I recently discussed the SmartFlare project at a joint Sée and Lévy Group Meeting, partially extolling the virtues of Open Science but mostly focussing on the data acquired so far and some of the proposed experiments.
Regarding the SmartFlare project, the three big questions that formed the basis of the talk were:
- Why do only some cells take up SmartFlares?
- Where are the SmartFlares?
- Why are the VEGF and Scrambled SmartFlares fluorescent in puncta?
Below are some of the interesting points that were made during the meeting. Replies at the time (as best as I can remember them) in green:
- In relation to the heterogeneity of SF uptake, it was asked whether the ‘dark’ cells may have taken up the SmartFlares but just lack detectable VEGF mRNA. This is a really interesting question as there’s not really any way to see the SmartFlares when they’re quenched using fluorescence microscopy. Multimode fluorescence and photothermal is a real option, however, even the Uptake Controls show heterogeneous uptake suggesting that this is unlikely to be a quenching/de-quenching effect.
- Could possible quenching effects explain the same phenomenon. Would lower concentrations of SmartFlares actually show more signal? Possible, as this could affect controls as well as VEGF SmartFlares. I think this is unlikely, but again, photothermal would be a great way to ‘image’ the distribution of gold nanoparticles in cells. At this point, we’re following the manufacturer’s guidelines, so presumably it’s been extensively tested at this concentration.
- Would longer time points show SmartFlares escaping from the vesicles? Again, a possibility and an easy experiment to do, but as stated above, we’ve been following the manufacturer’s guidelines which recommend 18-24h exposure.
- May be worth trying a shorter treatment with chloroquine once particles have been taken up. Absolutely. The chloroquine literature is quite conflicted as regard whether the drug destabilises endosomes, prevents maturation or inhibits endocytosis all together. (UPDATE: See “Interpretation of the Chloroquine results” for more information)
Some other interesting points and random thoughts that came up during the meeting:
- Microinjection of SmartFlares
- This is, in my opinion, a really important experiment to do. None of the experiments so far hint that the SmartFlares work as advertised. It’s important to know if this is a problem with delivery or with the product. Microinjecting directly into the cytosol bypasses delivery, but will probably lead to massive variability in loading cell to cell.
- Fluorescent spectrophotometer / FLiM measurements +/- DTT
- Similar to the microinjection is some basic in vitro work to make sure that changes in fluorescence associated with de-quenching are even detectable. Putting the SmartFlares in a cuvette +/- DTT to strip the oligos should give us the dynamic range.
- Quantitative texture analysis
- In looking at the data so far, we’ve been qualitatively assessing punctate versus diffuse distribution. At the OME User’s meeting in 2014, I was talking to a member of Robert Murphy’s Lab who have for some time been working on Subcellular Localisation Features. The idea is to turn images into more metrics than a spatial intensity map (IE an image) by expressing the shape, texture, position and edge characteristics. These can then be compared to positive controls for example: exogenously expressed fluorescent protein markers, or the 2D HeLa Cell Reference Library (see also this reference) to give a certainty of similarity. At some point, I’d like to try to quantify the texture or endosomal versus cytosolic ‘similarity’ of the images. Of interest is that there is an implementation of this technique available as an OMERO webapp (although I think it only runs on the older OMERO 4.x).
- Caveolin / Caveosomes
- This one is going to get a post all to itself. A fascinating if somewhat confusing subject.