London Metropolitan University

Researchers have deciphered the complex interplay of various enzymes around the innate immune receptor toll-like receptor 7 (TLR7), which plays an important role in defending our bodies against viruses. LMU researchers have deciphered the complex interplay of various enzymes around the innate immune receptor toll-like receptor 7 (TLR7), which plays an important role in defending our bodies against viruses. Toll-like receptor 7 (TLR7), located in the dendritic cells of our immune system, plays a crucial role in our natural defense against viruses. TLR7 recognizes single-stranded viral and other foreign RNA and activates the release of inflammatory mediators. Dysfunctions of this receptor also play a key role in autoimmune diseases, making it all the more important to understand, and ideally modulate, the exact activation mechanism of TLR7. Researchers led by Professor Veit Hornung and Marleen Bérouti from the Gene Center Munich and the Department of Biochemistry at LMU have now managed to gain deeper insights into the complex activation mechanism. It was known from earlier studies that complex RNA molecules have to be cut up first so that the receptor is able to recognize them. Using a wide range of technologies from cell biology to cryogenic electron microscopy, the LMU researchers have revealed how single-stranded foreign RNA is processed to be detected by TLR7. Their work has been published in the journal Immunity. Numerous enzymes are involved in the recognition of foreign RNA In the course of evolution, the immune system has specialized in recognizing pathogens from their genetic material. For example, the innate immune receptor TLR7 is stimulated by viral RNA. We can picture viral RNAs as long threads of molecules, which are much too large to be recognized as ligands for TLR7. This is where nucleases come in -- molecular cutting tools that chop the 'RNA thread' into small pieces. Endonucleases cut the RNA molecules through the middle like scissors, while exonucleases cleave the thread from one end to the other. This process generates various RNA snippets, which can now bind to two different pockets of the TLR7 receptor. Only once both binding pockets of the receptor are occupied by these RNA pieces a signaling cascade set in motion, which activates the cell and triggers a state of alarm. The researchers discovered that RNA recognition by TLR7 requires the activity of the endonuclease RNase T2 operating in conjunction with the exonucleases PLD3 and PLD4 (phospholipase D3 and D4). "Although it was known that these enzymes can degrade RNAs," says Hornung, "we have now demonstrated that they interact and thereby activate TLR7." Balancing the immune system The researchers also found that the PLD exonucleases have a dual role within immune cells. In the case of TLR7, they have a pro-inflammatory effect, whereas in the case of another TLR receptor, TLR9, they have an anti-inflammatory effect. "This dual role of PLD exonucleases points to a finely coordinated balance for controlling appropriate immune responses," explains Bérouti. "The simultaneous promotion and inhibition of inflammation by these enzymes could serve as an important protective mechanism for preventing dysfunctions arising in the system." What role other enzymes could have on this signaling pathway and whether the molecules involved are suitable as target structures for therapies are subjects for further investigations.

Machines can learn not only to make predictions, but also to handle causal relationships. An international research team shows how this could make therapies safer, more efficient, and more individualized. Machines can learn not only to make predictions, but also to handle causal relationships. An international research team shows how this could make therapies safer, more efficient, and more individualized. Artificial intelligence is making progress in the medical arena. When it comes to imaging techniques and the calculation of health risks, there is a plethora of AI methods in development and testing phases. Wherever it is a matter of recognizing patterns in large data volumes, it is expected that machines will bring great benefit to humanity. Following the classical model, the AI compares information against learned examples, draws conclusions, and makes extrapolations. Now an international team led by Professor Stefan Feuerriegel, Head of the Institute of Artificial Intelligence (AI) in Management at LMU, is exploring the potential of a comparatively new branch of AI for diagnostics and therapy. Can causal machine learning (ML) estimate treatment outcomes -- and do so better than the ML methods generally used to date? Yes, says a landmark study by the group, which has been published in the journal Nature Medicine: causal ML can improve the effectiveness and safety of treatments. In particular, the new machine learning variant offers "an abundance of opportunities for personalizing treatment strategies and thus individually improving the health of patients," write the researchers, who hail from Munich, Cambridge (United Kingdom), and Boston (United States) and include Stefan Bauer and Niki Kilbertus, professors of computer science at the Technical University of Munich (TUM) and group leaders at Helmholtz AI. As regards machine assistance in therapy decisions, the authors anticipate a decisive leap forward in quality. Classical ML recognizes patterns and discovers correlations, they argue. However, the causal principle of cause and effect remains closed to machines as a rule; they cannot address the question of why. And yet many questions that arise when making therapy decisions contain causal problems within them. The authors illustrate this with the example of diabetes: Classical ML would aim to predict how probable a disease is for a given patient with a range of risk factors. With causal ML, it would ideally be possible to answer how the risk changes if the patient gets an anti-diabetes drug; that is, gauge the effect of a cause (prescription of medication). It would also be possible to estimate whether another treatment plan would be better, for example, than the commonly prescribed medication, metformin. To be able to estimate the effect of a -- hypothetical -- treatment, however, "the AI models must learn to answer questions of a 'What if?' nature," says Jonas Schweisthal, doctoral candidate in Feuerriegel's team. "We give the machine rules for recognizing the causal structure and correctly formalizing the problem," says Feuerriegel. Then the machine has to learn to recognize the effects of interventions and understand, so to speak, how real-life consequences are mirrored in the data that has been fed into the computers. Even in situations for which reliable treatment standards do not yet exist or where randomized studies are not possible for ethical reasons because they always contain a placebo group, machines could still gauge potential treatment outcomes from the available patient data and thus form hypotheses for possible treatment plans, so the researchers hope. With such real-world data, it should generally be possible to describe the patient cohorts with ever greater precision in the estimates, thereby bringing individualized therapy decisions that much closer. Naturally, there would still be the challenge of ensuring the reliability and robustness of the methods. "The software we need for causal ML methods in medicine doesn't exist out of the box," says Feuerriegel. Rather, "complex modeling" of the respective problem is required, involving "close collaboration between AI experts and doctors." Like his TUM colleagues Stefan Bauer and Niki Kilbertus, Feuerriegel also researches questions relating to AI in medicine, decision-making, and other topics at the Munich Center for Machine Learning (MCML) and the Konrad Zuse School of Excellence in Reliable AI. In other fields of application, such as marketing, explains Feuerriegel, the work with causal ML has already been in the testing phase for some years now. "Our goal is to bring the methods a step closer to practice. The paper describes the direction in which things could move over the coming years."

Researchers have developed a method to determine how reliably target proteins can be labeled using super-resolution fluorescence microscopy. LMU researchers have developed a method to determine how reliably target proteins can be labeled using super-resolution fluorescence microscopy. Modern microscopy techniques make it possible to examine the inner workings of cells in astonishing detail. "We can now observe the arrangement and interaction of individual proteins under the microscope," says Professor Ralf Jungmann, Chair of Molecular Physics of Life at LMU and Max Planck Fellow at the MPI of Biochemistry. The biophysicist's team recently developed the revolutionary RESI (Resolution Enhancement by Sequential Imaging) method. This technique can be used to improve the resolution of fluorescence microscopy down to the Ångström scale -- far below the classical diffraction limit of light. DNA-conjugated marker molecules, which the researchers attach precisely to the molecules they want to understand better, are crucial for this. Jungmann's team has now presented a technique in the journal Nature Methods that can be used to quantify how well biomarker molecules bind to the target proteins. "This is absolutely crucial if you want to make quantitatively reliable statements," explains the physicist. If you know the labeling efficiency, you can carry out spatially resolved proteomics in this way. This allows you to find out not only what individual proteins do in a cell, but also to what extent they are present and how their quantity and behavior change under certain circumstances. "But this is only possible if we can assess how well the labeling has worked." This is because only labeled proteins emit flashes of light under the microscope and thus become visible. Reliable and versatile The method developed by Jungmann's team makes this assessment possible by adding a reference biomarker to the target proteins. This marker "glows" in a different color during microscopy, so that successfully marked proteins appear in two colors. Jungmann's team demonstrated this using the membrane protein CD86, among others: The reference produces a pink fluorescence, the actual marker a bluish one. This creates a pattern of innumerable pink and blue points of light. Where the marking did not work, only the reference lights up individually. The marking efficiency is calculated from the ratio of double and single illuminated molecules. The method offers several advantages compared to previous methods for determining binding efficiency: "It works not only in vitro, but also in vivo, i.e. in the context of intact cells," explains Jungmann. "The technique can also be applied to a variety of different target molecules, biomarkers, and samples and is compatible with a whole range of super-resolution methods." A reliable and widely applicable means of assessing marker efficiency is crucial to ensure accurate data evaluation and enable reliable comparisons between different binders, labeling conditions, and research laboratories. The authors of the study are certain that the new quantification method has paved the way for significantly expanding the potential of their super-resolution microscope method: "Now we can also consider specific biomedical applications in which the quantitative detection of proteins and processes is of great importance," says Jungmann. This includes cancer research, for example, where information about interactions between proteins on the cell surface and drugs with molecular resolution is essential for the development of new types of medication.