SmartFlare Maths

SmartFlare are nanoparticle sensors which are sold by Merck and are supposed to detect mRNA inside live cells. The technology has been developed by Chad Mirkin. In his papers, the nanoparticles are called Nano-Flares or Spherical Nucleic Acids. I am saying “supposed to” because the central question of how those sensors could possibly reach the target that they are supposed to detect has not been addressed by Mirkin nor by Merck.

After evaluating the SmartFlare, we published recently our conclusions at ScienceOpen. We ran this research as an open science project, sharing our experimental results, analyses and conclusions in quasi real time using an open science notebook. All of the imaging data can also be consulted via our online Open Microscopy Environment repository.

Gal Haimovich, who reviewed our paper, first on his blog and then at ScienceOpen, suggested we should do some SmartFlare Maths (point 4 of his list of comments). This had been at the back of my mind for some time. There are various ways to look at this problem, but all those I have tried lead to the same conclusion that the protocols, results and conclusion published do not add up. Here is what I believe the simplest way to think of the SmartFlare Maths problem. As usual, comments and corrections would be very much appreciated.

Estimation of the number of SmartFlares per cell

SF-figure adapted from Giljohann

Adapted from Giljohan et al, Figure 1b

Estimate 1. SmartFlares are added to cells at a final concentration of 0.1 nM (following Merck’s protocol). For 400,000 cells and 20 μL (following Merck’s protocol), this would result in 150,000 SmartFlares per cell, assuming that all nanoparticles are uptaken.


Estimate 2. Giljohann et al  (Mirkin’s group) published a quantitative study of uptake of SmartFlares in various cell lines in 2007. From their Figure 1b, we can see that in the lower concentration range tested, there is a linear correlation between SmartFlare concentration in the medium and number of particles per cell. For cells exposed to a medium concentration of 0.1 nM, this would result in an uptake of 75 000 SmartFlares per cell. In the following discussion, we will use this lower estimate. With ~50 oligo probes per SmartFlare, this would give 3,750,000 oligo probes per cell.

Oligo probes per cell versus mRNA per cell

The copy number of any specific mRNA per cell depends on sequence, cell types, signalling events etc, but typically it ranges from a few copies to a few thousands of copies. Our estimate above indicates an excess of oligo probes of at least three orders of magnitude over the most abundant mRNA.

If just 0.1% of these probes would bind their target, it would block 3,750 mRNA resulting in silencing. However, Merck and Mirkin both report that there is no silencing effect in the conditions of these experiments. It follows that more than 99.9% of the SmartFlares do not bind their target mRNA.

Fluorescence background


Reproduced from Seferos et al, Figure 1.

Seferos et al (2007, Mirkin’s group) show that in the absence of release of the probe, fluorescence value of ~30% of the total value after release is observed (in ideal test-tube conditions, i.e. in the absence of nucleases). This is presumably due to a non-complete quenching of the fluorescence. For the SmartFlares to work, we would therefore have to detect a variation of less than 0.1% over a background of ~30%.



  1. Can you estimate from your images how many particles are actually being taken by the cells? because to my eyes from the images it does not look like 75,000, but maybe I’m wrong.
    but if that is the case then indeed you are in very high excess over any target mRNA, essentially constantly at saturation, theoretically.


  2. Quantifying from the EM is extremely hard work because we are probing a 75 nm thick section so even at the lowest magnification (where it is hard/impossible to count nanoparticles) we are only probing a very small fraction of a single cell. To do this properly would therefore require systematic analysis of several thousands of images so that you have proper magnification and complete coverage of a reasonable number of cells. Given the observed cell-to-cell heterogeneity of uptake we would need an even larger number of images/cells to be on the safe side. As we mention in the response to the reviewers, here we use EM to look at intracellular localisation only.
    Images from the repository, e.g. this one show several hundred particles. Given that this is a tiny proportion of the cell volume, it suggest that the estimates above are not unreasonable.


  3. It’s been demonstrated that DNA-AuNP conjugates, after endocytosis, can perform gene knockdown via the antisense pathway. Therefore, some of the oligonucleotides must be making it to the cytosol. What are your thoughts on this?

    Liked by 1 person

    1. Good point. The smartflare maths calculation in this post suggests that if a tiny proportion of the particles were to escape and interact with their target, this would be enough to have an antisense effect; the immense majority in the endosome would make it impossible to use it as an mRNA sensor because it would dominate the fluorescence signal, but in principle it would not preclude antisense effect. However, I think each paper would have to be evaluated critically and the reproducibility by other teams should be checked too. There is also the possibility of off-target effect.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.