Bringing the World's Best Biology to You
00:00:11.12 So in this tip, I'm going to show you how to measure the
00:00:15.26 point spread function of your microscope. And so to see
00:00:21.15 how your microscope diffracts the light, you first need an object
00:00:25.24 that is smaller than the resolution of your microscope. And so that
00:00:30.16 object is given by these beads that have a size of, in this case,
00:00:35.18 100 nanometers. So this is an example of green fluorescent
00:00:41.11 beads. These ones here are very popular, these TetraSpeck beads,
00:00:46.10 because they fluoresce in multiple colors. But today, we're going to use
00:00:52.04 these yellow-green fluorescent beads. You can see here that you
00:00:56.12 buy a bottle with quite a few milliliters, however to get a good
00:01:03.29 spread of -- a good density of beads on the coverslip. All we need
00:01:09.09 is a dilution of 100,000 fold out of this one. And that means that,
00:01:15.22 so basically, we'll take 1 microliter of beads and dilute that
00:01:24.28 100,000 fold. And you can use -- people use various different
00:01:30.12 strategies to make these samples. What I like to do is dilute it
00:01:36.07 with ethanol, since the ethanol evaporates very rapidly. And that
00:01:43.15 gives us a very nice easy to make sample. So first dilute
00:01:50.05 the beads 1,000 fold. And then go on and dilute it another 100 fold.
00:02:05.22 Now since the beads are fluorescent, they also photobleach.
00:02:27.02 And that's why I'll put them away now so that they are stored
00:02:32.09 in the dark and don't photobleach. So here's our sample of beads diluted
00:02:38.24 at 100,000 fold. Now, to prepare the coverslip, you could in principle put them
00:02:46.29 straight down on the coverslip itself. But these coverslips are
00:02:51.15 relatively dirty. And so, what I tend to do is wash them in ethanol
00:02:57.27 and then flame them. So I'll switch on the bunsen burner, and when you do
00:03:04.19 this, be careful because ethanol is highly flammable. And you don't
00:03:16.11 want to set the whole thing on fire. So this is a relatively
00:03:25.02 crude way of cleaning the coverslip, but since these beads
00:03:29.07 are highly fluorescent, it is actually good enough.
00:03:55.09 So at this point, we can pipet roughly 4 microliters on one of these
00:04:09.07 18x18 mm coverslips. And that will spread then very quickly
00:04:18.24 over the coverslip and fill the whole surface area. And then
00:04:28.23 to both speed up the drying of the ethanol, but also to bake
00:04:35.04 the polystyrene beads onto the coverslip itself, what I like to do
00:04:40.28 is to put the coverslips, bead side up, onto a heating block. And that
00:04:48.08 seems to like melt the bottom of these polystyrene beads onto the
00:04:53.23 glass. You don't have to do this, the beads will also stick without it.
00:04:57.29 But this makes them wiggle less, and under certain circumstances that can be a good
00:05:04.11 idea. So now, an alternative strategy is to pipet the 4 microliters
00:05:10.21 onto the coverslip and use this little centrifuge, in which you can spin
00:05:18.19 the coverslip. And the advantage of this is that the solution then
00:05:24.10 spreads very nicely and dries in one go. So I'll put a coverslip into the
00:05:30.02 centrifuge, pipet on the beads solution, and spin it.
00:05:43.18 And then, so the coverslip is now dry. But I'll also heat that up
00:05:52.11 for a few seconds. Then what remains to be done is mounting
00:05:59.08 the coverslip onto a slide. I like to wipe the slide clean with a
00:06:07.10 Kimwipe. Now there are different types of mounting media you can
00:06:12.04 use. One of the simplest is 100% glycerol. I know this because it's
00:06:18.10 highly viscous. The refractive index is somewhat elevated over
00:06:23.27 that of water, and because of its viscosity and lack of oxygen,
00:06:31.25 it also happens to be a pretty good anti-fade reagent,
00:06:37.17 in its own right. So we now take the coverslip, bead side up,
00:06:44.01 turn it upside down so that the beads are facing the glycerol,
00:06:48.20 put it down, the glycerol spreads out. One important thing is
00:07:00.17 labeling the slides, so that you later still know what it is.
00:07:03.26 So these are 100 nm green beads. And we're done preparing
00:07:19.07 our beads slide.
00:07:20.28 Okay, so we have here at the fluorescent microscope. We have our
00:07:25.05 bead slide. We are using a 63X 1.4 NA oil objective. This
00:07:34.04 particular objective actually has a phase ring in it. So, we'll give it a
00:07:39.22 tiny drop of oil, of the appropriate immersion oil. And then put the
00:07:48.20 slide on it, coverslip side down. So it's advantageous to look at
00:07:54.25 the oil and to judge roughly where you are, where the objective is
00:08:01.04 with respect to the coverslip. Because these beads are tiny,
00:08:05.05 diffraction limited, and they're also not super bright, so they're actually
00:08:10.15 hard to find. And so being close is going to help you dramatically.
00:08:16.14 So then, we can now observe them. Okay, and I see my
00:08:27.10 tiny little beads. And this literally looks like stars at night. I see the
00:08:33.19 big dipper there. What you might already be seeing, what becomes
00:08:40.21 a little bit more obvious when we zoom in onto this image,
00:08:46.12 and then again go live. You see that as I slowly go through focus,
00:08:55.18 so here we see a single bead. And you see that from one side,
00:09:05.01 there is a bit of a line in this direction, and when I go through
00:09:10.20 focus, there's a bit of a line in that direction. And that is
00:09:16.17 a sign of astigmatism. So there's a tiny bit of astigmatism in this
00:09:22.00 optical system here. Now to collect the whole point spread function,
00:09:30.03 we're going to use the software to do that for us. So we're starting
00:09:37.28 out now here in the middle, and we tell the software to take a
00:09:43.17 z-stack. We want that z-stack to go from 5 microns below the
00:09:48.16 point of focus to 5 microns on top of the point of focus, and
00:09:54.12 steps of 0.2 microns. So when I now acquire, you see the
00:09:59.23 z-motor slowly stepping up and taking an image at every next z.
00:10:04.21 You now see that bead image forming as we go.
00:10:12.24 We now have the whole data set here, and we can quickly
00:10:26.09 manually go through focus. What we can do is to get the
00:10:33.15 cross-section of this image. The ImageJ command, image, stacks,
00:10:39.15 orthogonal views. We now get two extra images that display
00:10:47.13 this sideways view. And at the moment, we don't see anything there.
00:10:53.06 But if we put the cursors now near a spot, and then go through
00:11:05.25 focus here, we now see the side view of our point spread function.
00:11:12.02 And so here, you see that typical shape with the airy discs, so the same
00:11:22.25 you can see here on this side, where we see the airy discs in
00:11:28.12 cross-section. Now there's also auto scaling going on. And that makes
00:11:34.04 this all a little bit harder to see. And I'll now take off that auto scaling,
00:11:40.22 so you now can see that most of the light is actually contained in this
00:11:47.01 small central area here. We have the cross-section of one of these
00:11:53.11 z-slices. So z goes here, and this is x. And you now do see that there's
00:11:59.21 a bit of an asymmetry in this. So there's some energy going off
00:12:03.12 in this direction. But you also do see that this diffraction pattern
00:12:08.11 is you know, pretty nice. So here we are at a Nikon inverted
00:12:17.11 fluorescent microscope. This has a turret worked in here, and so
00:12:21.13 there are dichroic cubes sitting in here that we can bring in place.
00:12:27.28 We have an epi light source, and so we'll use it here to image
00:12:34.06 the beads. And then when I now open the shutter, we have blue light coming out.
00:12:40.03 I send it to the eyepiece. We can cover up the sample with like a
00:12:52.00 black lid or something dark, so the overhead light doesn't make
00:12:56.09 it into the camera. And then that way we can have some light on
00:13:00.04 while filming. I'm now manually going through focus, we have
00:13:06.22 the software on auto stretching. And you can now live see these
00:13:13.13 rings. And what is also highly apparent here is that when I go in
00:13:19.29 z in one direction, we have no rings, whereas if I go in the other
00:13:24.13 direction, we see beautiful rings forming there. And this is a
00:13:30.00 hallmark of spherical aberration. So now going in one direction,
00:13:34.15 and we see beautiful airy discs, although a little bit asymmetric.
00:13:38.09 Then when I go the other direction, I don't see them at all.
00:13:44.27 It's just fuzziness. So now, we will do the same as on that other
00:13:53.08 system. So we're going to do a z-stack, and again we'll go from -5
00:14:00.02 to +5, with 0.2 micron steps. Let's go. There we go. So we see
00:14:13.04 the point spread function as we speak. So we're going from the
00:14:18.07 glass to the beads and then we go through. Okay, now with
00:14:23.08 this set we can look at the orthogonal view. So what you
00:14:29.22 now can see very, very clearly, for instance, let's say in this side view
00:14:36.13 here, is how asymmetric the point spread function is. So it's nicely
00:14:43.17 defined here at this end, but it's kind of a blur here.
00:14:47.18 And that's really a hallmark of spherical aberration.
Nico Stuurman shows us the technique of measuring a point spread function – how to prepare a bead slide and image the beads in a microscope to obtain its point spread function.
Nico Stuurman is a Research Specialist at the University of California, San Francisco, in the lab of Ron Vale. Nico combines his expertise in computer programming and microscopy to advance many projects including the Open Source software, Micro-Manager. Continue Reading
iBiology and iBiology Courses are part of the Science Communication Lab (SCL). Our mission remains the same, to connect people to science. However, our focus has shifted to producing and evaluating cinematic films for education and the public, which you can find on the Science Communication Lab website. For more information, please see this blog post!
Leave a Reply