Tiny bubbles bursting in a drinker’s face and the bite of carbonation are all part of the experience when sipping champagne and sparkling wines. But how long can these drinks be stored in sealed bottles before they go flat? According to researchers reporting in ACS Omega, the answer depends on the container’s size. They estimate a 40-year shelf-life for 750-milliliter (25-ounce) bottles, and 82 and 132 years for 1.5-liter (50-ounce) and 3-liter (101-ounce) bottles, respectively.
Champagne and other sparkling wines get their bubbliness and tingly sensation from carbon dioxide, which is generated during a second round of fermentation that happens inside their bottles. Combining yeasts, sugar and wine launches the production of this gas and additional alcohol. Although the yeast die within a few months, complex aromas develop as the bottles age undisturbed for 15 months to several decades. But at the same time, the beverage is losing carbon dioxide, which is slowly escaping through the sealed metal caps or corks. So, Gérard Liger-Belair and colleagues wanted to answer the question: How does the size of the bottle influence how long you can age a champagne before it’s flat?
The researchers measured the carbon dioxide in different champagne vintages aged for multiple decades, and estimated the original amount of yeast-produced carbon dioxide. They found that the amount of gas inside the vessels, which were sealed with metal caps, decreased the longer the bottles aged. For example, the oldest vintage from 1974 lost the most carbonation, nearly 80%. Additionally, the team observed a correlation between the volume of a bottle and the carbon dioxide level, such that larger bottles retained gas substantially better than smaller ones.
In the end, the researchers developed a formula to calculate a bottle’s shelf life, or how long aged champagne would still spontaneously produce bubbles when poured in a glass. They predicted a shelf life of 40 years for standard 750-milliliter bottles, 82 years for 1.5-liter bottles and 132 years for 3-liter bottles, after which point the champagne would be flat. From their large selection of aged champagne, going back nearly 50 years, the researchers say they’ve shown how the drink’s bubbliness over time depends on the bottle’s size.
More information: Gérard Liger-Belair et al, Losses of Yeast-Fermented Carbon Dioxide during Prolonged Champagne Aging: Yes, the Bottle Size Does Matter!, ACS Omega (2023). DOI: 10.1021/acsomega.3c01812
More than 20 enzymes are involved in the production of the 21R-citrinadin A molecule. Credit: Gustavo Raskosky/Rice University
Many of the drugs we use to treat cancer and infectious disease are—or derive from— natural products, but it’s difficult to know exactly how nature assembles them.
Retracing nature’s steps, Rice University chemical engineer Xue Gao and her team mapped out the full series of enzyme-powered reactions a marine fungus uses to produce 21R-citrinadin A, a complex molecule with anticancer properties.
In the process, Gao and her collaborators identified a new enzyme, CtdY, which is the only one of its kind known to break an amide bond, according to the new study published in the Journal of the American Chemical Society.
“CtdY belongs to a large family of enzymes known as cytochrome P450s that perform a variety of different functions and are being studied for their potential use in industrial and pharmaceutical settings,” Gao said. “However, none of the P450s documented so far can break an amide bond.
“Amide bonds are found in all proteins—they’re the ones linking the amino acids together. It’s a fundamental, very stable type of bond.”
The enzyme’s ability to cleave amide bonds could make it a useful tool for creating new drugs.
“The fact that CtdY can do this is quite remarkable,” said Qiuyue Nie, a postdoctoral researcher in the Gao lab who is one of the lead authors of the study. “It holds significant promise for the pharmaceutical industry,” she said.
Qiuyue Nie (left) and Shuai Liu are lead co-authors on the study published in the Journal of the American Chemical Society. Credit: Gustavo Raskosky/Rice University
The enzyme is notable not only because it can break a highly-stable bond, but also because it does so for a very complex molecular structure.
“You want to maintain the rest of this structure and only want to break this single, hard-to-break bond ⎯ this is a very specific and difficult task,” Gao said.
Once CtdY breaks the amide bond—which has a circular 3D structure—a group of seven other enzymes intervene to complete the assembly of the 21R-citrinadin A molecule.
“Once it opens the ring, all the other enzymes are able to perform oxidation and install oxygen-hydrogen groups in a highly precise way,” Gao said. “It’s like CtdY brings the Christmas tree home, and then these other enzymes come together to decorate it.”
The Gao lab has been working for years to uncover all the steps involved in the production of the 21R-citrinadin A compound, which has been shown to be effective against leukemia in rats and human throat cancer cells, according to Shuai Liu, a Rice postdoctoral researcher who is the study’s lead co-author.
The newly identified enzyme is one of several discovered by the Gao lab that can perform singular catalytic functions such as controlling chirality and facilitating the Diels-Alder reaction.
“This really is a complete story,” Gao said. “We used gene knockout, heterologous expression, mutagenesis studies, enzymology and so on to solve nearly every single step in the biosynthesis of this compound. Over 20 enzymes assemble and coordinate to produce the molecule. I find it fascinating that enzymes work cooperatively in this way to produce this wonderfully complex molecule.”
More information: Shuai Liu et al, Fungal P450 Deconstructs the 2,5-Diazabicyclo[2.2.2]octane Ring En Route to the Complete Biosynthesis of 21R-Citrinadin A, Journal of the American Chemical Society (2023). DOI: 10.1021/jacs.3c02109
Depiction of the seeded supramolecular polymerization protocol (a) and the associated time-dependent changes (b). Supramolecular polymers obtained by open-ended and closed-ended seeds (c and d, respectively). Credit: Shiki Yagai from Chiba University
Supramolecular polymers are a new class of polymers that are currently being evaluated for material applications. These interesting compounds also play an important role in cellular activities in the body. “Supra,” as the name suggests, is attributed to some unique properties that go beyond those of conventional polymers. Unlike traditional polymers, which are held together by strong, irreversible covalent bonds, supramolecular polymers are held together by weaker, reversible hydrogen bonds.
Supramolecular polymers can reversibly assemble and disassemble, are highly versatile, and can be used for developing targeted drug delivery therapies, sensors to detect pollutants, diagnostic markers, energy storage devices, personal care products, and self-repairing and recyclable materials. While their excellent recyclability makes them wonderful candidate molecules for sustainable applications, there is one roadblock—researchers have yet to understand how to control their polymer growth.
However, there have been advancements in this aspect. Researchers are now able to build “unlikely” polymers by triggering their assembly with “seeds,” enabling control their polymer growth. There are two main mechanisms through which this seed-induced self-assembly occurs: primary nucleation or elongation, where the polymer grows from its end; and secondary nucleation, where new molecules join the polymer by sticking to its surface. The distinction between these processes is important because it enables researchers to better control and manipulate the growth of these unique polymers. Unfortunately, in most cases of seeded self-assembly, primary and secondary nucleation can be difficult to tell apart.
To tackle this issue, a group of researchers led by Professor Shiki Yagai from Chiba University aimed to compare and study the impact of these two processes while delineating the role of precisely controllable “seeded supramolecular polymerization.” Their goal was to figure out how different seed shapes affect the formation of new supramolecular polymers. Their findings are published in Chemical Communications.
Prof. Yagai tells us what motivated the team to pursue this topic of research: “Because of the difficulty in controlling polymerization, supramolecular polymers have not yet reached the point of practical application even though three decades have passed since their establishment as a concept.” He is convinced, however, that because of their versatility, further research in this area is likely to lead to widespread applications of these self-organizing polymers in our daily lives.
For their experiments, the researchers used two supramolecular polymers as “seeds.” While a closed-ended ring-shaped seed was used in a previous study, an open-ended, helicoidal seed was newly prepared. The researchers found that when the open-ended, helicoidal seed was used, it acted as a template for the target molecules to attach and grow longer. On the other hand, when the closed-ended ring-shaped seed was used, it did not elongate itself, but rather served as a surface where new molecules could attach and form clusters, like a platform for new structures.
This research shows that the type of seed used in self-assembling supramolecular polymers influences the way the molecules assemble, and the final shape of the formed structures. This opens up exciting possibilities for various applications, from self-repairing and more easily recyclable materials to more advanced drug delivery systems, sensing technologies, and energy storage devices.
Prof. Yagai states, “By understanding these assembly processes, we can design and develop the next generation of more precise and environmentally friendly polymers with tailored structures and properties. The practical application of supramolecular polymers will enable us to produce plastic materials with lower energy consumption and reduce the energy required for recycling.”
The ability to manipulate these versatile, self-assembling polymers at the molecular level offers great potential for addressing complex challenges and creating innovative, sustainable solutions in fields ranging from healthcare to environmental sustainability.
More information: Hiroki Itabashi et al, Distinct seed topologies enable comparison of elongation and secondary nucleation pathways in seeded supramolecular polymerization, Chemical Communications (2023). DOI: 10.1039/D3CC01587D
RNA, an essential biomolecule for life, has been used in environmental applications including monitoring microbial communities, developing pesticides, and quantifying the abundance of pathogenic viruses, such as SARS-CoV-2, in water and wastewater systems. Understanding how quickly RNA breaks down in given conditions is critical to harnessing the molecule in these and other emerging technologies.
According to a new study by researchers working with Kimberly Parker, assistant professor of energy, environmental & chemical engineering in the McKelvey School of Engineering at Washington University in St. Louis, RNA can undergo rapid hydrolysis when adsorbed into iron oxide minerals. This discovery unveils a previously unknown abiotic pathway for RNA degradation and sheds light on biogeochemical processes and environmental system dynamics. The results were published May 22 in Environmental Science & Technology.
“This is the first abiotic process we’ve found that causes RNA degradation in the environment on timescales that can compete with biotic degradation,” Parker said. “Instead of depending on biological agents like enzymes or microbes to break down RNA molecules, we found that RNA degradation catalyzed by minerals happens relatively quickly regardless of the biological context. This could be an important limit on how long RNA persists in the environment.”
First author Ke Zhang conducted the research in Parker’s lab while earning a doctorate in environmental engineering at WashU in 2022. Zhang found that RNA undergoes rapid hydrolysis on the timescale of hours when adsorbed to iron oxide minerals such as goethite and hematite. This hydrolysis process is uniquely facilitated by the presence of iron in the minerals, which chemically accelerates the structural breakdown of the RNA molecule. This finding challenges scientists’ previous assumptions about the environmental factors affecting RNA degradation, particularly in iron-rich soils and sediments, which account for approximately 10% of global ice-free land.
“This process could provide an important limit on how long RNA hangs around in the environment, but there are certain conditions that can block this breakdown pathway,” Parker said. “While we measured the reaction timescales and determined the reaction products in this research, we need to develop more insights into the reaction mechanism in the future. Understanding the mechanisms as well as timescales of RNA degradation is crucial for accurately interpreting relative amounts of DNA versus RNA, studying viruses and pesticides, and even exploring the origin of life.”
More information: Ke Zhang et al, RNA Hydrolysis at Mineral–Water Interfaces, Environmental Science & Technology (2023). DOI: 10.1021/acs.est.3c01407
Adam Hollerbach with a SLIM device created at Pacific Northwest National Laboratory. Credit: Andrea Starr | Pacific Northwest National Laboratory
The universe is awash in billions of possible chemicals. But even with a bevy of high-tech instruments, scientists have determined the chemical structures of just a small fraction of those compounds, maybe 1%.
Scientists at the Department of Energy’s Pacific Northwest National Laboratory (PNNL) are taking aim at the other 99%, creating new ways to learn more about a vast sea of unknown compounds. There may be cures for disease, new approaches for tackling climate change, or new chemical or biological threats lurking in the chemical universe.
The work is part of an initiative known as m/q, or “m over q”—shorthand for mass divided by charge, which signifies one of the ways that scientists measure chemical properties in the world of mass spectrometry.
“Right now, we can take a sample from soil, where, depending on soil type, there may be thousands of chemical compounds in just a teaspoon’s worth,” said Thomas Metz, who leads the m/q Initiative. “And we don’t know what most of them are in terms of their chemical structures. We simply have no idea what’s in there.”
Scientists typically rely on reference libraries that contain information about thousands of molecules to identify substances. Researchers sort their samples from soil, the body, or elsewhere and compare what they have measured experimentally to what’s in the library. While that’s helpful, it limits scientists to only structurally identifying molecules that have been seen before—for example, through analysis of standard compounds purchased from chemical suppliers.
In the latest development, a team led by scientist Adam Hollerbach has combined two high-resolution instruments into one system to size up molecules in unprecedented detail. The results were published June 12 in the journal Analytical Chemistry.
Now, scientists can make several important measurements about chemical compounds in one experiment, gaining important information faster, more conveniently, and more accurately than before.
Hollerbach’s technique applies to ions—molecules that have either a positive or negative charge. That makes them easier to control and possible to detect using mass spectrometry.
Mass spectrometry: Tool of the ion whisperers
Like the people who study them, ions have many features that distinguish one from another. In people, weight, hair color, size, shape, eye color, and many other characteristics help us know who’s who. For ions, identifying characteristics include mass, shape, size, electric charge, and chemical composition. Those not only serve as identifiers but also as guides to the associated molecules’ behavior—clues to their potential to cure disease or sop up pollutants, for example.
That understanding should help the efforts of scores of scientists at PNNL who focus on understanding the effect of microbes on climate. Microbes play a key role in transforming elements like carbon into other forms that are important for the planet. Their impact on warming or cooling the planet is mighty. But scientists have much to learn.
“There may be millions of microbes in just a gram of soil, and we don’t know who most of them are or what they do. There’s a lot of discovery still to happen,” said Metz. “From the viewpoint of challenging science, it’s either a worst-case scenario or one of our greatest opportunities, depending on how you look at it.”
The m/q scientists are seizing the opportunity. Instead of framing their questions within the relatively small number of compounds that can be identified in conventional mass spectrometry measurements, they’re trying to leapfrog current limitations and create a whole new way of identifying what is unknown today. It’s a bit like when a new telescope is deployed and reveals several distinct stars where before, just one blurry hodgepodge of celestial bodies was visible.
The work is both experimental, putting molecules through their paces in the laboratory, and on computers, where scientists model what they are seeing and predict what they will likely see.
In the experiments described in the Analytical Chemistry paper, Hollerbach and colleagues made sensitive measurements of peptides and lipids. The experiments combined two instruments with similar names but that provide different details about ions. Both are used in mass spectrometry, a field whose history is interwoven with discoveries by PNNL scientists.
The first instrument is a mass spectrometer, which measures an ion’s mass, electric charge, and how the ion breaks apart. In this study, the team used an Orbitrap developed by Thermo-Fisher Scientific. Such instruments sort molecules of different masses well, but two molecules with the same mass are difficult to separate. Think of two people, each weighing 180 lbs.—one is tall and thin while the other is short and stocky. On a scale alone, they would be impossible to separate.
A SLIM approach: Ion mobility spectrometry brings hefty results
The second instrument is known as SLIM: structures for lossless ion manipulations. SLIM, created by PNNL scientist Richard D. Smith and colleagues, is an ion mobility spectrometer that measures an ion’s size and electric charge.
SLIM, which is about the size of a laptop and stands at just one-quarter of an inch thick, is a hothouse of molecular activity. Dozens of long, winding paths transform the small device into a 42-foot-long molecular racetrack, with ions that are controlled tightly by electric fields racing round and round an oval obstacle course.
The “obstacles” are other, known molecules such as helium or nitrogen molecules. As the ions under study race through the SLIM device, they navigate around or through the other molecules, tumbling and swerving much like a football running back runs through and around opposing blockers. The term “ion mobility spectrometry” truly captures the action.
By recording how long it takes for the ions to complete the course—how deftly they navigate the blocking ions—scientists learn all kinds of things about ions’ shape and size. That information, which isn’t available from a standard mass spec instrument, is combined with data about the ion’s mass, electric charge, and fragmentation pattern. Altogether, the data yields the ion’s collision cross section, its molecular formula, and its fragmentation pattern, properties that are central to understanding a molecule’s structure.
“Two different molecules can have the same number of atoms, and the same mass and charge, but they could have very different structures and activity. That’s where SLIM comes in to tell the difference,” said Hollerbach. “Just one small change can mean the difference between a molecule that is indicative of a disease and one that’s not.”
The key to Hollerbach’s experiment was getting the two different instruments to play nicely together. While both standard mass spectrometry and ion mobility spectrometry analyze ions, they work on different time scales. Ions make their journey through SLIM and arrive at the Orbitrap faster than they can be processed.
So Hollerbach drew on an old technique, deploying “dual-gated ion injection.” He added gates to control the intake of ions into the system and to control their arrival at the Orbitrap, choosing to send some of the ions from SLIM into oblivion to keep the flow at a manageable rate.
“Really, the questions we ask are very simple,” said Hollerbach. “What is this, and how much is there? But the techniques we use are complex.”
Other m/q scientists are working on additional ways to identify or exploit unknown molecules. Some are creating ways to use data like that from Hollerbach’s experiment to predict an ion’s structure automatically, so drug makers and other scientists would know exactly what they’re working with. Others are scouting out the millions of possibilities for forms of compounds such as fentanyl, sorting out what’s unlikely from what might show up on the street one day. Then they predict how those compounds would behave inside a mass spectrometer—creating a way to identify them if and when they do show up.
More information: Adam L. Hollerbach et al, A Dual-Gated Structures for Lossless Ion Manipulations-Ion Mobility Orbitrap Mass Spectrometry Platform for Combined Ultra-High-Resolution Molecular Analysis, Analytical Chemistry (2023). DOI: 10.1021/acs.analchem.3c00881
Composition, structure and crystallography of the new type planar defects. Credit: the authors
In collaboration with the Institute of Physics at the Chinese Academy of Sciences (IPCAS), researchers from Beijing University of Technology (BJUT) have discovered a new type of grain-interior planar defect in a ceramic phase in TiC doped cemented tungsten carbides.
These planar defects were found to be a result of the ordered distribution of heteroatoms on specific crystal planes of tungsten (W) and carbon (C). Importantly, these newly identified defects display distinct characteristics that set them apart from known planar defects, such as phase boundaries, grain boundaries, twin boundaries, stacking faults and complexions.
The work is published in Advanced Powder Materials, and involved detailed characterizations on the atomic scale for the composition, structure and crystallography of the new type of planar defects. In addition, comprehensive model calculations were conducted to assess the defects’ energy state and stability, providing further insights into their nature.
The research team, led by Professor Xiaoyan Song from BJUT, found that the occurrence of titanium (Ti) monolayer on the basal planes of WC was caused by the destabilization of (W,Ti)Cx complexions, which formed at the WC/Co interfaces by dissolution-precipitation processes during sintering of the powder mixture. The stable Ti monolayer may provide nucleation sites for the growth of WC crystal along the [0001]WC direction. This possibility was confirmed by model calculations.
“We further explored the possibility for the formation of such planar defects in WC grains by doping of V, Zr, Nb, Mo, and Hf through modeling,” explained Song. “We found that the Ti-monolayer induced planar defects had the highest stability in the WC grain interior, and also were much easier to form in the cemented carbides.”
The work highlights the significance of planar defects with high stability hindering the long-distance motion of stacking faults and dislocations within grains, and act as obstructions against propagation of the transgranular cracks. Thus, the risk of transgranular fracture, which is the dominant failure mode of the covalent crystals in ceramics and ceramic matrix composites, can be significantly reduced.
“We conclusively demonstrated that adjusting the density of planar defects allows for optimal mechanical performance, striking an exceptional balance between strength and fracture toughness in the materials,” added Song. “Our study paves the way for improving the mechanical properties of materials through the deliberate introduction and customization of heteroatomic monolayer-induced grain-interior planar defects.”
Furthermore, the methodology described in this article, using cemented tungsten carbides as a representative case, can be extended to other materials. By carefully selecting dopants and controlling sintering parameters, it becomes possible to fine-tune the density of planar defects to achieve desired properties in various ceramic systems.
More information: Xingwei Liu et al, Grain-interior planar defects induced by heteroatom monolayer, Advanced Powder Materials (2023). DOI: 10.1016/j.apmate.2023.100130
UPLC analysis of polymerase/endonuclease catalyzed P1 formation under optimized conditions. UPLC trace showing crude product P1 formed during the KOD polymerase (2 µM) and TnEndoV endonuclease (2 µM) catalyzed reaction using T1 (1 µM), dNTPs (4 mM each), in 20 mM Tris-HCl (pH 8), 20 mM MgCl2, 100 mM arginine-HCl (pH 8), 100 mM glutamate-KOH (pH 8), DTT (10 mM), formamide (10%), trehalose (0.2 M), 1,2-propanediol (1 M) and AcBSA (0.01 mg/ml) following 12 hours incubation at 70 °C. Product was generated following 330 cycles of template extension and product cleavage to afford 0.33 mM (equivalent to 1.9 g/L) of P1. Credit: Science (2023). DOI: 10.1126/science.add5892
A team of biochemists at the Manchester Institute of Biotechnology has developed an isothermal biocatalytic process that can be used to manufacture therapeutic oligonucleotides in large volumes. In their paper published in the journal Science, the group describes their process and possible medical applications.
Oligonucleotides (short base pairs of RNA or DNA molecules) have been used to treat several rare diseases. But the process of manufacturing them in large quantities is difficult for more general use. More recently, the development of therapies such as siRNA inclisiran to reduce cholesterol in the bloodstream has put pressure on biochemists to develop a less difficult process.
In this new effort, the team developed a one-pot manufacturing process. They note that it can be conducted in an aqueous solution and uses polymerase enzymes to extend template strands and an endonuclease enzyme to release the final product, allowing for the templates used in the process to be used repeatedly.
The process starts by adding nucleoside triphosphates to an aqueous solution, which drives a template-dependent synthetic reaction. Adding an extended template results in product cleavage, leading to the creation of endonuclease V, part of which can be removed for use as it is. The other part is then exposed to yet another template made using inosines, and the material that makes it through the final template consists of the desired oligonucleotides.
The research team suggests that in addition to making the manufacture of therapeutic oligonucleotides less expensive, it also makes it more scalable. To prove their claims, they used their new method to manufacture several well-known therapeutic oligonucleotides, such as pegaptanib, which has been used to treat macular degeneration.
They were able to produce approximately 2g per liter, but suggest the process should allow for amounts over 10g per liter—and at some point, as much as 100g per liter. They plan to continue fine tuning the process to manufacture more kinds of oligonucleotides. The team has partnered with Novartis, which already produces inclisiran, to further develop the scaling process.
More information: E. R. Moody et al, An enzyme cascade enables production of therapeutic oligonucleotides in a single operation, Science (2023). DOI: 10.1126/science.add5892
Researchers from ETH Zurich and the University of Geneva have developed a new method that allows them to observe chemical reactions taking place in liquids at extremely high temporal resolution. This means they can examine how molecules change within just a few femtoseconds—in other words, within a few quadrillionths of a second. The method is based on earlier work done by the same group of researchers led by Hans Jakob Wörner, Professor of Physical Chemistry at ETH Zurich. That work yielded similar results for reactions that take place in gas environments.
To expand their X-ray spectroscopy observations to liquids, the researchers had to design an apparatus capable of producing a liquid jet with a diameter of less than one micrometer in a vacuum. This was essential because if the jet were any wider, it would absorb some of the X-rays used to measure it. The work is published in the journal Nature.
Molecular pioneer in biochemistry
Using the new method, the researchers were able to gain insights into the processes that led to the emergence of life on Earth. Many scientists assume that urea played a pivotal role here. It is one of the simplest molecules containing both carbon and nitrogen. What’s more, it’s highly likely that urea was present even when the Earth was very young, something that was also suggested by a famous experiment done in the 1950s: American scientist Stanley Miller concocted a mixture of those gases believed to have made up the planet’s primordial atmosphere and exposed it to the conditions of a thunderstorm. This produced a series of molecules, one of which was urea.
According to current theories, the urea could have become enriched in warm puddles—commonly called primordial soup—on the then lifeless Earth. As the water in this soup evaporated, the concentration of urea increased. Through exposure to ionizing radiation such as cosmic rays, it’s possible that this concentrated urea produced malonic acid over multiple synthesis steps. In turn, this may have created the building blocks of RNA and DNA.
Why this exact reaction took place
Using their new method, the researchers from ETH Zurich and the University of Geneva investigated the first step in this long series of chemical reactions to find out how a concentrated urea solution behaves when exposed to ionizing radiation.
It’s important to know that the urea molecules in a concentrated urea solution group themselves into pairs, or what are known as dimers. As the researchers have now been able to show, ionizing radiation causes a hydrogen atom within each of these dimers to move from one urea molecule to the other. This turns one urea molecule into a protonated urea molecule, and the other into a urea radical. The latter is highly chemically reactive—so reactive, in fact, that it’s very likely to react with other molecules, thereby also forming malonic acid.
The researchers also managed to show that this transfer of a hydrogen atom happens extremely quickly, taking only around 150 femtoseconds, or 150 quadrillionths of a second. “That’s so fast that this reaction preempts all other reactions that might theoretically also take place,” Wörner says. “This explains why concentrated urea solutions produce urea radicals rather than hosting other reactions that would produce other molecules.”
Reactions in liquids are highly relevant
In the future, Wörner and his colleagues want to examine the next steps that lead to the formation of malonic acid. They hope this will help them to understand the origins of life on Earth.
As for their new method, it can also generally be used to examine the precise sequence of chemical reactions in liquids. “A whole host of important chemical reactions take place in liquids—not just all biochemical processes in the human body, but also a great many chemical syntheses relevant to industry,” Wörner says. “This is why it’s so important that we have now expanded the scope of X-ray spectroscopy at high temporal resolution to include reactions in liquids.”
The researchers from ETH Zurich and the University of Geneva were assisted in this work by colleagues from Deutsches Elektronen-Synchrotron DESY in Hamburg, who performed calculations required to interpret measurement data.
Surface representation of SARS CoV-2 Mpro protein with fragment hits from Diamond XChem platform bound in active site (Green). Credit: Diamond Light Source Ltd
In a new study, researchers from IBM, Oxford University and Diamond Light Source show that IBM’s AI Model, MoLFormer, can generate antiviral molecules for multiple target virus proteins, including SARS-CoV-2, that can accelerate the drug discovery process and bolster our response to future pandemics.
The results are laid out in a new paper published in Science Advances, and at the time of the paper’s submission, the antiviral properties of eleven molecules were successfully validated by Oxford researchers. This breakthrough has the potential to get drugs to people faster in the next crisis and bring treatments for urgent, life-threatening illnesses within reach.
Early in the pandemic, a group of computer scientists at IBM wanted to explore if generative AI could be used to design never-before seen molecules to block SARS-CoV-2, the virus that causes COVID-19. David Stuart, Head of the Division of Structural Biology in the Department of Clinical Medicine at the University of Oxford and Life Sciences Director at Diamond Light Source, the UK’s national synchrotron who is an authority on pathogens HIV, SARS, and Ebola, among other viruses explains he was initially skeptical. “The idea that you could take a protein sequence and, with AI, pluck out of thin air chemicals that would bind to a 3D site on the virus seemed very unlikely,” he said.
However, he and Martin Walsh also an expert structural biologist and Life Sciences Deputy director at Diamond joined up with the IBM team and over the course of three years, demonstrated that generative AI could, “pluck viable starting points for antivirals out of thin air,” in collaboration with Enamine Ltd., a chemical supplier in Ukraine, and other researchers at Oxford.
Because the generative model was also a foundation model, pre-trained on massive amounts of raw data, it was versatile enough to create new inhibitors for multiple protein targets without extra training or any knowledge of its 3D structure.
The Stuart and Walsh groups had commenced working on two essential SARS-CoV-2 proteins, namely the spike protein and the main protease. Using these targets, the team hit on four potential COVID-19 antivirals in a fraction of the time it would have taken using conventional methods. The work then exploited Diamond’s high-throughput macromolecular crystallography beamlines to visualize how a subset of the AI generated compounds bound to the main protease.
Their work is showcased in their new paper in Science Advances and IBM has released a web-based interface for interacting with the model and chemical foundation models like it in IBM Cloud.
The team stated that the validated molecules have many more hurdles to clear, including clinical trials, before companies could potentially turn them into drugs. But even if the AI-generated “hits” never materialize into actual drugs, the work provides confirmation that generative AI has an important role to play in the future of drug development, especially in a time of crisis.
“It took time to develop and validate these methods, but now that we have a working pipeline in place, we can generate results much faster,” said study co-senior author, Payel Das, a researcher at IBM Research. “When the next virus emerges, generative AI could be pivotal in the search for new treatments.”
“Generating initial compounds that bind with high affinity to a drug target of interest accelerates the structure-based drug discovery pipeline and underpins our efforts to be better prepared for future pandemics,” said, Martin Walsh, who was co-senior author at Diamond
The researchers built their model, Controlled Generation of Molecules (or CogMol), on a generative AI architecture known as variational autoencoders, or VAEs. VAEs encode raw data into a compressed representation, and then decode, or translate, it back into a statistical variation on the original sample. Their model was trained on a large dataset of molecules represented as strings of text, along with general information about proteins and their binding properties. But they deliberately left out information about SARS-CoV-2’s 3D structure or molecules known to bind to it. Their goal was to give their generative foundation model a broad base of knowledge so that it could be more easily deployed for molecular design tasks it has never seen before.
Their goal was to find drug-like molecules that would bind with two COVID protein targets: the spike, which transmits the virus to the host cell, and the main protease, which helps to spread it. Though the 3D structures of both proteins had been discovered by that time, the IBM researchers chose to use only their amino acid sequences, derived from their DNA. By limiting themselves in this way, they hoped that the model could learn to generate molecules without knowing the shape of their target.
The researchers input only the amino acid sequence for each protein target into CogMol, which generated 875,000 candidate molecules in three days. To narrow the pool, the researchers ran the candidates through a retrosynthesis platform, IBM RXN for Chemistry, to understand what ingredients would be needed to synthesize the compounds. Based on the platform’s predicted recipes, they selected 100 molecules for each target. Chemists at Enamine further pared the list to four molecules for each target, selecting those deemed easiest to manufacture.
After synthesizing the eight novel molecules, Enamine shipped them to Oxford for testing their ability to disrupt the functions of the two protein targets in the labs of Prof Chris Schofield and PRof Gavin Screaton. . The intense X-ray beam generated from Diamond which are 10 billion times brighter than the sun were used to visualize how the compounds interacted with proteins to inactivate their function. The novel compounds were further tested in target inhibition and live virus neutralization tests. Two of the validated antivirals target the main protease; the other two not only targeted the spike protein but proved capable of neutralizing all six major COVID variants. “You get a map that shows exactly where things bind, and bang! you’ve got a confirmation,” said Stuart.
CogMol is one of several chemical foundation models that IBM has since developed. The largest, MoLFormer-XL, was trained on a database of more than 1.1 billion molecules and is currently being used by Moderna to design mRNA medicines. “We created valid starting points for accelerated development of antivirals using a generative foundation model that knew relatively little about its protein targets,” said the study’s co-senior author, Jason Crain, a researcher at IBM Research and professor at Oxford. “I’m hopeful that these methods will allow us to create antivirals and other urgently needed compounds much faster and more inexpensively in the future.”
Though the researchers focused on validating antivirals for COVID, they argue that these methods can be extended to existing viruses that continue to mutate, like the flu, or viruses that have yet to surface. “If you want to be prepared for the next pandemic, you want drugs that act on different sites of the protein,” concluded Stuart. “It becomes much harder for the virus to escape.”
More information: Vijil Chenthamarakshan et al, Accelerating drug target inhibitor discovery with a deep generative foundation model, Science Advances (2023). DOI: 10.1126/sciadv.adg7865
A team of researchers from the Universities of Amsterdam and Zurich together with the Swiss company Allocyte Pharmaceuticals have for the first time been able to discover allosteric sites in a type of cell surface receptor called integrin.
In a paper recently published in the Journal of Chemical Information and Modeling, they describe how this reveals previously inaccessible druggable integrin pockets. At the heart of the research is a novel computational approach for mixed-solvent molecular dynamics simulation developed by Dr. Ioana Ilie at the Computational Chemistry group of the Van ‘t Hoff Institute for Molecular Sciences at the University of Amsterdam.
Integrins are a family of cell surface adhesion receptors which are capable of transmitting signals bidirectionally across membranes. They are known for their therapeutic potential in a wide range of diseases. However, the development of integrin targeting medication has been impacted by unexpected downstream effects. In particular, these are observed with drugs targeting the native binding site of the integrin.
The so-called allosteric modulation of integrins is a promising approach to potentially overcome these limitations. Here, the drug binds elsewhere on the receptor, changing the conformation and thus impacting the activity of the protein. Allosteric modulation of receptors therefore creates opportunities for drug discovery and development which are potentially superior to classic orthosteric modulation.
Novel druggable pockets
The novel computational approach developed by Ioana Ilie relies on enriching the solvent with small organic molecules (benzene in small concentrations) to enable the gentle opening of the integrin α I domain. This revealed novel previously inaccessible druggable pockets within the integrins LFA-1, VLA-1, and Mac-1. This study thus offers structural and dynamic insight on the effect of small alterations in solvent conditions on the accessibility of novel potentially druggable pockets, which are validated via virtual screening. The study acts as proof of concept and sets the foundation for the design of the next-generation integrin-targeting drugs. Additionally, it opens new research avenues towards the identification of allosteric sites in other up to date undruggable protein targets.
Finally, the study goes beyond drug discovery as it demonstrates that minor changes in the solvent conditions can have a dramatic impact on the conformational space of the solvated molecule. This offers the opportunity to tune the solvent conditions in order to obtain a specific response of the solvated molecule (e.g., protein, material), which implicitly can aid in the development novel bio-inspired materials with responsive properties.
More information: Ioana M. Ilie et al, Decrypting Integrins by Mixed-Solvent Molecular Dynamics Simulations, Journal of Chemical Information and Modeling (2023). DOI: 10.1021/acs.jcim.3c00480
