The cells have been mostly ok- I lost one dish to bacteria, and another contracted a case of mold, which I’m treating with fungizone.

In the meantime, while the cells are doing their thing, I’ve updated my fluorescence setup.

Fluorescence is used as an imaging technique in my lab- basically, I can ‘paint’ specific molecules a certain color, and then use the microscope optics to only image those molecules. As an example, DAPI (“dap-ee”) binds to the DNA in a cell and thus, by looking only at DAPI, I can clearly see the nucleus of cells. As other examples, I can paint acetylated alpha-tubulin (the protein that makes up the cilia ‘skeleton’), actin (the cytoskeleton), ZO-1 (the junctions between cells), SSTR3 (a molecule thought to hang out on cilia), etc. etc. I should post some images…. The point is, fluorescence imaging lets us image a specific molecule in the cell.

The technique is very standard, and so is widely used in biomedical research. One of the easy pieces of information that is discovered by this technique is where proteins are located in a cell- many proteins can move from the membrane to the nucleus, and this is evidence that the cell is “doing something”. It’s possible to watch proteins being made-and once made, get moved around to their final destination- Jennifer Lippincott-Schwartz has taken some amazing videos.

But, there are some limitations. For one, there’s a limited ‘color palate’. That is, if one color is too close to another color (say green-blue and blue-green), then there’s no way to distinguish the two colors in the images, meaning I can’t tell which molecule is which. It’s fairly easy to use three colors- red, green, blue- and not much more effort is needed to squeeze in a fourth- either deep blue or deep red- meaning that usually we can only track 3 or 4 molecules at a time. It would be nice to watch 10 or 20!

The companies that make the color filters (there aren’t that many) have been making the filters better and better over time, and I just got a new filter that allows me to image 5 different colors. That means I need only 1 filter to look at 5 distinct colors: deep blue, blue, green, red, and deep red, and I can look at them all *simultaneously*.

There’s another advantage to this filter (a quad Sedat filter), in that I removed 3 filters that were occupying space in the microscope. So, not only can I image more colors than I could before, but I have additional space (which I filled already- yellow, Fura-2, and a laser tweezer cube) that lets me perform additional functions.

I’ll have to post some images… I’ll sacrifice some cells tomorrow or Friday and stain them.

The cells have been growing all week without any contamination…. I’ve moved some cells into the differentiation incubator, so there should be ciliated cells in another week- if everything remains sterile.

So, in the meantime, I took a few images demonstrating how a certain type of imaging works: differential interference contrast imaging (DIC).

Usually, DIC is performed in transmission- most of the wiki pics are in transmission mode. However, it’s possible to perform reflection DIC as well on surfaces that are highly reflective. Without getting into too much detail about how it works (the wiki article has it all), in reflection mode DIC converts the slope of the surface into color. For example, a steep slope may be red, but a gradual slope may be blue, and as the color turns from blue -> green -> yellow -> red the slope gets steeper.

Razor blades are the *perfect* object to show this effect: this was taken at 16X

This is a clean single-edge blade used in boxcutters. What you are seeing is where the tapered part of the blade starts- the vertical stripes are from the grinding process. Realize this is bare shiny metal, and the color comes from the lens (and other optical components)- it’s not a rainbow-hued Photoshop job.

Making images like this is fairly straightforward, but there is some ‘art’ to getting really colorful images. Here’s another example, imaging the front lens of a 16X Epiplan HD objective.

‘HD’ means the lens is made for darkfield imaging and the lens is actually similar to a catadioptric lens. The front lens is very flat- perfect for DIC imaging.

Good lenses have a special thin coating on them to prevent unwanted reflections from occurring- plain glass reflects about 7% of the light, and if you have 3 or 4 lenses, about 50% of the light is lost to these reflections. These coatings have special optical properties and on most coated lenses, if you look at the lens the right way with your eye, you can see a deep purple color.

I exploited the optical properties of the coating using DIC imaging to make the coating change color:

For some technical reasons, I had to correct the overall brightness of each image to make them all appear similar, but the effect is clear.

So, the cells I plated earlier have attached and started to grow. They are growing slowly, but growing nonetheless. This makes 1 week without contamination. Fingers crossed!

In microscopy (and macrophotography), there is an inherent trade-off between resolution (the ability to see fine detail) and depth of field- how much of the object is in focus. Now it is possible to collect a whole ‘stack’ of images and process the stack so that only the in-focus portions are retained, leading to an image with incredible detail *and* a large depth of field.

Here are two images, processed with the free program “Combine ZP” one of a butterfly wing and the other of a metal shaving. The butterfly wing was acquired with a 16x epiplan in transmission mode,, the metal shaving with the 25mm Luminar and reflected light.

Notice, in the wing picture, there are some regions with no ‘best focus’- I moved the stage too much- and in the metal shaving image, how the two sides of the shaving look ery different- one is smooth, the other crinkled.

As of today, the fresh media I made (w/o antibiotics) is still clear, after sitting in the incubator. The existing culture is also clear. So at least some things are right in the world.

The freshly-thawed cells were ok, but it was only a few hours. Tomorrow I’ll know for sure.

So this was a minor catastrophe, not complete destruction.

I had to throw out all my cultures today- there was evidence of a bacterial infection.

I suppose I was due; my technique has been a little sloppy lately.

The cultures don’t have an immune system- they are defenseless against all the garbage in the air- bacteria, virii, mold, yeast, etc. etc. Because of this, cell culture is carried out under sterile conditions, using a special culture hood- not only is the air circulated through a HEPA filter, but there’s also a germicidal bulb that kills via ultraviolet light. We also use *lots* of ethanol, bleach, biocides, etc. etc.

Even then, I was getting contamination every month or so- the lab is in the Physics Department of an old dusty building- so I opted for one final line of defense: antibiotics, added to the culture media.

We use a combination of penicillin/streptomycin and gentamicin. The pen/strep protects against one type of bacteria, while gentamicin protects against the other main type. I also have some fungicide for use as needed.

Adding antibiotics chronically can have several bad effects- one, it masks an infection. Bacteria can live just fine within a cell- they are competing for the culture media, they don’t ‘eat’ the cells- and when they erupt from a dead cell, they get killed by the antibiotics. So a culture can be infected and appear perfectly normal. Which is what happened to me.

So today, I tossed all the cultures, cleaned the incubators and hood, and made some fresh media- without antibiotics. A sample of that media is now spending the night in the incubators to determine if they are contaminated (or if the media itself has contamination), and then I’ll thaw recently frozen cells, incubate them with the antibiotic-free media, and see if the frozen cultures are clean. If so, we can get back on track. If there’s a problem with the frozen cells, well….

I got off to a slow start this week, so here’s another ‘Power of 10’ image series, this time of a computer chip

I was hoping to capture a timelapse auto-fluorescent series of our cells undergoing division, but for various reasons it didn’t work out- he cells divide once every 8 hours, and after 5 hours under the microscope, nothing had happened. Which in experimental science, is the case more often than not.

So instead, here’s another magnification series: one of the back of a $2 bill, since there’s a few tiny faces to explore, and the second of the front of a new-ish US nickle. First, the paper money- zooming in until the paper fibers are more prominent than the drawing, and at maximum magnification, individual dots of ink can be discerned (but possibly not in the de-scaled images I posted here).

Personally, I think the guy all the way on the left of picture #3 is freaky-looking.

Now for the nickel- again, a similar set of images. So far, the only real change occurs when I switch from oblique to epi-illumination- notice the contrast greatly increases.

But wait- there’s more. One of the powerful techniques in microscopy is Differential Interference Contrast (DIC), which also goes by several aliases: ICT, Interphako, etc. In reflection imaging, DIC converts height into color. Here three images taken with this technique, using the 4X and 16X epiplan objective- notice the brilliant color present.

Also of note are the two images taken at 16X- what was done is to slide one of the optical components ever so slightly; this changes the ‘bias’ of the image, and is used to emphasize particular parts of the image.

The reason the colors at 16X are so much more vivid than 4X, is that the height variation of the object matches the depth of field of the 16X lens better. At lower magnifications (and lower numerical apertures), the height variation is barely detectable. At higher magnifications, the height variations are significantly greater than the depth of field, and so most of the image appears blurry (which is also why I didn’t include any).

That’s mostly true. Glow-in-the-dark toys use phosphorescence, which a little different than fluorescence. But, living cells can have a lot of auto-fluorescence: naturally occurring chemicals in the cells fluoresce, which can make many microscopy techniques difficult. Some of these chemicals include tryptophan and tyrosine (two amino acids that make up proteins), vitamin B6, NADH, and many others. In our cells, there are lots of mitochondria which fluoresce due to these chemical compounds.

Here is an image of living cells auto-fluorescing a bright green. This is the first image taken of a time-lapse, again to verify that the cells can survive long periods of time under the microscope.

The main difference between this test and the brightfield tests earlier is the method of illumination. When excited, the fluorescent compounds can be broken apart, which over time leads to cell damage. It’s similar to getting a sunburn. The excitation light was turned to it’s minimum setting, and the EMCCD camera captured images with 1-second integration time and high levels of on-chip multiplication. Through the eyepiece, it was impossible to see anything- that’s why we use an EMCCD for live cell imaging.

The cells survived the 5 hour time-lapse (although they were not happy) and recovered overnight. So, by being smarter about turning on the light (for example, only when taking an image and not during the 1-minute interval), the cells can be kept more comfortable.

The next step is Fura-2 imaging. We have had problems getting the dye into the cells, but hopefully with fresh chemicals we can successfully carry out calcium imaging.

Here are some more images of a fish vertebrae, showing the amazing structure present, and two images of some seedpod. All images taken with the 100mm luminar.