Cortical Plasticity is associated with Blood-Brain-Barrier Modulation

Reviewer #1 (Public Review):

The goal of the current study was to evaluate the effect of neuronal activity on blood-brain barrier permeability in the healthy brain, and to determine whether changes in BBB dynamics play a role in cortical plasticity. The authors used a variety of well-validated approaches to first demonstrate that limb stimulation increases BBB permeability. Using in vivo-electrophysiology and pharmacological approaches, the authors demonstrate that albumin is sufficient to induce cortical potentiation and that BBB transporters are necessary for stimulus-induced potentiation. The authors include a transcriptional analysis and differential expression of genes associated with plasticity, TGF-beta signaling, and extracellular matrix were observed following stimulation. Overall, the results obtained in rodents are compelling and support the authors' conclusions that neuronal activity modulates the BBB in the healthy brain and that mechanisms downstream of BBB permeability changes play a role in stimulus-evoked plasticity. These findings were further supported with fMRI and BBB permeability measurements performed in healthy human subjects performing a simple sensorimotor task. There is literature to suggest that there are sex differences in BBB dysfunction in pathophysiological conditions and the authors have acknowledged the use of only males as a minor limitation of the study that should be addressed in the future. Future studies should also test whether the upregulation of OAT3 plays a role in cortical plasticity observed following stimulation. Overall, this study provides novel insights into how neurovascular coupling, BBB permeability, and plasticity interact in the healthy brain.

Reviewer #2 (Public Review):

Summary:
This study builds upon previous work that demonstrated that brain injury results in leakage of albumin across the blood brain barrier, resulting in activation of TGF-beta in astrocytes. Consequently, this leads to decreased glutamate uptake, reduced buffering of extracellular potassium and hyperexcitability. This study asks whether such a process can play a physiological role in cortical plasticity. They first show that stimulation of a forelimb for 30 minutes in a rat results in leakage of the blood brain barrier and extravasation of albumin on the contralateral but not ipsilateral cortex. The authors propose that the leakage is dependent upon neuronal excitability and is associated with an enhancement of excitatory transmission. Inhibiting the transport of albumin or the activation of TGF-beta prevents the enhancement of excitatory transmission. In addition, gene expression associated with TGF-beta activation, synaptic plasticity and extracellular matrix are enhanced on the "stimulated" hemisphere. That this may translate to humans is demonstrated by a break down in the blood brain barrier following activation of brain areas through a motor task.

Strengths:
This study is novel and the results are potentially important as they demonstrate an unexpected break down of the blood brain barrier with physiological activity and this may serve a physiological purpose, affecting synaptic plasticity.

The strengths of the study are:

1. The use of an in vivo model with multiple methods to investigate the blood brain barrier response to a forelimb stimulation.
2. The determination of a potential functional role for the observed leakage of the blood brain barrier from both a genetic and electrophysiological view point
3. The demonstration that inhibiting different points in the putative pathway from activation of the cortex to transport of albumin and activation of the TGF-beta pathway, the effect on synaptic enhancement could be prevented.
4. Preliminary experiments demonstrating a similar observation of activity dependent break down of the blood brain barrier in humans.

Weaknesses:
The authors adequately addressed most of my points. A few remain:

1. Although the reviewers have addressed the possible effects of anaesthesia on neuro-vascular coupling. They have not mentioned or addressed the possible effects of ketamine (an NMDA receptor antagonist) on synaptic plasticity. Indeed, the low percentage of SEP increase following potentiation (10-20%) could perhaps be explained by partial block of NMDA receptors by ketamine.
2. The experimental paradigms remain unclear to me. Now, it appears that drugs are applied for 50 minutes and that the stimulation occurs during the "washout period". The more conventional approach would be to have the drug application during the stimulation period to determine if the drugs occlude or enhance the effects of stimulation and then washout the drugs. The problem is that drugs variably washout at different rates depending upon their lipid solubility.
3. It is still not clear to what extent the experimenters and those doing the analysis were blinded to group. If one or both were blind to group, then please put this in the methods.

Reviewer #3 (Public Review):

Summary:

This study used prolonged stimulation of a limb to examine possible plasticity in somatosensory evoked potentials induced by the stimulation. They also studied the extent that the blood brain barrier (BBB) was opened by the prolonged stimulation and whether that played a role in the plasticity. They found that there was potentiation of the amplitude and area under the curve of the evoked potential after prolonged stimulation and this was long-lasting (>5 hrs). They also implicated extravasation of serum albumin, caveolae-mediated transcytosis, and TGFb signalling, as well as neuronal activity and upregulation of PSD95. Transcriptomics was done and implicated plasticity related genes in the changes after prolonged stimulation, but not proteins associated with the BBB or inflammation. Next, they address the application to humans using a squeeze ball task. They imaged the brain and suggest that the hand activity led to an increased permeability of the vessels, suggesting modulation of the BBB.

Strengths:

The strengths of the paper are the novelty of the idea that stimulation of the limb can induce cortical plasticity in a normal condition, and it involves opening of the BBB with albumin entry. In addition, there are many datasets and both rat and human data.

Weaknesses:

The conclusions are not compelling however because of a lack of explanation of methods. The explanation of why prolonged stimulation in the rat was considered relevant to normal conditions should be as clear in the paper as it is in the rebuttal. The authors need to ensure other aspects of the rebuttal are as clear in the paper as in the rebuttal too. The only remaining concern that is significant is that it is hard to understand the figures.

Author Response

Reviewer #1 (Public Review):

The goal of the current study was to evaluate the effect of neuronal activity on blood-brain barrier permeability in the healthy brain, and to determine whether changes in BBB dynamics play a role in cortical plasticity. The authors used a variety of well-validated approaches to first demonstrate that limb stimulation increases BBB permeability. Using in vivo-electrophysiology and pharmacological approaches, the authors demonstrate that albumin is sufficient to induce cortical potentiation and that BBB transporters are necessary for stimulus-induced potentiation. The authors include a transcriptional analysis and differential expression of genes associated with plasticity, TGF-beta signaling, and extracellular matrix were observed following stimulation. Overall, the results obtained in rodents are compelling and support the authors' conclusions that neuronal activity modulates the BBB in the healthy brain and that mechanisms downstream of BBB permeability changes play a role in stimulus-evoked plasticity. These findings were further supported with fMRI and BBB permeability measurements performed in healthy human subjects performing a simple sensorimotor task. While there are many strengths in this study, there is literature to suggest that there are sex differences in BBB dysfunction in pathophysiological conditions. The authors only used males in this study and do not discuss whether they would also expect to sex differences in stimulation-evoked BBB changes in the healthy brain. Another minor limitation is the authors did not address the potential impact of anesthesia which can impact neurovascular coupling in rodent studies. The authors could have also better integrated the RNAseq findings into mechanistic experiments, including testing whether the upregulation of OAT3 plays a role in cortical plasticity observed following stimulation. Overall, this study provides novel insights into how neurovascular coupling, BBB permeability, and plasticity interact in the healthy brain.

While there are many strengths in this study, there is literature to suggest that there are sex differences in BBB dysfunction in pathophysiological conditions. The authors only used males in this study and do not discuss whether they would also expect to sex differences in stimulation-evoked BBB changes in the healthy brain.

We agree with the reviewer regarding the importance of examining sex differences on stimulation-evoked BBB changes. To address this issue we have: (1) clarified in the methods section that the human study involved both males and females; (2) added a section to the discussion highlighting the male bias as a key limitation of our animal experiments; and (3) stated that future work should examine whether stimulation-evoked BBB changes differ between makes and females.

Another minor limitation is the authors did not address the potential impact of anesthesia which can impact neurovascular coupling in rodent studies.

We are grateful for this comment and agree with the reviewer that the potential effects of anesthesia should be discussed. We have added the following discussion paragraph:

“A key limitation of our animal experiments is the fact they were performed under anesthesia, due to the complex nature of the experimental setup (i.e., simultaneous cortical imaging and electrophysiological recordings). Anesthetic agents can affect various receptors within the NVU, potentially altering neuronal activity, SEPs, CBF, and vascular responses (Aksenov et al., 2015; Lindauer et al., 1993; Masamoto & Kanno, 2012). To minimize these effects, we used ketamine-xylazine anesthesia, which unlike other anesthetics, was shown to generate robust BOLD and SEP responses to neuronal activation (Franceschini et al., 2010; Shim et al., 2018).”

Reviewer #2 (Public Review):

Summary:

This study builds upon previous work that demonstrated that brain injury results in leakage of albumin across the bloodbrain barrier, resulting in activation of TGF-beta in astrocytes. Consequently, this leads to decreased glutamate uptake, reduced buffering of extracellular potassium, and hyperexcitability. This study asks whether such a process can play a physiological role in cortical plasticity. They first show that stimulation of a forelimb for 30 minutes in a rat results in leakage of the blood-brain barrier and extravasation of albumin on the contralateral but not ipsilateral cortex. The authors propose that the leakage is dependent upon neuronal excitability and is associated with an enhancement of excitatory transmission. Inhibiting the transport of albumin or the activation of TGF-beta prevents the enhancement of excitatory transmission. In addition, gene expression associated with TGF-beta activation, synaptic plasticity, and extracellular matrix are enhanced on the "stimulated" hemisphere. That this may translate to humans is demonstrated by a breakdown in the blood-brain barrier following activation of brain areas through a motor task.

Strengths:

This study is novel and the results are potentially important as they demonstrate an unexpected breakdown of the blood-brain barrier with physiological activity and this may serve a physiological purpose, affecting synaptic plasticity.

The strengths of the study are:

1. The use of an in vivo model with multiple methods to investigate the blood-brain barrier response to a forelimb stimulation.

1. The determination of a potential functional role for the observed leakage of the blood-brain barrier from both a genetic and electrophysiological viewpoint.

1. The demonstration that inhibiting different points in the putative pathway from activation of the cortex to transport of albumin and activation of the TGF-beta pathway, the effect on synaptic enhancement could be prevented.

1. Preliminary experiments demonstrating a similar observation of activity-dependent breakdown of the blood-brain barrier in humans.

Weaknesses:

There are both conceptual and experimental weaknesses.

1. The stimulation is in an animal anesthetized with ketamine, which can affect critical receptors (ie NMDA receptors) in synaptic plasticity.

We agree that the potential effects of anesthesia should be considered. The Discussion was revised to address this point: “A key limitation of our animal experiments is the fact they were performed under anesthesia, due to the complex nature of the experimental setup (i.e., simultaneous cortical imaging and electrophysiological recordings). Anesthetic agents can affect various receptors within the NVU, potentially altering neuronal activity, SEPs, CBF, and vascular responses (Aksenov et al., 2015; Lindauer et al., 1993; Masamoto & Kanno, 2012). To minimize these effects, we used ketamine-xylazine anesthesia, which unlike other anesthetics, was shown to generate robust BOLD and SEP responses to neuronal activation (Franceschini et al., 2010; Shim et al., 2018)”

1. The stimulation protocol is prolonged and it would be helpful to know if briefer stimulations have the same effect or if longer stimulations have a greater effect ie does the leakage give a "readout" of the stimulation intensity/length.

Thank you for this important comment. We are also very curious about the potential relationship between stimulation magnitude/duration and subsequent leakage and have added the following statement to the discussion:

“Future studies should also explore the effects of stimulation magnitude/duration on BBB modulation, as well as the stimulation threshold between physiological and pathological increase in BBB permeability.”

Our current findings indicate that a one-minute stimulation does not affect vascular permeability or SEP and we aim to test additional stimulation paradigms in future studies.

1. For some of the experiments (see below), the numbers of animals are low and the statistical tests used may not be the most appropriate, making the results less clear cut.

We appreciate this comment and have revised the statistical analysis of Figure 1J,K. We now use a nested t-test to test for differences between rats (as opposed to sections). The differences remain significant (EB, p=0.0296; Alexa, p=0.0229). The text was modified accordingly.

1. The experimental paradigms are not entirely clear, especially the length of time of drug application and the authors seem to try to detect enhancement of a blocked SEP.

Thank you for pointing this out. Figures 2&3 were revised for clarification and a ‘Drug Application’ subsection was added to the methods section.

1. It is not clear how long the enhancement lasts. There is a remark that it lasts longer than 5 hours but there is no presentation of data to support this.

Thank you for this comment. As the length of experiments differed between animals, the exact length could not be specifically stated. To clarify this point, we revised the text to indicate that LTP was recorded until the end of each experiment (between 1.5-5 hours, depending on the condition the animal was in). We also added a panel to figure 2 (Figure 2d) with exemplary data showing potentiation 60, 90, and 120 min post stimulation.

1. The spatial and temporal specificity of this effect is unclear (other than hemispheric in rats) and even less clear in humans.

Our animal experiments (using both in vivo imaging and histological analysis) showed no evidence of BBB modulation outside the cortical somatosensory area corresponding to the limbs. We looked at the entirety of the coronal section of the brain and found enhancement solely in the somatosensory area corresponding to limb. The right side of panels h and i in Figure 1 show an x20 magnification of the section, focusing on the enhanced area. The whole section was not shown, as no fluorescence was found outside the magnified area. Moreover, our quantification showed that the enhancement was specific to the contralateral and not ipsilateral somatosensory cortex (Figure 1 j-k).

We agree that temporal specificity needs to be further explored, and we have now stated that in the discussion: “Future studies are needed to explore the BBB modulating effects of additional stimulation protocols – with varying durations, frequencies, and magnitudes. Such studies may also elucidate the temporal and ultrastructural characteristics that may differentiate between physiological and pathological BBB modulation.”

We also agree that larger studies are needed to better understand the specificity of the observed effect in humans, and to account for potential inter-human variability in vascular integrity and brain function due to different schedules, diets, exercise habits, etc.

1. The experimenters rightly use separate controls for most of the experiments but this is not always the case, also raising the possibility that the application of drugs was not done randomly or interleaved, but possibly performed in blocks of animals, which can also affect results.

Thank you for pointing out this lack of clarity. We have now highlighted that drug application was done randomly.

1. Methyl-beta-cyclodextrin clears cholesterol so the effect on albumin transport is not specific, it could be mediating its effect through some other pathway.

We agree that the effect of mβCD may not be specific. To mitigate this issue, we used a very low mβCD concentration (10uM). Notably, this is markedly lower than the concentrations reported by Koudinov et al, showing that cholesterol depletion is observed at 5mM mβCD and not at 2.5mM/5mM (Koudinov & Koudinova, 2001). This point was added to the discussion.

1. Since the breakdown of the blood-brain barrier can be inhibited by a TGF-beta inhibitor, then this implies that TGFbeta is necessary for the breakdown of the blood-brain barrier. This does not sit well with the hypothesis that TGF-beta activation depends upon blood-brain barrier leakage.

Thank you for pointing out this lack of clarity. We have added a discussion paragraph that clarifies our hypothesis: “As mentioned above, albumin is a known activator of TGF-β signaling, and TGF-β has a well-established role in neuroplasticity. Interestingly, emerging evidence suggests that TGF-β also increases cross-BBB transcytosis (Betterton et al., 2022; Kaplan et al., 2020; McMillin et al., 2015; Schumacher et al., 2023). Hence, we propose the following two-part hypothesis for the TGF-β/BBB-mediated synaptic potentiation observed in our experiments: (1) prolonged stimulation triggers TGF-β signaling and increased caveolae-mediated transcytosis of albumin; and (2) extravasated albumin induces further TGF-β signaling, leading to synaptogenesis and additional cross-BBB transport – in a self-reinforcing positive feedback loop. Future research is needed to examine the validity of this hypothesis.

Reviewer #3 (Public Review):

Summary:

This study used prolonged stimulation of a limb to examine possible plasticity in somatosensory evoked potentials induced by the stimulation. They also studied the extent that the blood-brain barrier (BBB) was opened by prolonged stimulation and whether that played a role in the plasticity. They found that there was potentiation of the amplitude and area under the curve of the evoked potential after prolonged stimulation and this was long-lasting (>5 hrs). They also implicated extravasation of serum albumin, caveolae-mediated transcytosis, and TGFb signalling, as well as neuronal activity and upregulation of PSD95. Transcriptomics was done and implicated plasticity-related genes in the changes after prolonged stimulation, but not proteins associated with the BBB or inflammation. Next, they address the application to humans using a squeeze ball task. They imaged the brain and suggested that the hand activity led to an increased permeability of the vessels, suggesting modulation of the BBB.

Strengths:

The strengths of the paper are the novelty of the idea that stimulation of the limb can induce cortical plasticity in a normal condition, and it involves the opening of the BBB with albumin entry. In addition, there are many datasets and both rat and human data.

Weaknesses:

The conclusions are not compelling however because of a lack of explanation of methods and quantification. It also is not clear whether the prolonged stimulation in the rat was normal conditions. To their credit, the authors recorded the neuronal activity during stimulation, but it seemed excessive excitation. Since seizures open the BBB this result calls into question one of the conclusions. that the results reflect a normal brain. The authors could either conduct studies with stimulation that is more physiological or discuss the caveats of using a supraphysiological stimulus to infer healthy brain function.

The conclusions are not compelling however because of a lack of explanation of methods and quantification.

Thank you for this comment. In the revised paper, we expanded the Methods section to better describe the procedures and approaches we used for data analysis.

It also is not clear whether the prolonged stimulation in the rat was normal conditions.

We believe that the used stimulation protocol is within the physiological range (and relevant to plasticity, learning and memory) for the following reasons:

1. In our continuous electrophysiological recordings, we did not observe any form of epileptiform or otherwise pathological activity.

2. Memory/training/skill acquisition experiments in humans often involve similar training duration or longer (Bengtsson et al., 2005), e.g., a 30 min thumb training session performed by (Classen et al., 1998).

3. The levels of SEP potentiation we observed are similar to those reported in:

a) Rats following a 10-minute whisker stimulation (one hour post stimulation, (Mégevand et al., 2009)).

b) Humans following a 15 min task (McGregor et al., 2016).

This important point is now presented in the discussion.

Reviewer #1 (Recommendations For The Authors):

The discussion would benefit from additional discussion of the potential impacts of sex and anesthesia in their findings.

We agree with the reviewer and have added the following paragraph to the discussion:

“A key limitation of our animal experiments is the fact they were performed under anesthesia, due to the complex nature of the experimental setup (i.e., simultaneous cortical imaging and electrophysiological recordings). Anesthetic agents can potentially alter neuronal activity, SEPs, CBF, and vascular responses (Aksenov et al., 2015; Lindauer et al., 1993; Masamoto & Kanno, 2012). To minimize these effects, we used ketaminexylazine anesthesia, which unlike other anesthetics, was shown to maintain robust BOLD and SEP responses to neuronal activation (Franceschini et al., 2010; Shim et al., 2018). Another limitation of our animal study is the potentially non-specific effect of mβCD – an agent that disrupts caveola transport but may also lead to cholesterol depletion (Keller & Simons, 1998). To mitigate this issue, we used a very low mβCD concentration (10uM), orders of magnitude below the concentration reported to deplete cholesterol (Koudinov et al). Lastly, our animal study is limited by the inclusion of solely male rats. While our findings in humans did not point to sex-related differences in stimulation-evoked BBB modulation, larger animals and human studies are needed to examine this question.”

The figure text is quite small.

Thank you for pointing this out, we revised all figures and increased font size for clarity.

Including pharmacological concentrations within the figure legends would improve the readability of the manuscript.

Thank you for this suggestion, the figure legends were modified accordingly.

In methods for immunoassays the 5 groups could be more clear by stating that there are 3 timepoints for stimulation experiments. There is a typo in this section where the 24-hour post is stated twice in the same sentence.

Thank you for pointing this out, the text was modified accordingly.

Reviewer #2 (Recommendations For The Authors):

1. In Figure 1, J and K seem to indicate that in these experiments the statisitics were done per slice and not per animal. This is not a reasonable approach, a repeat measure ANOVA or averaging for each animal are more appropriate statistical approaches.

We thank the reviewer for pointing this out. The statistical analysis for Figure 1j,k was modified. We now use a nested ttest to test for differences between rats and not sections. The differences are still significant (EB, p=0.0296; Alexa, p=0.0229). The manuscript was modified accordingly.

1. In Figure 2, the protocol does not seem to give much idea about time course. There was a stimulation test for 1 minute before and then 1 minute after the 30-minute stimulation train. How was potentiation assessed for the next 5 hours and where are the data?

Potentiation was assessed by repeating 1min test stim every 30 min for the duration of the experiment, we added a panel to show late potentiation, see response above.

1. In Figure 2, there is a notable lack of controls eg the effect of sham stimulation and application of saline. These are important as the drift of response magnitude can be a problem in long experiments.

We did test for the potential presence of response drift, by examining whether SEPs of non-stimulated animals change over time (at baseline, 30 or 60 minutes of recording; n=6). No statistical differences were found. Our analysis focused on using each animal as its own control (i.e., comparing baseline SEP to SEP post albumin perfusion), because SEP studies highlight the importance of comparing each animal to its own baseline, due to the large inter-animal variability (All et al., 2010; Mégevand et al., 2009; Zandieh et al., 2003).

1. Figure 3 a is not clear – were the drugs applied throughout?

Thank you for pointing this out. We have revised Figure 3 a to show that the drugs were applied for 50 min before the stimulation.

1. In Figure 3 panel d is repeated in panel j. This needs correcting

Thank you. This mistake was fixed.

1. In LTP-type experiments usually the antagonist is applied during the stimulation and then washed out. This avoids the problem in this figure in which CNQX effectively blocks transmission and so it is not possible to detect any enhancement if it were there. Eg in panel e, CNQX block transmission, and then the assessment is performed when the AMPA receptors are blocked after 30 minutes of stimulation. If receptors are blocked no enhancement will be detectable. Moreover, surely the question is the ratio of the effect of 30-minute stimulation on the SEP in the presence of CNQX and so the statistics should be done on the fold change in the SEP following 30-minute stimulation in the presence of CNQX.

Thank you. The protocol might have been misrepresented in the original figure. We modified Fig 3a to clarify that the antagonists were indeed washed out upon stimulation start to make sure the receptors are not blocked during the test stimulation following the 30 min stimulation. In addition, we tested for the difference in fold change between 30 min stim, and 30 min stimulation following antagonists wash-in (Fig 3f and Fig S2a).

1. Interesting in Figure f, stimulation, albumin, and AP5 all seem to have the same enhancement of the SEP. Is the lack of effect of 30-minute stimulation in the presence of AP5, a ceiling effect ie AP5 has enhanced the SEP, and no further enhancement from stimulation is possible.

This is a very interesting point that will require further research.

1. SJN seems to block neurotransmission. What is the mechanism? The same analysis as for CNQX should be performed ie what is the fold change not compared to baseline but in the presence of SJN.

Our quantification showed that SJN did not significantly reduce the SEP max amplitude, and we therefore did not include this graph in the figure.

1. Please acknowledge that the effect of mbetaCD is non-specific. There is a large literature on the effects of cholesterol depletion on LTP.

We agree that the effect of mβCD may not be specific. To mitigate this issue, we used a very low mβCD concentration (10µM). Notably, this is markedly lower than the concentrations reported by Koudinov et al, showing that cholesterol depletion is only observed at a concentration of 5mM (Koudinov & Koudinova, 2001). This point is now discussed under the discussion paragraph describing the study’s limitations.

1. k&l seem to have used the same control in which case they should not be analysed separately (they are all part of the same experiment).

We agree with the reviewer and have revised the figure accordingly.

1. The difference in gene expression in Figure 4 would be more convincing if it could be prevented by for example a TGFbeta inhibitor.

We agree and acknowledge the impact such experiments could provide. We plan to incorporate these experiments into our future studies.

1. Figure 5 seems to indicate bilateral and widespread BBB modulation arguing that this may be a non-specific effect. Panel g should look at other neocortical regions eg occipital cortex.

We agree and thank the reviewer for this comment. We revised the figure to include other cortical areas, such as the frontal and occipital cortices (Figure 5g)

Minor comments

1. Paired data eg in Fig 2D are better represented by pairing the dots usually with a line.

1. Please correct the %fold baseline in axes in graphs which show % change for baseline.

1. Figure 4 is not correctly referred to in the text.

We agree with all the points raised by the reviewer and revised the figures and text accordingly.

Reviewer #3 (Recommendations For The Authors):

The conclusions are not compelling however because of a lack of explanation of methods and quantification. It also is not clear whether the prolonged stimulation in the rat was normal conditions. To their credit, the authors recorded the neuronal activity during stimulation, but it seemed excessive excitation. Since seizures open the BBB this result calls into question one of the conclusions. that the results reflect a normal brain. The authors could either conduct studies with stimulation that is more physiological or discuss the caveats of using a supraphysiological stimulus to infer healthy brain function.

Major concerns:

Methods need more explanation. Rationales need more justification. Examples are provided below.

Throughout many sections of the paper, sample sizes and stats are often missing. For stats, please provide p-values and other information (tcrit, U statistic, F, etc.)

Thank you, we added the relevant information where it was missing throughout the manuscript.

For transcriptomics, they might have found changes in BBB-related genes if they assayed vessels but they assayed the cortex.

We agree with the reviewer that this would be a very interesting future direction. The present study could not include this kind of analysis due to lack of access to vasculature isolation methods or single-cell RNA seq.

What were the inclusion/exclusion criteria for the subjects?

Thank you for pointing out this lack of clarity. The methods section (under ‘Magnetic Resonance Imaging’ – ‘Participants’) was expanded to include the following:

“Male and female healthy individuals, aged 18-35, with no known neurological or psychiatric disorders were recruited to undergo MRI scanning while performing a motor task (n=6; 3 males and 3 females). MRI scans of 10 sex- and age- matched individuals (with no known neurological or psychiatric disorders) who did not perform the task were used as control data (n=10; 5 males and 5 females.

Were they age and sex-matched?

They were, indeed, age and sex-matched. This was now clarified in the relevant Methods section.

Were there other factors that could have influenced the results?

Certainly. Human subjects are difficult to control for due to different schedules, diets, exercise habits, and other factors that may impact vascular integrity and brain function. Larger multimodal studies are needed to better understand the observed phenomenon.

Fig. 1. Images are very dim. Text here and in other figures is often too small to see. Some parts of the figures are not explained.

Our apologies. Figures and legends were revised accordingly.

Fig 2a, f. I don't see much difference here- do the authors think there was?

We agree that the difference may not be visually obvious. The quantification of trace parameters (amplitude and area under curve) does, however, reveal a significant SEP difference in response to both stimulation (panels X and y) and albumin (panels z and q).

Fig 3 d and j seem the same.

We thank the reviewer for noticing. This was a copy mistake that was now rectified.

Lesser concerns and examples of text that need explana9on:

Introduction

Insulin-like growth factor is transported. From where to where?

The text was edited to clarify that this was cross-BBB influx of insulin-like growth factor-I.

RMT that underlies the transport of plasma proteins was induced by physiological or non-physiological stimulation.

This was shown without stimulation, in normal physiology of young and aged healthy mice. The text was edited to clarify this point.

What was the circadian modulation that was shown to implicate BBB in brain function?

The text was edited for clarity.

Results

When the word stimulation is used please be specific if whiskers are moved by an experimenter, an electrode is used to apply current, etc.

We have now moved the ‘Stimulation protocol’ section closer to beginning of the Methods and emphasized that we administered electrical stimulation to the forepaw or hindlimb using subdermal needle electrodes.

Please explain how the authors are convinced they localized the vascular response.

The vascular response was localized via: (1) visual detection of arterioles that dilated in response to stimulation (due to functional hyperemia / neurovascular coupling) [figure 1 d]; and (2) quantitative mapping of increased hemoglobin concentration (Bouchard et al., 2009) [Figure 1 b]. This is now mentioned in the methods (under ‘In vivo imaging’) and results (under the ‘Stimulation increases BBB permeability’).

"30 min of limb stimulation" means what exactly? 6 Hz 2mA for 30 min?

Thank you. The text was revised for clarity (Methods under ‘Stimulation protocol’):

“The left forelimb or hind limb of the rat was stimulated using Isolated Scmulator device (AD Instruments) attached with two subdermal needle electrodes (0.1 ms square pulses, 2-3 mA) at 6 Hz frequency. Test stimulation consisted of 360 pulses (60 s) and delivered before (as baseline) and after long-duration stimulation (30 min, referred throughout the text as ‘stimulation’). In control and albumin rats, only short-duration stimulations were performed. Under sham stimulation, electrodes were placed without delivering current.”

Histology that was performed to confirm extravasation needs clarification because if tissue was removed from the brain, and fixed in order to do histology, what is outside the vessels would seem likely to wash away.

Thank you for pointing out the need to clarify this point. The Histology description in the Methods section was revised in the following manner:

“Albumin extravasacon was confirmed histologically in separate cohorts of rats that were anesthetized and stimulated without craniotomy surgery. Assessment of albumin extravasacon was performed using a well-established approach that involves peripheral injection of either labeled-albumin (bovine serum albumin conjugated to Alexa Flour 488, Alexa488-Alb) or albumin-labeling dye (Evans blue, EB – a dye that binds to endogenous albumin and forms a fluorescent complex), followed by histological analysis of brain tissue (Ahishali & Kaya, 2020; Ivens et al., 2007; Lapilover et al., 2012; Obermeier et al., 2013; Veksler et al., 2020). Since extravasated albumin is taken up by astrocytes (Ivens et al., 2007; Obermeier et al., 2013), it can be visualized in the brain neuropil after brain removal and fixation (Ahishali & Kaya, 2020; Ivens et al., 2007; Lapilover et al., 2012; Veksler et al., 2020). Five rats were injected with Alexa488-Alb (1.7 mg/ml) and five with EB (2%, 20 mg/ml, n=5). The injections were administered via the tail vein. Following injection, rats were transcardially perfused with…”

It is not clear why there was extravasacon contralateral but not ipsilateral if there are cortical-cortical connections.

Interpersonally, we also did not observe ipsilateral SEP in response to limb stimulation, with evidence of SEP and BBB permeability only in the contralateral sensorimotor region. This finding is consistent with electrophysiological and fMRI studies showing that peripheral stimulation results in predominantly contralateral potentials (Allison et al., 2000; Goff et al., 1962).

After injection of Evans blue or Alexa-Alb, how was it shown that there was extravasacon?

Extravasalon in cortical sections was visualized using a fluorescent microscope (Figure 1 h-i). Since extravasated albumin is taken up by astrocytes, fluorescent imaging can be used for visualizing and quantifying labeled albumin (Ahishali & Kaya, 2020; Ivens et al., 2007; Knowland et al., 2014). Here is the relevant methods excerpt:

“Coronal sections (40-μm thick) were obtained using a freezing microtome (Leica Biosystems) and imaged for dye extravasacon using a fluorescence microscope (Axioskop 2; Zeiss) equipped with a CCD digital camera (AxioCam MRc 5; Zeiss).”

How is a sham control not stimulated - what is the sham procedure?

In the sham stimulation protocol electrodes were placed, but current was not delivered. A section titled ‘Stimulation protocol’ was added to the methods to clarify this point.

What was the method for photothrombosis-induced ischemia?

The procedure for photothrombosis-induced ischemia is described under the Methods section ‘Immunoassays’ – ‘Enzyme-linked immunosorbent assay (ELISA) for albumin extravasalon’:

“Rats were anesthetilzed and underwent … photothrombosis stroke (PT) as previously described (Lippmann et al., 2017; Schoknecht et al., 2014). Briefly, Rose Bengal was administered intravenously (20 mg/kg) and a halogen light beam was directed for 15 min onto the intact exposed skull over the right somatosensory cortex.”

Fig 1d. All parts of d are not explained.

Thank you for pointing this out. In the revised manuscript, the panels of this figure were slightly reordered, and we made sure all panels are explained in the legend.

e. Is the LFP a seizure? How physiological is this- it does not seem very physiological.

Thank you for your comment. We believe that this activity is not a seizure because it lacks the typical slow activity that corresponds to the “depolarizalon shir” observed during seizures (Ivens et al., 2007; Milikovsky et al., 2019; Zelig et al., 2022).

f. Permeability index needs explanation. How was the area chosen for each rat? Randomly? Was it the same across rats?

We have now revised the Methods section to provide a clearer description of the permeability index calculation and the choice of the imaging area:

“Across all experiments, acquired images were the same size (512 × 512 pixel, ~1x1 mm), centered above the responding arteriole. Images were analyzed offline using MATLAB as described (Vazana et al., 2016). Briefly, image registration and segmentation were performed to produce a binary image, separating blood vessels from extravascular regions. For each extravascular pixel, a time curve of signal intensity over time was constructed. To determine whether an extravascular pixel had tracer accumulation over time (due to BBB permeability), the pixel’s intensity curve was divided by that of the responding artery (i.e., the arterial input function, AIF, representing tracer input). This ratio was termed the BBB permeability index (PI), and extravascular pixels with PI > 1 were identified as pixels with tracer accumulation due to BBB permeability.”

g. For Evans blue and Alexa-Alb was the sample size rats or sections?

Thank you for this question. We revised the statistical analysis for Figure 1j,k to appropriately asses the differences between rats. We used a nested t-test to test for differences between rats (and not sections). The differences remained significant (EB, p=0.0296; Alexa, p=0.0229) and the text was modified accordingly.

h, i, j need more contrast and/or brightness to appreciate the images. Arrows would help. The text is too small to read.

Thank you. This issue was addressed in the revised paper.

To induce potentiation, 6 Hz 2 mA stimuli were used for 30 min. Please justify this as physiological.

Thank you for the comment. We believe that the used stimulation protocol is within the physiological range (and relevant to plasticity, learning and memory) for the following reasons:

1. In our continuous electrophysiological recordings, we did not observe any form of epileptiform or otherwise pathological activity.

2. Memory/training/skill acquisition experiments in humans often involve similar training duration or longer (Bengtsson et al., 2005), e.g., a 30 min thumb training session performed by (Classen et al., 1998).

3. The levels of SEP potentiation we observed are similar to those reported in:

a. Rats following a 10-minute whisker stimulation (one hour post stimulation, (Mégevand et al., 2009)).

b. Humans following a 15 min task (McGregor et al., 2016).

We have revised the Discussion of the paper to clarify this important point.

The test stimulus to evoke somatosensory evoked potentials was 1 min. Was this 6 Hz 2 mA for 1 min? Please justify.

Yes. We chose these parameters as these ranges were shown to induce the largest changes in blood flow (with laserdoppler flowmetry) and summated SEP (Ngai et al., 1999), corresponding with our findings. We also show that these stimulation parameters do not induce changes in BBB permeability nor synaptic potentiation, therefore served as test control.

How long after the 30 min was the test stimulus triggered- immediately? 30 sec afterwards?

The test stimulus was applied 5 min afterwards to allow for BBB imaging protocol (now explained in the Methods section).

How were amplitude and AUC measured? Baseline to peak? For AUC is it the sum of the upward and downward deflections comprising the LFP?

Yes, and yes. This is now clarified in the ‘Analysis of electrophysiological recordings’ section in the Methods.

How was the same site in the somatosensory cortex recorded for each animal?
Potentiation was said to last >5 hrs. How often was it measured? Was potentiation the same for the amplitude and the AUC?

The location of the cranial window over the somatosensory cortex was the same in all rats. The location of the specific responding arteriole may change between animals, but the recording electrode was places around the responding arteriole in the same approaching angle and depth for all animals.

As the length of experiments differed between animals, the exact length could not be specifically stated. We therefore revised the text to clarify that LTP was recorded until the end of each experiment (depending on the animal condition, between 1.5-5 hours) and added a panel to figure 2 (Figure 2f) with exemplary data showing potentiation 120 min (2hr) post stimulation.

Why was 25% of the serum level of albumin selected- does the brain ever get exposed to that much? Was albumin dissolved in aCSF or was aCSF chosen as a control for another reason?

Yes, albumin was dissolved in aCSF and the solution was allowed to diffuse through the brain. The relatively high concentration of albumin was chosen to account for factors that lower its effective tissue concentration:

1. The low diffusion rate of albumin (Tao & Nicholson, 1996).

2. The likelihood of albumin to encounter a degradation site or a cross-BBB efflux transporter (Tao & Nicholson, 1996; Zhang & Pardridge, 2001).

Figure 2.

a. Please show baseline, the stimulus, and aftier the stimulus.

Please point out when there was stimulacon.

What is the inset at the top?

The inset on top is the example trace of the stimulus waveform, the legend of the figure was modified for clarity.

b. Please show when the stimulus artifact occurred. The end of the 1-minute test stimulus period is fine. Why are the SEPs different morphologies? It suggests the different locations in the cortex were recorded.

What is shown is the averaged SEP response over 1min test stimulus, each SEP is time locked to each stimulus. Regarding SEP waveform, it does indeed show different morphology between animals, as sometimes different arterioles respond to the stimulation, and we localize the recording to the responding vessel in each rat. However, in each rat the recording is only from one location. Once the electrode was positioned near the responding arteriole it was not moved.

d, e. What are the stats?

h, i. Add stats. Are all comparisons Wilcoxon? Please provide p values.

The comparisons were performed with the Wilcoxon test. We now state that and provide the exact p values.

j. What was selected from the baseline and what was selected during Albumin and how long of a record was selected?

What program was used to create the spectrogram?

What is meant by changes at frequencies above 200 Hz, the frequencies of HFOs?

The Method section (under ‘Electrophysiology – Data acquisition and analyses’) has been revised for clarification. Spectrogram was created with MATLAB and graphed with Prism. For analysis, we selected a 10 min recorded segment before starting albumin perfusion, and 10 min after terminating albumin perfusion.

When the cortial window was exposed to drugs, what were concentrations used that were selective for their receptor? How long was the exposure?

Was the vehicle tested?

We have revised the Methods section (under ‘Animal preparation and surgical procedures - Drug application’) to clarify the duration and concentration used and justification. All blockers were exposed for 50 min. The vehicle was an artificial cerebrospinal fluid solution (aCSF).

For PSD-95, what was the area of the cortex that was tested?

Were animals acutely euthanized and the brain dissected, frozen, etc?

We have revised the Methods section (under ‘Immunoassays’) for clarity.

What is mbetaCD?

The full term was added to the results section. It is also mentioned in the Methods.

Is SJN specific at the concentration that was chosen? Did it inhibit the SEP?

In the concentration used in our experiments, SJN is a selective TGF-β type I receptor ALK5 inhibitor (see (Gellibert et al., 2004)).

Fig. 3b. It looks like CNQX increased the width of the vessels quite a bit. Please explain.

For AP5, very large vessels were imaged, making it hard to compare to the other data.

The vascular dilation in response to the stimulation under CNQX was similar to that seen under “normal” conditions (i.e. aCSF). As for AP5, in some experiments the responding arteriole was in close proximity to a large venule that cannot be avoidable while imaging. For quantification we always measured arterioles within the same diameter range.

e. Sometimes CNQX did not block the response after 30 min stimulation. Why?

CNQX is washed out before the 30 min stimulation starts, so it is not expected to block the response to stimulation. However, in some cases the response to stimulation was lower in amplitude, likely due to residual CNQX that did not wash out completely.

Regarding DEGs, on the top of p 10 what are the percentages of?

In this analysis we tested in each hemisphere how many genes expressed differentially between 1 and 24 hours post stimulation (either up- or down- regulated). The results were presented as the percentages of differentially expressed genes in each hemisphere (13.2% contralateral, and 7.3% ipsilateral). The text was rephrased for clarity.

Please add a ref for the use of the JSD metric methods and support for its use as the appropriate method. Other methods need explanation/references.

References were added to the text to clarify. The Jensen-Shannon Divergence metric is commonly used to calculate the statistical pairwise distance among two distributions (Sudmant et al., 2015). From comparing a few different distance metric calculations including JSD, our results were similar irrespective of the distance metric applied. Therefore, we demonstrate the variability between paired samples of stimulated and non-stimulated cortex of each animal at two time points following stimulation (24 h vs. 1 h) using JSD.

What synaptic plasticity genes were selected for assay and what were not?

What does "largely unaffected" mean? Some of the genes may change a small amount but have big functional effects.

The selected genes of interest were taken from a large list compiled from previous publications (see (Cacheaux et al., 2009; Kim et al., 2017)) and are well documented in gene ontology databases and tools (e.g., Metascape, (Zhou et al., 2019)).

We agree that the term ‘largely unaffected’ is suboptimal, and we rephrased this section of the results to indicate that “No significant differences were found in BBB or inflammation related genes between the hemispheres”. We also agree that a small number of genes can have big functional effects. Future studies are needed to better understand the genes underlying the observed BBB modulation.

Please note that Slc and ABCs are not only involved in the BBB.

Thank you. We modified the text to no longer specify that these are BBB-specific transporters.

Please explain the choice of the stress ball squeeze task, and DCE.

DCE is a well-established method for BBB imaging in living humans, and it is cited throughout the manuscript. The ball squeeze task was chosen as it is presumed to involve primarily sensory motor areas, without high-level processing (Halder et al., 2005). This is now stated in the discussion.

What is Gd-DOTA?

Gd-DOTA is a gadolinium-based contrast agent (gadoterate meglumine, AKA Dotarem). Text was revised for clarity. Please see the Methods section under ‘Magnetic Resonance Imaging’ - ‘Data Acquisition’.

What does a higher percentage of activated regions mean- how was activacon defined and how were regions counted?

Higher percentage of activated regions refers to regions in which voxels showed significant BOLD changes due to the motor task preformed. The statistical approaches and analyses are detailed in the Methods section under ‘Magnetic Resonance Imaging - Preprocessing of functional data, and fMRI Localizer Motor Task’.

Figure. 4

Was stimulation 1 min or 30 min.?

30 min, Text has been revised for clarity.

What is the Wald test and how were p values adjusted-please add to the Stats section.

The Methods section under ‘Statistical analysis’ was revised to clarify this point.

Is there a reason why p values are sometimes circles and otherwise triangles?

The legend was revised to explain that ”Circles represent genes with no significant differences between 1 and 24 h poststimulation. Upward and downward triangles indicate significantly up- and down- regulated genes, respectively.”

How can a p-value be zero? Please explain abbreviations.

The p-value is very low (~10-10) and therefore appears to be zero due to the scale of the y-axis.

Fig. 5b.

There are unexplained abbreviations.

The x on the ball and hand is not clear relative to the black ball and hand.

Thank you for noticing. We revised the figure for clarity.

c. What was the method used to make an activator map and what is meant by localizer task?

The explanation of the “fMRI Localizer Motor Task” section in the methods was revised for added clarity.

f. What is the measurement "% area" that indicates " BBB modulation"?

Is it in f, the BBB permeable vessels (%)? f. Please explain: "Heatmap of BBB modulated voxels percentage in motor/sensory-related areas of task vs. controls."

The %area measurement indicates the percentage of voxels within a specific brain region that have a leaky BBB. See Methods.

Is Task - the control?

Yes.

Supplemental Fig. 2.

Why is AUC measured, not amplitude?

The amplitude, and now also the AUC are shown in Figure 3.

b. There is no comparison to baseline. The arrowhead points to the start of stimulation but there is no arrowhead marking the end.

In the revised paper we added a grey shade over the stimulation period to better visualize the difference to baseline. In this panel we wanted to show that NMDA receptor antagonist did not block the SEP, while AMPA receptor antagonist did.

c. In the blot there are two bands for PSD95- which is the one that is PSD95? There is no increase in PSD95 uncl 24 hrs but in the graph in d there is. In the blot, there is a strong expression of PSD95 ipsilateral compared to contralateral in the sham-why?

What is the percent change fold?

The PSD-95 is the top and larger band. The lower band was disregarded in the analysis. The example we show may not fully reflect the group statistics presented in panel d. Upon quantification of 8 animals, PSD-95 is significantly higher 30 min and 24 hours post stimulation in the contralateral hemisphere. No significant changes were found in sham animals. The % change fold refers to the AUC change compared to baseline. This panel was now incorporated in Figure 3 (panel h), and the title was corrected to “|AUC|, % change from baseline”.

Supplemental Fig. 4.

a. If ipsilateral and contralateral showed many changes why do the authors think the effects were only contralateral?

Our gene analysis was designed to complement our in vivo and histological findings, by assessing the magnitude of change in differentially expressed genes (DEGs). This analysis showed that: (1) the hemisphere contralateral to the stimulus has significantly more DEGs than the ipsilateral hemisphere; and (2) the DEGs were related to synaptic plasticity and TGF-b signaling. These findings strengthen the hypothesis raised by our in vivo and histological experiments.

Supplemental Fig. 5 includes many processes not in the results. Examples include dorsal cuneate and VPL, dynamin, Kir, mGluR, etc. The top right has numbers that are not mentioned. If the drawings are from other papers they should be cited.

The drawings of Figure 5 are original and were not published before. This hypothesis figure points to mechanisms that may drive the phenomena described in the paper. The legend of the figure was revised to include references to mechanisms that were not tested in this study.

Papers referenced in this letter:

Ahishali, B., & Kaya, M. (2020). Evaluation of Blood-Brain Barrier Integrity Using Vascular Permeability Markers: Evans Blue, Sodium Fluorescein, Albumin-Alexa Fluor Conjugates, and Horseradish Peroxidase. Methods in Molecular Biology, 2367, 87–103. https://doi.org/10.1007/7651_2020_316

Aksenov, D. P., Li, L., Miller, M. J., Iordanescu, G., & Wyrwicz, A. M. (2015). Effects of anesthesia on BOLD signal and neuronal activity in the somatosensory cortex. Journal of Cerebral Blood Flow and Metabolism, 35(11), 1819–1826. https://doi.org/10.1038/jcbfm.2015.130

All, A. H., Agrawal, G., Walczak, P., Maybhate, A., Bulte, J. W. M., & Kerr, D. A. (2010). Evoked potential and behavioral outcomes for experimental autoimmune encephalomyelitis in Lewis rats. Neurological Sciences, 31(5), 595–601. https://doi.org/10.1007/s10072-010-0329-y

Allison, J. D., Meador, K. J., Loring, D. W., Figueroa, R. E., & Wright, J. C. (2000). Functional MRI cerebral activation and deactivation during finger movement. Neurology, 54(1), 135–142. https://doi.org/10.1212/wnl.54.1.135

Bengtsson, S. L., Nagy, Z., Skare, S., Forsman, L., Forssberg, H., & Ullén, F. (2005). Extensive piano practicing has regionally specific effects on white matter development. Nature Neuroscience, 8(9), 1148–1150. https://doi.org/10.1038/nn1516

Betterton, R. D., Abdullahi, W., Williams, E. I., Lochhead, J. J., Brzica, H., Stanton, J., Reddell, E., Ogbonnaya, C., Davis, T. P., & Ronaldson, P. T. (2022). Regula/on of Blood-Brain Barrier Transporters by Transforming Growth Factor-β/Activin Receptor-Like Kinase 1 Signaling: Relevance to the Brain Disposition of 3-Hydroxy-3-Methylglutaryl Coenzyme A Reductase Inhibitors (i.e., Sta/ns). Drug Metabolism and Disposition, 50(7), 942–956. https://doi.org/10.1124/dmd.121.000781

Bouchard, M. B., Chen, B. R., Burgess, S. A., & Hillman, E. M. C. (2009). Ultra-fast multispectral optical imaging of cortical oxygenation, blood flow, and intracellular calcium dynamics. Optics Express, 17(18), 15670. https://doi.org/10.1364/oe.17.015670

Cacheaux, L. P., Ivens, S., David, Y., Lakhter, A. J., Bar-Klein, G., Shapira, M., Heinemann, U., Friedman, A., & Kaufer, D. (2009). Transcriptome profiling reveals TGF-β signaling involvement in epileptogenesis. Journal of Neuroscience, 29(28), 8927–8935. https://doi.org/10.1523/JNEUROSCI.0430-09.2009

Classen, J., Liepert, J., Wise, S. P., Hallett, M., & Cohen, L. G. (1998). Rapid plasticity of human cortical movement representation induced by practice. Journal of Neurophysiology, 79(2), 1117–1123. https://doi.org/10.1152/JN.1998.79.2.1117/ASSET/IMAGES/LARGE/JNP.JA47F4.JPEG

Franceschini, M. A., Radhakrishnan, H., Thakur, K., Wu, W., Ruvinskaya, S., Carp, S., & Boas, D. A. (2010). The effect of different anesthetics on neurovascular coupling. NeuroImage, 51(4), 1367–1377. https://doi.org/10.1016/j.neuroimage.2010.03.060

Gellibert, F., Woolven, J., Fouchet, M. H., Mathews, N., Goodland, H., Lovegrove, V., Laroze, A., Nguyen, V. L., Sautet, S., Wang, R., Janson, C., Smith, W., Krysa, G., Boullay, V., De Gouville, A. C., Huet, S., & Hartley, D. (2004). Identification of 1,5-naphthyridine derivatives as a novel series of potent and selective TGF-β type I receptor inhibitors. Journal of Medicinal Chemistry, 47(18), 4494–4506. https://doi.org/10.1021/jm0400247

Goff, W. R., Rosner, B. S., & Allison, T. (1962). Distribution of cerebral somatosensory evoked responses in normal man. Electroencephalography and Clinical Neurophysiology, 14(5), 697–713. https://doi.org/10.1016/0013-4694(62)90084-6

Halder, P., Sterr, A., Brem, S., Bucher, K., Kollias, S., & Brandeis, D. (2005). Electrophysiological evidence for cortical plasticity with movement repetition. European Journal of Neuroscience, 21(8), 2271–2277. https://doi.org/10.1111/J.1460-9568.2005.04045.X

Ivens, S., Kaufer, D., Flores, L. P., Bechmann, I., Zumsteg, D., Tomkins, O., Seiffert, E., Heinemann, U., & Friedman, A. (2007). TGF-β receptor-mediated albumin uptake into astrocytes is involved in neocortical epileptogenesis. Brain, 130(2), 535–547. https://doi.org/10.1093/brain/awl317

Kaplan, L., Chow, B. W., & Gu, C. (2020). Neuronal regulation of the blood–brain barrier and neurovascular coupling. In Nature Reviews Neuroscience (Vol. 21, Issue 8, pp. 416–432). Nature Research. https://doi.org/10.1038/s41583-020-0322-2

Keller, P., & Simons, K. (1998). Cholesterol is required for surface transport of influenza virus hemagglutinin. Journal of Cell Biology, 140(6), 1357–1367. https://doi.org/10.1083/jcb.140.6.1357

Kim, S. Y., Senatorov, V. V., Morrissey, C. S., Lippmann, K., Vazquez, O., Milikovsky, D. Z., Gu, F., Parada, I., Prince, D. A., Becker, A. J., Heinemann, U., Friedman, A., & Kaufer, D. (2017). TGFβ signaling is associated with changes in inflammatory gene expression and perineuronal net degradation around inhibitory neurons following various neurological insults. Scientific Reports, 7(1), 7711. https://doi.org/10.1038/s41598-017-07394-3

Knowland, D., Arac, A., Sekiguchi, K. J., Hsu, M., Lutz, S. E., Perrino, J., Steinberg, G. K., Barres, B. A., Nimmerjahn, A., & Agalliu, D. (2014). Stepwise Recruitment of Transcellular and Paracellular Pathways Underlies Blood-Brain Barrier Breakdown in Stroke. Neuron, 82(3), 603–617. https://doi.org/10.1016/j.neuron.2014.03.003

Koudinov, A. R., & Koudinova, N. V. (2001). Essen/al role for cholesterol in synaptic plasticity and neuronal degeneration. The FASEB Journal, 15(10), 1858–1860. https://doi.org/10.1096/r.00-0815re

Lapilover, E. G., Lippmann, K., Salar, S., Maslarova, A., Dreier, J. P., Heinemann, U., & Friedman, A. (2012). Periinfarct blood-brain barrier dysfunction facilitates induction of spreading depolarization associated with epileptiform discharges. Neurobiology of Disease, 48(3), 495–506. htttts://doi.org/10.1016/j.nbd.2012.06.024

Lindauer, U., Villringer, A., & Dirnagl, U. (1993). Characterization of CBF response to somatosensory stimulation: Model and influence of anesthetics. American Journal of Physiology - Heart and Circulatory Physiology, 264(4 33-4), 223–1228. https://doi.org/10.1152/ajpheart.1993.264.4.h1223

Lippmann, K., Kamintsky, L., Kim, S. Y., Lublinsky, S., Prager, O., Nichtweiss, J. F., Salar, S., Kaufer, D., Heinemann, U., & Friedman, A. (2017). Epileptiform activity and spreading depolarization in the bloodbrain barrier-disrupted peri-infarct hippocampus are associated with impaired GABAergic inhibition and synaptic plasticity. Journal of Cerebral Blood Flow and Metabolism, 37(5), 1803–1819. https://doi.org/10.1177/0271678X16652631

Masamoto, K., & Kanno, I. (2012). Anesthesia and the quantitative evaluation of neurovascular coupling. In Journal of Cerebral Blood Flow and Metabolism (Vol. 32, Issue 7, pp. 1233–1247). SAGE PublicationsSage UK: London, England. https://doi.org/10.1038/jcbfm.2012.50

McGregor, H. R., Cashaback, J. G. A., & Gribble, P. L. (2016). Functional Plasticity in Somatosensory Cortex Supports Motor Learning by Observing. Current Biology, 26(7), 921–927. https://doi.org/10.1016/j.cub.2016.01.064

McMillin, M. A., Frampton, G. A., Seiwell, A. P., Patel, N. S., Jacobs, A. N., & DeMorrow, S. (2015). TGFβ1 exacerbates blood-brain barrier permeability in a mouse model of hepatic encephalopathy via upregulation of MMP9 and downregulation of claudin-5. Laboratory Investigation, 95(8), 903–913. https://doi.org/10.1038/labinvest.2015.70

Mégevand, P., Troncoso, E., Quairiaux, C., Muller, D., Michel, C. M., & Kiss, J. Z. (2009). Long-term plasticity in mouse sensorimotor circuits after rhythmic whisker stimulation. Journal of Neuroscience, 29(16), 5326– 5335. https://doi.org/10.1523/JNEUROSCI.5965-08.2009

Milikovsky, D. Z., Ofer, J., Senatorov, V. V., Friedman, A. R., Prager, O., Sheintuch, L., Elazari, N., Veksler, R., Zelig, D., Weissberg, I., Bar-Klein, G., Swissa, E., Hanael, E., Ben-Arie, G., Schefenbauer, O., Kamintsky, L., Saar-Ashkenazy, R., Shelef, I., Shamir, M. H., … Friedman, A. (2019). Paroxysmal slow cortical activity in Alzheimer’s disease and epilepsy is associated with blood-brain barrier dysfunction. Science Translational Medicine, 11(521), eaaw8954–eaaw8954. https://doi.org/10.1126/scitranslmed.aaw8954

Ngai, A. C., Jolley, M. A., D’Ambrosio, R., Meno, J. R., & Winn, H. R. (1999). Frequency-dependent changes in cerebral blood flow and evoked potentials during somatosensory stimulation in the rat. Brain Research, 837(1–2), 221–228. https://doi.org/10.1016/S0006-8993(99)01649-2

Obermeier, B., Daneman, R., & Ransohoff, R. M. (2013). Development, maintenance and disruption of the blood-brain barrier. In Nature Medicine (Vol. 19, Issue 12, pp. 1584–1596). Nature Publishing Group. https://doi.org/10.1038/nm.3407

Schoknecht, K., Prager, O., Vazana, U., Kamintsky, L., Harhausen, D., Zille, M., Figge, L., Chassidim, Y., Schellenberger, E., Kovács, R., Heinemann, U., & Friedman, A. (2014). Monitoring stroke progression: In vivo imaging of cortical perfusion, blood-brain barrier permeability and cellular damage in the rat photothrombosis model. Journal of Cerebral Blood Flow and Metabolism, 34(11), 1791–1801. https://doi.org/10.1038/jcbfm.2014.147

Schumacher, L., Slimani, R., Zizmare, L., Ehlers, J., Kleine Borgmann, F., Fitzgerald, J. C., Fallier-Becker, P., Beckmann, A., Grißmer, A., Meier, C., El-Ayoubi, A., Devraj, K., Mittelbronn, M., Trautwein, C., & Naumann, U. (2023). TGF-Beta Modulates the Integrity of the Blood Brain Barrier In Vitro, and Is Associated with Metabolic Alterations in Pericytes. Biomedicines, 11(1), 1–19. https://doi.org/10.3390/biomedicines11010214

Shim, H. J., Jung, W. B., Schlegel, F., Lee, J., Kim, S., Lee, J., & Kim, S. G. (2018). Mouse fMRI under ketamine and xylazine anesthesia: Robust contralateral somatosensory cortex ac/va/on in response to forepaw stimulation. NeuroImage, 177, 30–44. https://doi.org/10.1016/J.NEUROIMAGE.2018.04.062

Sudmant, P. H., Alexis, M. S., & Burge, C. B. (2015). Meta-analysis of RNA-seq expression data across species, tissues and studies. Genome Biology, 16(1), 287. https://doi.org/10.1186/s13059-015-0853-4

Tao, L., & Nicholson, C. (1996). Diffusion of albumins in rat cortical slices and relevance to volume transmission. Neuroscience, 75(3), 839–847. https://doi.org/10.1016/0306-4522(96)00303-X

Vazana, U., Veksler, R., Pell, G. S., Prager, O., Fassler, M., Chassidim, Y., Roth, Y., Shahar, H., Zangen, A., Raccah, R., Onesti, E., Ceccanti, M., Colonnese, C., Santoro, A., Salvati, M., D’Elia, A., Nucciarelli, V., Inghilleri, M., & Friedman, A. (2016). Glutamate-mediated blood–brain barrier opening: Implications for neuroprotection and drug delivery. Journal of Neuroscience, 36(29), 7727–7739. https://doi.org/10.1523/JNEUROSCI.0587-16.2016

Veksler, R., Vazana, U., Serlin, Y., Prager, O., Ofer, J., Shemen, N., Fisher, A. M., Minaeva, O., Hua, N., SaarAshkenazy, R., Benou, I., Riklin-Raviv, T., Parker, E., Mumby, G., Kamintsky, L., Beyea, S., Bowen, C. V., Shelef, I., O’Keeffe, E., … Friedman, A. (2020). Slow blood-to-brain transport underlies enduring barrier dysfunction in American football players. Brain, 143(6), 1826–1842. https://doi.org/10.1093/brain/awaa140

Zandieh, S., Hopf, R., Redl, H., & Schlag, M. G. (2003). The effect of ketamine/xylazine anesthesia on sensory and motor evoked potentials in the rat. Spinal Cord, 41(1), 16–22. https://doi.org/10.1038/sj.sc.3101400

Zelig, D., Goldberg, I., Shor, O., Ben Dor, S., Yaniv-Rosenfeld, A., Milikovsky, D. Z., Ofer, J., Imtiaz, H., Friedman, A., & Benninger, F. (2022). Paroxysmal slow wave events predict epilepsy following a first seizure. Epilepsia, 63(1), 190–198. https://doi.org/10.1111/epi.17110

Zhang, Y., & Pardridge, W. M. (2001). Mediated efflux of IgG molecules from brain to blood across the blood– brain barrier. Journal of Neuroimmunology, 114(1–2), 168–172. https://doi.org/10.1016/S01655728(01)00242-9

Zhou, Y., Zhou, B., Pache, L., Chang, M., Khodabakhshi, A. H., Tanaseichuk, O., Benner, C., & Chanda, S. K. (2019). Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nature Communications, 10(1), 1–10. https://doi.org/10.1038/s41467-019-09234-6