Lymphocyte Membrane‐ and 12p1‐Dual‐Functionalized Nanoparticles for Free HIV‐1 Trapping and Precise siRNA Delivery into HIV‐1‐Infected Cells

Abstract Despite the success of small interfering RNA (siRNA) in clinical settings and its potential value in human immunodeficiency virus (HIV) therapy, the rapid clearance and absence of precise delivery to target cells still hinder the therapeutic effect of siRNA. Herein, a new system, which can escape immune recognition, has HIV‐1 neutralizing capacity, and the ability to deliver siRNA specifically into HIV‐1‐infected cells, is constructed by functionalizing siRNA delivery lipid nanoparticles with the lymphocyte membrane and 12p1. The constructed system is shown to escape uptake by the mononuclear phagocyte system. The constructed system exhibits strong binding ability with gp120, thus displaying distinguished neutralizing breadth and potency. The constructed system neutralizes all tested HIV‐1 pseudotyped viruses with a geometric mean 80% inhibitory concentration (IC80) of 29.75 µg mL−1 and inhibits X4‐tropic HIV‐1 with an IC80 of 64.20 µg mL−1, and R5‐tropic HIV‐1 with an IC80 of 16.39 µg mL−1. The new system also specifically delivers siRNA into the cytoplasm of HIV‐1‐infected cells and exhibits evident gene silencing of tat and rev. Therefore, this new system can neutralize HIV‐1 and deliver siRNA selectively into HIV‐1‐infected cells and may be a promising therapeutic candidate for the precise therapy of HIV.


Introduction
Despite decades of research, acquired immune deficiency syndrome remains a major public health problem worldwide, with replication. [2,5] Small interfering RNA (siRNA), short doublestranded RNA of 21-23 nucleotides, is the most commonly used RNAi tool. siRNAs can target particular genes to induce short-term highly specific silencing to block the production of the respective proteins. [6] Several siRNA therapeutics have been approved for the market. siRNAs silencing major HIV-1 regulatory genes, such as tat, rev, nef, and vif, are also being developed, and those genes simultaneously encoding two proteins are believed to be more potent. [4,7] However, siRNAs only play a therapeutic role in the cytoplasm, and their rapid degradation by RNase and the low cell permeability due to their strong hydrophilicity and inherently strong negative charge hamper their entry into cells. [2,4,6c] Therefore, considerable efforts have been made to develop effective siRNA delivery carriers, [2,4,6c] including nanocarriers, liposomes, polymeric nanoparticles, dendrimers, quantum rods, carbon nanotubes, and inorganic nanoparticles. [2,4,6c] Among these, lipid-based vectors are considered one of the most promising carriers and several lipid-based products have been applied in clinical settings.
Despite major achievements, some problems, such as nonspecific distribution of the delivery system, clearance by the mononuclear phagocyte system (MPS), and low endosomal escape efficiency, are known to seriously affect the efficiency of siRNA delivery into the cytoplasm of target cells. [2,6c] To increase the targeting ability to free HIV virions or HIV-infected cells, new delivery systems have been designed to encompass conjugation with PEGylated targeting moieties, such as antibodies, aptamers, and other ligands. [2,3,7a,8] Several studies have consistently shown that the conjugation of nanoparticles with a targeting ligand increases the association of the nanoparticles with target cells, thereby increasing the specificity of drug delivery and concurrently reducing off-target effects and toxicity. [7b,9] Although PEGylation is one of the major approaches for imparting stealthiness to these targeted drug delivery systems, the unexpected clearance of PEGylated materials in vivo after repeated administration remains a limitation to be overcome. [10] Biomimetic technology, especially coating with the cell membrane-an emergent alternative to PEGylationmeets these needs and is actively used in targeted nanomedicines. [1,6c,8,10b] Coating with cell membrane facilitates the transfer of membrane proteins, including receptors from source cells, onto the nanoparticle surface, thereby producing many distinct advantages, such as reduced MPS clearance, prolonged blood circulation, and improved accumulation at specific pathological sites. [1,6c,8,11] It has even been inferred that nanocarriers made with the host cell membrane will be able to neutralize the infection, provide a broad-acting countermeasure resistant to mutations, and protect against viruses as long as the target of the virus remains the identified host cells. [9] Several T cell-mimicking nanotraps and nanoparticles have been developed, and their neutralization against free HIV and targeting of HIV-infected cells have been authenticated. [1,3a,12] However, the infection-inhibiting efficacy of these cellular membranebased nanoscale vesicles is limited [13] due to the high number of host cells in vivo, which makes it difficult for the definite cellular membrane-based vesicles to competitively bind with free viruses or virus-infected cells on infinite host cells. Therefore, the targeting design of vesicles coated with cell membranes should be improved to enhance their competitive binding ability with free viruses or virus-infected cells.
Decoration with a certain density of ligands on cellular membrane-based nanoparticles might enhance their binding probability with free viruses or virus-infected cells, facilitate the precise delivery of siRNA into virus-infected cells, and finally improve the infection-inhibiting efficacy. The linear peptide,12p1 (RINNIPWSEAMM), has been shown to preferentially bind with gp120 prior to clusters of differentiation 4 (CD4) or chemokine receptor 5 (CCR5), and block the binding of gp120 to the native coreceptor CD4 or CCR5. [14] Currently, 12p1 has been used as a CD4-gp120 inhibitor and the peptide probe of the gp120 structure. [15] Therefore, 12p1 is considered a candidate targeting the ligand of gp120. The combination of 12p1 with the T cell membrane may predominantly enhance the nanoparticles' competitiveness of binding with gp120 compared to native T cells and may show satisfactory infection-inhibiting efficacy.
Accordingly, a lymphocyte membrane-and 12p1-dual functionalized siRNA delivery lipid nanoparticle system (MPLN) was constructed as shown in Figure 1A, and the schematic illustration of MPLN is proposed in Figure 1B. After injection, MPLN first escapes from MPS uptake and circulates throughout the whole body due to the existence of CD47 from the cell membrane and the inserted polyethylene glycol (PEG). Then, MPLN targets gp120 expressed by free HIV virions or HIV-infected cells and binds to them under the guidance of the T cell membrane and 12p1; as a result, MPLN can directly neutralize free HIV and inhibit gp120-induced killing of bystander cells. In parallel, siRNA encapsulated in MPLN can be specifically delivered into HIVinfected cells, where it serves to suppress HIV replication. Therefore, significantly improved HIV inhibition is expected for this dual-functionalized siRNA delivery system.
To validate the HIV-inhibition efficacy of MPLN, the human lymphoma (MT2) cell membrane and 12p1 functionalized lipid nanoparticles containing siRNA targeting HIV-1 tat and rev transcripts were constructed, and their characterization, ability to bind to gp120, HIV-1 neutralization, ability to deliver siRNA into the cytoplasm of HIV-1-infected cells, inhibition of gp120induced bystander cell killing, distribution, and safety in vivo were investigated.
To verify the transfer of proteins from the cell membrane to MPLN, the protein profile, specific proteins, and secondary www.advancedsciencenews.com www.advancedscience.com structure of proteins from the extracted MT2 cell membrane and different nanoparticles were detected, and the results are shown in Figure 2D-F. There were no significant differences in the protein profile of the extracted MT2 cell membrane and MPLN, and no protein signal was detected in LN. Specific proteins, including CD4, CCR5, chemokine (C-X-C motif) receptor 4 (CXCR4), and CD47, were displayed in the purified MT2 cell membrane and MPLN, but not in LN. There was a positive band at 196 nm and a negative band at 220 nm in all far-UV circular dichroism spectra (CD), and no significant shifts in the peaks or obvious changes in peak intensity.
The release of siRNA from LN, the MT2 cell membranemodified lipid nanoparticles (MLN), and MPLN was tested as shown in Figure 2G. It was obvious that siRNA was released slowly from MPLN or MLN, and the cumulative release of siRNA from MPLN could reach about 90% at 48 h while there was only about 30% of siRNA released from MLN at 48 h. In addition, there seemed to be no siRNA released from LN.
The colloidal stability of MPLN and its dilutions with phosphate-buffered saline (PBS) or cell culture media containing 10% fetal bovine serum (FBS) was monitored using Turbiscan Lab Expert, and the results are shown in Figure 2F. Additionally, the variations in both transmission and backscattering were less than 1% for all samples over a 72-h period.

In Vitro Cellular Uptake
The uptake of nanoparticles by human macrophages differentiated from human myeloid leukemia mononuclear (THP-1) cells was analyzed using both confocal laser scanning microscopy (CLSM) and flow cytometry (FCM), and the results are exhibited in Figure 3A,B. There was almost no fluorescence observed in cells treated with PBS control or free siRNA. In contrast, intensive fluorescence was found in cells treated with LN, and there was slight fluorescence displayed in cells treated with MLN or MPLN. The results from FCM were coincident with those from CLSM, in that macrophages treated with LN displayed the highest fluorescence intensity, followed by MLN, MPLN, and free siRNA or control. Significant differences in the fluorescence intensity were found between LN and MLN, LN and MPLN, free siRNA and MLN, free siRNA and MPLN, and MLN and MPLN.

Cytotoxicity
The cytotoxicity of the nanoparticles on TZM-bl cells, MT2 cells, and peripheral blood mononuclear cells (PBMC) was assessed by analyzing cell viability with the Cell Counting Kit-8 (CCK-8) assay. As shown in Figure 3C−E, there was no significant change found in cell viability for all three cell lines treated with MLN or MPLN compared with control while the viability of all three cell lines treated with LN significantly decreased compared with control when the concentration of LN used was more than 550 nM (siRNA concentration) or 1.84 mg mL −1 (nanoparticle mass concentration).

Binding of MPLN with gp120
The binding capability of MPLN with different HIV-1 gp120 recombinant proteins, including, X4-tropic HIV-1 MN gp120 and R5-tropic HIV-1 BaL gp120, was first investigated. The binding capability of various nanoparticles is displayed in Figure 4A, and the binding curves of MPLN with gp120 recombinant proteins are shown in Figure 4B. Very low fluorescence intensity was found in the samples treated with LN or the erythrocyte membranemodified lipid nanoparticles (EMLN) regardless of gp120 type. The fluorescence intensity was obviously improved for gp120 proteins incubated with MLN, the erythrocyte membrane, and 12p1 dual modified lipid nanoparticles (EMPLN) or MPLN, and the strongest fluorescence intensity was displayed in the group treated with MPLN. The fluorescence intensity of the captured MPLN increased gradually with increasing MPLN concentration, and the plateau was reached when the concentration of MPLN was approximately 4 mg mL −1 for both HIV-1 MN gp120 and HIV-1 BaL gp120. A Langmuir binary interaction model was used to fit the data curves, and the calculated dissociation constant was 0.83 ± 0.39 mg mL −1 for immobilized HIV-1 MN gp120 and 1.05 ± 0.22 mg mL −1 for immobilized HIV-1 BaL gp120.
The binding ability of MPLN with HIV-1 envelope glycoprotein was further evaluated by monitoring the characteristic changes in morphology, particle size, and zeta potential. X4tropic HIV-1 NL4-3 and R5-tropic HIV-1 AD8 were used as model viruses. As shown in Figure 4C, TEM revealed the aggregation of MPLN at the surface of virus particles. The particle size of the mixtures of HIV-1 NL4-3 and MPLN increased to 209.43 ± 42.87 from 77.97 ± 11.20 nm for HIV-1 NL4-3 and 134.40 ± 18.11 nm for MPLN ( Figure 4D), and the zeta potential of the mixtures of HIV-1 NL4-3 and MPLN changed to −10.24 ± 0.44 from −4.35 ± 0.60 mV for HIV-1 NL4-3 and −16.10 ± 0.61 mV for MPLN ( Figure 4D). Similar changes were found for the mixtures of HIV-1 AD8 and MPLN; that is, the particle size of the mixtures of HIV-1 AD8 and MPLN increased to 285.87 ± 85.42 from 94.67 ± 20.49 nm for HIV-1 AD8 and 134.40 ± 18.11 nm for MPLN, and the zeta potential of the mixtures of HIV-1 AD8 and MPLN changed to −10.90 ± 0.46 from −6.50 ± 0.78 mV for HIV-1 NL4-3 and −16.10 ± 0.61 mV for MPLN.
To evaluate the binding ability and the selectivity of MPLN to HIV-1-infected cells, HIV-1 NL4-3 -and HIV-1 AD8 -infected TZM-bl cells were used as model cells, and the binding efficacy was observed under CLSM, as shown in Figure 4E. As expected, negligible fluorescence was found in uninfected cells (control), and significantly stronger fluorescence was observed in both types of infected cells treated with MPLN.

HIV Neutralization Assay
To evaluate the neutralizing breadth and potency of MPLN against HIV-1, nine strains of geographically and genetically diverse envelope glycoprotein (env)-pseudotype viruses were used. The neutralizing breadth of MPLN was 100% against the combined nine-virus panel, and the geometric mean 50% inhibitory concentration (IC50) and the geometric mean 80% inhibitory concentration (IC80) of MPLN against all nine viral strains were 18.33 and 29.75 μg mL −1 , respectively, as shown in Figure 5A.
To further confirm the primary action of the MT2 cell membrane coating and 12p1 decoration, the neutralizing efficacy of LN, MLN, scrambled siRNA/MPLN, and the mixtures of various nanoparticles with antibodies was evaluated. As displayed in Figure 5D, there were almost no HIV-1 NL4-3 -neutralizing effects for free siRNA and LN, while significant HIV-1 NL4-3 -neutralizing efficacy was shown for MLN, MPLN, scrambled siRNA/MPLN, and antibodies. The most pronounced HIV-1 NL4-3 neutralization was www.advancedsciencenews.com www.advancedscience.com observed in MPLN and scrambled siRNA/MPLN, followed by antibodies, MLN, blocked MPLN, and blocked MLN. Significant differences in neutralizing efficacy were found between MLN and LN, MPLN and LN, scrambled siRNA/MPLN and LN, and MPLN and MLN. However, there were no significant differences in neutralizing efficacy between MPLN and scrambled siRNA/MPLN. Similar results were exhibited in terms of HIV-1 AD8 -neutralizing efficacy, as shown in Figure 5E.

Inhibition of HIV-1 gp120-Induced Bystander T-Cell Killing
The potential of MPLN for neutralizing the cytotoxicity of gp120 was also evaluated. Approximately 40% of the isolated CD4 + T cells died following treatment with X4-tropic HIV-1 MN gp120 recombinant proteins as displayed in Figure 5F. The introduction of free siRNA or LN in combination with HIV-1 MN gp120 recombinant proteins had no significant influence on the viability of CD4 + T cells. However, the viability of CD4 + T cells significantly improved following treatment with the mixtures of HIV-1 MN gp120 recombinant proteins and MLN or MPLN, and the highest viability of CD4 + T cells was found in the groups treated with the mixtures of HIV-1 MN gp120 recombinant proteins and MPLN regardless of siRNA. The inhibition efficacy of MPLN against R5-tropic HIV-1 BaL gp120 recombinant proteins was also tested, with the results shown in Figure 5G. Similar to the results against HIV-1 BaL gp120 recombinant proteins, MPLN and scrambled siRNA/MPLN displayed the strongest inhibition of gp120induced CD4 + T-cell death, followed by MLN, and neither free siRNA nor LN showed any evident inhibition of gp120-induced CD4 + T-cell death.

Endosomal Escape
The endosomal escape of MPLN was visualized using CLSM. The distribution of siRNA and endolysosomes were identified as green and red fluorescence, respectively, given that siRNA was labeled with cyanine 5 (Cy5) and endolysosomes were stained with LysoTracker Red before imaging. The confluence of green and red fluorescence was shown as yellow; that is, the presence of yellow fluorescence indicated that siRNA was entrapped within endolysosomes. As shown in Figure 6A, some green fluorescence at the cell surface and some yellow fluorescence were visible 1 h after the addition of nanoparticles. After 3 h, the green fluorescence at the cell surface almost disappeared, while the yellow fluorescence and the red fluorescence accounted for most of the fluorescent signal. The green fluorescence seemed to gradually dominate at 6 h in cells treated with MPLN.

In Vitro Gene Silencing
To verify the gene-silencing effect of siRNA, the tat and rev messenger RNA (mRNA) levels were analyzed with a real-time fluorescence quantitative polymerase chain reaction (PCR) system, and tat and rev protein expression levels were detected using western blot (WB) analysis. As shown in Figure 6B, there was no evident tat or rev mRNA suppression in HIV-1 NL4-3 -infected cells treated with free siRNA, while slight tat and rev suppression was shown in HIV-1 NL4-3 -infected cells treated with LN or scrambled siRNA/MPLN. Obvious tat and rev mRNA suppression was observed in other formulations. The efficiency of tat and rev mRNA suppression in HIV-1 NL4-3 -infected cells decreased in the following order: MPLN, blocked MPLN, MLN, blocked MLN, and LN. Consistent results were also exhibited in HIV-1 AD8 -infected cells, as displayed in Figure 6C.
Similar to the suppression of tat and rev mRNA, downregulation of tat and rev proteins was also found in both HIV-1 NL4-3infected cells and HIV-1 AD8 -infected cells treated with nanoparticles of different formulations, except those treated with free siRNA ( Figure 6D,E). The lowest tat and rev protein expression levels were found in cells treated with MPLN, regardless of HIV strains.

In Vivo Safety Evaluation
Histological sections of the main organs, basic metabolic panel (BMP), comprehensive metabolic panel (CMP), and body weight were examined, and the results are displayed in Figure 7A-C.
No pathological changes were observed in any of the organs, and there were no significant differences in BMP and CMP.

In Vivo Distribution
The distribution of siRNA in mice that received various formulations is displayed in Figure 8A. The mice that received physiological saline showed no fluorescence signal. In mice that received free siRNA, the fluorescence dispersed systematically soon after administration, accumulated gradually in the bladder and kidneys, and finally disappeared completely at 24 h. In mice that received LN, the fluorescence also dispersed systematically soon after administration and accumulated gradually in the bladder and kidneys. Regarding the difference in the distribution of siRNA between the mice that received free siRNA and those that received LN, there was still a weak fluorescence signal observed in the bladder even at 24 h in the mice that received LN. As for the mice that received MLN or MPLN, there was an evident fluorescence signal throughout the body during the entire detection period, and the fluorescence signal in the mice that received MPLN seemed stronger than that in the mice that received MLN.
The distribution of siRNA in the main organs is displayed in Figure 8B. A weak fluorescence signal still existed in the liver, spleen, and kidneys in the mice that received free siRNA, and an obvious fluorescence signal was observed in all of the main organs of the mice that received LN, MLN, or MPLN. As shown in Figure 8C, the fluorescence intensity in the major organs of the mice that received LN was stronger than that in the major organs of the mice that received MLN or MPLN, while the fluorescence intensity in the major organs of the mice that received MLN was higher than that in the major organs of the mice that received MPLN.

Discussion
MPLN was acquired by introducing 1,2-distearoyl-sn-glycero-3phosphorylethanolamine -N-methoxy(polyethylene glycol)-2000 www.advancedsciencenews.com www.advancedscience.com (DSPE-PEG2000)-12p1 and the MT2 cell membrane into LN. The insertion of DSPE-PEG2000-12p1 had no significant influence on the nanoparticle characteristics, including particle size, zeta potential, and siRNA encapsulation efficacy, while the addition of the MT2 cell membrane evidently raised the particle size and obviously decreased the zeta potential of the nanoparticles. These results are consistent with other literature. [6c] The membrane proteins of MPLN were coincident with those of the MT2 cell membrane given that no significant differences were observed in the protein profile, specific proteins, and the protein secondary structure. The results from the colloidal stability experiments showed that both MPLN and its dilutions were stable during the 72-h period, confirming the in vivo stability of MPLN.
MPLN was considered safe because no obvious cytotoxicity was found in vitro, and there were no evident changes in pathology, BMP, CMP, and body weight in mice treated with MPLN.
The immune-escape ability of MPLN was identified both in vitro and in vivo. In cellular experiments, the uptake of MPLN by human macrophages differentiated from THP-1 cells was evaluated both qualitatively and quantitatively. It was obvious that human macrophages more easily ingested LN than those disguised by the MT2 cell membrane. Therefore, the MT2 cell membrane plays an important role in cellular uptake, and the decoration of nanoparticles with the MT2 cell membrane might decrease the clearance by MPS. In the distribution experiments in humanized mice, the encapsulation of siRNA into nanoparticles prolonged the half-life of siRNA by protecting siRNA from rapid degradation. However, LN, the naked nanoparticles, were almost completely cleared from the humanized mice at 8 h, which was likely due to the nonspecific uptake by MPS because these nanoparticles were mainly distributed in the liver and kidneys at 4 h. In contrast, the nanoparticles disguised with the MT2 cell membrane displayed a longer circulating time, and there was still some siRNA detected in the whole body even at 24 h, especially for MPLN. The amount of siRNA accumulating in the liver and kidneys at 4 h was also significantly decreased for the nanoparticles disguised with the MT2 cell membrane. It is thought that this will be beneficial for the inhibition of HIV because the target cells of HIV are mainly CD4 + cells, which are mainly distributed in blood. Taken together, the constructed MPLN possesses obvious immune-escape ability, which could be attributed to the inherited proteins from the MT2 cell membrane, such as CD47, which is well known as a "don't-eat-me" molecule on the autologous cells' surface and can prevent macrophage-mediated phagocytosis in MPS. The introduction of PEG2000 was also validated to be bene-ficial for the immune escape of nanoparticles because PEGylation could form a physical barrier at the surface of the nanoparticles and prevent their subsequent clearance. [16] The binding ability and specificity of MPLN with gp120 were validated using HIV-1 gp120 recombinant proteins, free HIV-1 virions, and HIV-1-infected cells. To exactly reflect the binding ability and specificity of MPLN, two types of HIV-1 gp120 recombinant proteins and two strains of HIV-1, representing the X4 tropic strain and R5 tropic strain, were employed. While only slight fluorescence was detected for both LN and EMLN, the introduction of the MT2 cell membrane or 12p1 significantly improved the fluorescence intensity. It was considered as a matter of course that the strongest fluorescence was found in the samples treated with MPLN, which were simultaneously decorated with the MT2 cell membrane and 12p1. The binding specificity of MPLN with gp120 expressed by free HIV-1 virions was also verified given that the particle size of the mixtures obviously increased compared to that of MPLN or free HIV-1 particles, and the evident zeta potential changes of the mixtures were also detected. There was a visible overlap between MPLN and HIV-1 particles. Similarly, strong Cy5 fluorescence was detected in HIV-1infected cells treated with MPLN, while almost no Cy5 fluorescence was found in normal cells treated with MPLN. Therefore, MPLN could powerfully and specifically bind to gp120, regardless of its source and existing status.
MPLN showed a neutralizing breadth of 100% against the combined nine strains of the env-pseudotyped virus panel and different tropic HIV-1. The neutralization potency was robust against all nine strains of the env-pseudotyped virus and different tropic HIV-1 because the geometric mean IC80 of MPLN against the env-pseudotyped virus was 29.75 μg mL −1 , and the IC80 of MPLN was 64.20 μg mL −1 for X4-tropic HIV-1 NL4-3 and 16.39 μg mL −1 for R5-tropic HIV-1 AD8 . It was obvious that both the IC50 and IC80 of MPLN were much lower than those reported, [1,12] and were comparable to those of soluble CD4 and dimeric CD4immunoglobulin fusion proteins, which could only partly neutralize the tested env-pseudotyped HIV tested. [3a] It is well known that HIV-1 primarily infects CD4 + T cells, and the infection is initiated by binding the gp120 subunit of env with CD4, which triggers a conformational change in env that allows it to interact with a host cell coreceptor protein CCR5 or CXCR4. [3a] Therefore, CD4 + T cell-mimicking nanoparticles were designed for HIV inhibition, and their significant HIV-1 neutralization was confirmed. [1,12] However, these nanoparticles acquired their cell-mimicking ability by disguising themselves www.advancedsciencenews.com www.advancedscience.com with the membrane extracted from CD4 + T cells. Therefore, their evident HIV-1 neutralization in vivo was unexpected given the high number of CD4 + T cells in vivo, which makes it difficult for CD4 + T cell-mimicking nanoparticles to competitively bind with HIV-1 or HIV-1-infected cells.
In this research, 12p1, a peptide that can preferentially bind with gp120 prior to CD4 or CCR5, [14] was simultaneously introduced into the novel delivery system based on the MT2 cell membrane so as to improve the competitive strength compared to naive CD4 + cells in vivo and thereby enhance HIV-1neutralization ability. The env-pseudotyped HIV neutralization and HIV-1 neutralization results confirmed that the nanoparticles decorated with the MT2 cell membrane or 12p1 had significant neutralization abilities for both env-pseudotyped HIV and HIV-1. Additionally, MPLN decorated with both the MT2 cell membrane and 12p1 showed the strongest HIV-1 neutralization, indicating that the combination of the MT2 cell membrane and 12p1 significantly improved the HIV-1-neutralization ability. A higher gp120-binding ability of MPLN than that of MLN or EM-PLN was also observed. The enhanced HIV-neutralizing ability of MPLN may be attributed to the strengthened gp120 binding ability caused by the combination of the MT2 cell membrane and 12p1.
It is well known that siRNA plays its therapeutic role only in the cytoplasm.
[6c] However, most foreign substances including carriers encapsulating siRNA, are internalized into endocytic vesicles and finally degraded in lysosomes. In other words, endosomal escape is a significant bottleneck in the delivery of therapeutics within a cell, and failure to escape from a lysosome may lead to degradation in the lysosome, especially for siRNA. [17] Therefore, the endosomal escape of siRNA was also investigated in this research. Evident endosomal escape of siRNA was confirmed by the abundance of solitary green fluorescence and low levels of yellow fluorescence, presenting the confluence of cyanine 5 labeled siRNA (Cy5-siRNA) and endolysosomes. The endosomal escape of siRNA could mainly be attributed to the introduction of protamine, [18] a widely used and well-characterized cationic protein approved by the Food and Drug Administration, which has been extensively applied in condensing negatively charged nucleic acids and improving their transfection efficiency. [18,19] These capabilities of protamine are supposedly related to its compact region of positively charged arginine residues. [20] However, the addition of protamine might influence the release of siRNA from nanoparticles. It was displayed that there was almost no siRNA released from LN. It was different from above the 10% of siRNA released from the reported liposomes. [6] This was supposedly due to the formation of siRNA and protamine complex, who was stable in LN and the large size made these complexes difficult to pen-etrate the bimolecular film of LN. However, the release of siRNA from MLN and MPLN was significantly improved, which might be caused by the introduction of the MT2 cell membrane because the cell membrane possessed negative charges, could competitively bind with protamine, and thereof liberated siRNA from the complexes. The complete release of siRNA from MPLN was observed, and this was believed related to the further addition of DSPE-PEG2000-12p1, which had a large steric hindrance and might enhance the distance of molecules. All in all, the sustained and complete release of siRNA from MPLN was beneficial for the homing of MPLN and taking effects of siRNA in the target cells.
Tat and rev genes were silenced to different extents in HIV-1-infected cells treated with different formulations containing siRNA, except in the cells treated with free siRNA. [7] It is reasonable to conclude that there was no gene-silencing efficacy of free siRNA because it was almost impossible for it to enter into the cytoplasm to play a direct action. [6c] MPLN decorated with the MT2 cell membrane and 12p1 showed the strongest tat and rev gene silencing, followed by MLN and LN. Evident tat and rev gene silencing were also observed in cells treated with antibody-blocked formulations, although the strength of gene silencing was weakened, which was consistent for cells infected with X4-tropic HIV-1 NL4-3 or R5-tropic HIV-1 AD8 . The most distinguished gene silencing of MPLN was supposedly related to the decoration of the MT2 cell membrane and 12p1, which could promote the binding of the nanoparticles with HIV-1-infected cells and increase the amount of siRNA entering the target cells.
Although obvious gene silencing of tat and rev genes was exhibited after HIV-1-infected cells were treated with siRNA encapsulating MLN or MPLN, there was no significant difference in the neutralization or inhibition of HIV-1 for MPLN encapsulating tat/rev siRNA or scrambled siRNA. This may be due to the complex action mechanism of siRNA and the long lag time of effects. Indeed, the lag periods between the peak plasma concentrations and peak gene silencing or therapeutic efficacy have been found to range from several days to several months in the clinic for siRNA preparations, including patisiran, cemdisiran, and vutrisiran. [21] This lag time is believed to be mainly attributed to the slow RNA-induced silencing complex (RISC) loading process in target cells, which is essential for the specific binding of siRNA to the target complementary mRNA. The gradual loading of RISC is further considered to result from slow endosomal trafficking and the release of siRNA into the cytosol. These processes finally impact pharmacodynamic properties as RISC-loaded siRNA concentrations have been shown to be directly correlated with kinetics and the magnitude of target protein suppression. [21b,c] It is reasonable to suppose that MPLN encapsulating tat/rev siRNA did not have stronger neutralizing or in-  hibiting efficacy against HIV-1 compared to MPLN encapsulating scrambled siRNA because nanocarriers possessing gp120 receptors, such as MLN and MPLN, can rapidly bind to gp120 proteins and directly neutralize HIV-1 utilizing 12p1 and CD4 molecules at the surface of carriers, but has to spend several days to months to take effect for siRNA due to its complex action mechanism.
It is known that envelope glycoproteins including gp120 can bind to their cellular receptors and chemokine coreceptors prior to viral fusion and entry, and cause bystander CD4 + T-cell death. [12,22] This is considered a critical element of HIV pathogenesis given that it contributes to the selective depletion of CD4 + T cells and leads to immunodeficiency. [12,22] The constructed MLN, MPLN, and MPLN with scrambled siRNA markedly improved the viability of bystander T cells. MPLN and MPLN with scrambled siRNA showed stronger inhibition of bystander T-cell killing than MLN, while there was no significant difference in the inhibition efficacy caused by MPLN or MPLN with scrambled siRNA. These data indicated that the inhibition of HIV-1 gp120-induced bystander T-cell killing was mainly related to the binding ability with gp120, instead of siRNA.

Conclusion
In this study, we successfully constructed a lymphocyte membrane-and 12p1-dual-functionalized siRNA delivery lipid nanoparticle system. The constructed nanoparticles exhibited an appropriate particle size, good colloidal stability, and satisfactory lymphocyte membrane features. More importantly, this multifunctional nanocarrier achieved evident immune escaping capability, potent gp120 binding ability, robust HIV-1 neutralization, striking HIV-1 tat and rev gene silencing, and obvious inhibition of HIV-1 gp120-induced bystander T-cell killing on the premise of safety. Therefore, satisfactory HIV-1 therapy efficacy, including HIV-1 neutralization, specific siRNA delivery into HIV-1infected cells, and reduction of CD4 + T-cell depletion, is expected, and this new system holds potential for HIV-1 inhibition.

Experimental Section
Materials: Hydrogenated soybean phosphatidylcholine (HSPC), and DSPE-PEG2000 were bought from Shanghai Advanced Vehicle Technology Pharmaceutical LTD (Shanghai, China). DSPE-PEG2000-12p1 was prepared by Xi'an ruixi Biological Technology Co., Ltd (Xi'an, China). Extraction of Cell Membrane: The cell membrane of MT2 cells, a type of lymphocyte, was acquired in accordance with a previously reported method. [1,6c,12] In brief, the collected MT2 cells were first resuspended in cool PBS (pH7.4) containing ethylenediaminetetraacetic acid (1 μM) and phenylmethylsulphonyl fluoride (1%, W/V), before breaking into pieces in an ice bath by ultrasound with the power of 150 W for 2 min. After incubation at 4°C for 30 min, the suspension was centrifuged at 500× g for 10 min. The obtained supernatant was further centrifuged at 20 000× g for 20 min twice, the MT2 cell membrane was acquired after the supernatant was finally centrifuged at 100 000× g for 50 min, and the sediments were washed with PBS. The acquired precipitate was finally collected, identified, quantified, and stored at −20°C for use.
The erythrocyte membrane was isolated following a method similar to that for the lymphocyte membrane. [23] Briefly, the collected erythrocytes were treated with precooled PBS containing ethylenediaminetetraacetic acid (1 μM) and phenylmethylsulphonyl fluoride (1%, W/V) for 2 h, and then the suspension was centrifuged at 9050× g for 15 min. The acquired precipitate of the erythrocyte membrane was finally collected, identified, quantified, and stored at −20°C for use.
Preparation of Various Nanoparticles: LN was constituted with HSPC, cholesterol, and DSPE-PEG2000 (80:10:10, W/W/W), and prepared via a thin-film dispersion method with some modifications. [6c] Briefly, the lipids were dissolved in chloroform, and the solvent was evaporated under vacuum by a rotator RE-2000 (Ya Rong Biochemical Instrument Factory, China) at 40°C for 30 min. The formed thin film was then hydrated with a hydroxyethyl piperazine ethanesulfonic acid buffer solution (20 mM hydroxyethyl piperazine ethanesulfonic acid, 150 mM NaCl, pH 7.0) containing the complex of siRNA and protamine (1:1, W/W) at 55°C for 40 min, before dispersing the suspension with an ultrasound transducer at 427.5 W for 60 s. Finally, LN was acquired after free siRNA was removed by dialysis. MLN was acquired after mixing LN with the extracted MT2 cell membrane and subsequently extruding through polycarbonate membranes with pore sizes of 400 and 200 nm 20 times respectively. EMLN was prepared using the same method as the MLN. MPLN were finally obtained by inserting DSPE-PEG2000-12p1, whose synthetic route and identification are displayed in Figure S1, Supporting Information, into MLN at 37°C for 30 min, [6c] and EMPLN were acquired by inserting DSPE-PEG2000-12p1 into EMLN.
Morphology of Various Nanoparticles: Different nanoparticles were morphologically characterized using TEM (JEM-1010, JEOL, Tokyo, Japan) and AFM (Bruker Multimode 8, Bruker Daltonic, Billerica, MA, USA). [6c] The suspensions of various nanoparticles were dropped on a copper grid, dried at 25°C, and negatively stained with phosphotungstic acid (2%, W/V) before being observed under TEM. The suspensions of various nanoparticles were spread onto a mica sheet and dried at 25°C before observing under AFM.
Particle Size and Zeta Potential of Various Nanoparticles: The particle size and zeta potential of different nanoparticles were analyzed with a Malvern Zetasizer (Nano-ZS90, British) at 25°C. [6c] The measurements were conducted with 11 cycles, and the results of particle size were shown as Z-average diameter.
Encapsulation Efficiency: To calculate the encapsulation efficiency of siRNA in various nanoparticles, the amounts of total siRNA in these suspensions and free siRNA unencapsulated in the nanoparticles were both determined. [6c,24] To obtain the total amount of siRNA in various formulations, nanoparticle suspensions (200 μL) were first dissolved by adding Triton solution (200 μL, 10%, W/V), and the content of Cy5-siRNA was measured using a spectrofluorometer with 649 nm as the excitation wavelength and 680 nm as the emission wavelength. The amount of free Cy5-siRNA in the suspensions was detected with the spectrofluorometer after nanoparticle suspensions (200 μL) were diluted to 2 mL with distilled water, added into the ultra-filter (vivaspin2; Sartorius biotech, USA), and centrifuged at a speed of 14 000× g for 10 min. www.advancedsciencenews.com www.advancedscience.com The encapsulation efficiency was calculated according to Equation (1), where the W total drug represents the total amount of siRNA in nanoparticle suspensions and the W free drug indicates the amount of free siRNA unencapsulated.
Encapsulation efficiency (%) = W total drug − W free drug W total drug × 100% (1) Protein Detection: To validate the complete transfer of proteins from the MT2 cell membrane onto the nanoparticles, the protein profiles in the purified MT2 cell membrane and MT2 cell membrane-coated nanoparticles were analyzed using sodium dodecyl sulfate-polyacrylamide gel electrophoresis. [6c,25] In brief, the proteins were extracted from the samples with the cell total protein extraction kit, and then separated on a bistris 10-well minigel (4-12%, W/V) in a running buffer using a Bio-Rad electrophoresis system at 80 V for 0.5 h and thereafter at 120 V for 1 h. The resulting polyacrylamide gel was finally stained with SimplyBlue overnight for visualization. LN without the cell membrane was analyzed as a negative control.
Specific proteins including CD4, CCR5, CXCR4, and CD47 in the extracted MT2 cell membrane, LN, and MPLN were identified using WB analysis. Briefly, the extracted proteins were first quantified with the Pierce BCA protein assay (ThermoFisher, USA), before separating on an acrylamide gel (10%, W/V) and transferred to a polyvinylidene difluoride membrane (Millipore, USA). Thereafter, the samples were treated with primary antibodies against CD4, CCR5, CXCR4, and CD47 followed by incubating with HRP-labeled goat/anti-rabbit IgG. The protein signals were finally detected using a ChemiDoc MP imaging system (Bio-Rad, USA).
CD Analysis: To ensure the consistency of proteins from the MT2 cell membrane, MLN, and MPLN in the secondary structure, CD analysis was conducted on a Chirascan-Plus CD spectrometer with a 0.1-cm quartz cell. [25,26] The concentrations of proteins in different samples were maintained at approximately 0.3 mg mL −1 . The analysis was performed with a wavelength range of 190-300 nm and a bandwidth of 5 nm. Each spectrum was reported as the average of three scans.
In Vitro Release: The release of Cy5-siRNA from LN, MLN, and MPLN was inspected using a dialysis technique at 37°C. [6] Briefly, different nanoparticles (0.5 mL) were added into a dialysis bag (MWCO 50 kDa), immersed in PBS (pH 7.4, 35 mL), and stirred at 100 rpm. Then, the release sample (0.7 mL) was taken out and replaced with an equal volume of fresh release medium at 1, 2, 4, 7, 14, 24, and 48 h. The content of siRNA in the samples was determined using a spectrofluorometer with 649 nm as the excitation wavelength and 680 nm as the emission wavelength. The cumulative release of siRNA was calculated according to Equation (2), where V is the volume of release medium, C t represents the determined concentration of siRNA in the collected samples at time t, ∑C m denotes the concentration sum of the collected samples, V r corresponds to the volume of samples removed for analysis, and Dose is the amount of siRNA added into the release medium.
Colloidal Stability: The colloidal stability of MPLN and its dilutions with PBS (pH 7.4) or complete culture medium (RPMI 1640 containing 10% (W/V) FBS) was determined with Turbiscan Lab ® Expert at 37°C , and the evaluating indicators were the changes of transmission and backscattering. [6c,24] Cytotoxicity: The cytotoxicity of MPLN at different concentrations was evaluated in TZM-bl, MT2, and PBMC cell lines. [6c,25,27] Briefly, TZM-bl and MT2 cells (6 × 10 3 cells/well) were first seeded in 96-well plates, respectively. PBMC (1 × 10 4 cells/well) were seeded in 96-well plates and cocultured with the stimulating agent phytohemagglutinin (5 μg mL −1 ). [28] After incubation for 24 h and rinsing with PBS, the cells were treated with MPLN at siRNA concentrations of 50, 150, 250, 350, 550, and 750 nM for another 24 h. The nanoparticle mass concentrations switched from the siRNA concentration were 0.17 mg mL −1 , 0.50, 0.84, 1.17, 1.84, and 2.50 mg mL −1 , respectively. After washing with PBS three times, the cells in each well were further incubated with the cell culture media (100 μL) containing CCK-8 agent (10 μL) for 2 h. The absorbance of the samples at the wavelength of 450 nm was measured with 650 nm as the reference wavelength. The viability of the cells in the culture medium was defined as 100%.
In Vitro Cellular Uptake: The in vitro cellular uptake was evaluated in macrophages derived from THP-1 cells with CLSM (UltraVIEW Vox, PerkinElmer, USA) qualitatively and FCM (BD FACSCalibur, Franklin Lakes, NJ, USA) quantitatively. [6c,27] For CLSM analysis, macrophages, derived from THP-1 cells (2 × 10 5 cells/well) under the action of phorbol ester (200 ng mL −1 ), were seeded on a Petri dish, and cultured for 24 h. Then, the media were replaced with 2 mL of samples containing Cy5-siRNA (250 nM). After incubation for another 6 h, the cells were fixed with paraformaldehyde (4%, W/V) for 20 min and stained with Hoechst 33258 for 10 min at ambient temperature. Finally, fluorescent images were observed under CLSM.
For FCM analysis, macrophages derived from THP-1 cells were seeded, cultured, and treated with different samples as in the CLSM analysis. Finally, the cells were detached with Accutase solution and measured with FCM after resuspension in PBS (0.3 mL).
Binding of MPLN with gp120: To evaluate the binding ability of MPLN with HIV-1 gp120 molecules, solution (100 μL) containing R5-tropic HIV-1 BaL or X4-tropic HIV-1 MN gp120 recombinant proteins (2 μg mL −1 ) was added into each well of 96-well plates. [12,29] After incubating at 4°C overnight for coating, the solution in each well was removed, and the wells were washed with PBS containing Tween 20 (0.5%, W/V). Then, the wells were incubated with a blocking solution (200 μL) containing normal goat serum (4%, W/V) for 2 h. Thereafter, samples with different concentrations of Cy5-siRNA were added to the wells. After another incubation for 2 h, the samples in each well were removed, and the plates were rinsed. The fluorescence intensity of the samples at 649 nm with an emission wavelength of 680 nm was finally measured with a Spark microplate reader (Tecan, Männedorf, Switzerland).
To intuitively evaluate the binding ability of MPLN with gp120 on free HIV-1 virions, the morphology of MPLN, X4-tropic HIV-1 NL4-3 particles, R5-tropic HIV-1 AD8 particles, and the mixtures of MPLN with free HIV-1 NL4-3 or HIV-1 AD8 particles was observed using TEM, and their particle sizes and zeta potential were monitored with a Malvern Zetasizer. [12] To validate the binding ability of MPLN with gp120 expressed on HIV-1infected cells, TZM-bl cells were first infected with HIV-1 NL4-3 (0.01 multiplicity of infection, MOI) and HIV-1 AD8 (0.01 MOI). After identification of gp120 expression, the infected cells were incubated with different samples containing Cy5-siRNA (125 nM) for 2 h. After rinsing, the cells were fixed with paraformaldehyde (40 mg mL −1 ) for 20 min and stained with Hoechst 33258 for 10 min at ambient temperature. Finally, the fluorescence images were visualized under CLSM. [29] HIV-1 Neutralization: The neutralization breadth and potency of MPLN were first evaluated using pseudovirus and a single round of replication in TZM-bl cells. [12,29] The pseudovirus panel of nine geographically and genetically diverse env-pseudoviruses (AE34, B14, B16, B121, BC28, BC29, BC43, C11, and C15) representing the main subtypes and recombinant forms circulating in China were used in this paper. Pseudoviruses were obtained via the cotransfection of HEK 293T cells with an env-expressing plasmid and an env-deficient genomic backbone plasmid (pSG3ΔEnv) using polyethyleneimine. The 50% tissue culture infectious doses (TCID50) were measured with a luciferase-based assay in TZM-bl cells. In brief, serial dilutions of MPLN were incubated with pseudovirus (200 TCID50) in the presence of DEAE-dextran (15 μg mL −1 ) for 2 h, and the mixtures were further added into TZM-bl cells (1.0 × 10 4 cells/well) and incubated for 6 h. After washing with PBS, TZM-bl cells were incubated for another 48 h, and the neutralizing activity of MPLN was assessed by determining luciferase activity using a FilterMax F5 multimode microplate reader (Molecular Devices). Dose-response curves were fitted using nonlinear regression, and the IC50 and IC80 values were calculated using GraphPad Prism 8.
To further validate the HIV-1-neutralization potency of MPLN, two distinct HIV strains, that is, X4-tropic HIV-1 NL4-3 and R5-tropic HIV-1 AD8 , were also used for HIV-1-neutralization assay. Briefly, serial dilutions of www.advancedsciencenews.com www.advancedscience.com MPLN were incubated with 200 TCID50 of different viruses respectively in the presence of DEAE-dextran for 2 h, and the mixtures were further added into TZM-bl cells (2.5 × 10 4 cells/well) and incubated for 6 h. After rinsing, the cells were incubated for another 48 h, and the neutralizing activity of MPLN was assessed by measuring HIV-1 p24 production using the Alliance HIV-1 p24 Antigen ELISA kit (PerkinElmer).
To verify the enhanced HIV-1-neutralization activities of MPLN, the HIV-1 inhibition of LN, MLN, and scrambled siRNA/MPLN with a concentration equivalent to IC80 of MPLN was further evaluated. The HIV-1 inhibition of MPLN blocked with the antibody mixtures of CD4, CXCR4, and CCR5 (2:2:1, W/W/W) was also evaluated so as to confirm the possible binding site.
Inhibition of HIV-1 gp120-Induced Bystander T-Cell Killing: To investigate the effect of MPLN on HIV-1 gp120-induced killing of bystander T cells, X4-tropic HIV-1 MN or R5-tropic HIV-1 BaL gp120 recombinant proteins were incubated with LN, MLN, MPLN, and scrambled siRNA/MPLN at 25°C for 2 h. [12] Following incubation, the mixtures were added to human naive CD4 + T cells with a final nanoparticle concentration of 0.1 mg mL −1 and a final gp120 concentration of 1 μg mL −1 . The treated cells were incubated at 37°C for 24 h, and then further incubated with cell culture media (100 μL) containing CCK-8 (10 μL) for 2 h. The absorbance of the samples at 450 nm was measured with 650 nm as the reference wavelength. The viability of the cells in the culture medium was defined as 100%.
Endosomal Escape: To validate the delivery of siRNA into the cytoplasm, the endosomal escape behavior of MPLN was checked using CLSM. [30] Briefly, RAW264.7 cells (2 × 10 5 cells/well) were seeded on a Petri dish and cultured overnight. Then, the cells were incubated with MPLN at an siRNA concentration of 125 nM for 1, 3, and 6 h. After rinsing with PBS, the cells were further incubated with LysoTracker Red for 30 min. Thereafter, the cells were fixed with paraformaldehyde (4%, W/V) for 20 min and stained with Hoechst 33258 for 10 min at ambient temperature. The cells were finally observed under CLSM.
In Vitro Gene Silencing: To investigate the gene silencing of HIV-1 following treatment with MPLN, TZM-bl cells were first seeded and incubated overnight, and then the cells were attacked with X4-tropic HIV-1 NL4-3 (0.001 MOI) or R5-tropic HIV-1 AD8 (0.001 MOI) for 6 h. After washing with PBS, the cells were further incubated for 4 days and rinsed with PBS again to remove the free virus. Thereafter, the mixtures of HIV-1-infected cells and uninfected cells with the same cell number were incubated for 6 h, followed by treatment with the nanoparticles in different formulations at a siRNA concentration of 200 nM. After 8 h, the culture medium was replaced with fresh culture medium, and the cells were further incubated for 48 h (for mRNA assay) or 72 h (for WB assay). After washing, the cells were digested with trypsin, and the sediments generated by centrifugation were collected for analysis. [6c,24a] For mRNA assay, the total RNA was first extracted from the sediments using a Total RNA Extraction kit (DNase I) (GenePool, GPQ1801). Analysis was conducted on a real-time fluorescence quantitative PCR detection system (BIOER LineGene 9600Plus), and the relative gene expression was quantified using the 2 −ΔΔCt method. [7a,b] The primers for PCR amplification were as follows: glyceraldehyde 3-phosphate dehydrogenase forward: CCT CTG ACT TCA ACA GCG ACA C; glyceraldehyde 3-phosphate dehydrogenase reverse: TGG TCC AGG GGT CTT ACT CC; Tat forward: GGA AGC ATC CAG GAA GTC AG; Tat reverse: CTT GGC AAT GAA AGC AAC ACT; Rev forward: GAG ACA GAG ACA GAT CCA TTC G; Rev reverse: AGT TCC ACA ATC CTC GTT ACA A. The reaction parameters were as follows: 95°C for 5 s and then 60°C for 30 s for 45 cycles. The specificity was validated using melt curve analysis and agarose gel electrophoresis. SYBR Green was used in this section.
For WB analysis, the transfected cells were first collected and trypsinized. After centrifugation, the extracted proteins in the supernatant were quantified with the Pierce BCA protein assay and diluted to the same concentration. Thereafter, the proteins were separated on an acrylamide gel (10%, W/V) and transferred to a polyvinylidene difluoride membrane. Then, the samples were incubated with primary antibodies against tat and rev proteins, before incubating with HRP-labeled goat/anti-rabbit IgG. The protein signals were finally detected with a ChemiDoc MP imaging system (Bio-Rad, Hercules, CA, USA). [7] In Vivo Safety Evaluation: The safety of MPLN was also evaluated in PBMC humanized mice. [6c] In brief, the mice were first randomly divided into three groups (six per group), and administered physiological saline (control), free siRNA, or MPLN by intravenous injection every 2 days at a dose of 1.2 mg kg −1 converting to siRNA. Fourteen days later, approximately a venous blood sample (1.0 mL) was collected from the anesthetized animals for hemogram assay, before sacrificing the mice. The main tissues, including the heart, liver, brain, lung, and kidneys, were harvested and stained with hematoxylin and eosin for subsequent analysis.
In Vivo Distribution: To understand the behavior of MPLN in vivo, the distribution of MPLN was evaluated with fluorescence imaging in PBMC humanized mice. [6c] Briefly, the mice were randomly divided into five groups (three per group) and administered physiological saline (control) or different formulations containing Cyanine 7 labeled siRNA by tail vein injections at a dose of 1.2 mg kg −1 converting to siRNA. Subsequently, the fluorescence imaging was conducted with a NightOWL II in vivo imaging system (IVIS Lumina II, PerkinElmer, Waltham, MA, USA) at a predetermined time.
To intuitively observe the distribution of different formulations in the major organs, the other five groups of PBMC humanized mice were treated as mentioned above. After 4 h, the mice were sacrificed via cervical dislocation, and the major organs, including the heart, liver, spleen, lung, and kidneys, were excised and imaged. [6c] Statistical Analysis: Continuous variables are expressed as mean ± SD. Comparisons between the two groups were performed using the twosample t-test. In all cases, significance was defined as p < 0.05. Statistical analysis was carried out using EXCEL Software.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.