(T0930-05-28) Cellular Mechanisms Underlying Species-Related Differences in Effective Delivery of LNP-mRNA

Purpose: Pharmacological evaluation of lipid nanoparticle-mRNA (LNP-mRNA) relies on time- and resource-intensive in vivo studies in rodents and non-human primates, yet little is known about how well mechanisms of LNP mediated mRNA delivery are conserved across species. Importantly, it remains unclear if outcomes in preclinical models provide a reliable basis for the projection of potency and dose-response of LNP-mRNA therapies in humans. We hypothesized that species-related differences in functional delivery of LNP-mRNA could be recapitulated and mechanistically explained using physiologically relevant in vitro cellular systems derived from organs of biotherapeutic interest, such as the liver.

Methods: To compare LNP-mediated transfection efficiency in preclinical species and humans, we employed primary hepatocytes isolated from rat, cynomolgus monkey (cyno) and human livers. We adopted a “sandwich-culture” technique to maintain hepatocyte viability and ameliorate cell metabolic remodeling with time in culture (up to 3 days). To investigate the potential species-dependence of cellular processes governing LNP-mRNA transfection, we used LNPs that were formulated, in addition to phospholipids, cholesterol and a helper lipid, with D-Lin-MC3-DMA (MC3) ionizable lipid, due to its documented role in hepatic LNP delivery. Luciferase-encoding mRNA was encapsulated in the LNPs (MC3-Luciferase LNPs) since luciferase is commonly used in preclinical studies for characterizing transgene delivery. Doses of MC3-Luciferase LNP for in vitro transfection were selected to approximate typical in vivo exposure range, estimated as LNP-mRNA quantity per hepatocyte. For quantification of transgene expression in primary hepatocytes treated with MC3-Luciferase LNPs, we developed a calibration approach for correlating relative units of luminescence generated by luciferase enzymatic activity to luciferase protein concentration in hepatocyte lysates. The luciferase measurements were then normalized to the cell metabolic activity, determined using an AlamarBlue™ assay. To assess MC3-Luciferase LNP uptake in species hepatocytes, luciferase mRNA was quantified by dPCR and normalized to house-keeping gene expression, beta-actin. To distinguish between cell membrane-associated and intracellular LNPs, flow- and imaging flow- cytometry analyses were conducted with fluorescently labeled MC3-Luciferase LNPs, generated by incorporating rhodamine-labeled cholesterol (structural lipid) into particles’ formulation.

Results: Measurements of transgene level after LNP treatment of rat, cyno and human hepatocytes showed a linear dose-response in luciferase expression across all three species, however, the magnitude of expression and time kinetics varied as follows: In human and rat hepatocytes, luciferase protein was expressed at a comparably high level with biphasic kinetics, whereas only minimal luciferase signal was detected in cyno hepatocytes (Fig. 1). Additionally, in human and, to a lesser extent, cyno hepatocytes, donor-to-donor variability in the efficiency of LNP transfection was observed. Phenotypic differences in primary cells from different individuals are anticipated and, if better understood, might inform on the biological pathways impacting LNP interactions with target cells. Investigation of the initial steps required for effective LNP transfection uncovered species-dependence in cellular processes governing LNP-mRNA uptake in primary hepatocytes. Specifically, measurements of luciferase mRNA in cells shortly after LNP addition, reflected the differences seen in the subsequent protein expression, with a higher level in human and rat as compared to cyno samples, indicating less mRNA was successfully delivered to cyno hepatocytes (Fig. 2A). Indeed, experiments distinguishing between membrane associated and intracellular LNPs, showed that human and rat hepatocytes internalized LNPs to a higher extent as compared to cyno cells (Fig. 2B and 2C). However, surprisingly, there was no apparent correlation between the efficiency of LNP uptake and the propensity of LNP association to the cell membrane (Figs. 2 and 3). Cyno hepatocytes exhibited the highest magnitude of LNP membrane association, as determined by frequency of LNP-positive cells and the relative LNP signal per cell (Fig. 3). These data suggest there are species dependent differences in the membrane proteins that facilitate LNP binding and mediate the pathways of LNP internalization in hepatocytes. Proteomics studies are ongoing to identify species differences in the membrane protein composition underlying these critical processes.

Conclusion: Our data, obtained by comparing primary hepatocytes from rodents, non-human primates and humans, showed that the efficiency of LNP-mRNA transfection (dose- and time- dependency) substantially varied across species, and provides new evidence that cell-intrinsic mechanisms, such as differential membrane association and intracellular uptake, play a critical role in LNP delivery in preclinical species and humans. We believe these studies in physiologically relevant cellular systems, can generate key information in support of in vitro-to-in vivo extrapolation aiming to inform LNP design and fill a critical gap in the clinical translation of LNP therapies to patients.

Acknowledgements: All authors are employed by Pfizer, Inc. Primary hepatocyte samples (human, cyno, rat) were obtained from a 3rd party reputable sample repository.

Fig. 1: Species-specific difference in transgene expression in MC3-Luciferase LNP-transfected human, cyno and rat hepatocytes. Cells were treated with different doses of MC3-Luciferase mRNA LNPs for the indicated time periods, as specified in the figure. Data shown as mean and SD of technical duplicates. *, below lower limit of detection and #, below lower limit of quantification for this assay. Each row represents a different donor (lot of cells).

Fig. 2: Differences in LNP intracellular uptake in species hepatocytes. Primary hepatocytes from human, rat and cyno livers were treated with 2 pg/cell of unlabeled (A), or fluorescently-labeled MC3-Luciferase mRNA-LNPs, to differentiate between intracellular and membrane associated LNPs (B,C). (A) Luciferase mRNA was measured by dPCR at indicated time points after LNP addition, normalized to beta-actin house-keeping gene. (B) Representative images of hepatocytes incubated with LNPs (red) for 120 min at 37°C, generated by imaging flow cytometry. Note predominantly surface signal in cyno cells (arrow). (C) Time-dependent LNP association with cells was measured by flow cytometry by analyzing difference in fluorescence signal in samples incubated with LNPs on ice and at 37°C (allows active internalization). Graphs summarize time-dependent changes in the extent of LNP uptake, expressed as the ratio of the fluorescence signal (MFI) between the conditions. Data shown as mean and SD of biological replicates, n=2.

Fig. 3: Comparison of dose-dependent LNP binding in hepatocytes from different species. Cells were incubated with increasing doses of fluorescently-tagged MC3-Luciferase LNPs, as specified in the figure, for 60 min on wet ice to allow membrane association and minimize internalization. Frequency of LNP positive cells (A) and relative LNP signal per cell (B) of the labeled LNP was analyzed by flow cytometry.