Re-Evaluating the Spherical-Nucleic-Acid (SmartFlare) Technology

I co-authored a Correspondence pre-print article that puts into question the Smartflare technology. SmartFlares (the commercial name of NanoFlares) are gold nanoparticles covered in  oligos specific to a certain mRNA of interest (aka spherical nucleic acids). Supposedly, cells internalize these particles and, once the mRNA hybridize to the oligo, a complementary fluorecently labeled oligo is being unquenched and “flares”, indicating the presence of said mRNA. In this post I want to briefly mention the main topics of our pre-print, and expand on some points. I encourage readers to comment here or on Pubpeer on our article.

This article is a reponse to a recently published paper by Yeo et al that claims that SmartFlares can be used to diagnose certain scar types by detecting in the surrounding tissue the levels of a specific mRNA encoding CTGF (connective tissue growth factor).  The researchers claim up to 11-fold increase in fluorescence in keloidal fibroblasts compared to normal in vitro, 7-fold ex vivo and 2-3 fold in vivo.

There are two major points that we raise in our correspondence:

  1. The CTGF mRNA concentration is too low to be detected by the SmartFlares, based on the data provided by Yeo et al themselves in Fig S2. Any signal detected will be too low compared to the background of the not-100% quenched flares.  Unfortunately, there is no quantitative data on the actual number of mRNA molecules in these cells, just comparative levels by PCR.
  2. Most SmartFlares are actually found in endosomes, with only a minor fraction (2-3%) that escapes the endosomes (note the figure below, step “Enter”). The fraction retained in the endosomes is exposed to DNases and other harsh conditions that can release the Flares, thus getting a flase-positive signal that appears punctate. In Yeo et al‘s paper, by the way, the image resolution is too low to determine the patern of the fluorescnce.

Scheme of SmartFlare mode of action. Source: an article written by Merck-Millipore (distributers of SmartFlares – now discontinued).

There are also several other minor points raised there.

I wish to explore two points (which I already raised in a post several years ago):

Can SmartFlares actually bind mRNA?

Yeo et al used a short DNA oligo in an in vitro assay as a “positive control” to show the specificity and activation of SmartFlares, and the concentration range to get full release of the Flares. In fact, I have noticed than any paper that used NanoFlare/SmareFlare technology and decided to test the specificy of the probes used short DNA oligos. This was peculiar to me. The claim is that it can detect mRNA, but no-one, over a decade of use, has ever tested binding of SmartFlares to full-length mRNA in vitro? There could be multiple ways to test that – use synthetic RNA, in vitro transcribed mRNA or extracted RNA from cell lysates. How come no one tried it? Did people try and fail and did not report it?

Can a mRNA sterically reach the associated oligo? Note that in the scheme above the artist drew the 5′ end of the mRNA binds the oligo. But what happens if the target seqeunce is not at the exact 5′ or 3′ ends (e.g. as the case in Yeo et al, where the target sequence is in the CDS)? Furthermore, what if the RNA is folded or otherwise structured at or close to the target site? Last, can an endogenous mRNA, which is associated with multiple proteins (and RNAs) at the UTRs and is translated by multiple bulky ribosomes even approach this gold nanoparticle? I have not seen a single paper that directly asked or examined these questions.

We know from multiple studies that siRNA and miRNA are only binding a fraction of their target mRNAs – and these are assisted (and regulated) by cellular machineries. But we really don’t know anything about how these SmartFlares bind their target mRNA in vivo.

Speaking of translation – Yeo et al used SmartFlares against CTGF, which is a secreted protein. It contains a signal peptide, so most likely the translating mRNA resides close to the ER membrane, and it is associated both with the SRP receptor and possibly with other RNA binding proteins that target it to the ER. Can the SmartFlare approach such a mRNA easily to allow detection?

Can SmartFlares really be inert?

One of the claims pro-SmartFlares is that these have no adverse effects on cell physiology. Well, if >97% of the particles are segragated in endosomes and later secreted, and the other few % that are cytoplasmic fail to bind the mRNA, that makes sense.

But assuming SmartFlares bind a mRNA – what will happen to it?

I can see several options, none of which (to the best of my knowledge) were experimentally tested:

  1. Nothing – the mRNA is immediately release back to the cytoplasm unharmed (e.g. by tranlating ribosome, helicase or other factors). In that case, the mRNA is free to immediately associate with another oligo and release another Flare, thus boosting the signal in a non-proportional way to the actual mRNA level (perhaps leading to maximum release of Flares).
  2. The mRNA is degraded. This might be achived through several pathways:
    • RNaseH will degrade the DNA-RNA hybride.
    • A ribosome will be get stuck due to the DNA-RNA hybrid block. This will trigger the no-go decay.
    • Mis-regulation due to sterical blockade of protein factors will lead to translation inhibition and decay induction.
  3. The mRNA is not degraded, but also not properly translated. If the target sequence resides in the CDS (as in Yeo et al), then it is likely that the ribosomes will get stuck. Assuming it does not lead to no-go decay, we should detect a reduction in protein levels. If the nascent protein is stuck half-way through the ER translocon – how does that affect ER physiology? Does this initiate a stress response?

What about the released short oligo-Flare? Does it have any effect on the genomic DNA? On anti-sense RNA molecules of the target mRNA? Is is immediately degraded by nucleases, leaving just the fluorophore?

I will stop here, and again, I urge you to comment here or on Pubpeer.



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