In this interview, Lan Li, Susan Zondlo, Zamas Lam and Aurora VanAuken discuss the structure, function and analytical challenges of lipid nanoparticles (LNPs) used for nucleic acid delivery, covering their four-component composition, the difficulties in assessing their biodistribution and persistence in vivo, factors affecting off-target effects and premature payload release, and the major challenges and current approaches for quantifying LNP lipids using LC–MS methods.
Meet the Subject Matter Experts
Zamas Lam
SVP and Global Head of Bioanalytical & Preclinical Development
QPS Holdings, LLC (DE, USA)
Zamas is the Senior Vice President of Preclinical Development at QPS (DE, USA) and is trained in mass spectrometry, carbohydrates and glycoproteins. Zamas is one of the world’s few high-resolution mass spectrometrists by training and by trade, with a passion for biologics mass spectrometry and gene therapy. Since graduate school, he keeps thinking that it will be fun to work on a gene level in drug discovery and development versus at the protein level or at the metabolite level!
Susan Zondlo
Executive Director, Bioanalysis
QPS Holdings, LLC (DE, USA)
Susan Zondlo joined QPS, LLC in 2008 and is currently serving as an Executive Director of Bioanalysis in the Department of Translational Medicine. In this position, Susan serves as manager of teams for PCR Analysis, and Biomarkers. Prior to joining QPS, Susan held senior research scientist positions at Millennium Pharmaceuticals (MA, USA), Bristol-Myers Squibb (NJ, USA), and the University of Delaware (DE, USA), where she also served as an adjunct professor in the Department of Chemistry and Biochemistry. Susan received her BA in Biology from Rice University (TX, USA) and her PhD in Molecular, Cellular and Developmental Biology from Yale University (CT, USA).
Lan Li
Director, Bioanalysis
QPS Holdings, LLC (DE, USA)
Lan Li is Director of GLP Bioanalysis at QPS, LLC. After graduating from Cleveland State University (OH, USA) with a PhD in clinical-bioanalytical chemistry in 2011, she joined QPS, LLC and continued her focus on bioanalytical method development and validation. She is currently leading a team working on LC–MS/MS analysis of biologic molecules, including but not limited to lipids, oligonucleotides, peptides, proteins and conjugates.
Aurora VanAuken
Market Research Intern, GBA
QPS Holdings, LLC (DE, USA)
Aurora is currently a Market Research Intern for GBA at QPS and a fourth-year Biomedical Engineering student at the University of Delaware. She brings 3 years of preclinical research experience focused on the development of polymer and lipid nanoparticles as cancer therapeutics. Her work involved formulation optimization and pharmacokinetics, as well as supporting in vivo studies. At QPS, she contributes to data analyses, pipeline tracking, and identification of business development opportunities across global markets. Outside of science, Aurora enjoys distance running, fundraising and reading.
Structure of LNP
What are the components of LNPs and what are their roles?
Zamas & Aurora:
Lipid nanoparticles (LNPs) are nanocarrier delivery systems engineered to encapsulate, protect and transport nucleic acid therapeutics. They are composed of four components: cholesterol, phospholipids, ionizable lipids and PEG-conjugated lipids. Structurally, LNPs consist of a heterogeneous lipid assembly in which the genetic payload is encapsulated by ionizable lipid reverse-micelles and surrounded by an outer lipid monolayer. Each component has a critical role in delivering the payload to the proper tissue. Ionizable lipids are the key to LNPs, enabling encapsulation and adapting their charge during circulation, leading to an increased circulation period with decreased off-target effects. The payload is negatively charged, so it is important to have proper interaction between ionizable lipids internally to control release. Phospholipids provide the main structure to the outer layer, mimicking the cellular membrane, and facilitating cellular uptake. Cholesterol controls the structural integrity and stability of the monolayer by regulating its packaging and fluidity. With increased cholesterol, the rigidity of the monolayer will follow. Finally, PEG-conjugated lipids influence particle size and therefore affect stability and circulation time. Overall, LNPs are highly customizable and provide efficient delivery of nucleic acids to target tissues due to their four components.
How does the internal and external organization of lipids within LNPs contribute to their stability and function?
Zamas & Aurora:
Internally, the negatively charged nucleic acid interacts with ionizable lipids, forming stable reverse micelles, which protect the nucleic acid from degradation and premature release. Externally, the outer lipid monolayer, containing cholesterol, phospholipids and PEG-conjugated lipids, acts as a barrier preventing aggregation, premature release and increased stability during circulation. The overall structure contributes to the function of LNPs by allowing them to adapt to the microenvironment, such as changes in pH during cellular uptake. This organization enables proper delivery of the genetic payload by facilitating cellular uptake and endosomal escape, resulting in the desired changes to the cell’s protein expression.
Biodistribution of the LNP and payload, persistence, etc.
What are the major challenges in assessing the biodistribution and persistence of LNPs and their nucleic acid payloads in vivo?
Zamas & Aurora:
When assessing the biodistribution of LNPs and their payloads, there is no single method that simultaneously measures all factors at once. Biological factors, like uptake and degradation, can complicate estimations. Initial off-target uptake by the liver and spleen potentially masks the lower levels of LNP distribution in the target tissues. Assessing the biodistribution and persistence at the steady state of an LNP’s nucleic acid payload in vivo by organ can be challenging. Qualitative measurements of the payload may be inaccurate due to degradation of nucleic acid or lack of sensitivity. Currently, qPCR is the standard methodology to quantitate the mRNA payload in a tissue, assuming the mRNA has been stochastically delivered to the organ. In parallel, LC–MS quantitates the individual lipid components with the assumption that the molar ratio of the measured lipid is the same as the whole LNP. Overall, there is room to improve the efficiency and accuracy of biodistribution due to complex biological barriers with the current indirect bioanalytical methods.
What factors influence off-target effects and premature payload release in LNP systems, and how can they be controlled?
Zamas & Aurora:
The composition of LNPs is a balancing act that influences off-target effects and premature payload release. Factors such as charge, size, lipid stability and immune response are carefully considered when formulating the LNP systems. Engineering ionizable lipids that neutralize particles in serum, while enabling charge during intracellular delivery, is important to control premature payload release. If the ionizable lipid becomes charged during circulation, the payload may be released prematurely, limiting therapeutic effects. Off-target effects in the liver and spleen are controlled with PEG-conjugated lipids. Increased PEG-conjugated lipids will decrease immune response and cellular uptake. However, too many PEG-conjugated lipids can cause limited targeted delivery. To control cellular uptake inhibition, PEG-conjugated lipids are designed to dissociate from the external layer, increasing cellular uptake with increased circulation time. Overall, the composition of the four components can be advantageous in controlling off-target effects and premature payload release in LNP systems.
What are the major analytical challenges associated with quantifying LNP lipids by LC–MS?
Lan:
The major challenges of LC–MS quantification for LNP lipids stem from their extraordinary chemical diversity. Developing a single, unified reversed-phase LC method capable of accurately quantifying all components simultaneously is often difficult. For example, ionizable lipid and PEG frequently exhibit very different chromatographic behaviors. PEGs are relatively polar and typically require a low organic starting condition to achieve adequate retention on a reversed-phase column.
In contrast, ionizable lipids are often much more hydrophobic and strongly retained. Under the low-organic conditions needed for PEGs, highly retained ionizable lipids may not be adequately washed from the column, leading to significant carryover and compromised assay robustness. Additional chromatographic challenges arise from the highly basic nature of many ionizable lipids. These compounds can interact with residual silanols on silica-based stationary phases, resulting in strong secondary ionic interactions that cause peak tailing, poor peak shape and reduced quantitative performance.
When analyzing biological samples, matrix effects present another major obstacle. Endogenous lipids naturally present in plasma, tissues or cellular extracts can compete with analytes during electrospray ionization (ESI). Highly abundant matrix lipids may suppress or, in some cases, enhance the ionization of co-eluting analytes, potentially affecting assay accuracy and precision. Consequently, robust chromatographic separation and the use of stable isotope labelled internal standards are often critical for reliable quantification.
Finally, the various lipid components within an LNP formulation can exhibit dramatically different ionization efficiencies. Most ionizable lipids produce strong MS responses and are readily detected at low concentrations, whereas neutral lipids, such as cholesterols lack readily ionizable functional groups and often exhibit relatively poor sensitivity. As a result, a single LC–MS assay may face the challenge of detector saturation for highly responsive ionizable lipids while still struggling to achieve adequate sensitivity for cholesterol and other poorly ionizable components.
What are the current approaches for LC–MS method development of LNP lipids?
Lan:
Successful LC–MS method development for LNP lipid quantification begins with a thorough understanding of the physicochemical properties of each lipid class. Because LNP formulations typically contain lipids with widely varying polarity, hydrophobicity and ionization characteristics, method development often requires balancing the analytical needs of multiple components. For sample preparation, protein precipitation is frequently employed when a single assay is desired for the simultaneous analysis of multiple lipid classes. However, when lower quantification limits, improved specificity, or reduced matrix effects are required, biological samples are often further purified using liquid-liquid extraction (LLE) or solid-phase extraction (SPE). These approaches can significantly improve assay sensitivity and robustness by reducing endogenous interferences and concentrating the analytes of interest.
Chromatographically, reversed-phase UPLC methods utilizing C8 or C18 stationary phases remain the most common approach for separating LNP lipids based on hydrophobicity. Mobile phase optimization typically involves binary gradients of water and organic solvents, such as acetonitrile and alcohols, supplemented with volatile modifiers such as formic acid or ammonium acetate. These additives help control analyte ionization, promote desirable adduct formation, and minimize secondary interactions between ionizable lipids and the stationary phase.
For mass spectrometric detection, ESI is the predominant ionization technique. Highly ionizable lipids are commonly quantified using multiple reaction monitoring (MRM) on triple quadrupole mass spectrometers, providing excellent sensitivity, selectivity and quantitative linearity. In contrast, neutral, or poorly ionizable lipids, such as cholesterol, often require additional optimization. Strategies may include fine-tuning ESI source parameters, employing alternative ionization techniques such as atmospheric pressure chemical ionization (APCI), or introducing derivatization steps during sample preparation to improve ionization efficiency and overall assay sensitivity.
mRNA quantification by PCR or bDNA
In quantifying mRNA therapeutics, has one platform, such as RT-qPCR or bDNA, become the industry standard?
Susan:
Both RT-qPCR (reverse transcription quantitative PCR) and bDNA (branched DNA) platforms have been successfully used to measure mRNA therapeutics. When performing RT-qPCR, total RNA is usually extracted, quantitated, and then converted to cDNA by reverse transcriptase, followed by exponential amplification and quantitative measurement. Alternatively, in the bDNA method, there is a direct hybridization of capture primers to the assay plate or beads, and target probes with branching structures for linear signal amplification and detection. We have focused more on RT-qPCR, as this methodology is quite robust, has very high sensitivity, a wide dynamic range, and a high tolerance to background mRNA interference. Although the bDNA platform avoids the need for nucleic acid extractions, it is often noted that there are edge effects on the assay plates and high %CVs between replicates can occur.
Additionally, the dynamic ranges achieved for bDNA are smaller than those of RT-qPCR, and there are vendor limitations to bDNA assay design and reagents. In general, we have observed the sponsors we support at QPS moving towards using RT-qPCR more frequently, although the bDNA platform continues to be used for measuring mRNA therapeutics. A recent article authored by the AAPS Bioanalytical Community PCR Working Group discusses assay design, development and validation of RT-qPCR methods to measure mRNA therapeutics.
In addition to RT-qPCR and bDNA, does digital PCR have a role in the mRNA bioanalysis space?
Susan:
Although RT-digital PCR (RT-dPCR) has a smaller dynamic range compared to RT-qPCR, it does have a higher tolerance to PCR inhibition. For this reason, it can be useful in biodistribution studies, having more complex starting matrices that can be prone to such PCR inhibition (e.g., faeces, tissues). However, we have successfully supported many biodistribution studies using RT-qPCR, as inhibitors to PCR inhibition can easily be overcome by sample dilutions. Another consideration is cost, as digital PCR assay reagents are significantly more expensive than qPCR reagents. Going forward, it will be interesting to see if additional innovations to increase the dynamic range of dPCR to similar ranges experienced with RT-qPCR, such as Countable PCR by Countable Labs (CA, USA), will be widely adopted by the bioanalysis industry.
