Measurement

CAPTURE targets the multi-resistant pathogen Pseudomonas aeruginosa using antimicrobial peptides (AMPs). The AMP-encoding plasmid is delivered via one of the two distinct carrier systems that we have developed in parallel: lipid-based nanocarriers or outer membrane vesicles.

During the development phase, we encountered several challenges in measuring key characteristics of these carrier systems. As we progressed with our project, we faced critical questions:

• How can we verify and quantify plasmid encapsulation?
• How can we achieve the desired particle size and ultimately detect it?

These challenges pushed us to focus intensively on the measurement aspects of our project, leading to the development of new methods. While our primary focus was on Lipid Nanoparticles (LNPs), we aimed to create versatile techniques applicable to various lipid-based nanocarriers and potentially outer membrane vesicles.

The most common method for quantifying double-stranded DNA (dsDNA) in LNPs is the PicoGreen assay. While highly sensitive and well established, this method is expensive. Consequently, we devised a novel approach for rapid and cost-effective quantification of DNA in LNPs, employing the more commonly available DNA dye Midori Green in a straightforward plate reader assay. This new method represents a significant advancement in the field of nucleic acid quantification in lipid-based delivery systems.

In addition to developing a novel methodology, we have also conducted intensive research on dynamic light scattering (DLS), a widely used technique for detecting and characterizing nanoparticles. Our work has specifically focused on applying DLS to lipid nanoparticles (LNPs). In parallel, we have employed mathematical modeling to optimize the LNP production process. Moreover, we developed a new Python tool designed for the scientific visualization of empirical data obtained from DLS measurements.

Fluorescence-Based Nucleic Acid Quantification in Lipid Nanoparticles

The PicoGreen Assay - Benefits and Hurdles

LNPs are typically prepared by combining an organic phase containing the lipids with an aqueous phase containing the DNA. The mixing process leads to the formation of DNA-encapsulating LNPs. However, in addition to the formed particles, some non-encapsulated DNA also remains in the resulting solution. The PicoGreen assay (Figure 1) represents the current gold standard for the quantification of DNA in lipid-based nanocarriers. The PicoGreen stain is added to this solution, staining only the non-encapsulated DNA due to its impermeability to lipid membranes [1]. After initial fluorescence quantification, Triton X-100 is added to destroy the LNPs, releasing the encapsulated DNA. Consequently, all the DNA present in the sample, whether previously encapsulated or non-encapsulated, is now stained. A second measurement is then conducted, with the difference between the two values representing the quantity of DNA that has been encapsulated. The precise concentration is determined using a calibration curve.

The method has the advantage of being highly sensitive, particularly for small amounts of DNA. However, it is expensive and PicoGreen is not a common dye that is readily available in all laboratories. This inspired us to seek an alternative that would be more accessible, cost-effective, and user-friendly, not only for our team but also for future iGEM teams and researchers in the field.

Before we could develop our own method, it was essential to establish and validate the use of PicoGreen assay for our project.

Evaluation of the PicoGreen Assay

Considerations

The DNA solution used for the standard curve should be treated the same way as the experimental samples and contain similar levels of any potential contaminants. In our experiments we are using plasmid DNA encapsulated in LNPs. The release of DNA from the LNPs will be obtained by Triton X-100 treatment, an additional factor potentially influencing measurement.

To ensure accuracy in our quantification, we examined the suitability of various DNA standards for use in the PicoGreen assay with our pLNPs. The assay kit provides a Lambda DNA standard, which was initially employed as a reference for DNA quantification. However, inconsistencies were observed when calculating the concentrations of pLNPs encapsulating the plasmid pUC19 using the Lambda DNA standard, indicating that it may not be the most suitable reference for our specific application.

To address this, we tested the pUC19 plasmid, used for the production of our pLNPs, as a DNA standard. This experiment revealed significant differences in the behavior of the two standards (Figure 2). Specifically, the plasmid DNA standard produced a much flatter calibration curve compared to the Lambda DNA standard, highlighting that the two DNA types do not behave equivalently in the assay. This discrepancy demonstrates that the Lambda DNA standard does not accurately reflect the behavior of the plasmid DNA encapsulated in pLNPs, thus validating the importance of using a plasmid-specific standard for our applications.

Additionally, we measured the plasmid DNA standard in the presence of eLNPs to determine whether the lipids themselves might interfere with DNA quantification. The results showed that the presence of lipids had no significant influence on the measurements, suggesting that any differences observed are due to the DNA standards and not the lipid components.

Destroying LNPs with Triton X-100 is a crucial step in evaluating the encapsulation efficiency of our LNPs. This process allows us to release and quantify the encapsulated DNA. To investigate the effect of Triton X-100 on assay measurement and DNA quantification we did some control experiments: As shown in Figure 3, using Triton X-100 as a detergent might not result in complete destruction of all lipid/plasmid aggregates, making it problematic to assume the quantity of plasmid accessible for the PicoGreen stain. Please refer to our Results page for additional information.

With the process of destroying the LNPs to achieve quantification, the PicoGreen assay lacks opportunities for single particle analysis. It is solely suitable to determine the total DNA concentration in solution. It is thus interesting to investigate if similar assessments are possible with the particles remaining intact by labeling the DNA prior to encapsulation.

Evaluation of Midori Green for DNA Quantification

Our initial goal was to identify a DNA-binding dye that exhibits dose-dependent fluorescence across various DNA concentrations. In our first experiment, we tested Midori Green staining on different DNA concentrations, both before and after DNase degradation. Figure 4 illustrates the fluorescence intensity across DNA concentrations. Notably, the data revealed a robust logarithmic relationship from 10 to 100 ng/µL, suggesting a reliable calibration curve for the assay. However, we observed anomalous fluorescence readings at 0 and 5 ng/µL that exceeded the fluorescence at 10 ng/µL, raising concerns about the assay’s reliability at low concentrations. This unexpected behavior may be attributable to a quenching effect, whereby free Midori Green fluoresces more intensely in the absence of DNA than when binding to small amounts of DNA. As we have supporting data from the PicoGreen assay, we might be able to validate our findings at low DNA concentrations. However, we recommend that other laboratories employ this method only when DNA concentrations are 40 ng/µL or higher.

After DNase treatment, the results produced a nearly flat line across all initial DNA concentrations. This indicates that only background fluorescence remains after DNA degradation. This finding is advantageous, as it establishes a baseline for the background fluorescence of the DNase buffer, which we can utilize in subsequent analyses to improve accuracy.

Next, we aimed to determine whether the presence of LNPs affects the fluorescence of Midori Green-stained DNA. To investigate this, we compared samples of DNA stained with Midori Green in the presence of varying volumes of eLNPs. The measured fluorescence intensities were similar across different volumes of added eLNPs (Figure 5), indicating that the presence of eLNPs does not influence fluorescence measurements.

To further confirm this, we conducted an additional test using a range of DNA concentrations, comparing samples with and without the addition of 40 µL of eLNPs (Figure 6). Consistent fluorescent patterns were observed, reinforcing that LNPs do not interfere with the fluorescence signal in this assay.

To investigate the influence of encapsulation on the fluorescence of Midori-stained plasmid DNA, we compared the fluorescence intensity between pLNPs formed with pre-stained plasmid DNA and free plasmid DNA stained with Midori Green. We observed a reduction in fluorescence intensity post-encapsulation (Figure 7). This suggests that encapsulation alters the fluorescence properties of the Midori Green-stained plasmid, possibly due to shielding or quenching effects within the nanoparticle. As a result, the encapsulated DNA fluorescence was not directly comparable to that of free plasmid DNA. This discrepancy complicates the direct quantification of encapsulated DNA. To address this challenge, we revised the Midori Green assay protocol. Instead of directly measuring encapsulated DNA, our focus shifted to quantifying the unencapsulated DNA outside the nanoparticles. This approach allows for indirect quantification of encapsulated DNA by subtracting unencapsulated DNA from the total DNA used. The protocol for establishing a calibration curve and quantifying encapsulated DNA in pLNPs using Midori Green, is described in detail in the following section.

The Midori Green Assay - Refined Experimental Protocol

Our experimental setup consists of two parallel components: the calibration curve and the sample analysis.

1. Calibration Curve:

To establish a robust standard calibration curve

• Prepare plasmid DNA at a range of precisely measured concentrations using the same solvent employed in LNP preparation (typically a mixture of ethanol and aqueous buffer).
• Stain each concentration of plasmid DNA with a standardized amount of Midori Green fluorescent dye.
• Measure the fluorescence intensity of each sample using a plate reader (excitation: 490 nm, emission: 530 nm).
• Plot the fluorescence intensity against the known DNA concentrations to generate a standard calibration curve.

A precisely prepared calibration curve serves as a crucial reference point for subsequent quantification of plasmid DNA in LNP samples. By using the same solvent conditions as in LNP preparation, we ensure that the calibration curve accurately reflects the behavior of DNA in the experimental samples, enhancing the precision of our quantification method.

To accurately account for background signals in our assay, we implement a crucial step in our calibration process:

• Treat each DNA concentration used in the calibration curve with DNase enzyme.
• This enzymatic digestion destroys all DNA in these calibration samples, resulting in a consistent fluorescence signal across all samples, regardless of initial DNA concentration. The complete digestion of DNA in these samples is crucial, as it mirrors the destruction of non-encapsulated DNA in our actual LNP measurements.
• The uniform fluorescence observed post-DNase treatment primarily originates from the DNase buffer components. This uniform signal is essential for calculating a reliable Background Fluorescence value.

The standardized Background Fluorescence is calculated by the mean fluorescence intensity across all DNase-treated samples. This Background Fluorescence will be subtracted from all subsequent sample measurements.

The standard curve allows for the conversion of fluorescence measurements from unknown samples into DNA concentrations, forming the basis of our new quantification approach.

2. pLNP Analysis: Quantifying encapsulated DNA

Our novel method for analyzing plasmid-loaded Lipid Nanoparticles (pLNPs) involves a series of precise steps and calculations (Figure 10):

Sample preparation:

• The plasmid DNA is stained with Midori Green, and pLNPs are produced using the standard pipetting-mixing protocol. Notably, the ethanol is not removed from the sample using the spin concentrator, as this would disrupt the Midori concentration. This adjustment is why our standard curve is established using ethanol-containing solutions.

Initial fluorescence measurement:

• The pLNP solution contains both encapsulated and non-encapsulated Midori-stained plasmids. Fluorescence measurements are conducted at an excitation wavelength of 490 nm and an emission wavelength of 530 nm, referred to as Total Fluorescence. An important observation: encapsulated DNA shows reduced fluorescence compared to free DNA.

DNAse treatment and second measurement:

• Following this measurement, the sample is treated with DNAse to digest non-encapsulated DNA. A second measurement is taken, termed Post-DNAse Fluorescence.

Data analysis and calculations:

• Calculate encapsulated DNA fluorescence:
  Post-DNAse Fluorescence - Background Fluorescence = Encapsulated DNA Fluorescence
  This represents the fluorescence signal corresponding solely to the plasmid DNA within the pLNPs. However, this value cannot be directly compared to the standard curve due to the reduced fluorescence intensity of encapsulated plasmid compared to free plasmid.
• Instead, we determine free DNA fluorescence, which corresponds to non-encapsulated DNA in solution according the formula:
  Total Fluorescence - Encapsulated DNA Fluorescence = Free DNA Fluorescence
  This subtraction provides the fluorescence attributable exclusively to the surrounding unencapsulated DNA, allowing for a valid comparison with the standard curve. Finally, we can use the standard curve to convert the Free DNA Fluorescence to the concentration of non-encapsulated plasmid DNA.
• Subtracting this value from the total amount of DNA initially used in the preparation of the pLNP sample enables us to determine the concentration of encapsulated DNA:
  
  Encapsulated DNA = total DNA used - non-encapsulated DNA
  
  

Comparative Analysis of PicoGreen and Midori Green Assays

Finally we compared both assays to evaluate their performance and applicability in quantification of encapsulated DNA.

pLNP samples were prepared and divided into two portions. One portion was analyzed using the Midori Green assay, while the other was subjected to the PicoGreen assay for comparison. Comparable encapsulation efficiencies were determined by PicoGreen (70%) and Midori Green (63%) assay (Figure 11 A). The minor difference falls within an acceptable range of variation (Figure 11 B), validating Midori Green as a reliable alternative.

Cost Comparison of PicoGreen and Midori Green Assays

To provide a comprehensive comparison between the PicoGreen and Midori Green assays, we conducted a cost analysis of the components required for each method. The following table outlines the essential materials and their associated costs for both assays, offering a clear perspective on their economic implications.

Conclusion

Our experiments demonstrated the potential and efficacy of the newly developed Midori Green assay for DNA quantification in lipid nanoparticles (LNPs). The results obtained were comparable to those produced by the established PicoGreen assay, highlighting the validity of our method.

• We established a robust logarithmic relationship for DNA concentrations between 10-100 ng/µL, with optimal performance at ≥40 ng/µL.
• The assay effectively quantifies unencapsulated DNA, allowing indirect measurement of encapsulated DNA content.
• We confirmed that LNP presence does not interfere with Midori Green fluorescence, ensuring accurate measurements in complex formulations.
• Encapsulation efficiency results, which were obtained through comparative measurements, fall within an acceptable range of variation, validating the accuracy of our new method.

This assay offers several advantages: it is significantly more cost-effective, uses reagents commonly available in standard laboratories, and avoids the use of Triton X-100, which can interfere with plate reader measurements. Furthermore, its versatility extends beyond LNPs to other lipid-based nanocarriers and potentially even outer membrane vesicles (OMVs).

While the Midori Green assay may be slightly less sensitive than PicoGreen at very low DNA concentrations, it presents a more adaptable and robust alternative. The method is more reliable and user-friendly, reducing the likelihood of errors and allowing flexibility in experimenting with different DNA stains, as it minimizes disruptive factors like Triton X-100. We began exploring other stains, such as DAPI, but further testing is needed to optimize these alternatives.

In future iGEM projects, this method could be readily implemented for the characterization of a wide range of lipid-based nanocarriers, including LNPs and OMVs. By offering an accessible and affordable option for verifying and quantifying DNA encapsulation, this approach empowers future teams to tailor and optimize their nanocarrier systems without the financial burden of expensive kits like PicoGreen.

Dynamic Light Scattering

Why We Chose Dynamic Light Scattering

Dynamic Light Scattering, short DLS, is used to measure the size of particles ranging from less than 1 nanometer to several micrometers [2]. For our project focusing on LNPs, DLS offers several advantages:

• Size Range: Ideal for measuring particles in the 1-10,000 nm range, perfect for LNP characterization.
• Efficiency: Provides rapid measurements with minimal sample preparation.
• Non-destructive: Allows for sample recovery and further analysis.
• Versatility: Capable of measuring various types of nanoparticles in solution.

The Science Behind Dynamic Light Scattering

The particles are placed in a solution and scanned by a laser. The laser illuminates the particles and varies its laser intensity depending on the size of the measured particle. The particle size can be identified depending on its laser intensity [3].

Advantages	Disadvantages
good availability, short measurement times, easy to use and low material costs	intensity in percentage leads to misconceptions, no actual size number distributions or particle counting
if the sample fulfils the criteria, measurement is suitable to determine average size and polydispersity index	is limited to size measurement and can not always distinguish between different particle populations
size range of 1 - 10000 nm is optimal for LNP characterization and helps analyzing particles that are not visible in the microscope anymore	particle properties and sample concentration affect measurement, does not work well with very polydisperse samples and is limited to a maximum of 10000 nm

The ‘light’ scattering is caused by the Brownian motion of the particles, meaning the photon count fluctuations at the detector. This determines the translational diffusion coefficient, D, which produces the particle size, R, using the Stokes Einstein equation.

Depending on the ionic charge of the solution, the size may deviate in the DLS. The output comes in the form of an intensity that shows a weighted particle size distribution. As, the ionic strength of a solution shows the electrostatic interaction between particles, higher ionic interaction. Then the repulsion is smaller, the body is firmer and the scattering intensity is smaller. If the ionic strength is small, the repulsion is greater and therefore the intensity of the scattered light is higher. The measurement is therefore more accurate if there is a small ionic interaction between the particles. Depending on the particle solution composition, the intensity of the scattered light can vary and falsify the measurement.

Nevertheless, particles of all sizes are measured, but it is up to the user to analyze the data according to individual criteria.

Clear individual advantages of measuring with the DLS are that it is an available device. In addition, its efficiency in terms of fast and cost-effective measurements, its gentle measurement method and its ability to measure the smallest particles makes it a good option for measuring particles.

What are Particles?

The word particle usually describes the smallest elements of a homogeneous mass. A particle is therefore a part of a larger unit and does not differ in its basic building blocks from its neighboring particles in a sample. Size is a highly subjective unit of measurement, which characterizes a particle. In biology, we therefore speak of particles as building blocks that are in the micro- and nanometre range. These are, for example, proteins Lipid-based Nanocarriers and Outer Membrane Vesicles of CAPTURE.

DLS Analysis Tool

During our work on CAPTURE, we frequently utilized DLS to characterize our lipid-based nanocarriers. Through this process, we identified two significant challenges:

• As we are aware that the associated software for analyzing the DLS data is firstly expensive and secondly complicated, we wanted to simplify the process.
• We needed to determine the optimal number of extrusion cycles to consistently produce nanocarriers of ideal size for our project.

These challenges motivated us to develop a user-friendly, accessible tool that not only simplifies DLS data analysis but also optimizes the nanocarrier production process.

Features and Functionality

• Our DLS analysis tool assigns a numerical value to each measured intensity, enabling precise quantification of DLS data.
• It generates plots with nanocarrier size on the x-axis and intensity on the y-axis, providing a clear visual representation of size distribution for each sample.
• We developed the DLS-Extruding Model. This innovative feature calculates the optimal number of extrusion cycles required to achieve the ideal nanocarrier size for CAPTURE, enhancing process efficiency.
• For better illustration, we incorporated the acceptance-band-box into the plotting tool. This feature defines the ideal size interval for our nanocarriers, visually represented as a box on the plot.

Mathematical Model

Since we are dealing with a measured intensity of lipid-based nanocarriers of a certain size in the DLS, we have to take this into account in our calculation.

An intensity in percent stands in relation to all measured particles in a sample. This can of course vary with each measurement. You should be aware that, for example, 5 out of 10 and 500 out of 1000 lipid-based nanocarriers correspond to 50 %, although the numbers are very different.

Our tool uses a limited exponential growth model to fit the data within the acceptance band:

This model is based on the assumption that extrusion follows a limited exponential growth pattern. We assume this because we push lipid-based nanocarriers through a membrane with a pore size k. If nanocarriers cannot become smaller than k, the size of the vesicles will always approach k in order to obtain an ideal size.

Our model is implemented in Python, this function must be fitted via curve_fit from scipy.optimize. This results in a limited exponential growth function that converges towards a bound s. This convergence means that the lipid-based nanocarriers size does not decrease but rather stabilize after a certain number of extrusion cycles.

Application

Our DLS analysis tool offers practical benefits. By determining the optimal number of extrusion cycles, we can streamline our nanocarrier production process. Implementing the calculated number of extrusion cycles helps in working more efficiently. This approach saves valuable time and resources in nanocarrier development.

Contribution to the Scientific Community

This tool represents a significant contribution to the field of nanocarrier research and development. By simplifying DLS data analysis and optimizing the extrusion process, we have created a resource that can benefit the wider scientific community. Our DLS analysis tool not only facilitates our work on CAPTURE, but also demonstrates our team’s commitment to developing practical solutions that address common challenges in synthetic biology and nanotechnology research. By making this tool freely available, we are promoting open access to scientific resources.

In our experiments for CAPTURE, we made measurements with DLS. As described the device is suitable for measuring the size distribution of particles in a solution. We have measured and analyzed our lipid-based nanocarriers several times with DLS. As we are aware that the associated software for analyzing the DLS data is firstly expensive and secondly complicated, we wanted to simplify the process. With our DLS analysis tool, it is possible to quantify the DLS data by assigning a value to each measured intensity.

In the plot, the lipid-based nanocarrier size on the x-axis and the intensity on the y-axis show an intensity that can be displayed per sample for a specific size. In our case, we have extended our plotting tool. Since we had to reduce the size of our lipid-based nanocarriers produced by PVA swelling and minimized by extrusion, we asked ourselves how often we had to extrude on average until we got the ideal size for CAPTURE Lipid-based Nanocarriers and efficient of our process.

Our answer is the DLS-Extruding Model. This tool calculates based on our DLS data, how many times you should extrude a sample to get your individual ideal size for lipid-based nanocarriers. To illustrate this, we have built the acceptance-band-box into the first plotting tool. Therefore we form a box with the y and x-axis, which only defines an interval of the ideal size. CAPTURE’s Lipid-based Nanocarriers require 9 extrusion passages to reach the ideal size of 175 nm to 225 nm (see our Results page). Since we are talking about a measured intensity of lipid-based nanocarriers of a certain size in the DLS, we must take this into account in our calculation. An intensity in percent stands in relation to all measured particles of a sample. This can of course vary with each measurement. You should be aware that, for example, 5 out of 10 and 500 out of 1000 lipid-based nanocarriers correspond to 50%, although the figures differ greatly.

References

---