Inflammation in Alzheimer's disease

Video transcript

• 0:10 Hello, my name is Michael Hanneke.
• 0:12 Me and my team are interested in inflammatory mechanisms and various brain disorders including neurodegenerative diseases.0:21 We are have been studying the mechanisms that set the scene for inflammation such as locus psorius degeneration, the role of NSAID's and NSAID mediated protection in Alzheimer's disease and ultimately got very interested in innate immune activation through the NAP 3 inflammosome signaling pathway.
• 0:47 Inflammation in Alzheimer's disease.
• 0:49 What is the cause and what is the meaning and why?
• 0:52 This should be a future research focus.
• 0:56 So to consider the meaning of an inflammatory response in Alzheimer's disease, we have to go back and think about what the innate immune system which drives most of this response has been evolutionally developed for.
• 1:11 And actually this is the defense of our body to challenges from outside by fungi, by bacteria and by viruses.
• 1:21 So these innate immune cells, which we call myelin cells and which include microglia cells, which represent the brain's innate immune system, are actually equipped with pattern recognition receptors on the surface, which send signals that are foreign to them.
• 1:39 And then they elicit an inflammatory reaction which is one part releasing inflammatory mediators the other part which is a phagocytic clearance response.
• 1:51 It happens to be the fact that some of the bacteria we know very well, including Escherichia coli, Staphylococcus aureus, and Salmonella typhimurium, carry beta sheet structured amyloids very similar to those that they deposited on the surface of the brains of Alzheimer patients by nature.
• 2:11 Which means that if the myeloid cells by pattern recognition receptor ligation sends these structures, they respond in the same way because that's their evolutionary program.
• 2:23 Now if microglial cells become activated, they're not only releasing a lot of inflammatory mediators, they also stop doing a lot of beneficial things for our brain, including the generation and secretion of trophic factors, synaptic scaling, and pruning, which is necessary to keep neuronal networks going.
• 2:46 Now, inflammation has two major consequences.
• 2:50 One is a functional consequence.
• 2:52 We know that the release of these inflammatory mediators, including R1-beta, TNF alpha, nitric oxide or other cytokines is able to functionally impair neuronal network activity.
• 3:03 In a prominent example is the suppression of long term potentiation in the hippocampus.
• 3:10 While LTP is not the equivalent to memory, it's an important factor to memory consolidation and long term storage of information in our brain.
• 3:19 So if that's suppressed, we could have conclude that the presence of inflammation impairs the memory function.
• 3:28 Now after or in parallel to the functional impairment, there are structural consequences.
• 3:35 So, inflammation, of course, does not occur alone.
• 3:38 In the brains of these patients, there is excitotoxicity activity as well.
• 3:45 And paper a long time ago by Sandy Hewitt and Moses Chow and Neuron in 1996 nicely showed that inflammation and excitotoxicity stimuli acts synergistically and induced neuronal death.
• 4:01 There is an effect on synaptic structures which can be engulfed in a complement 3 receptor dependent mechanism and those synapses which are decor by C3 are then taken up by microglia.
• 4:18 And that's obviously one of the effects which also happens to take place, at least in models of cerebral amyloid doses, which are linked to Alzheimer's disease.
• 4:27 Now, there are other mechanisms by which inflammatory reactions can impair neuronal survival, including phagoptosis.
• 4:36 So if neurons do not present, do not emit signals on their surface, then the immediate reaction of microglia cells can be to take those neurons up, even alive, and digest them.
• 4:49 Beyond these direct effects on neurons, there are a variety of other factors.
• 4:56 So inflammatory molecules in some of the models have been shown to upregulate APP processing, especially by increasing beta secretase-1 levels, so fuelling into the pro-amyloidogenic pathway, which would lead to an increase in the concentrations of beta-amyloid peptides.
• 5:15 At the same time, inflammation is able to impair the phagocytic clearance of beta peptides by microglia cells and maybe other pathways as well.
• 5:25 An intriguing mechanism by which microglia can actually interfere with beta amyloid deposition is the activation of the NALP3 inflammasome, which leads to microglial paraptosis and the release of ASC Speck.
• 5:40 And ASC Specks are highly aggregating conglomerates which interact with beta amyloid, and actually seeding them is a fundamental mechanism involved in the seeding of beta amyloid pathology in response to the injection of APP brain lysate injection into the hippocampus.
• 6:04 So obviously this inflammatory molecule can change the propensity of beta amyloid to aggregate and form the seeding core of plaque deposits, and that could be replicated in a human plaque analysis as well.
• 6:24 Of course, we need to get more information about the ongoing neuroinflammation.
• 6:28 And if we consider the course of Alzheimer's disease, we first know that AD is not starting with the advent of clinical symptoms, but most likely is starting many decades prior to those clinical impairments.
• 6:46 We now know that a beta deposition as one of the things we can measure by PET imaging may start as early as 3 decades prior to any cognitive deficits.
• 6:58 If we now consider that beta amyloid is such a strong danger associated molecular pattern, we then would conclude that all the subsequent mechanisms are actually influenced by the presence of inflammation.
• 7:12 But inflammation normally runs in waves, so every inflammatory activation is followed by resolution and healing.
• 7:20 So it's likely that inflammation is not linear, exponential, or asymptotic, but runs in waves.
• 7:27 That's more complex because we now also know that in one hemisphere of an 80-year-old patient, there are hardly two brain areas that are in precisely the same stage of disease.
• 7:40 But the entorhinal cortex, for example, is long gone.
• 7:43 Another brain area, like the occipital cortex may not even know that there's so much going wrong up in the in the front part of the brain.
• 7:52 Now consider Alzheimer's disease as relay race, A relay race where a beta is the first runner and gives the stick after it's round to the next runner, which could be innate immune activation.
• 8:10 And when it made an immune activation has has run it's round in the stadium, it gives it over to a second, third mechanism, for example, mitochondrial dysfunction.
• 8:21 And finally the last runner could be Tau.
• 8:24 But you don't have one team running alone.
• 8:26 You have several teams and call one hippocampus, one entorhinal cortex, one parietal cortex, one cerebellum, one occipital cortex, and so on.
• 8:36 So, this means that we most likely have various types of inflammation and responses to different degrees and types of PDA amyloid deposition in the very same brain, which makes it complex.
• 8:50 An explains the need for precise biomarkers which tell us in which brain region which inflammatory mechanisms is most prevalent.
• 8:59 So we can consider biomarkers which inform us about microglial activation status.
• 9:07 You also will need biomarkers which address the function of microglia, for example, their capacity to still clear amyloid beta or the capacity to release traffic factors.
• 9:19 We also will need biomarkers which inform us about microglial proliferation and microglial cell death, paraptosis, so to say.
• 9:30 In general, we would like to seek a fingerprint after inflammation in biological fluids, for example, such as the CSF, or even better, in blood or samples, serum, or plasma, which would give us this information.
• 9:45 One way to think about fingerprints is to consider a post-translational modification.
• 9:50 An example could be to measure the final downstream effects of inducible nitric oxide synthase activity which results in nitroxylation and nitration of peptides and proteins.
• 10:05 Those patterns of protein and peptide nitration or nitroxylation could actually inform us about the degree of an inflammatory response in the brain, which could then be linked to clinic or other imaging information.
• 10:23 What we have right now is that we can measure beta amyloid peptides, Tau and phosphorylated Tau from the CSF of patients.
• 10:31 And while one of these markers alone is less indicative of a certain disease stage, we know that if we, for example, combine them and analyze this, we can clinically use them to make a judgment.
• 10:45 In which stage of the disease the patient currently is most informative are the ratios between A-beta-42 and A-beta-40, or between A-beta-42 and phospho-Tau, for example.
• 11:01 More recently, people have looked at neurofilaments, which are likely to give us an even better information with respect to treatment responses.
• 11:12 Neurofilaments can be measured in the peripheral blood and can give us a hint about the degree of damage which is ongoing in the brain and hopefully we can use neurofilaments one day to read out treatment responses when we use anti-inflammatory treatment strategies.
• 11:34 All in all, it's important that we need longitudinal information.
• 11:40 So most of the current studies are cross-sectional and give us information about a certain group of patients at a certain time point and stage of disease.
• 11:50 But what we really need, and based on the information I gave you before about the long course of the disease is longitudinal data which show us at which time point and which stage of disease connected to a certain a symptom, the respective inflammatory parameter comes up and gives us a sign for a mechanism or a certain stage of the innate immune system.
• 12:15 So what are the number one interesting targets in neuroinflammation?
• 12:18 And let me emphasize upfront that this is a personal list I will give you.
• 12:23 We can consider targets which come from genetic information which includes therapy, P, trend 2, PLC, gamma, CD33, and of course those have evidence by genetic risk polymorphisms which increase or decrease the risk to develop Alzheimer's disease and of and are all tightly linked to the immune system and in most cases to innate immunity itself.
• 12:54 Then there are targets which come from experimental studies, which include for example toll-like receptors, the NALP inflammasome, the ASK Specks Sigma, 1P part gamma and progranulin.
• 13:13 And we also have to consider single cytokines by more recent evidence, for example, showing that Enbrel may or may not be effective.
• 13:23 And those data have to be revisited, but would put TNF alpha upfront and like TNF alpha into leukein.
• 13:30 One beta could be an interesting target itself because it's the downstream inflammatory cytokine which results from the alpha inflammasome activation and microglial cells.
• 13:43  Now another point to discuss and consider is the innate immune memory.
• 13:51 We have assumed for decades that only adaptive immunity has memory cell and can keep information for a long time.
• 14:01 But it turned out by work from Mihai Netiya and others that innate immune cells also have a certain memory type of information storage.
• 14:12 This memory information could be a known target because it sets the scene for an even stronger immune response when a second or third challenge occurs.
• 14:25 Now one important point is that these targets I just told you may undergo changes when along the entire disease course.
• 14:38 So one of these or each of these targets may have its own therapeutic window or its own time where it can be targeted and therapeutically harnessed.
• 14:52 It may be that at a certain stage, modification or degradation, an inactivation by a phosphorylation, for example, as it is likely for people gamma completely shuts down the respective pathway and makes it inaccessible for a therapeutic intervention.
• 15:10 So if you consider emerging therapies which help us to study neuroinflammation in more detail in future, then let's distinguish between the cellular, the animal and the human level.
• 15:23 On a cellular level, I think what's urgently needed is more information on protein levels and protein change, changes by mass spectrometry or by mass cytometry.
• 15:35 And this is because a lot of transcriptomic data have been accumulated now, of which we know that only a minor percentage is actually being translated and biologically meaningful proteins.
• 15:49 So while all this wealth of transcriptomic and epigenetic information is really important and gives us a hint what's going on, I think we need a more detailed information on the protein level.
• 16:03 On the animal side, I believe that humanizing mouse models would be a big leap forward, as would more precise and sophisticated analysis by techniques such as three- or two-photo and laser scanning in vivo microscopy, which helps us really look into the living brains of mice.
• 16:29 And this even in anesthetized or non-anesthetized mice, which can conduct, for example, memory tasks while they are imaged for neuronal network activity or for inflammatory cell activity.
• 16:47 Those models need to be further developed.
• 16:51 And one step ahead is to use optogenetic means to modulate inflammatory signals, so to, for example, by the use of light responsive Cree elements which allow us to express or suppress single inflammatory molecules and study the effect for example around beta amyloid deposits or even prior or after the deposition.
• 17:25 So on the human level, we need information on the precise temporal and spatial resolution.
• 17:33 And one of the ways forward would be to develop PET ligands beyond TSP 4, for example, by developing ligands for the cannabinoid receptor 2, for CSF1 receptor for COX2, and there might be many other targets.
• 17:54 Another important way of going forward on the human level would be to use IPSC-derived microglia-like cells to study individual responses to treatment and move the diagnostics or the therapeutic interventions to a more personalized level.
• 18:12 So I think these are very exciting times for everybody working in this field and there are almost new findings every week or every day, so to say.
• 18:24 There's a lot more out there and we haven't understood the interaction between microglia cells and other brain cells in greater detail yet.
• 18:33 For example, their modulation of oligodendrocyte function, their modulation of astroglial function.
• 18:39 There's a lot more to learn, a lot lot more to study.
• 18:42 And for that reason, I'm wondering whether we should be more careful in suggesting very quickly clinical interventional trials using inflammatory targets.
• 18:55 I think the innate immune system and the adaptive immune system are very delicate systems.
• 19:02 We also need to consider that we might change things in the periphery to the harm of our patients.
• 19:08 So I think we need a lot more studies and a lot more insight before we actually can move into clinical trials.
• 19:15 But it's certainly worth doing it.