key: cord-1021197-kuxachyf
authors: Zheng, Jinpeng; Lu, Caihong; Ding, Yaning; Zhang, Jinbang; Tan, Fangyun; Liu, Jingzhou; Yang, Guobao; Wang, Yuli; Li, Zhiping; Yang, Meiyan; Yang, Yang; Gong, Wei; Gao, Chunsheng
title: Red blood cells-hitchhiking mediated pulmonary delivery of ivermectin: effects of nanoparticles properties
date: 2022-04-04
journal: Int J Pharm
DOI: 10.1016/j.ijpharm.2022.121719
sha: 423012539147232ebe879390f5107e011e4e4165
doc_id: 1021197
cord_uid: kuxachyf

Recent studies have demonstrated that ivermectin (IVM) exhibits antiviral activity against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative virus of coronavirus disease (COVID-19). However, the repurposing of IVM for treatment of COVID-19 has presented challenges primarily due to the low IVM plasma concentration after oral administration, which was well below IC(50). Here, a red blood cells (RBCs)-hitchhiking strategy was used for the targeted delivery of IVM-loaded nanoparticles to the lung. IVM-loaded poly (lactic-co-glycolic acid) (PLGA) nanoparticles (IVM-PNPs) and chitosan-coating IVM-PNPs (IVM-CSPNPs) were prepared and adsorbed onto RBCs. Both RBCs-hitchhiked IVM-PNPs and IVM-CSPNPs could significantly enhance IVM delivery to lungs, improve IVM accumulation in lung tissue, inhibit the inflammatory responses and finally significantly alleviate the progression of acute lung injury. Specifically, the redistribution and circulation effects were related to the properties of nanoparticles (NPs). RBCs-hitchhiked cationic IVM-CSPNPs showed longer circulation time, slower accumulation and elimination rates, and higher anti-inflammatory activity than RBCs-hitchhiked anionic IVM-PNPs. Therefore, RBC-hitchhiking provides an alternative strategy to improve IVM pharmacokinetics and bioavailability for repurposing of IVM to treat COVID-19. Furthermore, according to different redistribution effects of different NPs, RBCs-hitchhiked NPs may achieve various accumulation rates and circulation times for different requirements of drug delivery.

Coronavirus infectious disease 2019 , caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), an increasingly prevalent viral contagion at the global scale, causes acute respiratory distress syndrome (ARDS). (Mahmudpour et al., 2020; Mohanty et al., 2021) . Numerous clinical trials are being reviewed for the use of different drugs, biologics and vaccines in the treatment of COVID-19 (Mohanty et al., 2021) . However, there is currently no dependable treatment or suitable therapeutic for COVID-19 (Kaur et al., 2021) .

A recent study found that ivermectin (IVM) inhibits the replication of SARS-CoV-2 in vitro (Caly et al., 2020) . IVM is an FDA-approved broad-spectrum antiparasitic agent (González Canga et al., 2008) , which also exhibits antiviral activity against various RNA (COVID-19, Zika virus, HIV-1, Dengue virus) as well as DNA viruses (Equine herpesvirus type 1, Pseudorabies virus, Porcine circovirus 2) (Heidary and Gharebaghi, 2020; Li et al., 2021; Surnar et al., 2019) . IVM is believed to act through inhibition of viral proteins, including importin α/β1 heterodimer and integrase protein (Wagstaff et al., 2011; Yang et al., 2020) . Furthermore, IVM may have antiinflammatory properties and could play a supportive adjuvant role to prevent acute lung injury (ALI) in SARS-CoV-2 infection (Formiga et al., 2021; González Canga et al., 2008; Yan et al., 2011; Zhang et al., 2008) . However, despite its promising antiviral and preliminary anti-inflammatory activities, the development of IVM formulations presents challenges (Formiga et al., 2021) . Numerous IVM clinical trials have been registered and conducted to validate its use in the treatment of COVID-19. The results of some trails have shown a potential decrease in the length of hospital stay and survival benefit (Ahmed et al., 2021; Heidary and Gharebaghi, 2020) . In contrast, other trials have shown no significant difference between the IVM group and placebo group (or untreated group) (Krolewiecki et al., 2021; López-Medina et al., 2021) . There is evidence that the mean IVM plasma concentration levels correlate with antiviral effect (Krolewiecki et al., 2021) . The free plasma concentrations of IVM do not reach the IC 50 even for a dose level 10x higher than the approved dose due to its high binding rate to plasma proteins (93%) (Jermain et al., 2020; Peña-Silva et al., 2021; Schmith et al., 2020) . Moreover, excessive IVM causes dose-dependent adverse reactions and toxicity in animals, including ataxia, tremor, central nervous system depression, and coma, which usually lead to death Trailovic and Nedeljkovic, 2011) .

Considering these challenges, novel delivery strategies are important for repurposing of IVM for the treatment of COVID-19.

Compared with traditional drug delivery strategies, targeted drug delivery can greatly enhance the efficiency of drug delivery, increase local drug concentration in target organs/tissues, and enhance therapeutic efficacy, while reducing systemic exposure and side-effects of drug . Therefore, the delivery of IVM to the lungs may alter drug distribution, increase drug concentration in lungs to overcome the drawbacks of oral administration, and at the same time reduce its side-effects. Nanoparticles (NPs) offer advantages in controlled and sustained release, improved bioavailability and enhanced drug's therapeutic efficacy (Onoue et al., 2014) . Moreover, NPs can be easily modified for targeted drug delivery. However, ordinary NPs are rapidly removed from the bloodstream by the mononuclear phagocyte system (MPS) and the targeting capability is limited due to the complex environment in vivo Perry et al., 2012; Yang et al., 2021) .

Recently, a hybrid delivery strategy, namely red blood cells (RBCs)-hitchhiking, has offered an optimal blend of natural and synthetic systems to address these limitations (Anselmo et al., 2013; Villa et al., 2016a; Brenner et al., 2018) . NPs adsorb onto the surface of RBCs in a non-covalent manner to form NPs-adsorbed RBCs (RBC-NPs) in vitro. Then, NPs are scraped off the surface of RBCs because they fail to resist the shear stress between RBCs and small capillaries. These desorbed NPs mainly accumulate in the lungs, the first capillary bed where RBC-NPs pass through after intravenous injection (Brenner et al., 2021) . RBCs-hitchhiking, which combines NPs and RBCs carriers, improves drug solubility and enables parenteral delivery of insoluble drugs. Moreover, the sustained release ability of NPs and the long circulation property of RBCs can greatly prolong circulation and action time of drugs Glassman et al., 2020) . Importantly, RBCs-hitchhiking endows NPs with lung targeting abilities, changes the drug biodistribution, enhances drug accumulations in the lungs, and increases the drug concentration in regional lung tissue (Anselmo et al., 2013; Zelepukin et al., 2019) .

Through the RBCs-hitchhiking strategy, methylprednisolone has been 7 successfully delivered to the lungs in our previous research .

However, the effect of the properties of NPs on the delivery efficiency in vivo is unclear.

The lung targeting capability and circulation time of various NPs adsorbed on RBCs may vary greatly. The primary reason for these differences may be related to the binding force between the NPs and RBCs. In this study, various NPs with different materials, sizes and polymer coatings were prepared and the adsorption efficiency of NPs to RBCs was systematically investigated. Furthermore, in vivo circulation time, lung targeting capability, and anti-inflammatory activities of different RBCs-hitchhiked NPs were studied. prepared and adsorbed onto the RBC surface in a non-covalent manner to form a RBCsnanoparticles complex. The RBCs-hitchhiked nanoparticles were targeted delivered to the lungs to treat lung inflammation. The effects of the different properties of nanoparticles on their adhesion efficiency to RBCs, the circulation time, the lung targeting capability, and anti-inflammatory activities were investigated emphatically. 

PLGA nanoparticles (PNPs) with different materials and sizes were prepared using O/W emulsion solvent diffusion method (Li et al., 2019) . Briefly, PLGA (30 mg) was dissolved in acetone solution (10, 15, 20 and 30 mg/mL), then IVM (3 mg) or hydrophobic fluorescent probe (Cy5.5 or Cy7.5, 0.4 mg) was added and the mixture was used as the organic phase. When completely dissolved, the organic phase was 

Whole blood collected from the retro-orbital sinus was centrifuged (1000 g, 4C,

10 min) to separate plasma and RBCs. The packed RBCs were washed extensively with precooling phosphate buffered solution (PBS). RBCs at 10% hematocrit were incubated with NPs at 37C for 30 min, and unattached NPs were removed via centrifugation.

A standard hemagglutination assay was performed using V-shaped 96-well plates to examine the degree of hemagglutination after the adsorption of NPs (Zelepukin et al., 2019) . RBCs at 1% hematocrit were mixed with a known quantity of NPs. Then, RBC samples (50 μL) were added to V-shaped well and incubated at 37°C for 1 h. The aggregated RBCs form a 3D-network that prevents sedimentation to the bottom of Vshape instead of settling into a tight point in the center. Free RBCs without NPs were used as controls.

The RBC samples were dehydrated for imaging using a classic fixation technique (Zelepukin et al., 2019) . Briefly, RBC samples were fixed in 2.5% glutaraldehyde solution at 4°C overnight and transferred into a gradient ethanol solution for dehydration. Then, the samples were placed on silicon tape and sputter-coated with gold for observation by scanning electron microscopy (SEM, S-4800, Hitachi, Japan).

Besides, PNPs or CSPNPs labeled with Cy5.5 were adsorbed to RBCs (RBC-Cy5.5-PNPs and RBC-Cy5.5-CSPNPs, respectively) and then observed via a confocal laser scanning microscopy (CLSM, LSM 880, ZEISS, Germany).

The desorption of IVM-PNPs or IVM-CSPNPs from RBCs performed under shear stress was evaluated using a microfluid-controlled visual rheometer (Fluidicam Rheo, FORMULACTION, France) (Anselmo et al., 2013) . RBC samples were exposed to shear stress (1 Pa or 5 Pa) at 37°C for 15 min. 10% serum was added to RBC samples to prevent the re-adsorption of NPs. The adsorption efficiency of NPs to RBCs before or after the shear stress challenge was calculated.

To investigate the anti-phagocytosis ability of RBC-NPs, RAW264.7 cells (10 5 /dish) were seeded into a culture dish for 24 h. After incubation with Cy5.5-PNPs, Cy5.5-CSPNPs, RBC-Cy5.5-PNPs or RBC-Cy5.5-CSPNPs (10 μM) for 2 h, the cells were washed and fixed with 4% paraformaldehyde solution for 15 min and then incubated with Hoechst 33258 (10 µg/mL) for 10 min to stain the nuclei. The fluorescent images were measured via CLSM.

RBC-IVM-PNPs or RBC-IVM-CSPNPs of 1% hematocrit were resuspended in NaCl concentrations ranging from 0 mM to 154 mM for 30 min at 37°C. RBCs incubated in 154 mM NaCl or distilled water were used as negative (spontaneous lysis) or positive (total lysis) controls of hemolysis, respectively. The RBCs (1000 g, 5 min)

and NPs (18000 g, 20 min) in the samples were sequentially removed by centrifugation.

The hemoglobin in the supernatants was detected immediately with a plate reader by spectrophotometry at a wavelength of 540 nm.

RBC-IVM-PNPs or RBC-IVM-CSPNPs (50 μL) were diluted with PBS (5 mL)

containing 3 mM H 2 O 2 . Free RBCs subjected to the same condition were used as the negative control of hemolysis, and RBCs incubated in distilled water were used as the positive control (total lysis). All RBC samples were incubated at 37°C for 24 h.

Similarly, the RBCs (1000 g, 5 min) and NPs (18000 g 20 min) in samples were sequentially removed by centrifugation. The hemoglobin in the supernatants was detected immediately at 540 nm with the plate reader.

The Na + /K + -ATPase activity assay of free RBCs, RBC-IVM-PNPs, and RBC-IVM-CSPNPs was performed according to the protocol of the Na + /K + -ATPase kit (Zhang et al., 2018b) . All RBC samples were lysed in 50 × distilled water, and the activity of Na + /K + -ATPase in RBC lysate was measured. Meanwhile, the BCA protein quantification kit was used to measure the protein concentration in equal numbers of RBCs.

To determine the blood clearance of NPs-loaded RBCs, biotin-labeled RBC-IVM-PNPs and RBC-IVM-CSPNPs was prepared, and then injected into the donor mouse.

Then serial blood samples were collected from the retro-orbital sinus at certain time labeled RBCs at 0 h after injection was defined as 100% survival.

Healthy SD rats were randomly divided to five groups, free-IVM, IVM-PNPs, The plasma samples (100 μL) and IS (10 μL, 500 ng/ml) were mixed with ethyl acetate (900 μL) containing 10% (v/v) methyl tert-butyl ether. After vigorous vortexmixing (3 min) and centrifugation (14,000 rpm, 5 min), the organic layer was transferred and evaporated at 40°C for 30 min. The residue was then reconstituted with the mobile phase (100 μL). After centrifugation (14,000 rpm, 5 min), the supernatants (5 μL) were injected into the HPLC-MS/MS system for analysis.

The near-infrared fluorescent probe Cy7.5 labeled PNPs and CSPNPs (Cy7.5-PNPs and Cy7.5-CSPNPs) were applied to evaluate the lung targeting efficiency (Hu et al., 2020) . Healthy ICR mice were injected with Cy7.5-PNPs, Cy7.5-CSPNPs, RBC-Cy7.5-PNPs and RBC-Cy7.5-CSPNPs via the caudal vein (n = 3). After 0.5 h and 2 h, mice were anesthetized with 3% isoflurane and then observed (750/820 nm) via IVIS® Spectrum CT (IVIS® Spectrum, PerkinElmer, USA). After imaging, the mice were sacrificed and their major organs (liver, spleen, kidneys, heart, and lungs) were harvested and imaged.

Additionally, free IVM, IVM-PNPs, IVM-CSPNPs, RBC-IVM-PNPs, or RBC-IVM-CSPNPs were administered into healthy SD rats via the caudal vein (500 μg/kg, n = 3). The rats were anesthetized and sacrificed 0.5 h and 2 h after administration.

Then, the major organs were harvested, weighed, and homogenized. IVM concentrations in tissues were detected by HPLC-MS/MS as described above.

The ALI mouse model was established as reported previously (Fang et al., 2017; Ju et al., 2018) . LPS from Escherichia coli (5 mg/kg) or PBS (control) was instilled intratracheally, and 5 h after LPS challenge, mice were injected intravenously with different IVM-loaded formulations (500 μg/kg, n = 5). Mice were sacrificed 24 h after administration. After the right lung was ligated, the left lung bronchoalveolar lavage fluid (BALF) was collected and centrifuged (3000 rpm, 10 min). Subsequently, the cells in the pellet were counted with a hemocytometer, and the levels of total protein and pro-inflammatory cytokines (TNF-α and IL-6) in the supernatant were determined using the enhanced BCA protein assay kits and ELISA kits, respectively (Ju et al., 2018) .

A part of the right lung was used for weighing and drying (60°C, 3 d) to calculate the wet-to-dry weight ratio. The remaining right lungs were stained with hematoxylin and eosin (H&E) to assess the degree of lung damage.

The results are expressed as means ± standard deviation (SD), unless otherwise stated. Statistical analysis of the pharmacokinetic parameters was performed using unpaired two-tailed t-test. One-way analysis of variance (ANOVA) was performed to determine significant differences between different groups. A p value of < 0.05 was considered as statistically significant. P values < 0.05, < 0.01, < 0.001 are denoted as *, **, and ***, respectively, and NS: non-significant.

The NPs were successfully prepared and the mean sizes, zeta potential, EE, and DL were shown in Table 1 . Resulting IVM-PNPs exhibited mean particle sizes in the range 132.13 ± 1.08 to 201.05 ± 2.58 nm (Fig. S1 ). Compared to uncoated IVM-PNPs, S2 ). Increased particle sizes and positive potential indicated that IVM-PNPs were successfully coated with CS or PLL. The spherical IVM-PNPs cores and polymer outer rings could be clearly observed via TEM (Fig. 2) . The kinetics of IVM release was investigated in PBS to mimic in vivo environment. As shown in Fig. S3 , both IVM-PNPs and IVM-CSPNPs exhibited sustained release behavior within 48 h. At 24 h, 64%

and 59% of IVM were released from IVM-PNP and IVM-CSPNP, respectively.

Compared to IVM-PNPs, IVM-CSPNPs exhibited a slightly slower rate of IVM release.

The sustained release of IVM from formulations blocked the rapid peak of the drug, which might prolong the action time of the drug at the target site, thereby improving the therapeutic effect. Moreover, the EE of all NPs exceeded 90% and the DL was approximately 8%. 

The interaction forces between NPs and RBCs are mainly non-covalent interactions, such as electrostatic and hydrophobic interactions (Anselmo et al., 2013) . NPs size on the adsorption efficiency (Fig. 3B ). The adsorption efficiency increased with the IVM-PNPs sizes. Generally, small NPs had reduced surface with RBCs and were easy to desorb from the RBC surface during the washing process (Zelepukin et al., 2019) . However, once the size of NPs exceeded 450 nm, the number of NPs adsorbed on RBCs decreased rapidly as the size of NPs increased (Chambers and Mitragotri, 2004) . Therefore, IVM-PNPs with a particle size of 200 nm were selected for further research.

Next, the effect of polymer coatings on the adsorption efficiency was investigated using cationic IVM-CSPNPs or IVM-PLLPNPs. Interestingly, the adsorption efficiency of IVM-CSPNPs and IVM-PLLPNPs increased by 117.12% and 100.42%, respectively (Fig. 3C) . However, hemagglutination was observed under some experimental conditions (Fig. 3D) . IVM-PNPs caused hemagglutination at a ratio of 8 mg NPs per 1 mL RBCs (8 mg/mL RBCs), showing the lowest agglutination effect (Fig. 3E) . The RBCs agglutinated when the ratio of IVM-CSPNPs to RBCs exceeded 4 mg/mL RBCs. However, IVM-PLLPNPs caused hemagglutination at all experimental ratios, even at an excessively low concentration of 0.4 mg/mL RBCs. Generally, cationic NPs led to more severe hemagglutination than anionic NPs (Yu et al., 2011) .

Therefore, IVM-PLLPNPs and IVM-CSPNPs showed stronger hemagglutination effect than IVM-PNPs. Compared to PLL, CS had higher reliability due to its good biocompatibility and low toxicity (Ahmed and Aljaeid, 2016; Hoemann et al., 2022) .

Based on the above results, IVM-PNPs and IVM-CSPNPs were selected for further research and the ratio was selected as 4 mg/mL RBCs. 

The morphology of RBC-PNPs and RBC-CSPNPs was characterized using SEM and CLSM. Compared to IVM-PNPs, more IVM-CSPNPs were adsorbed to the surface of RBCs (Fig. 4A&C ), which agreed with the results of the adsorption experiment.

RBC-hitchhiked NPs desorbed from RBCs when exposed to the physiological shear stress. In capillaries, such as pulmonary capillaries, RBCs were subjected to high shear force (~5 Pa) because of vascular stenosis, while in normal blood vessels, RBCs were subjected to low shear force (~1 Pa) (Malek et al., 1999) . To simulate different stress environment in vivo, RBCs were subjected to low shear (~1 Pa) and high shear (~5 Pa). The percentage of NPs desorbed from RBCs at 5 Pa was higher than that of NPs desorbed from RBCs at 1 Pa for RBC-IVM-PNPs and RBC-IVM-CSPNPs (Fig.   4B ). Additionally, compared to IVM-CSPNPs, IVM-PNPs were easier to desorb from the RBC surface under high shear challenge.

NPs adsorbed to RBCs could avoid MPS uptake, which prolonged their circulation time. A macrophage uptake experiment was used to evaluate the anti-phagocytosis abilities of RBC-PNPs and RBC-CSPNPs (Yao et al., 2021) . The fluorescence intensity of Cy5.5-PNPs and Cy5.5-CSPNPs groups were significantly higher than those of the RBC-Cy5.5-PNPs and RBC-Cy5.5-CSPNPs groups, implying rapid clearance of PNPs and CSPNPs in vivo (Fig. 4E&F) . RBC-Cy5.5-PNPs and RBC-Cy5.5-CSPNPs significantly reduced the phagocytosis of RAW264.7 cells (Fig. 4E&F) . Overall, RBCs-hitchhiking inhibited the phagocytosis of RAW264.7 cells regardless of the surface charge of NPs. RBCs were negatively charged due to the sialic acid groups on the surface of the RBC membrane (Weiss and Zeigel, 1971) , which made it possible to form a strong electrostatic interaction with cationic CSPNPs (Anselmo and Mitragotri, 2014) . Therefore, compared to PNPs, CSPNPs exhibited a stronger interaction with RBCs, higher RBCs adsorption efficiency in vitro. In addition, cationic NPs were easier to escape from lysosome after internalization, whereas negatively and neutrally charged NPs tended to accumulate in lysosomes (Xia et al., 2008) . 

The adsorption of NPs to RBCs might cause damage to RBCs, and the clearance of damaged RBCs by the spleen will shorten circulation time and affect drug delivery ( Lee et al., 2011; Villa et al., 2016b; Villa et al., 2015) . Therefore, the damage of RBCs caused by NPs adsorption was evaluated in terms of osmotic fragility, oxidative fragility, Na + /K + -ATPase activity and survival time in vivo.

The osmotic fragility test was performed to evaluate the sensitivity of RBC-IVM-PNPs and RBC-IVM-CSPNPs to the hypotonic solutions (Foroozesh et al., 2011) . Free

RBCs, RBC-IVM-PNPs, and RBC-IVM-CSPNPs exhibited similar levels of hemolysis ( Fig. 5A) . Regarding oxidative fragility, H 2 O 2 challenge induced 23.21 ± 1.77%, 23.28 ± 0.68%, and 25.95 ± 2.39% hemolysis in free RBCs, RBC-IVM-PNPs, and RBC-IVM-CSPNPs, respectively. Na + /K + -ATPase is an important RBC membrane enzyme, which is mainly involved in the active transport of Na + and K + , and is closely related to the deformability of RBCs (Zhang et al., 2018b) . The adsorption of NPs on RBCs may lead to a decrease in Na + /K + -ATPase activity, thereby reducing the function and biological activity of RBCs carriers. The Na + /K + -ATPase activity of RBC-IVM-PNPs and RBC-IVM-CSPNPs (31.40 ± 2.34 μmolpi·gHb -1 ·h -1 and 31. 78 ± 2.24 μmolpi·gHb -These results indicated that the adsorption of IVM-CSPNPs exhibited certain side effects on RBCs under extreme conditions in vitro. To confirm the influences of NPs adsorption on RBCs, the circulation time of RBCs carriers in vivo was studied.

Compared to free RBCs, there were no significant differences in RBC-IVM-PNPs or RBC-IVM-CSPNPs (Fig. 5B) . Therefore, NPs adsorption may not significantly affect the circulation time of RBCs in vivo. The decrease of biotin-RBCs in the bloodstream was most likely due to multiple sampling and fluorescence attenuation (Chambers et al., 2004) . 

The average plasma concentrations of IVM versus time following a single dose of intravenous administration of free IVM, IVM-PNPs, IVM-CSPNPs, RBC-IVM-PNPs, and RBC-IVM-CSPNPs are exhibited in Fig. 6 . The PK parameters results (Table 2) showed that the area under curve (AUC), mean residence time (MRT) and t 1/2 of the IVM-PNPs and IVM-CSPNPs group were significantly higher than those of the IVM injection group. However, there was no significant difference in these parameters between IVM-PNPs and IVM-CSPNPs. Moreover, the MRT and t 1/2 of RBC-IVM-PNPs and RBC-IVM-CSPNPs groups were remarkably higher than those of the homologous NPs group. Importantly, compared to RBC-IVM-PNPs, the AUC, MRT and t 1/2 of RBC-IVM-CSPNPs group were significantly increased.

NPs can increase circulation time due to their unique physicochemical and biological properties (Bertrand et al., 2017) . However, when the NPs entered the body, a protein crown was quickly formed around the NPs (less than 0.5 min) (Tenzer et al., 2013; Zhang et al., 2018a) , resulting in rapid clearance by MPS in the bloodstream (Casals et al., 2010) . For IVM-CSPNPs, the hydrophilic surfaces may prevent the NPs from serum absorption and undergo relatively less opsonization and clearance by MPS (Acharya and Sahoo, 2011) . Therefore, compared to IVM-PNPs, IVM-CSPNPs could extend the circulation time in vivo. RBCs-hitchhiking could effectively reduce NPs uptake by MPS (Fig. 4D&E) and consequently prolong the circulation time (Chambers and Mitragotri, 2004; Zelepukin et al., 2019) . As expected, the PK profiles of RBC-IVM-PNPs and RBC-IVM-CSPNPs were significant different. The higher affinity with

RBCs and lower desorption rate of IVM-CSPNPs should contribute to a longer circulation time and higher bioavailability. RBCs-hitchhiking provides a new strategy for lung-targeting drug delivery. To evaluate the biodistribution of RBC-hitchhiked NPs, Cy7.5-labeled RBC-PNPs and RBC-CSPNPs were prepared for imaging (Fig. 7A&B) . The individual Cy7.5-PNPs and Cy7.5-CSPNPs were mainly concentrated in the liver, which was mainly attributed to the phagocytosis by the MPS (Gustafson et al., 2015) . RBC-Cy7.5-PNPs and RBC-Cy7.5-CSPNPs showed obvious lung-targeting effects in a short period of 0.5 h.

Interestingly, the fluorescence intensity of the lungs of RBC-Cy7.5-PNPs group was higher than that of RBC-Cy7.5-CSPNPs at 0.5 h. However, the fluorescence accumulation of RBC-Cy7.5-CSPNPs group in lungs was higher than that of RBC-PNPs group at 2 h. Similar results were obtained in biodistribution of RBC-IVM-PNPs and RBC-IVM-CSPNPs (Fig. 7C&D) . Additionally, 0.5 h after administration, the IVM concentrations in lungs of the RBC-IVM-PNPs and RBC-IVM-CSPNPs groups were remarkably higher than those of the IVM-PNPs and IVM-CSPNPs groups. Moreover, compared to RBC-IVM-CSPNPs, a significant increase in lung accumulation was observed for RBC-IVM-PNPs. However, the IVM concentrations in the lungs of RBC-IVM-PNPs group significantly decreased after 2 h, which was lower than that of the RBC-IVM-CSPNPs group. Furthermore, RBCs-hitchhiked IVM-CSPNPs in the liver decreased by 35% compared to IVM-CSPNPs at 0.5 h after administration. However, there was no significant difference in liver accumulation between RBC-IVM-PNPs and IVM-PNPs. Therefore, both RBC-IVM-PNPs and RBC-IVM-CSPNPs could significantly enhance IVM delivery to lungs and increase the concentration of IVM in the lungs.

However, due to differences in the properties of NPs (NPs materials, surface charge), the redistribution effects of RBC-IVM-PNPs and RBC-IVM-CSPNPs in the lungs were different (Zelepukin et al., 2019) . The NPs adsorbed on the RBCs were squeezed and sheared when passing through the pulmonary capillaries, before being desorbed from RBCs. Cationic IVM-CSPNPs possessed a stronger bonding force and adsorption efficiency with RBCs compared to anionic IVM-PNPs. Therefore, the desorption rate of IVM-CSPNPs might be slower than that of IVM-PNPs in vivo. Moreover, the adhesiveness of IVM-CSPNPs could promote retention of IVM-CSPNPs in lungs and enhance internalization. Consequently, the accumulation and elimination rates of RBCs-hitchhiked IVM-CSPNPs in the lung were slower than that of RBCs-hitchhiked IVM-PNPs. In contrast, anionic IVM-PNPs would easily desorb from RBCs and exhibit rapid accumulation and elimination in the lungs. 2 h (D) administration in rats, n = 3. The results are expressed as means ± SD. P-values < 0.05, < 0.01, and < 0.001 are denoted as *, **, and ***, respectively, and NS: nonsignificant.

In addition to antiviral activities, IVM also had anti-inflammatory activities, which may also be necessary to prevent ALI in SARS-CoV-2 infection. The antiinflammatory activities of various formulations were investigated in an ALI mouse model induced by LPS. ALI was mainly characterized by pro-inflammatory cytokines production, inflammatory cell infiltration, alveolar barrier disruption, and severe pathological damage, such as edema, hemorrhage, and thickened alveolar walls (Matthay et al., 2012; Mokra and Kosutova, 2015) . LPS challenge significantly increased the lung wet-to-dry ratio (an indicator of pulmonary edema, Fig. 8A CSPNPs) or free IVM mixed RBC-NPs did not exhibit significant therapeutic effects compared to control or free IVM, respectively (Fig. S4) . Notably, RBC-IVM-CSPNPs exhibited a stronger inhibitory effect on these indicators than RBC-IVM-PNPs.

Furthermore, the histopathological changes were investigated to determine the therapeutic efficacy of various IVM-loaded formulations at the tissue level (Fig. 8F ).

interstitial edema (dark purple) in ALI mice. The degree of improvement of alveolar structure in each administration group was consistent with the inhibitory trend of the inflammatory response. Compared to control or free IVM, IVM-free RBC-NPs (RBC-PNPs and RBC-CSPNPs) or free IVM mixed RBC-NPs did not attenuate inflammatory damage, respectively (Fig. S5) . Moreover, RBCs-hitchhiked IVM-PNPs and IVM-CSPNPs significantly reduced infiltrating inflammatory cells and relieved the severity of lung lesions compared to homologous IVM-PNPs and IVM-CSPNPs. For RBC-IVM-CSPNPs groups, the alveoli were almost intact (light purple) and inflammatory cells disappeared, exhibiting significant therapeutic effect on ALI.

LPS triggers the innate immune response and activates the redox-sensitive transcription factor nuclear factor kappa-B (NF-κB), which in turn induces the expression of various pro-inflammatory mediators, including TNF-α and IL-6 (Cohen, 2002) . IVM could inhibit LPS-induced production of pro-inflammatory cytokines by suppressing NF-κB pathway (Zhang et al., 2008) . Therefore, the administration of IVM-loaded formulations could inhibit the inflammatory response and ultimately lead to the alleviation of ALI. Notably, RBCs-hitchhiked IVM-PNPs and IVM-CSPNPs showed a significant therapeutic effect on ALI, mainly due to their extended circulation time (Fig. 6) and increased IVM concentration in the lungs (Fig. 7) . Furthermore, compared to RBC-IVM-PNPs, slower accumulation and elimination rates of RBChitchhiked IVM-CSNPs in the lung could improve the treatment effects. are expressed as mean ± SD. P-values < 0.05, < 0.01, and < 0.001 are denoted as *, **, and ***, respectively, and NS: non-significant. (G) Lung tissue from each experimental group was processed for histological evaluation. The black arrows indicate inflammatory cells infiltration. scale bar: 50 μm.

This study investigated the potential of an RBCs-hitchhiking strategy to deliver IVM to lungs. After modified PNPs with chitosan, the optimized CSPNPs could significantly improve the efficiency of adsorption of NPs to RBCs. RBCs-hitchhiking enhanced lung delivery and increased IVM accumulation in the lungs and prolonged the circulation time of NPs in vivo. Importantly, the redistribution and circulation effects were related to the properties of NPs. The accumulation and elimination rates of RBCs-hitchhiked IVM-CSPNPs in the lungs were remarkably slower than those of RBCs-hitchhiked IVM-PNPs. Therefore, RBC-hitchhiking provided an alternative strategy to improve IVM pharmacokinetics and bioavailability for repurposing of IVM for the treatment of COVID-19. Furthermore, according to different redistribution effects of different NPs adsorbed on RBCs, RBCs-hitchhiked NPs may achieve various accumulation rates and circulation times for different requirements of drug delivery. 

We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work. There is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the position presented in, or the review of the manuscript entitled "Red blood cellshitchhiking mediated lung-targeting delivery of ivermectin: effect of nanoparticles properties".

All authors agreed to submit this manuscript.

All data generated or analyzed during this study are included in this published article [and its supplementary information files].

PLGA nanoparticles containing various anticancer agents and tumour delivery by EPR effect

A five-day course of ivermectin for the treatment of COVID-19 may reduce the duration of illness

Preparation, characterization, and potential application of chitosan, chitosan derivatives, and chitosan metal nanoparticles in pharmaceutical drug delivery

Delivering nanoparticles to lungs while avoiding liver and spleen through adsorption on red blood cells

Cell-mediated delivery of nanoparticles: taking advantage of circulatory cells to target nanoparticles

Mechanistic understanding of in vivo protein corona formation on polymeric nanoparticles and impact on pharmacokinetics

Red Blood Cell Hitchhiking: A Novel Approach for Vascular Delivery of Nanocarriers

Red blood cell-hitchhiking boosts delivery of nanocarriers to chosen organs by orders of magnitude

The FDAapproved drug ivermectin inhibits the replication of SARS-CoV-2 in vitro

Time evolution of the nanoparticle protein corona

Prolonged circulation of large polymeric nanoparticles by non-covalent adsorption on erythrocytes

The immunopathogenesis of sepsis

Liposomal Systems as Nanocarriers for the Antiviral Agent Ivermectin

RBC-hitchhiking chitosan nanoparticles loading methylprednisolone for lung-targeting delivery

microRNA-1246 mediates lipopolysaccharideinduced pulmonary endothelial cell apoptosis and acute lung injury by targeting angiotensin-converting enzyme 2

Ivermectin: an award-winning drug with expected antiviral activity against COVID-19

Preparation and in-vitro characterization of tramadol-loaded carrier erythrocytes for long-term intravenous delivery

Vascular Drug Delivery Using Carrier Red Blood Cells: Focus on RBC Surface Loading and Pharmacokinetics

The pharmacokinetics and interactions of ivermectin in humans--a mini-review

Nanoparticle Uptake: The Phagocyte Problem

Ivermectin: a systematic review from antiviral effects to COVID-19 complementary regimen

Chitosan coatings with distinct innate immune bioactivities differentially stimulate angiogenesis, osteogenesis and chondrogenesis in polycaprolactone scaffolds with controlled interconnecting pore size

Phagocyte-membrane-coated and laser-responsive nanoparticles control primary and metastatic cancer by inducing anti-tumor immunity

Nanoparticle-based topical ophthalmic formulations for sustained celecoxib release

MicroRNA-27a alleviates LPS-induced acute lung injury in mice via inhibiting inflammation and apoptosis through modulating TLR4/MyD88/NF-κB pathway

Ivermectin as a potential drug for treatment of COVID-19: an in-sync review with clinical and computational attributes

Antiviral effect of high-dose ivermectin in adults with COVID-19: A proof-of-concept randomized trial

Mechanism for phosphatidylserine-dependent erythrophagocytosis in mouse liver

Size Dependency of Circulation and Biodistribution of Biomimetic Nanoparticles: Red Blood Cell Membrane-Coated Nanoparticles

Quantitative proteomics reveals a broad-spectrum antiviral property of ivermectin, benefiting for COVID-19 treatment

Macrophage-mimic shape changeable nanomedicine retained in tumor for multimodal therapy of breast cancer

Effect of Ivermectin on Time to Resolution of Symptoms Among Adults With Mild COVID-19: A Randomized Clinical Trial

COVID-19 cytokine storm: The anger of inflammation

Hemodynamic shear stress and its role in atherosclerosis

The acute respiratory distress syndrome

Red blood cell volume can be independently determined in vitro using sheep and human red blood cells labeled at different densities of biotin

Targeting Some Enzymes with

Repurposing Approved Pharmaceutical Drugs for Expeditious Antiviral Approaches Against Newer Strains of COVID-19

Biomarkers in acute lung injury. Respiratory physiology & neurobiology

5-Fluorouracilloaded PLA/PLGA PEG-PPG-PEG polymeric nanoparticles: formulation, in vitro characterization and cell culture studies

Nanodrugs: pharmacokinetics and safety

The Effect of Polymeric Nanoparticles on Biocompatibility of Carrier Red Blood Cells

Pharmacokinetic considerations on the repurposing of ivermectin for treatment of COVID-19

PEGylated PRINT nanoparticles: the impact of PEG density on protein binding, macrophage association, biodistribution, and pharmacokinetics

The Approved Dose of Ivermectin Alone is not the Ideal Dose for the Treatment of COVID-19

Orally Administrable Therapeutic Synthetic Nanoparticle for Zika Virus

Rapid formation of plasma protein corona critically affects nanoparticle pathophysiology

Ivermectin effects on motor coordination and contractions of isolated rat diaphragm

Central and peripheral neurotoxic effects of ivermectin in rats

Red blood cells: Supercarriers for drugs, biologicals, and nanoparticles and inspiration for advanced delivery systems

Delivery of drugs bound to erythrocytes: new avenues for an old intravascular carrier

Drug delivery by erythrocytes

An AlphaScreen®-based assay for high-throughput screening for specific inhibitors of nuclear import

Tumor-and mitochondria

Red blood cell-hitchhiking chitosan nanoparticles for prolonged blood circulation time of vitamin K(1)

Cell surface negativity and the binding of positively charged particles

Cationic polystyrene nanosphere toxicity depends on cell-specific endocytic and mitochondrial injury pathways

Antiinflammatory effects of ivermectin in mouse model of allergic asthma

The application of nanoparticles in cancer immunotherapy: Targeting tumor microenvironment

The broad spectrum antiviral ivermectin targets the host nuclear transport importin α/β1 heterodimer

Aging erythrocyte membranes as biomimetic nanometer carriers of liver-targeting chromium poisoning treatment

Impact of silica nanoparticle design on cellular toxicity and hemolytic activity

Nanoparticle-based drug delivery via RBC-hitchhiking for the inhibition of lung metastases growth

Ligand Size and Conformation Affect the Behavior of Nanoparticles Coated with in Vitro and in Vivo Protein Corona

Autologous Red Blood Cell Delivery of Betamethasone Phosphate Sodium for Long Anti-Inflammation

Ivermectin inhibits LPS-induced production of inflammatory cytokines and improves LPS-induced survival in mice

Writing -Original Draft, Writing -Review & Editing. Caihong Lu: Investigation, Formal analysis, Data curation

Jingzhou Liu: Data curation. Guobao Yang: Data curation. Yuli Wang: Data curation. Zhiping Li: Data curation. Meiyan Yang: Data curation. Yang Yang: Data curation. Wei Gong: Conceptualization, Methodology, Writing -Original Draft, Writing -Review & Editing, Supervision, Project administration. Chunsheng Gao: Supervision

This study was supported by the National Science and Technology Major Projects for "Major New Drugs Innovation and Development" (Grant No. 2018ZX09711003-008-001).

The authors declare no conflicts of interests.