A human cell is a precisely regulated system that relies on the complex interaction of molecules. Structural insights into the cellular machinery at the atomic level allow us to understand the underlying regulatory mechanism and provide us with a roadmap for the development of novel drugs to fight diseases. Facilitated by recent technological breakthroughs, the Nobel prize-winning technique electron cryomicroscopy (cryo-EM) has become a versatile and extremely powerful tool to solve routinely near-atomic resolution three-dimensional protein structures. Consequently, it has become the focus of attention for structure-based drug design. In this review, we describe the basics of cryo-EM and highlight its growing role in biomedical research. Furthermore, we discuss latest developments as well as future perspectives. . . . . Keywords Electron cryomicroscopy Cryo-EM Biological macromolecule Biomedical research Drug design Basics of single-particle cryo-EM and filaments for decades. It does not depend on crystals, and protein structures can be determined rather quickly. The function of cellular macromolecules is closely related to However, until recently cryo-EM structures were limited their three-dimensional architecture. Detailed structural in- in resolution and did not allow building atomic models. sights help us not only to understand the mechanism underly- Due to important technological advances, single-particle ing cellular processes, but provide us with a topological map cryo-EM has caught up rapidly in recent years . Now, for the development of potential therapeutic compounds. high-resolution structures can be obtained routinely, ren- Over the last 50 years, the determination of atomic pro- dering cryo-EM to a major technique in structural biology. tein structures has been dominated primarily by X-ray In transmission electron microscopy (TEM), accelerated crystallography and, to some extent, by NMR spectrosco- electrons pass through and interact with the specimen. The py. Although both methods have contributed enormously interference between scattered and non-scattered electrons re- to our current molecular understanding of biological pro- sults in the so-called phase contrast and image formation. cesses, they also come with drawbacks. For X-ray crystal- Because electron microscopes require high vacuum, living lography, biomolecules have to be crystallized. Obtaining cells or more generally hydrated samples cannot be examined well-diffracting crystals, however, is often challenging, by this method at room temperature. time-consuming, and, in several cases, impossible. In cryo-EM, this problem is solved by embedding the sam- Although NMR does not require crystals, it is only practi- ples in amorphous ice through plunge-freezing in liquid eth- cable for molecules with low molecular weight, usually ane, a sample preparation method developed by Dubochet and below 50 kDa. Single-particle cryo-EM has been an alter- colleagues in the early 1980s . When imaged at cryogenic native for solving the structure of large protein complexes temperatures (77 K), the vapor pressure of the so-called vitri- fied sample is low and the proteins can therefore be imaged in their hydrated state. The contrast increases with increasing atomic number. Therefore, TEM images of biological matter, which is com- * Stefan Raunser posed mainly of light elements, such as carbon, oxygen, and stefan.raunser@mpi–dortmund.mpg.de hydrogen, have relatively low contrast. Biological samples are also very vulnerable to electron-beam damage . Therefore, Department of Structural Biochemistry, Max Planck Institute of the cumulative electron dose has to be kept low during image Molecular Physiology, Otto-Hahn-Str. 11, acquisition in order to reach atomic resolution, even further 44227 Dortmund, Germany 484 J Mol Med (2018) 96:483–493 decreasing the single-to-noise ratio of the images. Because of the low contrast of biological specimens in Consequently, the images of many proteins have to be record- cryo-EM, the thickness of the surrounding amorphous ice ed and overlaid to increase the signal-to-noise ratio. In addi- should be as thin as possible. At the same time, there should tion, since cryo-EM images represent two-dimensional (2-D) be as many particles as possible in an image in order to in- projections of a three-dimensional (3-D) object, many images crease the size of the dataset and decrease the amount of mea- of a protein in different orientations are needed to computa- suring time. The sample preparation technique is therefore tionally reconstruct its 3-D structure  (Fig. 1a). Joachim crucial for the success of the analysis and normally comprises Frank has been the driving force behind the development of the following workflow. Initially, the quality and suitability of computer-based image processing over the years. the sample is checked using negative-stain EM . For this, Fig. 1 The most common cryo-EM techniques. Schematic drawing in (a) small section of the diffraction image with individual diffraction spots to (d) illustrates principles of the most popular cryo-EM methods in the at higher magnification. The electron diffraction image was kindly upper panel and shows corresponding raw data in the lower panel. (a) provided by T. Gonen, Janelia Research Campus. (d)Cryo-ET: the Single-particle cryo-EM: particles are embedded in a thin layer of specimen is tilted within the microscope and images at different angles amorphous ice. Resulting representative class averages are shown as are recorded. A tomographic slice shows the cellular periphery with insets on the right. Scale bar, 50 nm. (b) Single-particle negative-stain microtubule bundles (black arrows) and plasma membrane (green EM: particles are embedded in a layer of heavy metal salts to increase the arrows). (e) Resolution range coverage of various methods in structural weak contrast of biological materials. Resulting representative class biology. Color code used for the TEM-based methods corresponds to averages are shown as insets on the right. Scale bar, 50 nm. (c) Micro- (a)–(d). Yellow: single-particle analysis; orange: electron ED: small 3-D crystals are hit with a focused electron beam and crystallography/micro-ED; red: electron tomography diffraction patterns are recorded at different tilt angles. Inset shows a J Mol Med (2018) 96:483–493 485 the protein is embedded into a layer of heavy metal salts, a high read-out speed, allowing the recording of movies in- mostly uranyl formate (Fig. 1b). This increases the contrast stead of single images. Since the signal-to-noise ratio of the and is easy to use. Although the sample is dehydrated and single movie frames is relatively high, they can be aligned to flattened during sample preparation, negative-stain EM is ide- correct the movement of the particles in the amorphous ice al for the initial examination of protein samples. during recording . The result is a tremendously increased Suitable samples - i.e., homogeneous, pure, and stable pro- resolution. tein complexes - are then vitrified for cryo-EM. The samples In 2015, the structure of β-galactosidase with a molecular are applied to a carbon-coated copper, molybdenum, or gold weight of 465 kDa was resolved at an overall resolution of grid. The carbon film contains holes of regular size, 2.2 Å (Fig. 2a). In this structure, the binding mode of the over which the sample is spread. Before plunging into liquid known inhibitor PETG could be studied in molecular detail, ethane, the excess protein solution is blotted away using filter highlighting the growing role of cryo-EM in drug discovery. paper. The parameters that have to be optimized during vitri- The 2-Å barrier was broken 1 year later with a 1.8-Å structure fication are the blotting time; humidity; type of filter paper; of the 334-kDa protein glutamate dehydrogenase, currently protein concentration; different grid types (hole size, metal still the record holder . support, support layer); the addition of additives, such as de- Higher resolution, however, has not been the only im- tergents to facilitate spreading of the sample; multiple blotting provement. The proteins studied by single-particle cryo- rounds to saturate the surrounding carbon with protein; and EM have also become increasingly smaller. The lower size the buffer composition [6–8]. Some proteins cannot be obtain- MW limit broke the 100-kDa barrier in 2016 with a struc- ed at sufficiently high concentration to directly embed them ture of the 93-kDa-sized cancer-associated isocitrate dehy- into amorphous ice. In this case, an additional support layer of drogenase in complex with its inhibitor ML309 at 3.8 Å carbon or graphene oxide can be used that spreads over the . This remarkable achievement was surpassed in 2017 holes and to which the proteins adhere . This decreases the with the elucidation of the structure of hemoglobin at 3.2 Å necessary protein concentration by up to two orders of mag- with a size of only 64 kDa. These studies were made pos- nitude. The optimization of freezing conditions is often labo- sible through the use of the Volta phase plate . This rious but key to successful structure determination at high device enhances the phase contrast and thereby improves resolution by cryo-EM. the signal-to-noise ratio in the low-resolution range and Images are taken under cryogenic conditions using a state- makes it possible to solve the structures of relatively small of-the-art high-end transmission electron microscope proteins (Fig. 2c) . equipped with a field emission gun and direct electron detec- Beside these latest developments in hardware, also new tors (DED). Single particles are digitally extracted from TEM software releases have made it easier for scientists to obtain high-resolution structures using cryo-EM. Software for high- images and processed resulting in 3-D reconstructions at up to 1.8-Å resolution . Reconstructions at near-atomic resolu- throughput data collection is under continuous development, tion (a term used to describe cryo-EM densities with resolu- clearly steering in the direction of Bon-the-fly^ processing. tions between ~ 2.5 Å to ~ 4 Å) allow de novo building of With a number of comprehensive and streamlined image pro- atomic models and their biochemical interpretation. cessing suites, such as, for example, Relion, EMAN2, cryoSPARC, or SPHIRE, even scientists with little prior knowledge in image processing can solve cryo-EM structures New developments paved the way [16–19]. The field also profits immensely from already avail- for near-atomic resolution able software hitherto used exclusively in X-ray crystallogra- phy, e.g. COOT, Rosetta, or similar programs for de novo A number of developments over the years enabled cryo-EM to building of atomic models . enter the current era of near-atomic structure determination. Apart from this, the emergence of new and the refinement The performance of transmission electron microscopes has of already established sample preparation methods has also steadily improved, resulting in, for example, high- greatly contributed to the resolution revolution, but its detailed performance field emission guns, stable stages, and automated description would exceed the scope of this review [21, 22]. data acquisition over extended periods of time. Especially the development of direct electron detectors (DED) has been a game-changer. In contrast to previous detection technologies, Merits of single-particle cryo-EM namely photographic film and charge-coupled device (CCD) as high-resolution structural tool cameras, in DEDs, primary electrons are directly converted into electrical signals, increasing their detection quantum ef- Various techniques exist in structural biology and should ficiency (DQE) in all frequencies and consequently the signal- be regarded as complementary. Each technique offers to-noise ratio of recorded images . In addition, DEDs have unique advantages and in combination with other methods 486 J Mol Med (2018) 96:483–493 Fig. 2 High-resolution cryo-EM as tool for structure-based drug design. causing parasite Plasmodium falciparium, demonstrating the potential of (A) Electron density map of the 1.8-Å structure of glutamate dehydroge- cryo-EM in structure-based drug design. The actin filament is shown in nase, showing that single-particle cryo-EM is capable of achieving atomic light blue with central subunits colored in dark blue, magenta and cyan. resolution. Subunits of the homo-hexameric enzyme are colored in [EMD-3805]. (C) The Volta phase plate has revolutionized the EM field magenta, pink, cyan, and three different green hues. [EMD-8194]. (B) by providing unprecedented contrast for biological specimen without the Visualization of the density for the cyclic peptide jasplakinolide (yellow) need of defocusing. The introduction of additional phase shift greatly in the cryo-EM map of the non-canonical actin PfAct1 from the malaria- enhances the phase contrast. Scale bar, 10 nm they offer an optimal tool kit for determining molecular be directly studied. Very flexible parts of the protein aver- structures. The advantages of single-particle cryo-EM are age out during image processing and do therefore not im- as follows. First, only little sample is required; 3–5 μlof pede the analysis. In general, single-particle cryo-EM can protein at a concentration of 0.05–5 μMissufficient for deal with a certain amount of sample heterogeneity. obtaining a structure at near-atomic resolution. However, Different conformations can be separated and processed mostly more protein at the same concentration is needed to using unbiased computational classification procedures, screen for optimal freezing conditions (100–200 μl). oftenreferredtoas in silico purification. Second, proteins can be directly prepared and imaged im- mediately after purification. Therefore, even low-abundant and unstable complexes isolated from endogenous sources Bridging the gap between cells and atoms can be analyzed. Third, buffer conditions can be almost freely chosen offering a large range of conditions to study Cryo-EM covers a large range of applications. As de- the protein of choice. Fourth, while in X-ray crystallogra- scribed above, it is mostly used to solve high-resolution phy the likelihood of crystal formation drops with increas- structures of protein complexes by single-particle cryo- ing molecular weight of the proteins/multiprotein com- EM. Alternatively, protein structures, especially of small plexes, there is virtually no upper size limit for protein proteins, can be obtained by electron crystallography or complexes in EM. micro electron diffraction (micro-ED) (Fig. 1c, e). Cryo Flexible parts of proteins are still a major challenge in electron tomography (cryo-ET) is applied to image pro- structural biology. In X-ray crystallography the protein of teins and protein complexes in their cellular environment, interest often has to be extensively re-engineered by re- cellular compartments, whole cells, and tissues [24, 25] moval of loop regions, termini, or glycosylation sites to (Fig. 1d, e). obtain well-diffracting crystals. It is therefore difficult to Electron crystallography requires 2-D protein crystals. study proteins that only properly fold and function when They are imaged under cryogenic temperatures in the TEM glycosylated . In single-particle cryo-EM, however, and high-resolution information is derived from electron the engineering of proteins is normally not needed and diffraction patterns [26–28]. This technique enabled the full-length, post-translationally modified proteins can Henderson and colleagues in 1990 to generate the first J Mol Med (2018) 96:483–493 487 high-resolution structure of bacteriorhodopsin . The Single-particle cryo-EM of biomedically structure of AQPO determined by electron crystallography relevant proteins reached a resolution of 1.9 Å and allowed the visualization of water molecules and annular lipids interacting with the Single-particle cryo-EM has been very successful in elucidat- protein . In general, this method is ideally suited to ing the structure of a large variety of disease-related macro- study the structure of membrane proteins in a lipidic envi- molecules and cellular machines. In the following, we high- ronment . However, de novo 2-D crystallization of light a number of medically important targets that were struc- membrane proteins proved to be difficult, and most of the turally studied by cryo-EM. structures obtained by electron crystallography have only Traditionally, high-resolution cryo-EM structures could on- reached medium resolution. ly be obtained from large objects with a high degree of sym- The recent development of micro-ED uses tiny 3-D crys- metry, such as icosahedral viruses [36, 37] or filamentous tals instead of 2-D crystals. The crystals are smaller than viruses . While this has been a tedious and difficult pro- those used for conventional X-ray crystallography, making cess, the structure determination of icosahedral and filamen- the technique especially useful for the structure determina- tous viruses by cryo-EM became straightforward after the in- tion of the plethora of proteins that do not readily form large troduction of the DEDs. Recent important structures include crystals [30, 31]. The crystals are tilted in the TEM and that of the thermally stable Zika virus , which belongs to diffraction images are collected at defined tilt angles to later the family of Flaviviridae and is linked to congenital micro- reconstruct the 3-D volume (Fig. 1c). Micro-ED has the cephaly and the Guillain-Barré syndrome . The 3.7-Å potential of becoming a high-throughput method to deter- structure provides an important topology map of the virus mine the structures especially of small and difficult-to- helping to understand its mechanism of infection and to de- crystallize proteins. velop potent vaccines against it. The determination of high- In cryo-ET, the specimen is also tilted in the TEM, but resolution structures of filamentous viruses such as the tobac- images instead of diffraction patterns are recorded at defined co mosaic virus (TMV), a plant pathogen, also profited from tilt angles to later reconstruct the 3-D volume. The spectrum of the new technology (Fig. 3a) . samples used is broad, ranging from large complexes, such as Especially glycoproteins on the surface of envelope viruses the nuclear pore complex or envelope viruses, to prokaryotes are difficult to study in their native state but are of high med- and thin sections of eukaryotic cells and tissue. Importantly, ical interest as they are the most promising targets for vaccine the specimen should be as thin as possible to allow the passage design. Since they are asymmetrical, envelope viruses are of electrons and reduce beam damage (Fig. 1d). more difficult to study by single-particle cryo-EM. Initially developed as an alternative to cryo-sectioning, Therefore, these viruses have been mostly analyzed using cryo-focused ion-beam (FIB) milling has been recently shown cryo-ET, giving us a conceptual idea of how viral particles to be the best method so far to obtain thin slices of larger enter cells [42, 43]. With the recent introduction of DEDs, samples . In this method, a focused ion-beam cuts out a subtomogram averaging became very powerful and enabled thin lamella with high precision. the determination of high-resolution structures of the essential The resolution of tomograms is normally not better trimeric envelope glycoproteins that mediate the cell entry of than 15–20 Å. However, if a tomogram contains several HIV and Ebola virus. In addition, the structures of these pro- copies of a protein complex, the corresponding teins in complex with neutralizing antibodies/corresponding subvolumes can be extracted and averaged. This process, receptors were determined (Fig. 4a) [44–46]. called subtomogram averaging, increases the signal-to- Filamentous actin (F-actin) and microtubules (MTs) are noise ratio and the resolution of the reconstruction. For major components of the cytoskeleton and play essential cel- symmetrical particles even near-atomic resolution can be lular roles in cell division and muscular contraction. They reached using this technique [33, 34]. interact with a plethora of proteins and are of central impor- The combination of FIB milling and cryo-ET is just one tance for the homeostasis of the cell. Consequently, MT- or F- example that highlights the integrative nature of cryo-EM. actin-related disturbances are connected to many diseases, in Another demonstrative example is correlative light and elec- particular cancer, neurodegeneration, and myopathies tron microscopy (CLEM), in which cryo-EM is combined [47–49]. Since MTs and F-actin complexes, as well as other with fluorescence microscopy, allowing the identification of filamentous protein complexes, are notoriously difficult to complexes by means of fluorescence labeling in combination analyze in their respective filamentous state using crystallog- with EM-based subcellular localization studies . raphy, high-resolution structures and thus their understanding In summary, cryo-EM, with all its sub-disciplines and ap- at the atomic level has remained enigmatic until recently. plications, is a very powerful method and greatly extends the Profiting from the new detector technology, near-atomic res- toolbox of structural biologists to understand malfunctions in olution structures of F-actin and MTs have now been solved complex biological systems at molecular level. (Fig. 2b, 3b) [50, 51]. In addition, recent cryo-EM structures 488 J Mol Med (2018) 96:483–493 Fig. 3 Cryo-EM structures of filamentous proteins and their biomedical neurons, constituting the track for cargo-transporting motor proteins. relevance. (A) Tobacco mosaic virus is one of the most widespread [EMD-8322]. (C) Tau filaments are neurodegenerative deposits that are viruses around the world, being a prime example for plant pathogens that found in the brains of AD patients. [EMD-3741]. (D) The interaction of can have far-reaching consequences for the economy as well as food F-actin and myosin filaments is responsible for muscle contraction. supply. [EMD-2842]. (B) Microtubules are not only core components Malfunctions can cause myopathies. [EMD-8165] of the cytoskeleton, but are also essential in axonal transport along vastly contributed to the understanding of MT dynamics in- senile plaques in AD, namely amyloid-β (1–42) fibrils, were cluding regulation by MT-associated proteins (MAPs) and also solved by cryo-EM . Based on these findings, novel revealed the interaction with small molecules that have the therapeutics might be tailored to reverse the formation of AD potential of becoming anti-cancer drugs [52, 53]. Our lab deposits. Relatedly, the highly glycosylated membrane- gained valuable structural insights into the interaction between embedded 130-kDa protease γ-secretase that is responsible F-actin and myosin (Fig. 3d) . Furthermore, we solved the for the formation of β-amyloid plaques, has also been studied structure of the non-canonical actin filaments of Plasmodium by cryo-EM . Two hotspots for disease-evoking mutations falciparum, the parasitic microorganism causing malaria, in in presenilin 1, the catalytic component of γ-secretase, have complex with the naturally occurring cyclic peptide been identified, bringing researchers a step closer to under- jasplakinolide, exemplifying the potential of cryo-EM for standing the dysregulation of this crucial protein complex. structure-based drug design (Fig. 2b) . We identified that Membrane proteins represent the target of ~ 50% of subtle but significant differences are responsible for the inher- market-approved small-molecule drugs. The determination ent instability of the filaments, which is an essential feature for of their structure is key for directed drug design, but their proper host cell invasion of the pathogen. crystallization is often challenging, hampering a fast progress Alzheimer’s disease (AD), the most common neurodegener- in drug discovery. In the past 5 years, many high-resolution ative disease, is characterized by large filamentous deposits in structures of membrane proteins have been determined by the brain . Single-particle cryo-EM studies allowed for the single-particle cryo-EM. Especially for large membrane pro- first time the ability to resolve the structure of aberrantly folded tein complexes, the technique proved to be superior to X-ray tau filaments isolated from AD patient-derived material to a crystallography. Indeed, one of the first near-atomic resolution resolution of 3.4 Å . The structure revealed how two iden- structures that were solved making use of the new DED tech- tical protofilaments adopt a cross-β/β-helix structure (Fig. 3c). nology was that of the transmembrane protein TRPV1, a The filamentous deposits constituting the main component of member of the transient-receptor-potential (TRP) family, J Mol Med (2018) 96:483–493 489 Fig. 4 Selected examples of biomedically relevant cryo-EM structures. 2751]. Individual subunits are depicted in various colors. The hetero- (A) Cryo-EM structure of the trimeric envelope glycoprotein of HIV. It trimeric HIV envelope glycoprotein in A can be divided into the gp120 was solved in complex with two neutralizing antibody Fab fragments. trimer (yellow, orange, and red) and gp41 trimer (pink, blue, and cyan). [EMD-3308]. (B) Cryo-EM structure of the anthrax protective antigen Bound Fab fragments (green) are indicated by black arrows. Heptameric pore from Bacillus anthracis. [EMD-6224]. (C) Cryo-EM structure of anthrax protective antigen pore (yellow, orange, light red, dark red, TcdA1 from Photorhabdus luminescens in its prepore state. [EMD- magenta, cyan, and green) in (B), pentameric TcdA1 (yellow, orange, 3645]. (D) Cryo-EM structure of the ryanodine receptor RyR1. Map at red, green and blue) in (C), and tetrameric RyR1 (yellow, orange, red, low threshold (transparent) is shown to visualize the nanodisc (blue and green) in D arrow), which stabilizes the transmembrane helices of RyR1. [EMD- which mediates a wide range of sensational input such as pain, newly developed, such as amphipols, SMALPs, or the natural 2+ taste, pressure, and warmth . The Ca channel was also scaffold-based saposins, all with the objective to stabilize the resolved in complex with the spider peptide toxin DkTx and a respective membrane protein, especially the transmembrane small vanilloid agonist . The small molecules were clearly region [72–74]. resolved in the structures, demonstrating that cryo-EM is in- In regard to pharmacological application of cryo-EM, it deed able to visualize ligands . should be mentioned that it was recently possible to solve The list of all relevant membrane proteins that have recent- the cryo-EM structure of a B-class G-protein-coupled receptor ly been solved by cryo-EM is long, and even a very extended (GPCR), belonging to the family of the most abundant cell review would not suffice to name and describe them in detail. surface receptors and implicated in chronic diseases such as Many structures were determined for a large number of med- diabetes and obesity. The development and application of the ically important channels and transporters that are closely Volta phase plate (see above)  was key to solve the struc- linked to various diseases [63–65]. In addition, extraordinarily ture of these relatively small receptors that were recalcitrant to large membrane protein complexes, in particular those ones crystallization so far. involved in respiration and photosynthesis, were structurally Most multiprotein complexes are highly dynamic in order characterized by cryo-EM [66, 67]. A good example of both a to execute their respective functions. We have already men- channel and a large membrane protein complex is the tioned above that cryo-EM is able detect and separate different ryanodine receptor, a key mediator of calcium release from conformational states during image processing of a single the sarcoplasmic reticulum, initiating muscle contraction. dataset. A good example for this is the ATP-synthase. Three groups in parallel determined the cryo-EM structure Different conformational states were identified in one dataset 2+ of RyR1 giving novel insights into the 2.2-MDa Ca channel and separately processed. The resulting structures represented [68–70]. We reconstituted the channel into small disc-shaped different conformational states and allowed the correlation of membrane patches, known as lipid nanodiscs, to provide a mechanistic information with structural snapshots . close-to-native lipid environment (Fig. 4d) . A similar A long list of mostly asymmetric macromolecular ma- study with TRPV1 reconstituted into lipid nanodiscs revealed chines were shown to yield structures, the majority of them a density corresponding to annular lipids interacting with the reaching near-atomic resolution, including the ribosome , channel . In recent years, several other lipid mimetic sys- spliceosome , proteasome , dynein/dynactin [80, 81], tems based on amphipathic polymers were rediscovered or transcription (pre-) initiation complex , inflammasome 490 J Mol Med (2018) 96:483–493 , signalosome , and exosome , all of them indis- Bsnapshots^ could be ultimately used to reconstitute molecu- pensable for cellular performance. lar movies of biological processes. Bacterial toxins, which are self-containing agents, can The bottleneck in obtaining atomic structures is now have profound effects on human health. They are found shifting from former technical limitations towards sample pro- in a variety of human pathogenic microorganisms, such duction and preparation, requiring strong biochemistry. as, for example, the bacterium Bacillus anthracis.The Recent breakthroughs, in particular the ability to solve the structure of the anthrax protective antigen pore was re- structures of many membrane proteins, have opened up the solved by single-particle cryo-EM to a resolution of way for structure-based drug design. Consequently, pharma- 2.9 Å (Fig. 4b) . The cryo-EM structures of a num- ceutical companies are becoming increasingly interested in ber of pore-forming toxins that assemble into large pores cryo-EM and hopefully invest in its further development. All to disturb essential cellular gradients, usually featuring > in all, the field of cryo-EM is constantly evolving and has a 10 subunits, were also determined, giving important in- bright future ahead. sights into conformational changes during membrane in- Acknowledgements Open access funding provided by Max Planck sertion [87, 88]. Cryo-EM studies on Tc toxin complexes Society. We thank T. Gonen for kindly providing the diffraction image from Photorhabdus luminescens revealed a unique depicted in Fig. 1. syringe-like injection and translocation mechanism for membrane permeation and delivery of the toxic compo- Funding information This work was supported by the Max Planck nent into host cells (Fig. 4c) [89, 90]. The energy re- Society (to S.R.), the European Council under the European Union’s Seventh Framework Programme (FP7/ 2007–2013) (grant no. 615984) quired for this process is provided by the compaction (to S.R.). D.Q. is a fellow of Fonds der Chemischen Industrie. of an internal entropic spring . Furthermore, cryo- EM and cryo-ET led to unprecedented insights into the Open Access This article is distributed under the terms of the Creative architecture of large bacterial secretion systems and the Commons Attribution 4.0 International License (http:// creativecommons.org/licenses/by/4.0/), which permits unrestricted use, delivery mechanism of important effectors, causing hu- distribution, and reproduction in any medium, provided you give appro- man diseases [92–94]. priate credit to the original author(s) and the source, provide a link to the The large number of various medically relevant protein Creative Commons license, and indicate if changes were made. structures that have been solved by cryo-EM highlights the importance of this technique as a structural tool for biomedical research. Cryo-EM has the potential of becoming the major structural biology technique to study larger protein References complexes. 1. Cheng Y (2015) Single-particle Cryo-EM at crystallographic reso- lution. Cell 161:450–457 2. Dubochet J, Lepault J, Freeman R et al (1982) Electron-microscopy of frozen water and aqueous-solutions. J Microsc 128:219–237 Challenges and opportunities 3. Glaeser RM, Taylor KA (1978) Radiation damage relative to trans- mission electron microscopy of biological specimens at low tem- The recent revolution in cryo-EM has led to an increased perature: a review. J Microsc 112:127–138 4. De Rosier DJ, Klug A (1968) Reconstruction of three dimensional demand for equipment, with smaller labs often having diffi- structures from electron micrographs. Nature 217:130–134 culties to afford costly state-of-the-art instrumentation. To 5. De Carlo S, Harris JR (2011) Negative staining and cryo-negative counteract this trend, in the last 2–3 years, cryo-EM facilities staining of macromolecules and viruses for TEM. Micron 42:117– at many universities and research institutions have been 131 6. Wilson NR, Pandey PA, Beanland R, Young RJ, Kinloch IA, Gong founded. In addition, central facilities have been established L, Liu Z, Suenaga K, Rourke JP, York SJ, Sloan J (2009) Graphene to also allow access and professional support for structural oxide: structural analysis and application as a highly transparent biologists without high-end EM equipment. The required time support for electron microscopy. ACS Nano 3:2547–2556 from data collection to the final structure has been tremendous- 7. Russo CJ, Passmore LA (2014) Electron microscopy: ultrastable gold substrates for electron cryomicroscopy. Science 346:1377– ly reduced over the past years, facilitated by abovementioned improvements on both hardware and software. Although the 8. Cheung M, Kajimura N, Makino F, Ashihara M, Miyata T, Kato T, resulting throughput is not yet comparable with that in X-ray Namba K, Blocker AJ (2013) A method to achieve homogeneous crystallography, a cryo-EM structure can be obtained in less dispersion of large transmembrane complexes within the holes of carbon films for electron cryomicroscopy. J Struct Biol 182:51–56 than a week. 9. Pantelic RS, Meyer JC, Kaiser U, Baumeister W, Plitzko JM (2010) One direction of research that the field is slowly but steadi- Graphene oxide: a substrate for optimizing preparations of frozen- ly taking is time-resolved cryo-EM. The aim is to capture hydrated samples. J Struct Biol 170:152–156 short-lived states within non-equilibrium systems to monitor 10. Merk A, Bartesaghi A, Banerjee S, Falconieri V, Rao P, Davis MI, Pragani R, Boxer MB, Earl LA, Milne JLS, Subramaniam S (2016) conformational changes over time . These time-resolved J Mol Med (2018) 96:483–493 491 Breaking cryo-EM resolution barriers to facilitate drug discovery. 29. Raunser S, Walz T (2009) Electron crystallography as a technique to study the structure on membrane proteins in a lipidic environ- Cell 165:1698–1707 ment. Annu Rev Biophys 38:89–105 11. McMullan G, Faruqi AR, Clare D, Henderson R (2014) Comparison of optimal performance at 300keVof three direct elec- 30. Shi D, Nannenga BL, Iadanza MG, Gonen T (2013) Three- tron detectors for use in low dose electron microscopy. dimensional electron crystallography of protein microcrystals. Ultramicroscopy 147:156–163 eLife Sci 2:e01345 12. Li X, Mooney P, Zheng S, Booth CR, Braunfeld MB, Gubbens S, 31. Rodriguez JA, Ivanova MI, Sawaya MR, Cascio D, Reyes FE, Shi Agard DA, Cheng Y (2013) Electron counting and beam-induced D, Sangwan S, Guenther EL, Johnson LM, Zhang M, Jiang L, motion correction enable near-atomic-resolution single-particle Arbing MA, Nannenga BL, Hattne J, Whitelegge J, Brewster AS, cryo-EM. Nat Methods 10:584–590 Messerschmidt M, Boutet S, Sauter NK, Gonen T, Eisenberg DS (2015) Structure of the toxic core of α-synuclein from invisible 13. Bartesaghi A, Merk A, Banerjee S, Matthies D, Wu X, Milne JLS, crystals. Nature 525:486–490 Subramaniam S (2015) 2.2 Å resolution cryo-EM structure of β- galactosidase in complex with a cell-permeant inhibitor. Science 32. Mahamid J, Schampers R, Persoon H, Hyman AA, Baumeister W, 348:1147–1151 Plitzko JM (2015) A focused ion beam milling and lift-out approach for site-specific preparation of frozen-hydrated lamellas from mul- 14. Khoshouei M, Radjainia M, Baumeister W, Danev R (2017) Cryo- ticellular organisms. J Struct Biol 192:262–269 EM structure of haemoglobin at 3.2 Å determined with the Volta phase plate. Nat Commun 8:16099. https://doi.org/10.1038/ 33. Schur FKM, Obr M, Hagen WJH, Wan W, Jakobi AJ, Kirkpatrick ncomms16099 JM, Sachse C, Kräusslich HG, Briggs JAG (2016) An atomic mod- el of HIV-1 capsid-SP1 reveals structures regulating assembly and 15. Danev R, Buijsse B, Khoshouei M, Plitzko JM, Baumeister W maturation. Science 353:506–508 (2014) Volta potential phase plate for in-focus phase contrast trans- mission electron microscopy. Proc Natl Acad Sci U S A 111: 34. Turoňová B, Schur FKM, Wan W, Briggs JAG (2017) Efficient 3D- 15635–15640 CTF correction for cryo-electron tomography using NovaCTF im- proves subtomogram averaging resolution to 3.4Å. J Struct Biol 16. Moriya T, Saur M, Stabrin M, Merino F, Voicu H, Huang Z, 199:187–195 Penczek PA, Raunser S, Gatsogiannis C (2017) High-resolution single particle analysis from electron cryo-microscopy images 35. de Boer P, Hoogenboom JP, Giepmans BNG (2015) Correlated using SPHIRE. J Vis Exp e55448. https://doi.org/10.3791/55448 light and electron microscopy: ultrastructure lights up! Nat Methods 12:503–513 17. Tang G, Peng L, Baldwin PR, Mann DS, Jiang W, Rees I, Ludtke SJ (2007) EMAN2: an extensible image processing suite for electron 36. Böttcher B, Wynne SA, Crowther RA (1997) Determination of the microscopy. J Struct Biol 157:38–46 fold of the core protein of hepatitis B virus by electron cryomicroscopy. Nature 386:88–91 18. Scheres SHW (2012) A Bayesian view on cryo-EM structure de- termination. J Mol Biol 415:406–418 37. Yu X, Jin L, Zhou ZH (2008) 3.88 A structure of cytoplasmic poly- hedrosis virus by cryo-electron microscopy. Nature 453:415–419 19. Punjani A, Rubinstein JL, Fleet DJ, Brubaker MA (2017) cryoSPARC: algorithms for rapid unsupervised cryo-EM structure 38. Sachse C, Chen JZ, Coureux P-D, Stroupe ME, Fändrich M, determination. Nat Methods 14:290–296 Grigorieff N (2007) High-resolution electron microscopy of helical specimens: a fresh look at tobacco mosaic virus. J Mol Biol 371: 20. Wang RY-R, Song Y, Barad BA, Cheng Y, Fraser JS, DiMaio F 812–835 (2016) Automated structure refinement of macromolecular assem- blies from cryo-EM maps using Rosetta. eLife Sci 5:352 39. Kostyuchenko VA, Lim E, Zhang S, Fibriansah G (2016) Structure of the thermally stable Zika virus. Nature 533:425–428 21. Stark H, Chari A (2016) Sample preparation of biological macro- molecular assemblies for the determination of high-resolution struc- 40. Cao-Lormeau VM, Blake A, Mons S, Lastère S, Roche C, tures by cryo-electron microscopy. Microscopy (Oxford) 65:23–34 Vanhomwegen J, Dub T, Baudouin L, Teissier A, Larre P, Vial AL, Decam C, Choumet V, Halstead SK, Willison HJ, Musset L, 22. Arnold SA, Müller SA, Schmidli C, Syntychaki A, Rima L, Chami Manuguerra JC, Despres P, Fournier E, Mallet HP, Musso D, M, Stahlberg H, Goldie KN, Braun T (2018) Miniaturizing EM Fontanet A, Neil J, Ghawché F (2016) Guillain-Barré Syndrome sample preparation: opportunities, challenges and Bvisual outbreak associated with Zika virus infection in French Polynesia: a proteomics^. Proteomics 18:1700176 case-control study. Lancet 387:1531–1539 23. Li Y, Luo L, Rasool N, Kang CY (1993) Glycosylation is necessary 41. Fromm SA, Bharat TAM, Jakobi AJ, Hagen WJH, Sachse C (2015) for the correct folding of human immunodeficiency virus gp120 in Seeing tobacco mosaic virus through direct electron detectors. J CD4 binding. J Virol 67:584–588 Struct Biol 189:87–97 24. Mahamid J, Pfeffer S, Schaffer M, Villa E, Danev R, Kuhn Cuellar 42. Sougrat R, Bartesaghi A, Lifson JD, Bennett AE, Bess JW, L, Forster F, Hyman AA, Plitzko JM, Baumeister W (2016) Zabransky DJ, Subramaniam S (2007) Electron tomography of Visualizing the molecular sociology at the HeLa cell nuclear pe- the contact between T cells and SIV/HIV-1: implications for viral riphery. Science 351:969–972 entry. PLoS Pathog 3:e63 25. Kosinski J, Mosalaganti S, Appen v A et al (2016) Molecular ar- 43. Fu C-Y, Johnson JE (2011) Viral life cycles captured in three- chitecture of the inner ring scaffold of the human nuclear pore dimensions with electron microscopy tomography. Curr Opin complex. Science 352:363–365 Virol 1:125–133 26. Henderson R, Baldwin JM, Ceska TA, Zemlin F, Beckmann E, 44. Gong X, Qian H, Zhou X, Wu J, Wan T, Cao P, Huang W, Zhao X, Downing KH (1990) Model for the structure of bacteriorhodopsin Wang X, Wang P, Shi Y, Gao GF, Zhou Q, Yan N (2016) Structural based on high-resolution electron cryo-microscopy. J Mol Biol 213: insights into the Niemann-Pick C1 (NPC1)-mediated cholesterol 899–929 transfer and Ebola infection. Cell 165:1467–1478 27. Kühlbrandt W, WANG DN (1991) 3-dimensional structure of plant 45. Misasi J, Gilman MSA, Kanekiyo M, Gui M, Cagigi A, Mulangu S, light-harvesting complex determined by electron crystallography. Corti D, Ledgerwood JE, Lanzavecchia A, Cunningham J, Muyembe- Nature 350:130–134 Tamfun JJ, Baxa U, Graham BS, Xiang Y, Sullivan NJ, McLellan JS 28. Gonen T, Cheng Y, Sliz P, Hiroaki Y, Fujiyoshi Y, Harrison SC, (2016) Structural and molecular basis for Ebola virus neutralization by Walz T (2005) Lipid-protein interactions in double-layered two- protective human antibodies. Science 351:1343–1346 dimensional AQP0 crystals. Nature 438:633–638 492 J Mol Med (2018) 96:483–493 46. Lee JH, Ozorowski G, Ward AB (2016) Cryo-EM structure of a 67. Guo R, Zong S, Wu M et al (2017) Architecture of human mito- native, fully glycosylated, cleaved HIV-1 envelope trimer. Science chondrial respiratory megacomplex I2III2IV2. Cell 170:1247– 351:1043–1048 1257.e12 47. Parker AL, Kavallaris M, McCarroll JA (2014) Microtubules and 68. Yan Z, Bai X, Yan C, Wu J, Li Z, Xie T, Peng W, Yin CC, Li X, their role in cellular stress in cancer. Front Oncol 4:153 Scheres SHW, Shi Y, Yan N (2015) Structure of the rabbit ryanodine receptor RyR1 at near-atomic resolution. Nature 517:50–55 48. Dubey J, Ratnakaran N, Koushika SP (2015) Neurodegeneration and microtubule dynamics: death by a thousand cuts. Front Cell 69. Efremov RG, Leitner A, Aebersold R, Raunser S (2015) Neurosci 9:343 Architecture and conformational switch mechanism of the ryanodine receptor. Nature 517:39–43 49. Costa CF, Rommelaere H, Waterschoot D et al (2004) Myopathy 70. Zalk R, Clarke OB, Georges d A et al (2015) Structure of a mam- mutations in alpha-skeletal-muscle actin cause a range of molecular malian ryanodine receptor. Nature 517:44–49 defects. J Cell Sci 117:3367–3377 71. Gao Y, Cao E, Julius D, Cheng Y (2016) TRPV1 structures in 50. Alushin GM, Lander GC, Kellogg EH, Zhang R, Baker D, Nogales nanodiscs reveal mechanisms of ligand and lipid action. Nature E (2014) High-resolution microtubule structures reveal the structur- 534:347–351 al transitions in αβ-tubulin upon GTP hydrolysis. Cell 157:1117– 72. Tribet C, Audebert R, Popot JL (1996) Amphipols: polymers that keep membrane proteins soluble in aqueous solutions. PNAS 93: 51. von der Ecken J, Müller M, Lehman W et al (2015) Structure of the 15047–15050 F–actin–tropomyosin complex. Nature 519:114–117 73. Frauenfeld J, Löving R, Armache J-P, Sonnen AFP, Guettou F, 52. Zhang R, Alushin GM, Brown A, Nogales E (2015) Mechanistic Moberg P, Zhu L, Jegerschöld C, Flayhan A, Briggs JAG, Garoff origin of microtubule dynamic instability and its modulation by EB H, Löw C, Cheng Y, Nordlund P (2016) A saposin-lipoprotein proteins. Cell 162:849–859 nanoparticle system for membrane proteins. Nat Methods 13:345– 53. Kellogg EH, Hejab NMA, Howes S, Northcote P, Miller JH, Díaz JF, Downing KH, Nogales E (2017) Insights into the distinct mech- 74. Postis V, Rawson S, Mitchell JK, Lee SC, Parslow RA, Dafforn TR, anisms of action of taxane and non-taxane microtubule stabilizers Baldwin SA, Muench SP (2015) The use of SMALPs as a novel from cryo-EM structures. J Mol Biol 429:633–646 membrane protein scaffold for structure study by negative stain 54. EckenvdJ, HeisslerSM, Pathan-ChhatbarSet al (2016) Cryo-EM electron microscopy. Biochim Biophys Acta 1848:496–501 structure of a human cytoplasmic actomyosin complex at near- 75. Liang Y-L, Khoshouei M, Radjainia M, Zhang Y, Glukhova A, atomic resolution. Nature 534:724–728 Tarrasch J, Thal DM, Furness SGB, Christopoulos G, Coudrat T, 55. Pospich S, Kumpula E-P, Ecken v d J et al (2017) Near-atomic Danev R, Baumeister W, Miller LJ, Christopoulos A, Kobilka structure of jasplakinolide-stabilized malaria parasite F-actin re- BK, Wootten D, Skiniotis G, Sexton PM (2017) Phase-plate veals the structural basis of filament instability. Proc Natl Acad cryo-EM structure of a class B GPCR–G-protein complex. Sci U S A 114:10636–10641 Nature 546:118–123 56. Pospich S, Raunser S (2017) The molecular basis of Alzheimer’s 76. Zhao J, Benlekbir S, Rubinstein JL (2015) Electron plaques. Science 358:45–46 cryomicroscopy observation of rotational states in a eukaryotic 57. Fitzpatrick AWP, Falcon B, He S et al (2017) Cryo-EM structures of V-ATPase. Nature 521:241–245 tau filaments from Alzheimer’s disease. Nature 547:185–190 77. Khatter H, Myasnikov AG, Natchiar SK, Klaholz BP (2015) 58. Gremer L, Schölzel D, Schenk C, Reinartz E, Labahn J, Ravelli Structure of the human 80S ribosome. Nature 520:640–U338 RBG, Tusche M, Lopez-Iglesias C, Hoyer W, Heise H, Willbold 78. Bertram K, Agafonov DE, Liu W-T, Dybkov O, Will CL, Hartmuth D, Schröder GF (2017) Fibril structure of amyloid-β(1-42) by cryo- K, Urlaub H, Kastner B, Stark H, Lührmann R (2017) Cryo-EM electron microscopy. Science 358:116–119 structure of a human spliceosome activated for step 2 of splicing. 59. Bai X-C, Yan C, Yang G, Lu P, Ma D, Sun L, Zhou R, Scheres Nature 542:318–323 SHW, Shi Y (2015) An atomic structure of human γ-secretase. 79. Schweitzer A, Aufderheide A, Rudack T, Beck F, Pfeifer G, Plitzko Nature 525:212–217 JM, Sakata E, Schulten K, Förster F, Baumeister W (2016) 60. Liao M, Cao E, Julius D, Cheng Y (2013) Structure of the TRPV1 Structure of the human 26S proteasome at a resolution of 3.9 ang- ion channel determined by electron cryo-microscopy. Nature 504: strom. PNAS 113:7816–7821 107–112 80. Urnavicius L, Zhang K, Diamant AG, Motz C, Schlager MA, Yu 61. Cao E, Liao M, Cheng Y, Julius D (2013) TRPV1 structures in M, Patel NA, Robinson CV, Carter AP (2015) The structure of the distinct conformations reveal activation mechanisms. Nature 504: dynactin complex and its interaction with dynein. Science 347: 113–118 1441–1446 62. Merino F, Raunser S (2017) Electron cryo-microscopy as a tool for 81. Zhang K, Foster HE, Rondelet A, Lacey SE, Bahi-Buisson N, Bird structure-based drug development. Angew Chem Int Ed Engl 56: AW, Carter AP (2017) Cryo-EM reveals how human cytoplasmic 2846–2860 dynein is auto-inhibited and activated. Cell 169:1303–1314.e18 63. Wu J, Yan Z, Li Z et al (2015) Structure of the voltage-gated calci- 82. Louder RK, He Y, López-Blanco JR, Fang J, Chacón P, Nogales E um channel Cav1.1 complex. Science 350:aad2395 (2016) Structure of promoter-bound TFIID and model of human 64. Matthies D, Dalmas O, Borgnia MJ, Dominik PK, Merk A, Rao P, pre-initiation complex assembly. Nature 531:604–609 Reddy BG, Islam S, Bartesaghi A, Perozo E, Subramaniam S 83. Zhang L, Chen S, Ruan J, Wu J, Tong AB, Yin Q, Li Y, David L, Lu (2016) Cryo-EM structures of the magnesium channel CorA reveal A, Wang WL, Marks C, Ouyang Q, Zhang X, Mao Y, Wu H (2015) symmetry break upon gating. Cell 164:747–756 Cryo-EM structure of the activated NAIP2-NLRC4 inflammasome 65. Jin P, Bulkley D, Guo Y, Zhang W, Guo Z, Huynh W, Wu S, reveals nucleated polymerization. Science 350:404–409 Meltzer S, Cheng T, Jan LY, Jan YN, Cheng Y (2017) Electron 84. Cavadini S, Fischer ES, Bunker RD, Potenza A, Lingaraju GM, cryo-microscopy structure of the mechanotransduction channel Goldie KN, Mohamed WI, Faty M, Petzold G, Beckwith REJ, NOMPC. Nature 547:118–122 Tichkule RB, Hassiepen U, Abdulrahman W, Pantelic RS, 66. Wei X, Su X, Cao P, Liu X, Chang W, Li M, Zhang X, Liu Z (2016) Matsumoto S, Sugasawa K, Stahlberg H, Thomä NH (2016) Structure of spinach photosystem II-LHCII supercomplex at 3.2 Å Cullin-RING ubiquitin E3 ligase regulation by the COP9 resolution. Nature 534:69–74 signalosome. Nature 531:598–603 J Mol Med (2018) 96:483–493 493 85. Liu J-J, Niu C-Y, Wu Y, Tan D, Wang Y, Ye MD, Liu Y, Zhao W, 91. Meusch D, Gatsogiannis C, Efremov RG, Lang AE, Hofnagel O, Vetter IR, Aktories K, Raunser S (2014) Mechanism of Tc toxin Zhou K, Liu QS, Dai J, Yang X, Dong MQ, Huang N, Wang HW (2016) CryoEM structure of yeast cytoplasmic exosome complex. action revealed in molecular detail. Nature 508:61–65 Cell Res 26:822–837 92. Low HH, Gubellini F, Rivera-Calzada A, Braun N, Connery S, 86. Jiang J, Pentelute BL, Collier RJ, Zhou ZH (2015) Atomic structure Dujeancourt A, Lu F, Redzej A, Fronzes R, Orlova EV, Waksman of anthrax PA pore elucidates toxin translocation. Nature 521:545– G (2014) Structure of a type IV secretion system. Nature 508:550– 87. Law R, Lukoyanova N, Voskoboinik I (2010) The structural basis 93. Whitney JC, Quentin D, Sawai S et al (2015) An interbacterial for membrane binding and pore formation by lymphocyte perforin. NAD(P)+ glycohydrolase toxin requires elongation factor Tu for Nature 468:447–451 delivery to target cells. Cell 163:607–619 88. Bokori-Brown M, Martin TG, Naylor CE, Basak AK, Titball RW, 94. Worrall LJ, Hong C, Vuckovic M, Deng W, Bergeron JRC, Savva CG (2016) Cryo-EM structure of lysenin pore elucidates Majewski DD, Huang RK, Spreter T, Finlay BB, Yu Z, Strynadka membrane insertion by an aerolysin family protein. Nat Commun NCJ (2016) Near-atomic-resolution cryo-EM analysis of the 7:11293. https://doi.org/10.1038/ncomms11293 Salmonella T3S injectisome basal body. Nature 540:597–601 89. Gatsogiannis C, Lang AE, Meusch D, Pfaumann V, Hofnagel O, 95. Fu Z, Kaledhonkar S, Borg A, Sun M, Chen B, Grassucci RA, Benz R, Aktories K, Raunser S (2013) A syringe-like injection mech- Ehrenberg M, Frank J (2016) Key intermediates in ribosome anism in Photorhabdus luminescens toxins. Nature 495:520–523 recycling visualized by time-resolved cryoelectron microscopy. 90. Gatsogiannis C, Merino F, Prumbaum D, Roderer D, Leidreiter F, Structure 24:2092–2101 Meusch D, Raunser S (2016) Membrane insertion of a Tc toxin in near-atomic detail. Nat Struct Mol Biol 23:884–890
Journal of Molecular Medicine – Springer Journals
Published: May 5, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera