And again the microglial cells

"Only true artists are attracted to science", said Ramon y Cajal, the father of neuroscience, whose drawings, rich in incredible details, can be seen even in today's anatomy textbooks, too. He was always certain of the autonomy and diversity of the nerve cells he drew as individuals, "mysterious butterflies of the soul". His neuron hypothesis was confirmed in 1956, 22 years after his death, by one of the first electron micrographs taken by Palay, which showed two nerve cells a few times 10 nanometres apart. The site was later called a synapse.

While we are on the importance of electron microscopic identification of synapses, let us mention another major discovery from our institute published in Nature (Gulyás et al., 1993 Nature). In collaboration with Richard Miles (†2024), the Freund group was the first to perform a physiological study of the correlated morphology of a synapse, i.e. to investigate the physiological properties of an electron microscopically identified synapse.  

Talent, creativity, perseverance - and some passion, some personal connection with what the researcher is working on. A lot has changed since Cajal, but it remains the same. Today, there is no substitute for a researcher who can put his or her whole soul into a problem, searching for the answer to a question.

The 'first authors' of the article in Nature Communications by the Neuroimmunology group of Ádám Dénes, who led the research, look such this type. 

Let the conversation begin with PhD student Péter Berki and senior researchers Csaba Cserép PhD and Zsuzsanna Környei PhD. 

- Likely the greatest achievement of your group so far was when you described and demonstrated the direct, somatic connection between microglial cells and neurons. You applied several state-of-the-art microscopic techniques and presented fascinating images as proof. Your current paper details how microglial phenotypic changes affect the function of neural networks in an acute brain slice model of injury. Why has such an experiment only just taken place?

Peti

- It's a fair question since slice preparation as a test method has been available since 1966. Its great advantage is that we can extract only the area of the brain to be studied while keeping the cells alive, making it much easier to study the cells than in a living animal. The disadvantage, however, is that the artificially created environment does not perfectly reflect the brain milieu, connections from other brain areas are lost, and the exact consequences of the damage are not yet well understood. So far, in brain slice preparations, most researchers have been interested in the function of neurons and neuronal networks. 

However, since the 2000s, research on microglial cells has grown fast and their key role in regulating and protecting the neuronal network has become clear. Microglial cells respond extremely sensitively to environmental changes. It was reasonable to assume that slicing would trigger defense mechanisms in them that would affect the structure of the whole neural network and its processes. So the slice model which is also a model of injury, has essentially given itself away. We were surprised to realize that no one had looked at it before.

- Making such a preparation could not have been easy, and we also had to strive for reproducibility! 

Peti

- Speed, sterility, and precision are essential requirements for the preparation of slides, the quality of the slides depends as much on the expertise of the researcher preparing the slides as on the method used and the quality of the solutions and supplies. The slicing method itself has evolved a lot since its invention. It is now possible to produce slides of a quality in which the cells produce a pattern of activity that can be observed in the living animal. KOKI researchers have made significant contributions to the development of the method, and several of our laboratories have also produced internationally excellent quality research on slices. Although our experiments were built on these traditions, we also tried other techniques. Besides, the samples were obtained by independent laboratories. We use slices between 300 and 450 micrometers thick because experience has shown that in these slices the local neuronal network can regenerate in such a way that it can generate activity patterns similar to physiology.

- To what extent were your results with the model comparable to those observed in brain lesions? What and how generalizable are the conclusions?

Csabi

- This is an interesting question. Most of the observed microglial parameters showed similar changes in the acute slice model as we usually see in different pathological conditions. This is true speaking about changes in microglial P2Y12 receptor expression or for the morphological response. 20 minutes after slice cutting, the microglial cell is in a similar state to what we see in cerebral infarction near the lesion. It is becoming more and more convincing that there is a general injury/pathology-induced microglial phenotype change that results in a reduction in the functional capacity of the microglial cell, which is nevertheless still able to work to preserve as much as possible the integrity and functionality of the brain tissue. This shows quite amazing stamina and strength! 

- I assume that the tissue necrosis did not stop during the measurements, it just happened less and more slowly than when the brain slice was made.

Csabi

- This has not been specifically examined. It would be difficult to measure it reliably because apoptotic cells are cleared at an incredibly high rate. In such a measurement, it is almost impossible to say what has changed, whether the rate/pace of clearance increased or the cell death getting slower. However, the decay of ATP release over time supports the idea that a quasi-equilibrium state is reached relatively quickly after the initial storm of cell death, with a large contribution from the glial 'shell' that forms on the surface of the slice. In addition to astrocytic cells, the migration of microglial cells to the surface also plays a role. 

-What can be the biggest difference between your model injury and a real brain injury?

Peti

- Of course, the damage caused by slicing is unrealistically large compared to a typical brain injury, such as a concussion or stroke. What is surprising, however, is that the behavior of the microglial cells we observed agreed with the results observed in mice modeling in vivo injury, or with lesions measured in human Alzheimer's patients, for example. In addition, the kinetics and spatial distribution of both passive and active ATP events that trigger the phenotypic transformation of microglial cells in living animals are very similar. Based on the results obtained so far, I think it is reasonable to assume that the processes we have described are relevant and that similar lesions can be expected in the injured area and its vicinity in the case of a brain injury. 

Zsu

 - This is supported by the in vivo observations of our Chinese colleagues, who have seen/measured similar activity and flashes of ATP release after focal laser injury or stroke (Chen et al., 2024). It would be interesting to know whether similar ATP activity appears or is persistent in chronic neurodegenerative diseases!

- You have already mentioned ATP so much that I have to ask, what is the role of the nucleotide constituent of nucleic acids known as adenosine triphosphate, in its other role as the main energy storage molecule in cells, not least as a neurotransmitter/neurotransmitter, in all this?

Peti

- Injury causes ATP to flow into the intercellular space by two different mechanisms. One is a passive process where ATP flows out directly from the injured cells, and the other is an as-yet-unknown endogenous mechanism where ATP is actively released into the intercellular space, most likely by the astroglial cell. The latter phenomenon is seen in our 2-photon/1-photon/epifluorescence microscopy images as rapid focal flashes of the fluorescent ATP sensor. 

Csabi

- ATP is also of particular interest, because, as we previously demonstrated, nerve cells can constantly signal to microglia that they are "fine" by releasing tiny amounts of ATP through the somatic connection. If there is a problem, we can talk about ATP concentrations that are orders of magnitude higher. It is so special how the same molecule signals healthy and damaged states depending on the location and concentration of release.

- With this so-called fluorescent ATP sensor, how long could you measure how long it was reliably sensitive to ATP?

Peti

- We asked ourselves this question because we also wanted to be sure that the decrease in intensity observed immediately after the slice cut, after a large passive ATP release, was an indication of a deterioration in the sensitivity of the sensor, or a gradual decrease in ATP concentration as a result of the activity of ATP-degrading enzymes. In our experience, the sensor worked perfectly! During the experiments, we did not notice any drastic change in the sensitivity of the sensor, and we were able to record up to 40-50 minutes of continuous recordings without a systematic decrease in intensity. 

Zsu

- Preliminary experiments in crypto nests have shown that this time can be up to 90 minutes. As long as the cells producing the fluorescent ATP biosensor remain healthy, there is no obstacle to the sensor's functionality. Depending on the presence of the ligand, it can remain functional for much longer.

- After the ATP spill after the injury, was there a subsequent higher ATP release? If so, what is the reason?

Peti

- The dynamic ATP release detected after passive ATP efflux can be considered significant, with the potential to influence the phenotypic transformation of the microglial cell. This is known from the fact that our calibration measurements show that the intensity of the sensor increases linearly as a function of ATP concentration and is sensitive to detect nanomolar changes.  From this, we have been able to estimate the concentration of these ATP events quite accurately. During flash-like events, we have measured local ATP concentration increases of 50-100 nM in the intercellular space, and up to 1 µM (ten to twenty times higher concentrations) during larger area events.

Zsu

This concentration is one or two orders of magnitude lower than the average ATP concentration of ~2-5mM inside the cell. The ATP released by the cells is sufficient for signaling (signal transduction), however, extracellular ATPases can relatively easily "cope" with this amount, freeing the surrounding cells from a continuous purinergic effect. 

Peti

- Further investigation will be needed to understand the cause and function of this mechanism. As I have already mentioned, it is currently hypothesized that this endogenous process is primarily the result of injury to the nervous system and plays an important role in its spatial and temporal coding.

- Besides passive and endogenous, you mentioned dynamic ATP release. What does this mean?

Peti

- We wanted to refer to this newly discovered mechanism and to highlight the importance of the dynamic presence of ATP in the extracellular space in the acute slice. It is also worth mentioning that the properties of the focal ATP events we observed range widely and, based on the three parameters measured (area, intensity, duration), we have clearly distinguished two groups whose incidence varies after injury. One is that of a small area, short duration, faint flashes, and the other is that of a large area, long duration, and bright events/phenomena. The latter increases significantly in frequency within 1-2 hours after injury and then gradually decreases in frequency.  Thus, the injury is encoded not only spatially, but presumably also temporally, by an endogenous mechanism involving dynamic ATP release.

Zsu

- I would add that larger ATP flashes indicate ATP release from whole astroglial cells, while smaller ones from smaller astroglial cell compartments (e.g., a particular glial extension). This allows for an infinitely precise regulation, which will be of particular importance if we can detect these ATP flashes in other, possibly chronic, diseases after traumatic injury or stroke, as studied by our Chinese colleagues.

Peti

- Based on the evidence, this phenomenon appears to be an endogenous process in neural tissue whereby ATP events actively released by astroglial cells can encode the location and severity of injury in space and time, recruiting microglial cells to the injured area.

Zsu

- These injury-induced focal ATP events average ~85 seconds in duration, ranging from ATP release to extracellular degradation. However, the injury-induced phenomenon itself persists for long hours, giving rise to the possibility of even severe tissue rearrangements. No similar extracellular event sequence with similar spatiotemporal dynamics has been described before, so I think it is safe to say that we have witnessed the discovery of a new brain signaling modality. 

- What did you call "specific ATP events mediating the effects of injury on the microglial cell" and how did you distinguish them from other ATP events?

Peti

- We thought it was important to investigate how these injury-induced ATP events affect microglial cells. Previous experiments have shown that a high concentration of ATP solution, exogenously injected into the tissue, can rapidly shift microglial cell elongations towards increased ATP concentration. We have now demonstrated that endogenous ATP events can also recruit microglial cell protrusions to the site of the events, essentially "mediating" the purinergic signal generated by injury towards the microglial cell. The rapid appearance of microglial protrusions in a given area following a flash (Figure 2) was observed in 70% of events, indicating that they effectively mediate the injury coding information to microglial cells. To find out whether this process plays a role in microglial cell activation, for example, requires further investigation. 

- Besides morphological changes, Csabi mentioned changes in the expression of the P2Y12 receptor expressed in microglial cells. We have talked so much about purine-based ATP that it might be worth saying a few words about this important receptor with a special role!

Csabi

- P2Y12R is one of the major microglial purinergic receptors. It has been shown to play a crucial role in microglial cells sensing their environment and during the signal transduction between neurons and microglial cells in a healthy state. A decrease in cell surface expression appears to be a common phenomenon in both injury and pathological conditions. Based on our current data, we can only speculate that activation of the receptor itself may lead to a compensatory decrease in expression. These changes mainly suggest the possible existence of a stereotyped but complex "phenotype change package" in microglial cell lines upon injury. This is also supported by the fact that, as mentioned earlier, acute brain cell death is also a valid model of brain injury.

-  What do you think is the greatest achievement of this work?

Csabi

- For me, one of the greatest achievements would be to succeed in drawing attention to the fact that it is not enough to use well-established models, they need to be re-evaluated from time to time based on new aspects that were not known before or neglected until then. Our work has also demonstrated that the ex vivo brain slice can be used as a general injury model. I find it quite extraordinary how the microglial cell tries to maintain the functionality of the network until the end, even in the extreme case when its environment is reduced to a thin slice of brain tissue bounded by open wound surfaces.

Peti

- Microglial cells are an exciting, active part of the nervous system, and may have many surprises in the future, too. Likely, we learned a bit about the processes that trigger their defense mechanisms and how they affect the neuronal network. To my knowledge, we have succeeded for the first time in observing the complex response of microglial cells following injury in process, and in following the chain of events step by step from changes at the molecular level to network mechanisms in a given system.

Zsu

Like Peti, I find the unprecedented spatio-temporal dynamics of ATP activity to be the most surprising result, whose function is probably more multifaceted than we currently think. In particular, it is interesting to see how ATP, one of the most important - most common - most simple - most ancient molecules, is involved in such complex processes outside the cell.

It is compensatory that in this widely used model, we have also described glial activity with due care, and called the attention that in a trauma model, the activity of cells specialized in response to injury is involved in tissue reorganization and can even influence network activity. 

- What was your experimental work and what do you feel was your most important contribution to this study?

Csabi 

- While completing the project describing somatic junction, I asked Adam if we could look at what happens to microglial cells in the slice as a "side project". There used to be an initiative using an acute slice model, but focusing on interleukin signaling, only unfortunately it stalled. To avoid the gap I suggested exploring the temporal changes systematically. We then approached Peti, and Zsu joined in with ex vivo imaging. It was a well-coordinated team effort, with lots of consultation on the experiments and later on the design of the figures. In addition, the methodological development for the anatomical measurements was the part I would highlight as a creative process.

Peti

- I brought the acute slice methodology to the paper and was mainly responsible for the implementation of the electrophysiological measurements. As this article is the basis of my PhD dissertation, I tried to be present in all experimental and evaluation processes and to contribute to the anatomical and imaging parts. As Csabi mentioned, everyone had a part in every process, from the design of the experiments to the publication, we thought together all the way. 

Zsu

Years ago, I started to study neuron-microglial cell interactions ex vivo in cell and slice cultures in the context of Adam's Momentum and ERC grants. I contributed to the first version of this article by imaging the dynamic changes of microglial cells by 2-photon microscopy. A recent Sino-Hungarian TÉT grant contributed to the early acquisition of ATP sensors, and I was able to test different ATP sensor constructs in parallel with our Chinese partners. Using these experiences, we moved on to acute slice model imaging/imaging, which ultimately added new data and discoveries to what was initially a more morphologically oriented paper. 

- What was the hardest part?

Csabi

- For me, this is obvious: the lengthy review process, especially the part where we received unprofessional and contradictory opinions from one reviewer. I think much more serious editorial quality assurance would be needed when selecting potential reviewers!

Zsu 

- I'm also thinking about the review process. There is no more difficult than interpreting new and newer supplementary experiments in a way that is intended to transform an established text without creating a "patch on the back of a patch" feeling. 

Peti

- Surely every researcher remembers the primary"first-authored" article, which until its publication requires solving a series of unknown situations and overcoming unexpected difficulties. This was no different in my case, but in hindsight, I look back on them positively, as solving all the problems increased the quality of the paper.

- And how satisfied are you with the results?

Csabi

- We have managed to produce a strongly missing piece of material and to draw attention to the curious fact that we often don't know enough about the experimental models we have been using for decades. I consider it of particular value that we have "taken on" such a delicate and high-risk project.

Peti

- We tried to give our best. The results have set an exciting new direction for microglial and acute slice experimentation, and the future will decide how far we have succeeded in enabling other researchers to see further based on our work.

Zsu

- I am very pleased to have the opportunity to work with a new, sensitive, and reliable instrument like the GRAB ATP sensor. I have been able to see phenomena that very few people have been able to observe. I am also pleased that we could add our observations to our common knowledge of the acute slice model.

However, due to many unanswered questions, my curiosity is matched by a sense of satisfaction.