Breaking the Rules - Light Edition

Today's post is going to be short.  I've been thinking for a while about a conversation I once had with a physicist about some fascinating new technology (new to that time) in the field of microscopy.  You might assume that looking at something under a microscope is straightforward: you point the microscope at it and look.  And while that's basically true, it turns out those two ideas, point and look, can get really complicated depending on what you want to look at.

For example, let's say you want to tag something very small, so you can see it.  The DNA inside a cell is small, and there are ways to make it light up, but let's say instead you want to tag a specific sequence of DNA.  The problem with regular light microscopes is that all DNA looks essentially the same.  So one thing you can do is create a short sequence of single-stranded DNA that corresponds to the sequence you're interested in - and that happens to 'glow' when it becomes double-stranded.  Put a bunch of this your culture, and you can tell which cells have a certain gene based on whether they glow or not.  This has been around for a long time.

Let's get more ambitious.  It's nice to look at something in an artificial environment, but what about live-action microscopy?  Is it possible to tag certain cells and watch them circulate through a living animal?  For example, maybe you want to watch dendritic cells pick up antigen, migrate to the lymph node, and present that antigen to T-cells.  To do that, you'll need to be able to see through the outer layers of tissue.  But light will hit everything on the way in and on the way out.  It'll be a big mess, unless you can find a way to penetrate deep into the tissue.  All your resolution will disappear, no matter how tight the laser you use.  There's a way to do that, too.  You take two lasers, each without enough energy to cause the cell to 'light up' on their own.  But once the lines cross, because light is a wave, the two light waves add together to be enough to light up the cell you want to see.  The light doesn't get absorbed on the way in, but it does at the place the two lasers contact each other.  That's how you get deep-tissue imaging.

Now let's say you want to see something smaller, and you want to see the details.  The problem with light is the wavelength is larger than many of the structures we're trying to see.  Imagine I had a wood-block relief with an important message on it.  You want to read it, but you can't just look at the relief itself, you need something to press it into in order to get an image of the relief.  So you grab some Play-Doh and make a quick impression.  Done.

Now imagine I have another relief message for you to read, but this time the message is smaller than the grains of flour your Play-Doh is made of.  Suddenly, you can't use Play-Doh to read the message anymore, because the grains of flour can't get into the grooves to make an impression.  Normally this isn't a problem.  Most stuff is big enough for you to use regular Play-Doh, it's just when you want to look at something really small that it won't work.  Just as there's a lower limit to how small you can make impressions with Play-Doh, there's a lower limit to how small you can visualize something with light.  We call this the diffraction limit.

In the past, when you wanted to look at something below the diffraction limit of light you used an electron microscope.  That's because the wavelength of an electron is much smaller than that of light.  It's like using high-quality Play-Doh with extra-small grains of flour.  There's still a lower limit to how small you can get, but you can still see really small things you never could before.

There's a problem with electron microscopy, though: it's horrible for visualizing living things.  That's because with most electron microscopes you have to have a very high vacuum, or else the high-energy electrons you're shooting at your sample will just get absorbed on their way down and you won't see anything.  Under very high vacuum life ... does poorly.  (Actually, it's not just the vacuum problem, you also need something that conducts electricity well.  Normally when you prepare a sample on the electron microscope you have to coat it with a few nanometers of gold in order to see it, or you 'stain' it with something.  Either way, it has to be dead to get it to work.)  Even special electron microscopes (ESEM) still require some vacuum to make them work.  If you're looking for something really tiny in a living cell you can't see it with an electron microscope.

Back to my conversation with this physicist.  I told him about a new kind of light microscopy that happens below the diffraction limit of light.  Immediately he was skeptical, "That's not possible, even in theory.  The laws of physics don't allow for it."

I smiled and explained to him how it's done.  (For those interested in how this works, here's a taste of one way it's accomplished: using one laser for excitation and one for suppression.  You can 'tune' the spot size of your excitation laser around a donut-shaped suppression laser.)  I finished with, "... so there's now no theoretical limit to how small we can visualize something with light using this method.  There are still practical limitations, of course-"

"Wow, that would work."  He contemplated this for a while, grinning at the very idea of 'breaking' a hard-and-fast universal rule - the diffraction limit - without really breaking any universal laws at all.

Partly I'm sharing this because I love the myriad ways different types of microscopes allow for some amazing images to be captured.  But in the interest of philosophy of science, there are some other ideas I'd like to bring out:

  1. Often the built-in limitations given to us by the universe appear to be immutable - until they aren't.  For decades people will operate under the assumption that something can't be done because the laws of the universe deem it impossible.  Then one day someone comes along and says, "I think there's a way to go around this limitation" and suddenly it's just a matter of engineering, not of absolute limitations.
  2. You don't get to just 'ignore' those universal laws, though.  It's easy to see some end-run around a universal law and think, "it's just a matter of time until we get around all the other natural laws as well.  Nothing is truly impossible."  And that's not true either.  In the case of super-resolution microscopy we didn't skirt the diffraction limit of light; we used other aspects of the nature of light to work within that limit (in this case the wave nature of light and destructive interference).  When you look into the details of how it's done we didn't break any universal laws.  We just assumed something was impossible because of a universal law, but that assumption was wrong.
  3. The engineering trick that allows us to 'get around' the previous limitations comes with new limitations.  Yes, it's true there's no theoretical limit anymore to how small you can visualize something with light.  But there's a practical limit: the size of the fluorophore (the thing the laser hits that allows us to see it).  This isn't magic.  The engineering seems like magic when you first see it, because suddenly it can do things we couldn't before.  But there are still new limits.  Whether and how we work to expand those limits within natural law is an open question, but natural laws were never broken.  We just thought they prevented us from doing something, when all they prevented was us doing it in a certain way.


Popular posts from this blog

A better addiction

Covid-19: Epidemiology is useful

Open Questions: The Origin of Life