Sharper knives for sharper images, or, the ‘Deli slicer’ portion of your biology PhD

I’m doing a bunch of staining experiments at work right now. That means that, hopefully, in the next couple weeks I’ll get a bunch of pretty pictures to analyse. It also means that I had the good fortune to spend much of yesterday at the cryostat (basically a meat slicer designed to make incredibly thin, frozen, sections – or slices). Sectioning is a zen experience for me, much akin to knitting: fairly repetitive, relatively simple movements that need to be done precisely and with care. Although in the end, you get slides with tissue on them, rather than a scarf. (Side note: dissections, in this analogy, would be like drawing or painting: they require much more fine-motor control and focus.) It seemed like one of those times where the practicalities of being a scientist were very similar to the practicalities of being an artist (or at least a maker). And that made me think more about the whys, hows, and wherefores of the whole process.

Why do we do this, anyway? What’s the purpose of cutting up already tiny pieces of tissue into even tinier bits? It comes down to two big considerations: experimental design and optics.

Obviously, getting a cross-section of a sample rather than just looking at the outside is incredibly useful for seeing the shape of the structures within. That’s what cross-sections are for, after all. But in addition, biological samples tend to be relatively precious and have a good degree of variation between individuals. Taking five or ten or twenty or one hundred neighboring cross-sections gives you multiple technical replicates and allows you to do multiple experiments on the same sample: which is great as a way to conserve your resources and also as a way to control for many confounding effects of individual biology.

On top of that, the optics of microscopy basically mean that the thinner a slice you’re looking at, the crisper an image you’ll get. That’s because of how the microscope works with depth of field. Clearly, there are other labbers who know a lot more about how to control depth of field than I do. But a brief overview is as follows: depth of field refers to the range of space within which an object will appear reasonably focused. It’s controlled by two things: the size of the aperture and the focal distance. Bigger apertures and shorter focal distances mean smaller depth of field, and in microscopy we’re pretty much limited to having really short focal distances (simply by the geometry of the microscope) and really big apertures. This latter one is a little bit more complicated to explain – a small aperture means less light, and fluorescent samples are often so dim that if we were to use a small aperture, we wouldn’t get a crisp image – we would get no image at all.

So most fluorescent microscope images have very shallow depth of field. But everything above or below that field is still illuminated, and is still fluorescent, so it results in a diffuse blur. Not good for getting high-quality images.

We get around all of that by cutting thin slices of tissue: if the entire sample is only 10 microns thick, you can get away with having a depth of field only 10 microns thick. Note that there are other ways to do this as well: confocal microscopes only illuminate a tiny slice, creating an ‘optical slice’. These can only go so far, though, because while many tissues are translucent few are truly transparent. Generally confocal scopes are used in conjunction with sectioning, not instead of sectioning. Cell culture systems with adherent cells can be great: grow cells into a sheet one cell thick and image that. But if you aren’t working in cell culture, you’re pretty much stuck with sectioning.

How does it work?

You start by embedding your tissue in some relatively stiff matrix or another: essentially, a block of wax. (Many people use paraffin, in fact.) The idea is that as the knife shaves off a bit, the structure of the wax will prevent usually squishy tissue from being squished. It also provides something to stick to the slide later on. I use frozen sections, mostly because for the later steps it means fewer hazardous chemicals. The matrix is actually a mix of polyvinyl alcohol and polyethylene glycol with some other stuff, but when you freeze it, it acts like wax.

Next step is the knife. Microtome blades (used on paraffin sections, at room temperature) look like beefed up razor blades. But cryostat knives (used on frozen sections, in a climate-controlled cabinet) are big hunks of metal like the one on the left. My best guess as to why cryostats use such bulky knives is that it’s largely so that the heat created by the friction of sectioning won’t significantly heat up the blade. Since metals are generally good conductors, a big piece of metal heated in one spot and cooled elsewhere will maintain temperature. We keep the knife sharp, clean, and in the freezer. It needs to stay cold in order to not melt our sample: a hot knife goes through butter quickly, but it also turns that butter into a gooey mess. And gooey messes do not happy biological images make. Usually.

The machine is pretty similar to a deli slicer, when it all comes down to it. You move the sample against the blade to get slices. The biggest difference I can tell is that your hands are far away in this case – which is good for the fine sectioning that most biological applications require. A not-terribly-thin section on this thing is 10 microns – one one hundredth of a millimeter, about half the diameter of the finest human hair. You move the sample with a wheel on the right of the machine. Unfortunately, tiny variations in your movements – things like the speed or force with which you turn the wheel – and other factors – for example, the humidity or barometric pressure – can have big effects on whether your sections come out right. The cabinet that all of this sits in is a freezer itself, to keep the sample from melting, so there’s some control there. But never enough.

So, you make a few slices, and they come out as a little ribbon of wax-encrusted tissue. I think people who are better at sectioning than I am get these beautiful, totally square, ribbons. Mine almost always curl to the left. You can use a paintbrush to try to straighten them out, and that works sometimes.

This is probably my favorite part, because to get the sections off of the knife and onto a slide you just touch them with a warm slide: since the section is so thin, it will melt right onto the slide, and stick. And then you’re ready to actually do an experiment.