Microscopes have long helped scientists peer into the microscopic world to image tissues and cells and their organelles. In the 17^th century, Antonie van Leeuwenhoek significantly improved the simple microscope and discovered microbes and microscopic structures by cell imaging. He has since become known for his pioneering work in microscopy and contributions to microbiology, which is honored by the eponymous microbiology journal Antonie van Leeuwenhoek. The microscope has continued to evolve, with higher quality lens that capture clearer images, greater magnification, and diverse illumination sources, to better image tissues and cells. Technical advances, such as the super-resolution microscope, have pushed resolution past the diffraction limit of visible light, facilitating greater detail in tissue and cell images. Today, all biology laboratories have microscopes, from the basic school laboratory to advanced research facilities. Innovations have even made possible an inexpensive, foldable microscope capable of imaging cells.

Cell imaging is a mainstay of basic and biomedical research through several experimental techniques, such as colony, migration, invasion, and cell-cycle dynamics, and lineage assays. Immunocyto- and immunohistochemistry leverages antibody-labeling to image localization of specific target proteins in cells and tissues, respectively. Alternatively, cells can be imaged from the emission of overexpressed fluorescent proteins. The addition to the microscope of an imaging chamber to regulate temperature, humidity, and atmospheric composition enables time-lapse, live-cell imaging to study dynamic processes, such as vesicle formation, trafficking, and cell-cell interactions. 

The cell imaging field has come a long way and continues to evolve. Subcellular organelle and single-molecule tracking is emerging as one exciting avenue in cell imaging technologies. Also, like many fields experiencing massive increases in automated and large scale experimentation, multiplexed and high-throughput cell imaging is similarly making its debut.

Peering in more closely: Subcellular organelle and molecular structure

As microscopes have improved, so too has their resolution and their ability to image cellular substructure, even down to the single-molecule level. Researchers can now investigate the carefully orchestrated trafficking and interaction of organelles, cargo, and biomolecules that sustain cellular life. Examples span on the larger scale of cargo transport along neurons and mitochondrial-lipid droplet interactions, to the smaller scale intracellular visualization of protein dynamics and of nascent mRNA synthesis by single-molecule imaging.

“Advances in high resolution, single-molecule microscopy is offering us the opportunity to track and count single events in imaged cells. We can profile the distribution of organelles and cargo trafficking, not just the bulk whole ensemble. This allows us to discern any heterogeneity present in cellular trafficking and interaction events,” explained Professor

Kai Zhang, at the Department of Biochemistry, Neuroscience Program, and Center for Biophysics and Quantitative Biology at the University of Illinois Urbana-Champaign. Professor Zhang is a veteran of imaging, and started his scientific career in physical chemistry and photophysics, building microscopes. As his career progressed, he was lured to neuroscience, with a focus on neuronal axon trafficking. He has used live-cell imaging to study the transport of neurotrophic factors, such as nerve growth factor (NGF) and brain-derived neurotrophic factor (BDNF), as well as using super-resolution imaging to profile the axon tracks themselves, the microtubules.

“In neurons, one can think of the axon as a highway, and the cargo traveling from one end to the other as vehicles. It is essential that neurons coordinate cargo transport, so that essential molecules, such as growth factors, reach the sites where they are needed, e.g., growing neurites. Sustained, functioning cargo trafficking ensures neuron survival,” Professor Zhang elaborated. “Unfortunately, in disease states like neurodegenerative illnesses, the highway is damaged, which creates a roadblock. One of my earliest papers used live-cell imaging to look at

mitochondrial transport defects in neuronal axons. We expressed red fluorescent protein in mitochondria and were able to track them individually. We found that the Alzheimer’s disease (AD) peptide, amyloid-β, impaired anterograde mitochondrial transport (i.e., transport away from the neuron body), which was rescued by reducing expression of tau, another key AD protein. Intriguingly, in other types of neurological disorders such as the Charcot-Marie-Tooth type 2B neuropathy, axonal transport of the small GTPase, Rab7, could be accelerated, which could lead to premature degradation of the survival signal carried by NGF. These studies demonstrated how cell imaging could lend insight into disease pathology,” Professor Zhang continued of his work. Indeed, cell imaging has a multitude of applications for studying organelle and biomolecule dynamics in development, e.g., X chromosome inactivation, and disease, e.g., RNAs in cancer, mitochondrial trafficking in diabetes, and trafficking in misfolded protein or neurodegenerative diseases.

“Now that we know transport is defective in disease, can we rescue it? Can we switch transport on and off and study the effects?” Professor Zhang frames one of his research missions. “

Optogenetic control in cells is one of my lab’s main accomplishments. In this technique, we express in cells a light-responsive channel or protein, which we can activate by illumination. By controlling when and where the light is shone, we have temporal and spatial control in our experiments, and we image the cellular response.” Professor Zhang and his team have imaged the effects in cells of optogenetic regulation of receptor tyrosine kinase and RAF on cellular differentiation, modulation of protein activity, and organelle and molecular motor distribution. “We are now moving the technique in vivo,” Professor Zhang discussed his lab’s more recent projects. “By modulating kinases, we can influence vertebrate development of Xenopus laevis and image cells as they differentiate and move in situ in their microenvironments.” Moving the technique to mice is also on Professor Zhang’s radar, although there are technical barriers to imaging larger, non-transparent organisms in vivo. “It is possible to use a fiber optic, biodegradable implanted light-emitting device, or up-converting nanoparticles. We’re investigating these possibilities.”

Small cells, big high-throughput technologies

Automated, large scale methods that generate large datasets are a recurrent, recent trend in many areas of science. Cell imaging is no exception, and numerous high-throughput techniques have recently been developed. Automated processes, such as deep learning, apply computational image analysis and mining methods to imaged cells or tissues for cellular classification, increasing throughput. Whole slide imaging of tissue, but also of cells, can lead to clinical applications for automated histopathology for classifying tumors, such as lung cancers and their mutations, esophageal adenocarcinomas, or breast cancer lymph node metastases.

Another exciting high-throughput avenue is

imaging cell cytometry (IFC), which is an adaptation of flow cytometry that incorporates imaging. “IFC is a really powerful technique that combines ‘the best of both worlds’. It benefits from the single-cell imaging capabilities of microscopy along with the high-throughput abilities of conventional flow cytometry,” explained Professor Yu-Hwa Lo, at the Department of Electrical and Computer Engineering, University California San Diego. “Conventional flow cytometry analyzes cell populations for proteins of interest, which are detected by binding to fluorescently tagged antibodies or endogenous fluorescent protein expression. For each analyzed cell, flow cytometry only determines whether the proteins of interest are present or not. IFC extends capabilities by adding a detection modality for imaging, which provides an intracellular spatial distribution of the proteins of interest in analyzed cells.” This added spatial dimension has many useful applications. In research, IFC can be used for cellular morphological classification, cell-cell interactions, antigen presentation, extracellular vesicles characterization, and intracellular analysis of pathogens, among many others. IFC also has clinical diagnostic applications, such as for assessing acute leukemia, cellular damage from radiation (biodosimetry), and erythroid analysis (e.g., erythroid maturation, sickle cell, infectious diseases such as malaria).

Presently, imaging flow cytometers only produce 2-dimensional (2D) images of cells, although imaging in

3D is an emergent theme. “This is a current disadvantage of IFC because 2D images contain far less information than 3D images,” explained Professor Lo. “One analogy for 2D versus 3D cell images is the difference between an X-ray image and a CT (computed tomography) scan. An X-ray is the net 2D projection of all organs onto one image, so information is missing where organs overlap and may conceal important anatomical features. In contrast, the 3D CT scan contains more information by imaging a person in serial sections, resolving organs from one another and providing depth to the image.” Another challenge facing IFC is the limited ability for real-time sorting of imaged cells, in a manner analogous to fluorescence-activated cell sorting (FACS) in conventional flow cytometry. “IFC generates an extremely massive amount of data that are difficult to computationally process in real-time, which limits our ability to sort cells in real-time,” Professor Lo discussed of another IFC weakness. “However, sorting capabilities would generate so many potential research avenues. Sorting provides the unique ability for downstream molecular analysis of these isolated cells, connecting cellular phenotypic features to their molecular and genetic characteristics.”

Therefore, to further advance IFC, Professor Lo and his team are developing a 3D Image-Activated Cell Sorter. “We need to be able to both image cells in 3D and sort them for downstream analysis. The high information content produced by 3D cell imaging would improve cell classification and cell type discovery, better resolve protein localization, facilitate cell cycle studies, cell-cell interactions, cell marker discovery, and

cellular responses to drugs, toxins, and environmental stress, etc. Some of the applications can be greatly enhanced by deep learning and artificial intelligence. Moreover, the ability to then isolate cells based on their 3D image would be a major leap for the system, enabling downstream analysis.” Professor Lo is very excited about the 3D Image-Activated Cell Sorter he is developing, but is aware of some obstacles he needs to overcome. “First, for high-quality 3D images, we will need to enhance image resolution. Second, to deal with the large amount of generated data, we will need to improve data processing speed so that the instrument can image a cell and decide in real-time how to sort it. We are working very hard on both these aspects and hope to bring the 3D Image-Activated Cell Sorter to fruition.”

“It is a really fascinating multidisciplinary field, a collaboration of biophysicists, material engineers, and biologists,” Professor Zhang described of the cell imaging field. “When you bring all these areas of expertise together, you make major advances that unlock cells’ secrets.”