Materials

Cholesterol (Cat. No. 57-88-5) and DSPC (Cat. No. 816-94-4) were purchased from Lipoid GMBH (Ludwigshafen, Germany). ALC-0315 (Cat. No. 06040008600) and ALC-0159 (Cat. No. 06020112302) were acquired from SINOPEG (Xiamen, China). Ferric chloride hexahydrate (Cat. No. 701122), ammonium thiocyanate (Cat. No. 221988), and NH2-PEG10000-NH2 (Cat. No. 8218815000) were obtained from Sigma-Aldrich (St. Louis, MO, USA). 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS, Cat. No. ST1145), 3,3′,5,5′-tetramethylbenzidine dihydrochloride hydrate (TMB 2HCl, Cat. No. ST1708) and nonfat powdered milk (Cat. No. P0216) were purchased from Beyotime Biotechnology (Shanghai, China). 1,1-dioctadecyl-3,3,3,3-tetramethylindotricarbocyaine iodide (DiR, Cat. No. 100068-60-8) was purchased from Shanghai Maokang Biotechnology Co., Ltd. (Shanghai, China). Maxisorp 96-well microplates (Cat. No. 44-2404-21) were acquired from Nalge-Nunc International (Rochester, NY, USA). D-Luciferin (Cat. No. 88293) were purchased from Thermo Fisher Scientific (Waltham, MA, USA). Firefly luciferase mRNA (Cat. No. L-7702-1000) was obtained from Trilink Biotechnologies (San Diego, CA, USA). Rat anti-PEG IgM (Cat. No. rAGP6-PABM-A; Clone No. rAGP6) and rat anti-PEG IgG (Cat. No. r33G-PABG-A; Clone No. r33G) were acquired from Academia Sinica (Taipei, China). Peroxidase-conjugated affinipure rabbit anti-rat IgM µ-chain specific (Cat. No. 312-035-020; Clone No. 312-035-020) and peroxidase-conjugated affinipure donkey anti-rat IgG (H + L) (Cat. No. 712-035-150; Clone No. 712-035-150) were obtained from Jackson ImmunoResearch Laboratories Inc (West Grove, PA, USA).

Preparation of LNP, DiR-LNP, and DiR-LU@LNP

LNP, DiR-LNP and DiR-LU@LNP were formulated according to a previously reported protocol29. First, the ethanol phase was prepared by dissolving ALC-0315, DSPC, cholesterol, and ALC-0159 at a molar ratio of 46.3:9.4:42.7:1.6. Specifically, DiR was added into the ethanol phase at 0.4% mol for preparation of DiR-LNP and DiR-LU@LNP. Regarding the aqueous phase, it was prepared using 20 mM citrate buffer (pH4.0) for LNP and DiR-LNP formulations, with additional firefly luciferase mRNA added for DiR-LU@LNP formulation. Subsequently, the ethanol phase was mixed with the aqueous phase at a flow rate ratio of 1:3 (ethanol:aqueous) through a microfluidic mixer (Precision Nanosystems Inc., Canada). Afterward, the obtained nanoparticle solutions were dialyzed against 10 × volume of PBS (pH7.4) through a tangential-flow filtration (TFF) membrane with 100 kD molecular weight cut-off (Sartorius Stedim Biotech, Germany) for at least 18 h. Finally, nanoparticle solutions were concentrated using Amicon ultra-centrifugal filters (EMD Millipore, Billerica, MA, USA), passed through a 0.22 µm filter and stored at 2–8 °C until use.

Characterization of LNP, DiR-LNP, and DiR-LU@LNP

LNP, DiR-LNP, and DiR-LU@LNP were examined for their hydrodynamic size (Z-average), polydispersity index (PDI), and zeta potential with DLS (Zetasizer Nano ZS, Malvern Instruments Ltd, Malvern, UK) equipped with a solid state HeNe laser (λ = 633 nm) at a scattering angle of 173˚. Nanoparticles were either added into PBS (pH7.4) for Z-average and PDI measurements, or added into ultrapure water for determination of zeta potential. Three independent experiments were conducted, with each type of LNP examined at 25 ℃ for 10 s (pre-equilibration for 2 min) and repeated at least 10 times in disposable cuvettes (for Z-average and PDI) or zeta cuvettes (for zeta potential). The obtained data were presented as “mean ± standard deviation”. To further assess their stability in serum (simulating in vivo environment in this study), LNP, DiR-LNP, and DiR-LU@LNP were diluted to 1:100 with PBS containing 10% rat serum and then incubated at 37 °C for 24 h. Subsequently, 1 mL of diluted LNP, DiR-LNP, and DiR-LU@LNP were respectively collected at designated time points (1 h, 6 h, 12 h, and 24 h post-incubation), followed by characterization of Z-average and PDI with DLS. Three independent experiments were conducted, with each type of LNP examined at 37 ℃ for 10 s (pre-equilibration for 2 min) and repeated at least 10 times in disposable cuvettes. The obtained data were presented as “mean ± standard deviation”. Furthermore, the morphological characteristics of LNP, DiR-LNP, and DiR-LU@LNP were observed with Cryo-TEM. In brief, 3 μL of each LNP sample was deposited onto a holey carbon grid that was glow-discharged (Quantifoil R1.2/1.3) and vitrificated using a Vitrobot Mark IV System (FEI/Thermo Scientific, Waltham, MA, USA). Cryo-TEM imaging was performed on a Talos F200C device (FEI/Thermo Scientific, Waltham, MA, USA) equipped with a 4k × 4k Ceta camera at 200 kV accelerating voltage in the Center of Cryo-Electron Microscopy, Zhejiang University.

In addition, the phospholipid (DSPC) concentrations of LNP, DiR-LNP, and DiR-LU@LNP solutions were quantified via Steward’s assay for further calculation of LNP doses30. Briefly, ammonium ferrothiocynate was prepared by dissolving 27.03 mg ferric chloride hexahydrate and 30.4 mg ammonium thiocyanate in 1 mL of distilled water. 10 μL of the lipid sample was added to 990 μL of chloroform, followed by addition of 1 mL of ammonium ferrothiocynate. The obtained mixture was vortexed for 60 s and then centrifuged at 300 × g for 15 min at room temperature. The bottom chloroform layer was transferred to a glass cuvette and the absorbance was measured at 470 nm using a Unicam UV500 Spectrophotometer (Thermo Electron Corporation, USA). Standard curves for DSPC lipid were obtained and used for calculation of the phospholipid concentrations of LNP, DiR-LNP, and DiR-LU@LNP solutions. Eventually, the various doses of LNP tested in the animal experiments were calculated based on the phospholipid (DSPC) exposure amount per dose of related drug (see below for details).

Determination of LNP dosing protocols

According to the official drug information and clinical protocols, the mPEG2000 exposure amount for each injection of Comirnaty® in adults is 0.0406 mg18, while that for Onpattro® is 10.6434 mg17 (262 folds of that of Comirnaty®). Different from Comirnaty® and Onpattro®, the detailed LNP composition of Spikevax® including the molar lipid ratios is not included in the official drug information published in 202219. With a postulation that PEG2000-DMG is the only lipid contained in LNP of Spikevax®, we estimated that the possible “maximum” mPEG2000 exposure for each injection would be 1.542 mg (37.98 folds of that of Comirnaty®) referred to the clinical protocols of Spikevax®19 (Supplementary Methods; Supplementary Table 1).

Based on the above calculation and estimation, three mPEG2000 dosages including 0.0406 mg/dose (low dose), 1.542 mg/dose (middle dose), and 10.6434 mg/dose (high dose) were obtained. These dosages not only have an appropriate gradient ratio of 1:38:262, but cover the broad range of PEG exposure amount upon each injection of approved LNP-delivered drugs.

Next, we calculated the phospholipid (DSPC) contained in LNP of Comirnaty®, which is 0.09 mg for each injection in adults18. According to the animal-human dose exchange algorithm: animal equivalent dose = human dose × Km ratio (6.2 for rat)31, clinically relevant LNP doses for rats were as follows: low dose (L-LNP), 0.009 mg phospolipid/kg (0.09 mg/60 kg × 6.2); middle dose (M-LNP), 0.342 mg phospholipids/kg (0.009 × 38); high dose (H-LNP), 2.358 mg phospholipids/kg (0.009 × 262).

The clinical protocols of Comirnaty® were essentially simulated in this study. That is, LNP was administrated through intramuscular injection for two separate injections, with a 21-day interval (same as routine Comirnaty® vaccination).

Animals

10–12-week-old female Wistar rats were purchased from Hangzhou Medical College (Hangzhou, China), and maintained in the Laboratory Animal Center of Zhejiang University under controlled environmental conditions at constant temperature, humidity, and a 12-h dark/light cycle. Rats were given ad libitum access to a standard rat chow and water, and were acclimated for at least 7 days. All animal experiments were approved by the Laboratory Animal Welfare and Ethnics Committee of Zhejiang University and carried out in accordance with the guidelines of the committee (approval No. ZJU20210071).

Administration of LNP simulating clinical protocols and collection of serum samples for ELISA

Wistar rats were randomly divided into a Control group (n = 8) and three LNP-treated groups (n = 15). At Day 0, LNP-treated groups were intramuscularly injected with 0.009 mg phospholipids/kg LNP (L-LNP group), 0.342 mg phospholipids/kg LNP (M-LNP group), and 2.358 mg phospholipids/kg LNP (H-LNP group), respectively, while the Control group only received PBS. At Day 21, rats in each experimental group received same treatment as the initial injection. Peripheral blood samples of each rat were collected successively via the retro-orbital venous plexus at Day 0, 3, 5, 7, 14, 21, 24, 26, 28, 35, 42, and 49. All blood samples were centrifuged at 2000 × g for 15 min at 4 °C, and the serums were immediately stored at −80 °C for further quantification of anti-PEG antibody. At the experimental endpoint, the animals were humanely euthanized through intraperitoneal administration of 150 mg/kg of sodium pentobarbital salt (Cat. No. BCP07810, Biochempartner, Shanghai, China)32.

Quantification of anti-PEG IgM and anti-PEG IgG antibodies with ELISA

Maxisorp® 96-well microplates were coated with 5 μg/well NH2-PEG10000-NH2 in 100 µL of PBS overnight at 4 °C. Subsequently, plates were gently washed with 350 μL of washing buffer (0.05% (w/v) CHAPS in DPBS) for three times, followed by incubation with blocking buffer (5% (w/v) skim milk powder in DPBS, 200 μL/well) at room temperature for 1.5 h. Afterward, plates were rinsed with washing buffer for three times again. Then 100 μL of rat serum samples diluted 1:150 with dilution buffer (2% (w/v) skim milk powder in DPBS), together with seven serial dilutions of rat anti-PEG IgM standards (1.37, 4.12, 12.35, 37.04, 111.11, 333.33, and 1000.00 ng/mL) or rat anti-PEG IgG standards (0.05, 0.15, 0.46, 1.37, 4.12, 12.35, and 37.04 ng/mL), were added into anti-PEG IgM or anti-PEG IgG detection plates in duplicate and further incubated for 1 h at room temperature. After five successive washes, 50 µL of diluted peroxidase-conjugated affinipure rabbit anti-rat IgM µ-chain specific and peroxidase-conjugated affinipure donkey anti-rat IgG (H + L) antibodies were respectively added at 0.08 μg/mL to the corresponding plates and incubated for 1 h at room temperature. Again, unbounded antibodies were removed by five washes, followed by incubation with 100 µL of TMB for 30 min at room temperature. Finally, HRP-TMB reaction was stopped with 100 μL of 2 N H2SO4, and the absorbance was measured at 450 nm with a microplate reader (Thermo Fisher Scientific, Waltham, MA, USA), using 570 nm as a reference wavelength. Anti-PEG IgM and anti-PEG IgG standard curves in each batch of ELISA were constructed by plotting the average corrected absorbance values (OD450 nm-OD570 nm) and corresponding antibody concentrations with Four Parameter Logistic (4PL) curve fit using Origin 2021 software (OriginLab Corporation, Northampton, Massachusetts, USA). The goodness of fit of each standard curve was measured by the coefficient of determination (R2)33. Concentrations of anti-PEG IgG and IgM antibodies in serum samples were calculated based on corresponding standard curves. In addition, intra-assay precision was evaluated by calculating the Coefficient of Variation (CV% = (Standard deviation/Mean) × 100%) for all detectable standards and samples in all batches of ELISA23,24. Inter-assay precision was determined by calculating the Coefficient of Variation for serially diluted anti-PEG antibody standards among all batches of ELISA23,24. The acceptance criteria for mean intra-assay and inter-assay Coefficient of Variation (CV%) of ELISA are < 20% and < 25%, respectively23,24.

Biodistribution of PEGylated LNP in major organs of Wistar rats

Wistar rats were randomly divided into a Control group and three DiR-LU@LNP-treated groups (n = 6). At Day 0, LNP-treated groups were intramuscularly injected with 0.009 mg phospholipids/kg DiR-LU@LNP (L-LNP group), 0.342 mg phospholipids/kg DiR-LU@LNP (M-LNP group) and 2.358 mg phospholipids/kg DiR-LU@LNP (H-LNP group), respectively, while the Control group only received PBS. At Day 21, rats in each experimental group received same treatment as the initial injection. Six hours after the first and second injections, three rats in each group were administered intraperitoneally with D-luciferin at a dose of 150 mg/kg. Rats were sacrificed by cervical dislocation 15 min after D-luciferin administration and immediately dissected for collection of several primary organs, including heart, liver, spleen, lung, kidneys, draining lymph node, and muscle at the injection site. Whole-organ/tissue imaginings for DiR fluorescence (Excitation/Emission: 748 nm/780 nm) and firefly luciferase bioluminescence were performed with IVIS Spectrum imaging system and analyzed with Living Image software (Caliper Life Sciences, Waltham, Massachusetts, USA).

Blood clearance of PEGylated LNP in Wistar rats

Wistar rats were randomly divided into a Control group and three DiR-LNP-treated groups (n = 3). At Day 0, LNP-treated groups were intramuscularly injected with 0.009 mg phospholipids/kg DiR-LNP (L-LNP group), 0.342 mg phospholipids/kg DiR-LNP (M-LNP group) and 2.358 mg phospholipids/kg DiR- LNP (H-LNP group), respectively, while the Control group only received PBS. At Day 21, rats in each experimental group received same treatment as the initial injection. Peripheral blood samples were respectively collected from the retro-orbital venous plexus at 5 min, 30 min, 1 h, 3 h, 6 h, 10 h, 24, and 48 h after the first and second injections. Then blood samples were centrifuged at 2000 × g at 4 °C for 15 min, and serum samples were isolated and immediately stored in dark at −80 °C. DiR fluorescence associated with LNP in serum samples was detected by fluorescent spectroscopy on a Spectramax ID5 (Molecular Devices, San Jose, California, USA) at excitation/emission wavelengths of 748/780 nm. At the experimental endpoint, the animals were humanely euthanized through intraperitoneal administration of 150 mg/kg of sodium pentobarbital salt (Cat. No. BCP07810, Biochempartner, Shanghai, China)32.

Data presentation and statistical analysis

All data were presented as “mean ± standard deviation”. Concentrations (ng/mL) of anti-PEG IgM and anti-PEG IgG were analyzed after log10 transformation, and their differences among various groups at each time point were analyzed with Mann–Whitney U test using R 4.0.5 (R Software, Boston, MA, USA), with P values adjusted with FDR (false discovery rate) method. Changing curves of average level of anti-PEG antibody over time for various doses were fitted by the R package called “ggalt”. Profile analysis was performed to examine whether the overall trends of changing curves of average level of anti-PEG antibody over time between every two groups were equal. The analysis included two parts: parallel test and coincidence test. Only when the two changing curves of average level of anti-PEG antibody met both parallel and coincidence test (P > 0.05), the overall trend of two changing curves of average anti-PEG antibody level was considered as no statistical difference. According to factorial design (group × time) and repeated measures of antibody level, linear mixed models (LMM) were conducted to compare the change rates and average levels of anti-PEG antibody across groups, with all time points included. Several variables, including group (indicating mean differences in the average levels of anti-PEG antibody), time, time2, number of injections, and interaction term of group and time (indicating mean differences in the change rates of anti-PEG antibody) as fixed effect and subject as random effect were considered in LMM.

In addition, ▲Anti-PEG IgM (Log10 CONC) was defined as Anti-PEG IgM (Log10 CONC2nd injection) (log10-transformed concentration of anti-PEG IgM induced during the second injection cycle) subtracting corresponding Anti-PEG IgM (Log10 CONC1st injection) (log10-transformed concentrations of anti-PEG IgM induced during the first injection cycle). Similarly, ▲Anti-PEG IgG (Log10 CONC) was calculated by subtracting Anti-PEG IgG (Log10 CONC1st injection) (log10-transformed concentrations of anti-PEG IgG induced during the first injection cycle) from the corresponding Anti-PEG IgG (Log10 CONC2nd injection) (log10-transformed concentration of anti-PEG IgG induced during the second injection cycle). Differences in ▲Anti-PEG IgM (Log10 CONC) or ▲Anti-PEG IgG (Log10 CONC) among various groups at each time point were analyzed with Mann–Whitney U test using R 4.0.5, with P values adjusted for FDR (false discovery rate). Changing curves of average level of ▲Anti-PEG IgM (Log10 CONC) or ▲Anti-PEG IgG (Log10 CONC) over time for various doses were fitted by the R package called “ggalt”. Profile analysis was performed to examine whether the overall trends of changing curves of average level of ▲Anti-PEG IgM (Log10 CONC) or ▲Anti-PEG IgG (Log10 CONC) over time between every two groups were equal. The analysis included two parts: parallel test and coincidence test. Only when the two changing curves of average level of ▲Anti-PEG IgM (Log10 CONC) or ▲Anti-PEG IgG (Log10 CONC) met both parallel and coincidence test, the overall trend of the two changing curves of average level of ▲Anti-PEG IgM (Log10 CONC) or ▲Anti-PEG IgG (Log10 CONC) was considered as no difference. According to factorial design (group × time) and repeated measures of antibody level, LMM were conducted to compare the change rates and average levels of ▲Anti-PEG IgM (Log10 CONC) or ▲Anti-PEG IgG (Log10 CONC) across groups, with all time points included. Several variables, including group (indicating mean differences in the average levels of ▲Anti-PEG IgM (Log10 CONC) or ▲Anti-PEG IgG (Log10 CONC)), time, time2, and interaction term of group and time (indicating mean differences in the change rates of ▲Anti-PEG IgM (Log10 CONC) or ▲Anti-PEG IgG (Log10 CONC) levels) as fixed effect and subject as random effect were considered in LMM. After performing the Shapiro-Wilk test to check for normality and the F test to check for variance homogeneity, data obtained in the biodistribution and blood clearance study were analyzed using multiple unpaired t tests with correction for multiple comparisons using Prism 9.2.0 (GraphPad Software, San Diego, USA). P < 0.05 was considered statistically significant.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.