As mentioned before, one of the things that has been bugging me about the SmartFlares is that they’re not where anyone expects them. EMD expect them in the cytosol (we don’t see them there), I expected them in endosomes and they weren’t really there either. So what could this mystery compartment be?
There are not many papers dealing with the uptake of gold nanoparticle-conjugated nucleic acids. One from the Mirkin group suggested that they are taken up via scavenger receptors, however the mechanism was unclear. On page 2253:
These experiments suggest that the broader classes of endocytotic process such as caveolae-mediated endocytosis, macropinocytosis, or phagocytosis are not the primary modes of entry of DNA-Au NPs into HeLa cells.
No other mechanism was really offered other than the “scavenger receptor pathway”. Given that and somewhat confusingly (my emphasis):
…RNAi down-regulation of class A (SR-A) or class B (SR-B1) scavenger receptors (Supporting Information, Figure 10) did not result in inhibition of nanoparticle uptake (Supporting Information, Figure 5)…
The authors argue that the pharmacology does support Scavenger Receptors as the means of uptake. Still, some of the treatments are pretty harsh. In Figure 3A a 60% inhibition of endocytosis is recorded in the presence of 2mg/mL fucoidan. This concentration seems eyewateringly high to me.
Indeed, a quick search through the Sigma Aldrich references list show cell-based assays using 8μg/mL, 100μg/mL and 1-20μg/mL. While I was on the Sigma website, I noticed the huge range of things this prep is used for including:
- antitumor, antiangogenic, immunomodulatory, anti-inflammatory, anticoagulant, antithrombotic, antioxidant, induces apoptosis, induces macrophage activation, activates mitogen activated protein kinases (MAPKs) and causes natural killer cell proliferation, non-selective selectin blocker
Regardless, when we contacted him for a comment on How Smart Are SmartFlares, Dr. Mirkin’s feeling was that SmartFlares are taken up by “Class A scavenger-receptor, caveolin-mediated endocytosis”.
To try to get to the bottom (or at least to break the surface) of this, I thought it would be interesting to probe the samples with an anti-caveolin 1 antibody that was generously offered by the Dart Group.
The antibody is SC-894 from Santa Cruz. Before wasting SmartFlare reagent, I wanted to test the antibody in HeLa cells without any SmartFlares. The rough protocol was:
- Seed cells as for normal experiments, let adhere for 4h
- Incubate with 10kDa FITC dextran for 18h
- Wash 2x PBS
- Fix in 4% PFA at RT for 30min
- Wash 2x PBS
- Block and Permeabilise with 2% BSA: 0.1% TX-100:PBS for 60min at RT
- Primary 1:50 in 6% BSA:PBS for 60min
- Wash 3 x 15min PBS
- Secondary A647 anti-rabbit, 1:500 for 60min in 6% BSA:PBS
- Wash 3 x 15min PBS
The caveolin signal was mostly punctate (which is reassuring) but with some signal appearing more membrane associated (see yellow arrows in the example below), which is to be expected as Caveolin 1 forms the main scaffolding protein of the plasma-membrane caveolae.
Where it gets really interesting is that there is basically no co-localisation between caveolin positive compartments and dextran-containing compartments.
A closer view shows very little overlap between the two channels and possibly exclusion although quantitation will help with the analysis.
I’m now very interested to see the comparison between Cav-1 and SmartFlares (the experiment will be imaged later this week, so watch this space!).