Although the microscopic world is intimately part of our daily lives, visualizing viruses, bacteria, and even the molecular structures that mediate their interactions – among themselves and with the human body – remains one of science’s major challenges. In recent decades, however, this barrier has been progressively overcome with advances in cryo-electron microscopy, or cryo-EM, a tool that makes it possible to observe in detail structures thousands of times smaller than a bacterium that sustain life at the molecular scale. In Brazil and worldwide, this approach has been applied to the investigation of fundamental processes, with impacts ranging from a basic understanding of biological systems to the development of new therapeutic strategies with transformative potential for health.
Cryo-EM can reveal everything from proteins to entire viruses
Cryo-EM is an advanced microscopy technique that combines the ultra-rapid freezing of biological samples, their visualization in high-precision microscopes, and the computational reconstruction of structures, resulting in extremely high-resolution images. The approach, recognized with the Nobel Prize in Chemistry in 2017, makes it possible to observe molecules close to their natural state, without the need for extensive modifications and with a lower risk of damage. For this reason, it is a powerful tool for understanding ongoing biological processes and unraveling the three-dimensional organization of the structures that support them.
The advantages of cryo-electron microscopy drive its application across a wide range of contexts, from the observation of isolated proteins to complex systems involved in bacterial motility and secretion, as well as the analysis of complete viruses. In this scenario, different research groups have explored the technique’s potential to answer increasingly specific questions about how biological systems function. This is the case of the Center for Research in Biology of Bacteria and Bacteriophages (CEPID B3), a Brazilian research network based at the University of São Paulo (USP) and funded by the São Paulo Research Foundation (FAPESP).
Gabriel Araujo, a postdoctoral researcher at CEPID B3 and a specialist in cryo-EM techniques, explains that several groups affiliated with the Center have been using this approach in their projects, especially in the analysis of bacteriophage proteins, or phages – special types of viruses that infect only bacteria and can be used as alternatives to antibiotics. “In one of the projects, we are advancing the use of cryo-EM to reconstruct phage tail proteins that bind to receptors on the bacterial cell”, the researcher notes. “This may help us understand which hosts they can recognize and infect, as well as provide clues about the development of bacterial resistance through mutations and changes in these receptors”, he explains.
Samples are frozen to nearly −200 degrees Celsius
The path from obtaining a sample of interest to interpreting the results involves several meticulous steps that begin well before the microscopes. First, the materials undergo a series of biochemical preparations, followed by a delicate freezing process. “We need to test different freezing conditions to obtain the appropriate parameters for each sample”, the researcher explains. “This is done by applying the sample solutions onto what we call grids, which are very thin circular meshes only 3 millimeters in diameter”, he adds. The grids are coated with a special carbon film full of tiny holes, within which the proteins in the sample must be distributed.
Freezing must be almost instantaneous to prevent the water in the mixture from crystallizing and creating interference that could hinder observations. “To obtain so-called vitreous ice, the grid must be immediately plunged by the device into a cryogenic bath that is extremely cold”, Araujo says. From that moment on, all material is kept at temperatures close to −196 degrees Celsius. The resulting ice layer must be extremely thin, smooth, and as transparent as possible.
With the samples prepared, frozen, and confirmed to be suitable, the work proceeds to observation in powerful, specialized microscopes, where each protein present in the sample is imaged hundreds of times from different angles and positions. Some of the instruments used at this stage can reach nearly 3 meters in height. “Even the microscope room itself requires special care to isolate external interference; there is extensive engineering in the flooring to ensure greater stability”, Araujo says. The analyses conducted by CEPID B3 take place at the Brazilian Nanotechnology National Laboratory (LNNano), in Campinas. Data acquisition is automated and can take days. Thousands of images are generated, amounting to terabytes of information. “But for data processing, this is only the beginning”, the researcher emphasizes.
The next step depends on intensive computational work involving identification, alignment, image averaging, and the integration of information until three-dimensional reconstructions of the analyzed samples are formed. Only after this stage can specialists like Araujo begin to interpret the data, refining molecular models, identifying the position and interactions of amino acids – the building blocks of proteins – present in the sample, and investigating the implications of the new information for the biological systems under study.
Combining new technologies expands the potential to generate rich and comprehensive data
Amino acids could, in principle, adopt many different conformations during protein formation. However, due to their interactions with one another and with the aqueous environment, they ultimately organize into specific three-dimensional structures. Today, computational methods can predict which of these structures are likely to form under given conditions. These predictions facilitate the use of data obtained through cryo-electron microscopy. “With a good predicted 3D model, such as those generated by the artificial intelligence system AlphaFold, it becomes much easier to fit protein chains into the maps obtained and refine the structures with experimental data”, he explains.
Despite its complexity, cryo-electron microscopy offers important advantages over other structural biology techniques. “In general, it requires a smaller amount of sample”, Araujo notes, which is especially useful for materials that are difficult to obtain. In addition, it allows proteins to be observed in solution, under conditions closer to their natural environment.
Another advantage is the ability to visualize different conformations of the same protein within a single sample. “Instead of isolating a single state, we can computationally separate different particles and obtain distinct reconstructions”, the researcher explains. This significantly expands the possibilities for analysis.
The future of cryo-EM
In an interview with Jornal da Unicamp, Leonardo Talachia Rosa, a CEPID B3 researcher who also conducts studies using the technique, highlights that cryo-EM already has practical impacts that extend to human health. According to him, understanding how proteins are organized in space and perform their biological functions makes it possible to design more precise drugs that can be programmed to bind to specific regions of proteins, inhibiting or modifying their activity and combating mechanisms such as bacterial resistance.
Araujo emphasizes that, even with recent advances, the field remains open to innovation and suggests that in the coming years the technique may become even more efficient. “One of the major current bottlenecks of the technique is sample preparation, because it is difficult to obtain regions on the grids with ice thin enough for imaging”, he says. “Many proteins also tend to exhibit undesirable behaviors during freezing, such as sticking to the carbon or failing to distribute within the ice”, he adds. He argues that one proposed way to address some of these issues is to automate grid preparation, making the process much faster, on the scale of fractions of a second. “The idea is basically to freeze the sample before these undesirable phenomena have time to occur”, he explains. The approach is already being explored by several research groups worldwide, and some commercial models are available.
“We also expect advances in cryo-electron tomography, bringing structural biology into much more complex situations within cellular environments”, Araujo notes. This approach, a variation of cryo-EM, allows large complexes to be reconstructed directly inside cells. The main difference lies in how images are acquired: “while in cryo-EM we obtain three-dimensional information by combining particles that appear in different orientations in the images, in cryo-electron tomography we repeatedly record the same objects while progressively tilting the sample to view them from different angles”, the researcher says. “Brazil is still at an early stage in this type of application, but some groups are already advancing in data collection using this methodology”, he adds. For him, the field is still far from reaching its limits. “There is considerable room for progress, both in microscopy and in processing strategies”, he concludes, underscoring the potential to expand what we will be able to see with cryo-EM, inside and outside cells, in the coming years.
