The last post highlighted some of the results of using chloroquine to investigate the localisation and functionality of the SmartFlares. This post will deal solely with the interpretation of those results.
Below is my attempt to explain the experimental results, based on my interpretation of the data and reading of the literature (which I’ve tried to cite as I go). I’m always happy for people to suggest other interpretations or thoughts in the comments!
Preamble: On the mechanism of action of chloroquine
Many compounds accumulate in low pH compartments and lead to vacuolation in cells (see here, here and here for references). Largely these are weak bases that are highly permeable to cell and lysosomal membranes but become significantly less permeable when protonated.
This can lead to a huge accumulation of these compounds in highly acidic compartments like lysosomes (which is thought to be part of the mechanism by which chloroquine kills the malaria-causing Plasmodia). The buffering of the protons raises the pH, leading to a further influx of protons and counter-ions which consequently causes osmotic swelling, visible as vacuolation in the cells.
1) Signal dilution by osmotic swelling
PREMISE: SmartFlare uptake is unaffected, but the dilution of the SmartFlares by osmotic swelling leads to the apparent lower signal.
DISCUSSION: Given that we see uptake of dextran and SmartFlares as early as 2h (although we’ve not looked at any earlier time point), it’s seems likely that a good portion of the signal that we see at 18h is in lysosomes (although this assumption may not hold…more on this when we start to talk about caveolae in another post). That said, only some of the swollen compartments are SmartFlare and dextran positive:
Even the ‘normal’ size compartments (~1μm diameter) show a decrease in signal meaning that it’s likely not swelling of the compartments that account for the decreased signal.
2) Inhibition of endocytosis and/or maturation
PREMISE: Chloroquine is inhibiting endocytosis in the cells
DISCUSSION: This idea supports the data, as both the dextran and SmartFlare intensity is decreased. While endocytic inhibition is not (as far as I’ve found) a direct effect of chloroquine, it’s conceivable that endocytosis could be inhibited if the available pool of membrane available for internalisation is restricted (for instance by inhibiting maturation and recycling of vesicles back into the endocytic pathway).
This is a real possibility. It’s even feasible that some dextran/SmartFlare would be internalised early on, depleting the pool of available membrane, at which point everything ceases. Importantly, the probes would never mature to the lysosomes, explaining why we don’t see signal in the large (presumably pre-existing lysosomal) vacuoles.
Off the top of my head, a great way to test this would be to pre-load the cells with (for example) red dextran, then treat cells with a green dextran in the presence of chloroquine. If the hypothesis holds, you would expect to see swollen red vacuoles and small green puncta. I think this one’s going on the to-do list.
3) Endocytic disruption
PREMISE: Chloroquine is disrupting endosomes releasing their contents into the cytosol
DISCUSSION: Surprisingly, for a drug that has been around for so long, there is no real consensus about what the cellular effects of chloroquine treatment are. As mentioned above, there is plenty of evidence for chloroquine-mediated inhibition of acidification and even stabilisation of the lysosomes. One effect that we were especially interested in, was the use of chloroquine to disrupt endosomes (two published examples involve enhancing the delivery of cell penetrating peptides and gold nanoparticle assisted transfection of cells).
If indeed the chloroquine in our system were disrupting endosomes, the expectation would be that the SmartFlares would actually be exposed to VEGF mRNA1. For this reason, the experiment included an optional DMOG treatment. Somewhat disappointingly, there was qualitatively no difference between chloroquine treated cells with or without DMOG (despite an expected 13x increase in mRNA levels by qPCR … albeit in a different cell line).
Assuming that the chloroquine where disrupting endosomes, you would expect the SmartFlares to respond to the increased VEGF levels in the cell. If we do a very quick estimate (caveats as in the previous post), there is no difference between the two conditions (shown is mean ± standard deviation).
If we assume a null hypothesis of there being no effect of DMOG treatment, the P-value is 0.365 (for a non-paired, two-tailed T-Test on data with the same estimated variance). This, combined with the still punctate distribution of the SmartFlares after chloroquine treatment, leads me to believe that the chloroquine is not disrupting SmartFlare-containing endosomes in our system.
Given the caveats expressed in points (1) and (3), the most likely explanation is that chloroquine treatment is preventing the maturation of the dextran and SmartFlare-containing compartments in an indirect fashion. As mentioned above there is quite an easy way to test this. This also means that I’m going to have to find another way to deliver the SmartFlares into the cytosol (likely by microinjection), to assess the activity of the SmartFlares without having to worry about the efficiency of delivery.
1 – Typically the nomenclature is to use italics for a gene name (vegf) and regular text for a protein name (VEGF). I’m still not sure whether VEGF mRNA (that is, mRNA encoding for VEGF protein) or vegf mRNA (mRNA transcribed from the vegf gene) is correct. Answers on a postcard!