that the visual circuit from the eyes to V1 remained functional, albeit altered in a dose-dependent manner.
4.2 Focal radiation durably impacts the excitation/inhibition balance
While the reduction in spontaneous activity at high doses of 60 Gy and beyond might be expected, the disinhibitory effect of irradiation with 40 Gy, both regarding the spontaneous activity and the VEP peak time was surprising. The cellular assay, while not directly translatable to the in vivo situation in terms of absolute doses, may provide evidence on the cellular underpinnings of the disinhibitory effect. The reported higher sensitivity of interneurons is indeed in line with one earlier study, demonstrating that inhibitory cells are more sensitive to injury [
]. Earlier studies have shown that irradiation could affect GABA receptors and GABAergic inhibition, and further modulate the development and function of the central nervous system [
Melatonin exerts a neuroprotective effect against γ-radiation-induced brain injury in the rat through the modulation of neurotransmitters, inflammatory cytokines, oxidative stress, and apoptosis.
], there has been a great deal of study of CNS effects of various doses of radiation delivered to the whole body, or to fetal rodents in the uterus of the mother animal [
]. Such approaches are well suited to answering questions about the consequences of environmental radiation exposure with the presence of intertwined systemic, humoral, developmental, as well as neurological factors. For purposes of understanding the utility of focal radiation as a brain neuromodulation tool, however, a different approach is required.
Focusing more closely on an anatomical level, some studies concern radiation limited to the head. For example, Nair et al. (1963) described the acute anti-convulsant effects of 100 Gy X-ray delivered to rat heads within the 8–10 days prior to the animal's deaths [
]. Grossly apparent pathological changes in these brains were noted. Sherwood et al. (1967) described both convulsive and anti-convulsive effects of radiation delivered to the heads of rats [
]. Specifically, all doses (5, 50 and 100 Gy) were reported to lower seizure threshold relative to control. However only the higher doses (50 and 100 Gy) were shown to reduce the magnitude of the seizure clonus. These studies focused on early radiation effects occurring during the first week post-radiation.
On the receptor level, Kuznetsov et al. (1989) studied the effect of ionizing radiation on the binding of muscimol (a GABA agonist) by synaptic membranes in the cortex of rat brains [
]. 2 Gy was shown to reduce muscimol binding, suggesting GABAergic effects ensuing from the radiation. Daynac et al. (2013) noted in mouse brains that quiescent neural stem cells in the subventricular zone exited dormancy in response to changes in GABAA receptor signaling 2 days following irradiation at 4 Gy [
]. In an in vivo model, Dagne et al. (2017) reported that the selective reduction of GABAA-mediated inhibition occurred 24 h following head irradiation and disappeared by 48 h [
]. The animals were studied with MRI up to 9 months and sacrificed at different time intervals for histological and histochemical analysis. Increases in glial fibrillary acidic protein immunoreactivity, tyrosine hydroxylase, serotonin transporter, and the GluR1 subunit of the alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate glutamate receptor in penumbral regions of the lesion were found and suggested to be of compensatory neuronal adaptations.
In a functional PET study, again on minipig receiving focal irradiation, doses around 40 Gy led to an increase of local metabolism, contrasted with a decrease of local metabolism of 60 Gy or higher [
Effects of focal radiation on [18F]-Fluoro-D-Glucose Positron emission Tomography in the brains of miniature pigs: preliminary findings on local metabolism.
]. Together, there is a growing body of evidence suggesting that low doses of irradiation may indeed be capable of increasing the local excitability, reaching a new stable set point.
4.3 The disinhibitory effect of low doses of radiation might open novel avenues for the treatment of circuit disorders
To harness the clinical potential of non-ablative SRS a significant change in our perception of the underlying mechanism is required. For starters, it posited that sub-ablative doses of irradiation may provide means for achieving durable, disinhibitory upregulation as a form of neuromodulation. The present study provides some neurophysiological basis for approaching low-dose-irradiation-mediated neuromodulation from a clinical perspective. Meanwhile, modern medicine needs to finally acknowledge both the severity and treatment resistance of many behavioral disorders. This is especially true for treatment-resistant depression (TRD) major depressive disorder (MDD) and addiction. At the same time, ever better animal models and the understanding that comes from them provide contemporary medicine with more refined targets and related avenues for improving the treatment of behavioral diseases. The present study builds upon our early understanding of ionizing radiation-mediated neuromodulation effects and their neurophysiological basis. These principles could, in the future, lead to therapies for brain circuit-based disorders, achieving resilient neuronal networks. While it seems implicit that circuit-based disorders should be treated by circuit-based treatments, several impediments keep us from gaining an ideal understanding. These include the need to know what anatomical targets should be treated, and what functional alteration of these targets is required.
While focal radiation is not “reversible” like DBS, it does not share the same surgical risks as passing an electrode deep within the brain. With a better understanding of effective doses and targets, focal radiation may prove to be an effective neuromodulation tool and proved non-invasive treatment of dysfunctional brain circuits in human disorders.
Funding
Zap Surgical Systems provided project support to Aarhus, to NeuCyte and the AS Laboratory. Authors were supported by the Boehringer-Ingelheim Foundation and the German Research Foundation .
Data availability statement
The authors confirm that the data supporting the findings of this study are available upon reasonable request.
CRediT authorship contribution statement
Hamed Zaer: Methodology, Investigation, Visualization, Project administration, Writing – original draft, Writing – review & editing. Wei Fan: Methodology, Software, Investigation, Visualization, Writing – original draft, Writing – review & editing. Dariusz Orlowski: Methodology, Investigation, Visualization, Writing – original draft, Writing – review & editing. Andreas N. Glud: Investigation, Writing – review & editing. Morten B. Jensen: Investigation, Writing – review & editing. Esben S. Worm: Investigation, Writing – review & editing. Slávka Lukacova: Investigation, Writing – review & editing. Trine W. Mikkelsen: Investigation, Writing – review & editing. Lise M. Fitting: Investigation, Writing – review & editing. Liisa M. Jacobsen: Investigation, Writing – review & editing. Thomas Portmann: Methodology, Investigation, Writing – review & editing. Jui-Yi Hsieh: Investigation, Writing – review & editing. Christopher Noel: Investigation, Writing – review & editing. Georg Weidlich: Investigation, Writing – review & editing. Woody Chung: Investigation, Writing – review & editing. Patrick Riley: Investigation, Writing – review & editing. Cesare Jenkins: Investigation, Writing – review & editing. John R. Adler: Conceptualization, Writing – review & editing. M. Bret Schneider: Conceptualization, Methodology, Investigation, Funding acquisition, Supervision, Writing – review & editing. Jens Christian H. Sørensen: Conceptualization, Methodology, Investigation, Funding acquisition, Supervision, Writing – review & editing. Albrecht Stroh: Conceptualization, Methodology, Investigation, Funding acquisition, Supervision, Writing – review & editing.
Declaration of competing interest
GW, WC, PR, CJ, MBS, and JRA are employees of, and own stock in Zap Surgical Systems, Inc, a company that has patents in the field of stereotactic radiosurgery and neuromodulation. TP, JYH and CN are employees of NeuCyte, Inc. The above listed interests did not have influence on the results of the experiments. Other authors declare that they have no competing interests.
Acknowledgments
We gratefully acknowledge the administration assistance by Ms. Anne Sofie Møller Andersen, the staff at Påskehøjgaard Animal Facility, and the Department of Radiation Oncology Aarhus University Hospital. We would also express our gratitude to the ZAP Surgical System Inc., the Boehringer Ingelheim Foundation , and the German Research Foundation for their support.
Appendix A. Supplementary data
The following are the Supplementary data to this article:
Cost-effectiveness of stereotactic radiosurgery with and without whole-brain radiotherapy for the treatment of newly diagnosed brain metastases: clinical article.
Melatonin exerts a neuroprotective effect against γ-radiation-induced brain injury in the rat through the modulation of neurotransmitters, inflammatory cytokines, oxidative stress, and apoptosis.
Effects of focal radiation on [18F]-Fluoro-D-Glucose Positron emission Tomography in the brains of miniature pigs: preliminary findings on local metabolism.
An intracortical implantable brain-computer interface for telemetric real-time recording and manipulation of neuronal circuits for closed-loop intervention.
Optogenetic modulation of a minor fraction of parvalbumin-positive interneurons specifically affects spatiotemporal dynamics of spontaneous and sensory-evoked activity in mouse somatosensory cortex in vivo.
Experimental evidence to support the hypothesis that damage to vascular endothelium plays the primary role in the development of late radiation-induced CNS injury.
00061-4&id=gr1.jpg){kind=link}
00061-4&id=gr2.jpg){kind=link}
00061-4&id=gr3.jpg){kind=link}
00061-4&id=gr4.jpg){kind=link}
00061-4&id=gr5.jpg){kind=link}