I’m a firm believer that a picture tells a thousand words, and in biomedical sciences advances in microscopy can only add to the story that every captured image can portray. As we work through the big reveal of our top ten papers, the biological resolution made possible by the advance detailed in this next paper honestly just astounds me in its physical capacity. The resulting overwhelming beauty of what we can image is awesome, and as a Brit who used to live in the US, I don’t use that word lightly.
The problem with microscopy, from Hooke and van Leeuwenhoek onwards, is that most living things are inconveniently transparent. That is, photons (for light microscopy) or electrons (for electron microscopy) tend to pass through tissues without breaking a stride, lending very little contrast to the picture. So many crucial advances in microscopy have involved developing ways of staining tissue samples to enhance their interaction with light or electrons.
But what if you want to see specific proteins, instead of cells or organelles? The ingenious exploitation of antibodies opened up the fields of immunohistochemistry, immunofluorescence and immunoelectron microscopy, but these all suffer from the problem that you’re staining the tissue after it’s dead. Back in the 90’s, Roger Tsien and colleagues developed a way of engineering genes so that their protein product was spliced to a jellyfish protein, green fluorescent protein (GFP). When looked at under a high powered microscope those specific proteins (and nothing else) would glow green. GFP has since been used in tens of thousands of studies, and earned Tsien a share in the 2008 Nobel Prize for chemistry.
In 2011 Tsien published in PLOS Biology an analogous tagging system for protein visualization by way of electron microscopy (EM). These tags are called miniSOG (mini-Singlet Oxygen Generator – I won’t go into much detail, read the paper for more info), and were obtained by re-purposing and mutating a plant protein called phototropin. The miniSOG tag is half the size of GFP, and not only fluoresces (allowing it to be seen in fluorescent light microscopy), but if you zap it with blue light, it generates a highly reactive chemical called singlet oxygen. This can be used to make an insoluble deposit that can be stained and then seen by EM. As noted in their abstract, the authors had high hopes for their new method, claiming that: “MiniSOG may do for EM what Green Fluorescent Protein did for fluorescence microscopy”.
The power of this method from Roger Tsien and colleagues is best exemplified with some images from their paper, just to try and capture your imagination. For an idea of relative size, the scale bar in the top panel shows 50 μm (micrometres), the width of an average human hair. This is a low-resolution fluorescent image of miniSOG that is targeted to the mitochondria of a nematode worm. You could easily have done this with good ol’-fashioned GFP, but the strength of miniSOGs is that you can switch to EM and zoom in much further. The scale bar in the middle panel shows 500 nanometres (nm; 1000 nm = 1μm), so we’ve zoomed in a hundred fold. Now we can see individual mitochondria, and that human hair would be wider than the computer screen you’re using. Yet other images in the paper (like the bottom panel here) use miniSOG to reveal the ultrastructural locations of specific individual labeled cell adhesion molecules – yes, individual proteins – in synapses of an intact mouse brain. As one of our editorial board members, Franck Polleux, puts it this is “a beautiful illustration of the power of combining chemistry, biology and electron microscopy. This new technique dramatically improves the ability of biologists to locate specific proteins inside cells and organelles at nanoscales.”
Xiaokun Shu, first author of the study who now runs a lab at UCSF told me that the success of this project was “a fruit of thinking outside the box.” Shu notes that many people were trying to engineer GFP to be an efficient SOG, but his insights into GFP’s biochemistry led him to conclude it would be a dead-end. And so they hit upon flavin, a cellular chemical that binds to an Arabidopsis protein, phototropin. But the drawback was that unfortunately there was no activation. Shu put his background in studying protein structure and function to good use and engineered this protein to function as an efficient SOG. And the rest, as they say, is history.
Actually, that’s not quite true. Since its publication, this method has already been widely applied to very good use. In another recent PLOS Biology paper from Ben Nichols and colleagues, miniSOG was applied to determine the molecular composition and ultrastructure of the caveolar coat complex. The incredible resolution made possible using the method – and how it compares with super resolution light microscopy – was also discussed in a really nice primer written by Jacomine Krijnse Locker and Sandra L. Schmid.
The landmark EM work of another Nobel laureate, George Palade, can arguably be seen to be the start of cell biology as we know it, but a resurgence in EM may now be underway as our ability to fully harness the technique is brought into a much sharper focus by way of canny biochemical tricks. We can now visualize some of the smallest molecular components of the cell truly at home in their natural habitats, and the combination of live cell light microscopy imaging with such high-resolution EM surely will provide a much clearer picture of cells than we have ever known before.
Shu X, Lev-Ram V, Deerinck TJ, Qi Y, Ramko EB, Davidson MW, Jin Y, Ellisman MH, & Tsien RY (2011). A genetically encoded tag for correlated light and electron microscopy of intact cells, tissues, and organisms. PLoS biology, 9 (4) PMID: 21483721