Dr. Valerio Pereno

Oxford Nanoimaging (ONI)

Dr. Valerio Pereno’s background spans the biomedical engineering and drug delivery fields, with specific experience in ultrasound-mediated delivery of therapeutics to solid tumors. As part of the Business Development team at ONI, he focuses on investigating novel applications and establishing strategic partnerships across different fields. Valerio completed his DPhil and MSc at the University of Oxford and received an MEng in Mechatronics and AKC from King’s College London.

Aug 4, 2021 Presentation at the Global Conference for Lipid Nanoparticles & Other Non-viral Nanocarriers

Dr. V. Pereno: Vesicles Below the Diffraction Limit: Imaging with Super-Resolution

Talk Transcript:

T_T Scientific_ Dr. V. Pereno_ Vesicles Below the Diffraction Limit_ Imaging with Super-Resolution 


Nima Tamaddoni, Ph.D.

Yes. Thank you very much for coming.


Dr. Valerio Pereno:

Awesome! Well, I will just start by thanking Nima and the entire team at TNT for organizing this conference, it is very timely, and I am really excited by the talks. Secondly, thank you for inviting me, and thank you for the introduction, just a quick correction, I did my PhD in drug delivery at the University of Oxford and currently I work at a five-star company based in oxford UK, called ONI, and our mission is to basically make the ultimate imaging technologies available to as many people as possible. We realize that there are some massive improvements in optical characterization and we want to make sure that these technologies get into the hands of the scientists and not simply in core facilities. So, I am just going to share a few slides of, you know, the work we do and some, not very novel data of some supervised imaging of lipid nanoparticles. So, just share my screen right now, please let me know if you can see it properly. Can everyone see my screen?


Nima Tamaddoni, Ph.D.

Yes, Valerio, we do see our slide and we do hear you very well. Please go ahead.


Dr. Valerio Pereno:

Wonderful! So, as I said earlier ONI was founded as a university of oxford spin-out with the aim of making a technology that won the Nobel prize in 2014, available to basically everyone in terms of, not only in terms of pricing but also in terms of ease of use because we thought that giving access to everyone to 20 nanometers in resolution in fluorescence had really the potential of changing the way science is done and ultimately will impact, you know, clinical outcomes. So, our aim as a company is to make accessible technologies that are, you know, clinically translatable and basically empower scientists from academia to industry with a tool that can lead them to make discoveries that were previously invisible. So, just to give you some background, so we have developed a microscope which you can see in the picture which is about, you know, the size of a toaster and this is a full-fledged super’s microscope that allows you to stain your cells and visualize them with 20 nanometer resolution and just as a reference, these kind of machines typically occupy an entire room require optical tables, you know, they require air conditioning and specialized knowledge. So, we want to completely scrap that and allow every single person both in the R&D but also in process development and manufacturing access to this technology.

So, just to give you some backgrounds, this is on the right, just a quick example of a cytoskeleton of a cell that was visualized and captured using our microscope and the idea is that, you know, with higher resolution, not only you get prettier images but of course you get more information that then can lead to actionable insights without the need necessary of an extra microscope and just to give you some context, this is an example of a B-cell, an entire B-cell that was stained with CD20 and if you were to image the cell with a white field microscope, this is what you would see, right? You would see pixelated image where you can just about discern the presence of the receptor on the surface. With our microscope you are actually able to look at the spatial distribution of the molecules on the surface of the cell and not only has it allowed you to count the number of proteins that are expressed on each cell. So, this is, we think that this kind of technology can really have an impact not only on cellular imaging but we took it a step further to use this as a way to characterize lip and nanoparticles in a much deeper way. So, just to give you some even more contexts in terms of the application that we have been working on.

The first example is iron channels, imaged in two colors, so they have the sodium potassium ion channels and you can see the single receptors on the cell surface, when they come together where they are, you know, single ones but also you can look at their size to determine whether they are open or closed. Another structure that I quite like is nuclear pores where you can actually see single nuclear pores and their nuclear pore complex where you can see the eight proteins that make up the nucleophile complex. But also functionally like we also infected cells with Ebola virus and we actually tracked where the virus was actually being assembled but also being transported in the cell. So, this kind of this kind of information is simply not accessible using a conventional fluorescence confocal microscope. But super resolution allows you to go this deep. So, I am not going to spend too much time on the benefits of Nano carriers, I guess that everyone knows because they are here. But one key thing that we see on our side that there are very key limitations in how people characterize their particles simply because many of the approaches are bulk approaches rather than single vesicles with single nanoparticles and we hope to address that by using our technology. So, this is an example just to illustrate how the technology actually works. This is an example of an extracellular vesicles because in fact this was one of our first applications for the microscope, where you can see different receptors, you can see different look like as content within the actual vesicle. If you were to image this using a conventional microscope, this is what you would see, you would see maybe a single pixel or a few pixels but you would not be able to determine the spatial distribution of the components that make up your particle. But the moment you supervise you can actually look at the membrane of the lipid nanoparticle but you can also look at the distribution and the quantities, the relative quantities of the biomarkers that are present on a single particle and you can do this across, you know, tens of thousands of particles in one single image because they are so small. But how does the technology work, right?

So, imagine you have a ring structure, imagine your lipid nanoparticle or your liposome that is composed of fluorescent molecules. If you shine your laser at them in one single time, they will all fluoresce concurrently. So, you will get like a massive blur and it is very hard, you know, it is impossible to determine the structure, the underlying structure and this is called the diffraction limit which was described by Abby in a very simple equation and now the way we circumvent this limitation is by basically switching on each of the molecules one by one and if you do this, if you take a center of each blur in time, you are actually able to determine the structure that was actually emitting the fluorescence in this place. But, just to give you like even more deep understanding of how the technology works is like you take a beads for example and you, you know, you put your fluorescent antibodies on the beams, if you use conventional microscopy, you know, you see blobs and, you know, you see a pixelated image and if the concentration is high it is very hard to distinguish whether you are looking at a single vesicles, a couple of vesicles, different vesicles coming together. But if you use storm and you shine high power laser, these fluorophores start blinking.

Therefore, they switch on and off in time and our algorithm is able to detect center of each of these blinks and if you do this and then you zoom into your picture you are actually able to see single beads and look at the distribution of your fluorescent ant bodies on the surface of each and single beads. And, you know, up until now you, you know, you must say okay cool these are beats but, you know, how it works in practice with EVs or IMPs. So, this is what I am going to show you in the next few slides. So, just as an example, these are three vesicles that are using this technique and you can actually see that you are able to distinguish directly by imaging the music of the different sizes but not only, right? You are looking at also the distribution of your molecule on the surface whether it is uniformly distributed such as in this example but also you can look at clustering or you can even look at events that are in very low copy numbers. So, if you have for example, you know, a single protein or you have a single copy of a nucleic acid if the structure is tagged using a fluorophore we would be able to detect it using our turf microscope and this is one of my favorite slides because it really relates to the work I did back in my research days. So, these are lipids particle formulations that were encapsulated in different ways. So, in the first formulation you can actually see individual particles and when you use storm you can see that, you know, you have your circular morphology, you have the two biomarkers and this white circle represents the diffraction limit. So, with any other microscope we would not be able to see these details and if you take this the example on the right, you know, also in this case you have the two biomarkers, you have the purple, and you have the cyan. So, if you use a conventional microscopy method, you would still determine that this is an intact NMP when in fact it is probably lipid debris and then if you use, you know, alternative method you can also see that, you know, there are lots of different structures within the sample that are probably not desirable in our preps.

So, this is in terms of the lipids but we also quantify the DNA and RNA cargo in single vesicles, right? So, we actually stained the… in this case the DNA content and we actually able to see how much of the DNA was the outside of the vesicle and how much was inside and we were able to quantify the number of localizations in each of these conditions to determine that on extracellular vesicles, there is DNA, not only in the inside but also on the outside by fluorescent imaging and of course the moment you use single molecule imaging, you do not have pixels anymore but you have a point cloud. So, by counting the localizations you can determine what computes, a triple positive, a double positive, single positive particle in a very quantitative way. And, you know, we have developed online software that allows you to quantify the tens of thousands of vesicles at the same time and actually look at their structure and look at different parameters such as circularity, skew, amounts in terms of localizations of your biomarker of interest and solid support. So, this is just an example where we are counting the localizations from the CD63 and the CD81, the CD9, so these are the individual photon emission events that we are detecting using our localizer.

And, you know, our vision is to really embed this technology in the NMD manufacturing process where… and not only an impedables of viral vectors and RT’s where we are able to actually visualize the single particles. And, you know, across tens of thousands of them and determine, you know, how many are loaded? How many are empty? What are their morphology parameters? What are the quantity relative localization quantities in each of these particles? So, this is what we are like really moving towards and I am really keen to get some feedback from the audience even later on by email on what they think about this approach and one of the things that I have not mentioned that is this level of resolution allows you to visualize the receptor ligand interactions. So, this is a cell that was seen for transferring ligands and receptors. So, you can see the receptor in green and the ligands, actually you can see the interaction between the two on the surface of the cell. So, imagine you can actually quantify the interaction between receptor ligands on the surface of, you know, tens and tens of cells in one go in a quantitative manner. So, I am thinking of designing an NMP that has targeting capabilities, this would be one way you could actually determine whether the NMP is actually interacting with the cell. But, the microscope is as I told you earlier, you know, very small, it is heated to 37 degrees, so we can actually do live imaging and not only can we visualize the NMPs as they get up taken by the cell but we can also quantify their dynamics. So, for example here we can see that there is a diffusion coefficient that is lower on the left hand side but also on the right hand side you can see how the diffusion coefficient is much higher because your periphery of the cytoplasm.

Another example is this one here where we are tracking clothing if P being transported to the cell junction, then we can actually track single clothing proteins being transported across the cell and then use our algorithms to actually quantify on a bunch of different parameters and so all these data that I have shown you were mostly on extracellular vesicles, we started with a new program to really apply this technology to nanoparticles because we think that this can have a massive impact in the development, but also potentially in the QC space in terms of the nanoparticles. So, these are NMPs that were loaded with RNA. So, the two colors you see is, the blue is the body-P, so it is the lipid component while the purple is the RNA and you might say, you know, I see like a lot of debris here but the moment you actually zoom into the image and you have got look at, you know, single particles, you are able to see the RNA that is sort of like wrapping around the particle while the lipid component is only like right in the center of the lipid nanoparticle and then, you know, you can see like tens of thousands of them pictures, so we use an algorithm to basically determine that the air that is composed by the single particle. We look at their morphology.

So, we can actually look at not only like their sides, we can also look at whether they are oval shaped whether they are round and so on, whether they are aggregating. So, these two for example might be two particles and we can actually quantify this. so, we can look at the size distribution of the particles, we can look at the circularity how round they are, we can also look at the number of localizations that we detect per cluster. The way for to somewhat relate to the quantity of your biomarker within the individual particles and this is just a snapshot of our software that allows you to actually visualize single particles; you can actually click on each of these dots to actually visualize the parts that you are quantifying. So, you have this like visual feedback from the actual software and here you are actually like quantifying the amount of particles that are positive for quality, positive for our MRNA but, you know, our particles that have both, how many cases where the positive is alone. So, where there is no encapsulation at all or, you know, MRNA that has not been encapsulated within the particle and this process that is incredibly easy to use and intuitive and can be done in three steps online using a cloud-based platform and as I said earlier not only we can look at the single particles, we can also see how they get up taken. So, these are the same particles that were delivered to the cell. And, you know, with a few tracking algorithms, you can actually like track the single particles that were up taken into the cell, and then recreate the tracks of where they are going into the cell. You can also use a second marker to look at your endosomes or you can like examine their fate and so on and so forth and lastly, I just wanted to show you; like this is just an example we tried with mitochondria. So, we tracked the mitochondria using master tracker, the vesicles are still in body P and here you can actually see the interactions between the particles and the mitochondria and this is just like a nice fun example that we did in lab just last week made, just to give you this sense of the power of this technique, both on the supers side but also on the lifestyle image in quantification. And, you know, we were founded, you know, about five years ago these are some of the organizations that we work with our customers and if anyone has any cool samples, I think that this technology could benefit a lot from, do send me an email, it is super easy, that is my name at oni.bio. And, yeah, so I will open the floor to any questions that anyone might have.


Nima Tamaddoni, Ph.D.

Thank you very much. Yeah, this was great. Anyone has a question; this meeting should have been a webinar. So, if you have a question, we can even turn your video on or audio on to ask a question or you can directly email Dr. Pereno at the email, he shared


Dr. Valerio Pereno:

I am sorry. I think I missed it. What is the smallest nanoparticle that you were able to distinguish from the others? I guess it is the resolution. I think I missed that part. I am sorry I came in a little late. If you were to try to differentiate between because we get a lot of requests for, you know, of course smaller than 200 nanometers, so that can be sterile filter but I was wondering if there was a minimum to be able to distinguish between the individual particles.



Dr. Valerio Pereno:

So that is a good question. So, the technical resolution of our microscope is less than 20 nanometers, so this allows you to basically distinguish whether the two particles are individual particles when they are Nano meters or so apart. Whether you can distinguish between a single and multiple particles will be determined by their size, by the how efficiently they have been labeled and also their morphological parameters, right? You would be able to… for example see two parties are really close to each other are circular they are right in the right size range and, you know, that is the way you could tell that there are two separate particles. But actually pointing to two particles and saying that there is a single cluster whether that is useful is an open question. I think you can quantify and determine how the level of aggregation particles. I can bring up the site if that helps.


Nima Tamaddoni, Ph.D.

That would be great to see. Thank you. I appreciate that. Oh, you are muted to a Valerio, Valerio you are, and we are not hearing you once.


Dr. Valerio Pereno:

Wow! I spoke for 10 minutes. I was saying that, so these are the lipid Nano parts that are loaded with MRNA. We run a clustering algorithm that, basically allows you to look at the errors of highest density of localizations to determine what the particles actually are and this is called sigma molecular clustering; this is a well published method online. So, basically we do you run the clustering and this allows you to look at the… basically the shape of your particles, right? So, you can determine, if you compare this one that looks a bit like the UK to this particle here that is not more or less round, you can you can actually start making, you can looking at the differences in your place, not only visually but you can actually quantify this using circularity, for example and in this example here you might see, you know, two particles are really close together but they might be two distinct parts or for example you might have, like this is not the case in in this particular example but you might have a big blob of liquid debris for whatever reason and, you know, this technology will allow you to see that in comparison to the other particles or maybe next to it in the same image and by looking at the quantifications of, you know, different parameters, it will allow you to understand of the whole population in terms of the size, the skew, the circularity, the area, the length, we have about 10 parameters that we can extract from each particle. So, this is a bulk approach, you are, like we are actually pumped on a single particle basis.


Nima Tamaddoni, Ph.D

Very cool! Thank you. I really appreciate that, very cool. I think another question came in Valerio, it says you obviously need to label, we die both LMP and MRNA, which is specific dies where you used for both and can be distinguished different lipid components using different dyes and interaction with the MRNA, today.


Dr. Valerio Pereno:

Okay. Let us just break that up into single quest, in this case the MRNA was already labeled before the production of the LMPs, they were then labeled with body-P. However, we are developing methods to do the labeling, let us say absolutely, so after the… we are formed. So, you can do it in both ways, the pre-labeled is relatively easy and this is what we have shown here. Labeling afterwards requires a little bit of optimization something that we are developing for specifically for this technique and based on our experience in the past with extracellular vesicles, we found that many of the dyes, particularly the lipids lipid-based dyes do form aggregates. But in the case of the EVs were in very much in the same area and in the same size range. So, it is very important that these parameters are optimized specifically for MPs. I cannot remember the second question you might repeating it please.


Nima Tamaddoni, Ph.D.

Absolutely! So, the second one can we distinguish different lipid components using different dyes and their interaction with the MRNA? So, lipid component and the MRNA is a question that can we distinguish those and their insights.


Dr. Valerio Pereno:

So, I think this is very much a research question to be very frank. I guess that one way that one could do this is to incorporate already fluorescent lipids, labels for example a CY5 dye in for example two populations of lipids MBMRNA, so we can do imaging in three colors, so that will allow you to look at the spatial distribution of the two different lipids plus the MRNA. However, this is something that is very much in the research and development right now. Yeah.


Nima Tamaddoni, Ph.D. 

Okay. So, there is another question which type, sorry, which type, questions are coming in and I have to expand the window. So, one of the questions which type of contrast agent is used for this imaging and have you tried quantum dots AIE? I guess this is, okay that is from our next speaker, obviously, and nanoparticle or gold nanoparticle.


Dr. Valerio Pereno:

Right. So, let me start from the first question. So, the contrast agent is, you know, fluorescent dyes that are commercially available. So, in this case it was CY5 for the MRNA, body P for the lipids but you could use, you know, Alexa floor dyes. The best one is probably six four seven, you could use dye light, you could even use for example GST but then you have to go in with a secondary antibody than antibody for the GFP. So, as long as you can label their components fluorescently, using photoactivatable dyes, photo switchable dyes, you would be able to use this technique.


Nima Tamaddoni, Ph.D.

Yes, go ahead.


Dr. Valerio Pereno:

In terms of quantum dots we do not have any specific experience using them, we are already developing the fluorescence modes first but it is something that we would be willing to explore 


Nima Tamaddoni, Ph.D.

Wonderful! You already answered the next question which was does GFPMRNA work? Does it give auto fluorescence?


Dr. Valerio Pereno:

No, I mean, yeah…


Nima Tamaddoni, Ph.D.

Go ahead 


Dr. Valerio Pereno:

In terms of the GFP, I think GFP is really cool because you could, so for GPF is really good for the live cell imaging, if you are, especially if you already have it tagged. In terms of the super SU, you know, GFP does blink. So, you would be getting an image, it would not be at 20 nanometer resolution unless you use an antibody for the GFP that is conjugated to a storm compatible. So, those are options that we work around with, there are ways, so we can use super SU to look at GFP.


Nima Tamaddoni, Ph.D.

Okay and the last question I keep here and again the audience can ask questions in Q&A forums that are the last question for this session, it says concerns about dye label perturbing the intrinsic LMPA structure and causing artifacts 


Dr. Valerio Pereno:

I think that is a really good point, it is something that we are exploring. I think it would depend on at what point the dye is incorporated whether it is upon formation or afterwards and also the relative concentration of, you know, die lipids and so on and so forth. But this is something that will be explored and it also depends on the size and the… basically the states of the lipids themselves whether it is a little more liquid disorder, the liquid ordered, LMP and so on and so forth. So, all these have effects on that factor, yeah. 


Nima Tamaddoni, Ph.D.

Alright, wonderful! Thank you so much, Valeria. I am sure everybody enjoyed this talk, very excited, will talk soon. 


Dr. Valerio Pereno:

Wonderful! You have any more, just send me an email. Cool!


Nima Tamaddoni, Ph.D.

Yeah, absolutely! You guys in the other side, we have one starting right now.

Contact Us