I would like to know a relatively simple thing, but something I wasn't finding the answer to. When fluorescent protein is used in a relatively large, non-transparent animal, like a mouse, or theoretically a dog or a human, how is the protein generally incorporated into the tissue? I suppose that it would bind to elastin and otherwise it could be transfected into different cells in and throughout tissue, both around the skin and deeper. However, wouldn't it only be useful near to the skin, since ultraviolet light and visible light would penetrate a limited distance through tissue? Then again, a relatively bright light source could possibly produce light from deeper in tissue. I'm not quite sure. It's a relatively simple question, but I appreciate it.

(Edit): I may have been a little vague with the question, a more specific version of the question I asked would be; given the brightness of a fluorescent light source that is directed at fluorescent protein filled tissue (with UV assumed to be close to the visible spectrum) and the portion of that light which is converted to visible light, approximately how could you calculate the relative brightness of the protein at a certain depth in the tissue. Of course it probably is very difficult to get accuracy with that because that would depend on the portion of fat, muscle, collagen, etc in the tissue. But if there's something rough and general, like the depth of the protein would reduce its relative brightness proportionally to the natural log per millimeter or something, that would be fine.


1 Answer 1


Distribution of GFP

When used as a reporter usually GFP is expressed by the cells themselves or attached to an antibody and injected. It ends up everywhere, is what I mean.

Localization of GFP

There are three solutions to the hidden-behind-tissue problem for fluorescence.

  1. Use a functionally 2D sample (nematodes, mouse limbs, etc)
  2. Sacrifice the organism(if it's not dead already) and use optical clearing
  3. Cleverness

Of these cleverness is the most interesting. Types of cleverness include time-resolution imaging(Light scatters instantly but most fluorescent proteins take a few nanoseconds to relax) to separate fluorescence from backscatter. Excite a slice of the sample and look at it from a slight angle 3 ns later, and you should see a depth-map. Repeat.

You could also use optical tomography to essentially use computers and many cameras instead of time-delay imaging to solve for the 3d distribution of signal. Computed tomography is much cheaper, mostly because you don't need a nanosecond-scale laser + capture system(those are expensive).


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .