A method of preparing extracellular vesicles with reduced non-specific binding / endocytosis

Abstract

The present application relates to a method for preparing extracellular vesicles with reduced non-specific binding / endocytosis, which is obtained by knockout and deglycosylation of extracellular vesicle-related adhesion molecules, and the extracellular vesicles with reduced non-specific binding / endocytosis have a prolonged residence time in blood, further, IL12 and antibodies are loaded on the extracellular vesicles with reduced non-specific binding / endocytosis, to achieve the effect of IL12 "multimer" mode enhancement, further, the antibodies achieve targeting, reduce the effective concentration of IL12, improve safety, and also achieve high-efficiency inhibition of tumor growth under certain safety.

Description

Technical Field

[0001] This invention relates to the technical field of extracellular vesicle modification, including exosomes, and more specifically, to the technical field of knocking out and deglycosylating extracellular vesicle-related adhesion molecules, and co-loading IL12 with an antibody onto extracellular vesicles that reduce nonspecific binding / endocytosis. Background Technology

[0002] Extracellular vesicles (EVs) are vesicle-like structures secreted during cell growth. They often carry intracellular proteins and nucleic acids during their formation, and also carry extracellular proteins on their vesicle membranes or extracellularly, similar to the cell membrane. The extracellular vesicle membrane contains a wide variety of proteins, allowing EVs to bind to and endocytose almost all cells to varying degrees.

[0003] Cancer treatment typically combines chemotherapy, radiotherapy, surgery, and targeted therapy. Targeted therapy, at the cellular and molecular level, targets known carcinogenic sites. Corresponding therapeutic drugs can be designed, and once inside the body, these drugs specifically bind to and act on these carcinogenic sites, causing specific cell death. Extracellular vesicles are commonly used and highly efficient drug delivery systems; however, their use as drug delivery systems can lead to unnecessary and non-specific binding / endocytosis by cells, accelerating clearance. Therefore, ensuring stable drug delivery through extracellular vesicles in vivo and preventing their clearance remains a significant technical challenge in this field.

[0004] Interleukin-12 (IL12) is a type of cytokine. Biologically active IL12 is activated by inflammatory cells, and therefore, appropriate activation of these cells helps to eliminate tumor cells. However, overactivation can produce serious side effects, such as cytokine release syndrome (CRS). Low concentrations of IL12 are not active enough to activate the immune system and thus promote the elimination of tumor cells. Therefore, new methods are needed to exert the immune system activation ability within a safe concentration range. Improving the retention time of extracellular vesicles in the blood and enabling them to reach lesions more effectively is also an important problem that needs to be solved in the field of drug delivery. Summary of the Invention

[0005] To improve the retention time of extracellular vesicles in the blood and enable them to reach lesions more effectively, this invention obtains extracellular vesicles with reduced nonspecific binding / endocytosis by knocking out and deglycosylating extracellular vesicle-related adhesion molecules, thus prolonging their retention time in the blood. Furthermore, IL12 and antibodies are co-loaded onto the extracellular vesicles with reduced nonspecific binding / endocytosis. The structure of the vesicles achieves a "multimer" mode of IL12 to enhance drug efficacy. The antibody then targets the vesicles, reducing the effective concentration of IL12, improving safety, and also inhibiting tumor growth under certain safety conditions.

[0006] Beneficial effects:

[0007] This invention significantly reduces the non-specific binding / endocytosis of extracellular vesicles by knocking out ITGB1 alone or in combination with deglycosylation, thereby reducing the clearance caused by extracellular vesicles when used as a drug delivery system, thus improving drug delivery efficiency and final efficacy, and enabling extracellular vesicles to better target lesions.

[0008] This invention achieves efficient loading of the cytokine IL12 onto extracellular vesicles, thereby enhancing the immune system activation function of these low-clearance vesicles and efficiently achieving targeted therapeutic goals. Furthermore, by loading antibodies, the extracellular vesicles can be targeted more precisely. Specifically, by combining ITGB1 knockout and deglycosylation of extracellular vesicles, loading the cytokine IL12, and loading antibodies, IL12 can be targeted to the tumor site with higher delivery efficiency, exerting a local immune system activation function. IL12 expression on the EV membrane surface exhibits an aggregation effect compared to single-molecule IL12, significantly increasing its stimulatory function. It can exert its function with a lower dose and can exert a certain tumor-suppressive effect in vivo, particularly against hepatocellular carcinoma with high GPC3 expression. Attached image description:

[0009] Figure 1 : Flow cytometry detection of ITGB1 knockout diagram of the present invention;

[0010] Figure 2 The Western Blot diagram of this invention identifies the expression of EV ITGB1 and CD81;

[0011] Figure 3 : The EV standard curve diagram of this invention;

[0012] Figure 4 : Verification diagram of NanoLuc loading and positioning in this invention;

[0013] Figure 5The binding / endocytosis of WT EV and ITGB1-EV with different receptor cells in this invention: a EV binds / endocytosed with different receptor cells for 1 hour, and b EV binds / endocytosed with different receptor cells for 4 hours.

[0014] Figure 6 a) EV plasma pharmacokinetic curve of the present invention; b) EV tissue distribution time curve of the present invention;

[0015] Figure 7 The Western Blot diagram of EV deglycosylation in this invention;

[0016] Figure 8 The present invention describes the deglycosylation treatment of ITGB1-EV to determine its binding / endocytosis with different receptor cells. a EV binds / endocytosed with different receptor cells for 1 hour, and b EV binds / endocytosed with different receptor cells for 4 hours.

[0017] Figure 9 The present invention includes: a) plasma pharmacokinetic curve of EV and b) tissue distribution pharmacokinetic curve of EV.

[0018] Figure 10 The following figures illustrate the following: a) Immunofluorescence detection of the binding of ITGB1-mscIL12+HN3+EV to HepG2 cells; b) The binding of ITGB1-mscIL12+HN3+EV to different cells.

[0019] Figure 11 The diagram of hscIL12+EV stimulating human PBMCs to secrete IFNγ in this invention;

[0020] Figure 12 The diagram of IFNγ secretion by mouse spleen cells stimulated by mscIL12+EV according to the present invention;

[0021] Figure 13 The targeting properties of the ITGB1-mscIL12+HN3+Deg EV of the present invention in tumor-bearing mice are as follows: a. EV particle concentration per gram of tumor tissue, b. ratio of particle concentration per gram of tumor tissue to EV particle concentration in plasma, c. tumor size at the time of the experiment;

[0022] Figure 14 The tumor growth curve diagram of this invention;

[0023] Figure 15 HE staining image of the tumor in this invention. Detailed Implementation

[0024] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Example 1: Construction and collection of EVs derived from the ITGB1 knockout HEK293F suspension cell line, and in vitro and in vivo functional verification.

[0026] 1. Method:

[0027] (1) ITGB1 knockout and cell line sorting: HEK293F suspension cells were transiently transfected with ITGB1 CRISP / Cas9 / gRNA plasmid using PEI transfection reagent. After 72 h, the cells were co-incubated with APC anti-human ITGB1 antibody, and APC-negative cell populations were sorted by flow cytometry. The sorted cell populations were expanded and cultured to form the ITGB1 knockout HEK293F cell line (ITGB1-HEK293F).

[0028] (2) Validation of ITGB1 knockout cell lines: 2e5 HEK293F cells and 2e5 ITGB1-HEK293F cells were taken. The mixed cells were 1E+5 HEK293F cells and 1E+5 TGB1-HEK293F cells. After centrifugation, each cell was resuspended in 100 μl PBS, and 1 μl APC anti-human ITGB1 antibody was added. The cells were incubated at 20℃±5℃ in the dark for 30 min. After washing with PBS 3 times, the cells were resuspended in 300 μl PBS and analyzed. At the same time, after purifying EVs, the changes in ITGB1 expression on ITGB1-EVs were identified by Western blotting.

[0029] (3) NanoLuc Loading: NanoLuc was constructed using VB220306-1137jmq (Yunzhou Biotechnology, Sleeping Beauty expression vector) as a vector, and the pFc-TMD(NPTN-TMD)-ICD(EWI-F-ICD)-NanoLuc plasmid was co-transfected into cells with the Sleeping Beauty (SB) transposase expression vector pCMV-(CAT)T7-SB100X. After 72 h, Zeocin antibiotic was added to a final concentration of 100 μg / ml for pressurization until all cells in the unstable transfection group died (approximately 3 days). Cells were expanded and cultured, and the cell supernatant was collected by centrifugation at 5000 rpm for 30 min. EVs were then purified from the supernatant. A standard curve of EV particle number concentration-NanoLuc enzyme activity was established to verify the NanoLuc activity of EVs. EVs were treated with Triton X-100 and Proteinase K to verify the loading site of NanoLuc in EVs.

[0030] The Fc amino acid sequence involved (SEQ ID NO.1):

[0031] AESKYGPPCPPCPAPEAAGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSQEDPEVQFNWYVDGVEVHNAKTKPREEQFNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKGLPSS IEKTISKAKGQPREPQVYTLPPSQEEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSRLTVDKSRWQEGNVFSCSVMHEALHNHYTQKSLSLSLG

[0032] The NPTN-TMD sequences involved are:

[0033] LWPFLGILAEIIILVVIIVVY

[0034] EWI-F-ICD sequences involved:

[0035] SSHWCCKKEVQETRRERRRLMSMEMD

[0036] The relevant NanoLuc amino acid sequence is (SEQ ID NO.2):

[0037] MVFTLEDFVGDWRQTAGYNLDQVLEQGGVSSLFQNLGVSVTPIQRIVLSGENGLKIDIHVIIPYEGLSGDQMGQIEKIFKVVYPVDDHHFKVILHYGTLVIDGVTPNMIDYFGRPYEGIAVFDGKKITVTGTLWNGNKIIDERLINPDGS LLFRVTINGVTGWRLCER

[0038] (4) Cellular-level verification of the binding / endocytosis changes of EVs derived from ITGB1 knockout HEK293F cells with different cells: Recipient cells were seeded at 0.5E+5 cells per well in a 24-well plate. On the second day, the EV concentration was adjusted to 1E+10 cells / ml, and then diluted to 1E+9p / ml with complete culture medium. The supernatant was removed from the cells, and 0.5ml of EV was added. The cells were incubated at 37℃ for 1h or 4h. After the incubation time point, the supernatant was gently removed, and the cells were washed once with 1ml of PBS. 1ml of EDTA was added, and the cells were incubated for 10min to digest them. The cells were collected, centrifuged at 1000rpm for 5min, washed once with 1ml of 2% FBS / PBS, and 200μl of 2% FBS / PBS was resuspended. 50μl of the resuspended cells was added to a pure white 96-well plate, and 50μl of NanoLuc substrate was added to detect the NanoLuc activity of the sample.

[0039] (5) In vivo validation: To validate the effect of ITGB1 knockout on the tissue distribution of EVs in major organs, EVs were injected via tail vein into mice at a dose of 5E+10 mice per mouse. Peripheral whole blood was collected at 5, 15, 30, 60, 120, and 240 min to obtain plasma, liver, spleen, lung, and kidney tissue homogenates, and the NanoLuc enzyme activity in the plasma and tissue homogenates was measured. A standard curve was established using EV particle number concentration versus NanoLuc enzyme activity, and the EV particle number concentration in the tissues was calculated retrospectively.

[0040] 2. Results:

[0041] like Figure 1-6 As shown in Tables 1-4, wild-type HEK293F cells highly expressed ITGB1. After ITGB1 knockout, flow cytometry analysis showed a significant decrease in APC fluorescence in the cells after binding with the APC anti-human ITGB1 antibody. In contrast, a 1:1 mixture of wild-type and ITGB1 knockout cells clearly separated into a strongly APC-positive cell population and an ITGB1 knockout cell population, indicating successful ITGB1 knockout. Western blotting analysis showed that the EV marker CD81 was expressed in both WT EV and ITGB1-EV samples. WT EV expressed ITGB1, while ITGB1-EV did not express ITGB1 protein, indicating successful ITGB1 knockout at the EV level.

[0042] WT EV per 10 9 EV particle number concentration level enzyme activity is approximately 1E+05, ITGB1-EV per 10 9 The enzyme activity is approximately 1.6E+05 at EV particle number concentration. When the particle number concentration is between 1E+5 and 1E+10, the enzyme activity shows a linear relationship with the EV particle number concentration.

[0043] Proteinase K alone digests free NanoLuc in the system or outside the EV membrane. Adding Triton X-100 can disrupt the EV membrane structure, allowing Proteinase K to enter the EV and digest NanoLuc inside the membrane. Experimental results show that the enzyme activity of EV treated with Proteinase K alone decreases only slightly over time, while the enzyme activity of EV treated with both Triton X-100 and Proteinase K is almost undetectable at 90 min, indicating that NanoLuc is loaded inside the EV, which meets the loading target.

[0044] Following ITGB1 knockout, at 1 hour, the binding / endocytosis of EVs by human liver-derived HepG2 cells or mouse liver-derived Hepa1-6 and AML12 cells decreased, while mouse monocytes RAW264.7 cells, regardless of LPS activation, showed either increased or unchanged EV binding. At 4 hours, the binding / endocytosis of ITGB1-EVs by HepG2 and Hepa1-6 cells remained lower than that of wild-type HEK293 EVs, but the binding / endocytosis of both types of EVs by AML12 cells was similar. The results for RAW264.7 at 4 hours were similar to those at 1 hour. ITGB1 knockout promoted EV binding / endocytosis, indicating that ITGB1 knockout has different effects on binding / endocytosis at the in vitro cellular level depending on the recipient cell type. However, ITGB1 knockout can reduce the binding / endocytosis of EVs by liver-derived cells to a certain extent.

[0045] In vivo validation: The results showed that ITGB1 knockout did not significantly alter the distribution of EVs in the major organs (liver, spleen, lung, and kidney) and blood of mice.

[0046]

[0047] Table 1. NanoLuc Loading and Positioning Verification

[0048]

[0049]

[0050] Table 2. Binding / endocytosis of WT EV and ITGB1-EV with different receptor cells. a EV binds to / endocytosed different receptor cells for 1 hour, b EV binds to / endocytosed different receptor cells for 4 hours.

[0051] WT EV ITGB1-EV 5 3.25E+08 7.65E+08 15 3.65E+07 4.36E+07 30 1.53E+07 3.45E+07 60 7.47E+06 8.08E+06 120 1.16E+06 1.24E+06 240 BQL 1.19E+06

[0052] Table 3. EV plasma pharmacokinetic curve data

[0053]

[0054] Table 4. EV tissue distribution time-lapse curve data

[0055] Example 2: De-N-glycosylated HEK293F EV exhibits significantly improved in vitro and in vivo pharmacokinetic properties.

[0056] Most EV surface proteins are glycosylated, such as CD63, a classic EV marker, which is a glycoprotein. Glycoproteins participate in intracellular and extracellular signal transduction, intercellular interactions, and endocytosis by regulating cell adhesion, ligand binding, and receptor dimerization. In other words, glycoproteins are a crucial pathway for EV binding and endocytosis with receptor cells. N-glycosylation is one of the main types of protein glycosylation modification. Therefore, to reduce non-specific binding of EVs, we used PNGase F to remove N-glycosylation on the EV surface and verified the changes in EV binding / endocytosis at the in vitro cellular level after deglycosylation, as well as changes in pharmacokinetics and tissue distribution in vivo.

[0057] 1. Method:

[0058] (1) Deglycosylation of EVs (DegEV) and validation: 10 μl of PNGase F and 100 μl of 10×Reaction Buffer were added to every 2E+10 EVs, and the volume was increased to 1 ml with PBS. The EV concentration was 2E+10 p / ml. The mixture was incubated at 37℃ for 1 h. CD63 and GAPDH were detected by Western blotting.

[0059] (2) Cellular-level verification of changes in EV binding / endocytosis with recipient cells after deglycosylation: 0.5E5 cells were seeded into 24-well plates. On the second day, EVs were counted, and the concentration was adjusted to 1E10 cells / ml with PBS, then diluted to 1E9p / ml with 10% DMEM medium. The supernatant was removed from the cells, and 0.5ml of complete DMEM medium containing EVs was added. The cells were incubated at 37°C for 1h or 4h. After the incubation time point, the supernatant was gently removed, the cells were washed once with 1ml PBS, and 1ml EDTA was added. The cells were incubated for 10min to digest them. The cells were collected, centrifuged at 1000rpm for 5min, and washed once with 1ml 2% FBS / PBS. 200μl of the cells were resuspended in 2% FBS / PBS, and 50μl of each cell was added to a 96-well plate. 50μl of NanoLuc substrate was added to each well. The NanoLuc activity of the samples was measured.

[0060] (3) In vivo validation: To verify the effect of deglycosylation on the tissue distribution of EVs in major organs, EVs were injected via the tail vein at a dose of 5E+10 mice. Peripheral whole blood was collected at 5, 15, 30, 60, 120, and 240 min to obtain plasma, liver, spleen, lung, and kidney tissue homogenates. NanoLuc enzyme activity in the plasma and tissue homogenates was measured. A standard curve was established using EV particle number concentration versus NanoLuc enzyme activity, and the EV particle number concentration in the tissues was calculated retrospectively.

[0061] 2. Results

[0062] like Figure 7-9 As shown in Table 5-7, the CD63 and GAPDH of the deglycosylated EVs were detected by Western Blot: The results showed that the detection results of the two EVs were consistent. CD63 showed a diffuse band before deglycosylation due to the different degrees of glycosylation modification, while the band was relatively concentrated after deglycosylation, indicating that the deglycosylation was successful.

[0063] In vitro cellular validation: The binding / endocytosis experiments of ITGB1 knockout combined with deglycosylated EVs (ITGB1-DegEV) and ITGB1 knockout EVs (ITGB1-EV) with different receptor cells showed that deglycosylation promoted the binding of EVs with HepG2 and Hepa1-6 cells. However, for AML12 and RAW264.7 cells, less ITGB1-DegEV was bound / endocytosed than ITGB1-EV at 1 h, but more DegEV was bound / endocytosed at 4 h. LPS-stimulated RAW264.7 cells bound / endocytosed more DegEV than ITGB1-EV at 1 h, but less at 4 h. This indicates that deglycosylation treatment has an inconsistent effect on the binding / endocytosis ability of EVs with different receptor cells.

[0064] In vivo validation: In vivo results showed that deglycosylation alone (Deg EV) or ITGB1 knockout (ITGB1-EV) did not significantly change the in vivo tissue distribution and plasma retention time of EV, while deglycosylation combined with ITGB1 knockout significantly increased the plasma retention time of EV and correspondingly reduced its accumulation in major organs (liver, spleen, lung, kidney).

[0065]

[0066]

[0067] Table 5. Data on binding / endocytosis of deglycosylated ITGB1-EV with different receptor cells

[0068] WT EV Deg EV ITGB1-EV ITGB1-Deg EV 5 3.25E+08 6.27E+08 7.65E+08 3.87E+09 15 3.65E+07 1.47E+08 4.36E+07 1.26E+09 30 1.53E+07 2.27E+07 3.45E+07 6.48E+07 60 7.47E+06 9.09E+06 8.08E+06 4.08E+07 120 1.16E+06 1.18E+06 1.24E+06 4.54E+06 240 BQL 1.38E+06 1.19E+06 3.78E+06

[0069] Table 6. EV Plasma Pharmacokinetic Data

[0070]

[0071] Table 7. EV tissue distribution and time-lapse curve data

[0072] Example 3: Constructing ITGB1-scIL12+HN3+Deg EV and verifying its in vivo and in vitro functions

[0073] By loading antibodies onto EVs, they are given a certain targeting function, and by simultaneously loading scIL12, they can activate the immune system. EVs loaded with both antibodies and scIL12 can achieve a certain in vivo tumor-suppressing effect.

[0074] 1. Method

[0075] (1) hscIL12, mscIL12 Sleeping Beauty expression vector, and HN3 lentivirus were constructed respectively: hscIL12 and mscIL12 were constructed using VB220306-1137jmq (Yunzhou Biotechnology, Sleeping Beauty expression vector) as vectors, and pIL12-Fc-TMD(NPTN-TMD)-ICD(EWI-F-ICD)-NanoLuc plasmid was constructed and co-transfected into cells with Sleeping Beauty (SB) transposase expression vector pCMV-(CAT)T7-SB100X. HN3 lentivirus with the structure pLenti-HN3-Fc-ITGB1_TMD-ITGB1_ICD-PGK-BlasticidinS-WPRE was constructed. HN3 lentivirus and the IL12 Sleeping Beauty plasmid were used to infect / transfect ITGB1 knockout HEK293F cells, either alone or co-infected. Cells were cultured for 72 hours after transfection, followed by selection with puromycin and / or Blasticidin S antibiotics for another 72 hours. Cell supernatant was collected by centrifugation at 5000 rpm for 30 minutes, and EVs were purified from the supernatant. ITGB1-hscIL12+EV (or ITGB1-mscIL12+EV) and ITGB1-hscIL12+HN3+EV (or ITGB1-mscIL12+HN3+EV) were obtained. Further deglycosylation yielded ITGB1-hscIL12+HN3+DegEV (or ITGB1-mscIL12+HN3+Deg EV). The aforementioned EV is referred to as hscIL12 in in vitro human PBMC cell experiments. + EV, in vitro hGPC3 + In cell binding assays, it is abbreviated as HN3 + EV, abbreviated as mscIL12 in the mouse spleen lymphocyte stimulation assay. + EV.

[0076] The mscIL12 sequence involved:

[0077] mscIL12-P35 (SEQ ID NO.3):

[0078] MCQSRYLLFL ATLALLNHLS LARVIPVSGPARCLSQSRNLKTTDDMVK TAREKLK HYSCTAEDIDHEDI TRDQTSTLKT CLPLELHKNE SCLATRETSSTTRGSCLPPQ KTSL MMTLCL GSIYEDLKMYQTEFQAINAALQNHNHQQIILDKGMLVAIDELMQSLNHNGET LRQKPPVG EADPYRVKMKLCILLHAFSTRVVTINRVMG YLSSA

[0079] mscIL12-P40 (SEQ ID NO.4):

[0080] MCPQKLTISWFAIVLLVSPLMAMWELEKDVYVVEVDWTPDAPGETVNLTCDTPEEDDITWTSDQRHGVIGSGKTLTITVKEFLDAGQYTCHKGGETLSHSHLLLHKKENGIWSTEILKNFKNKTFLKCEAPNYSGRFTCSWLVQRNMDLKFNIKSSSSSPDSRAVTCGMASLSAEKVTLD QRDYEKYSVS CQEDVTCPTAEETLPIELALEARQQNKYENYSTSFFIRDIIKPDPPKNLQ MKPLKNSQVEVSWEYPDSWSTPHSYFSLKFFVRIQRKKEKMKETEEGCNQKGAFLVEKTSTEVQCKGGNV CVQAQDRYYN SSCSKWACVP CRVRS

[0081] The hscIL12 sequence involved:

[0082] hscIL12-P35 (SEQ ID NO.5):

[0083] RNLPVATPDPGMFPCLHHSQNLLRAVSNMLQKARQTLEFYPCTSEEIDHEDITKDKTSTVEACLPLELTKNESCLNSRETSFITNGSCLASRKTSFMMALCLSSIYEDLKMYQVEFKTMNAKLLMDPKRQIFLDQNMLAVIDELMQALNFNSETVPQKSSLEEPDFYKTKIKLCILLHAFRIRAVTIDRVMSYLNAS

[0084] hscIL12-P40 (SEQ ID NO.6):

[0085] IWELKKDVYVVELDWYPDAPGEMVVLTCDTPEEDGITWTLDQSSEVLGSGKTLTIQVKEFGDAGQYTCHKGGEVLSHSLLLLHKKEDGIWSTDILKDQKEPKNKTFLRCEAKNYSGRFTCWWLTTISTDLTFSVKSSRGSSDPQGVTCGAATLSAERVRGDNKEYEYSVECQEDSACPAAEESLPIEVMVDAVHKLKYENYTSSFFIRDIIKPDPPKNLQLKPLKNSRQVEVSWEYPDTWSTPHSYFSLTFCVQVQGKSKREKKDRVFTDKTSATVICRKNASISVRAQDRYYSSSWSEWASVPCS

[0086] The ITGB1 amino acid sequence involved (SEQ ID NO.7):

[0087] QTDENRCLKANAKSCGECIQAGPNCGWCTNSTFLQEGMPTSARCDDLEALKKKGCPPDDIENPRGSKDIKKNKNVTNRSKGTAEKLKPEDITQIQPQQLVLRLRSGEPQTFTLKFKRAEDYPIDLYYLMDLSYSMKDDLENVKSLGTDLMNEMRRITSDFRIGFGSFVEKTVMPYISTTPAKLRNPCTSEQNCTSPFSYKNVLSLTNKGEVFNELVGKQRISGNLDSPEGGFDAIMQVAVCGSLIGWRNVTRLLVFSTDAGFHFAGDGKLGGIVLPNDGQCHLENNMYTMSHYYDYPSIAHLVQKLSENNIQTIFAVTEEFQPVYKELKNLIPKSAVGTLSANSSNVIQLIIDAYNSLSSEVILENGKLSEGVTISYKSYCKNGVNGTGENGRKCSNISIGDEVQFEISITSNKCPKKDSDSFKIRPLGFTEEVEVILQYICECECQSEGIPESPKCHEGNGTFECGACRCNEGRVGRHCECSTDEVNSEDMDAYCRKENSSEICSNNGECVCGQCVCRKRDNTNEIYSGKFCECDNFNCDRSNGLICGGNGVCKCRVCECNPNYTGSACDCSLDTSTCEASNGQICNGRGICECGVCKCTDPKFQGQTCEMCQTCLGVCAEHKECVQCRAFNKGEKKDTCTQECSYFNITKVESRDKLPQPVQPDPVSHCKEKDVDDCWFYFTYSVNGNNEVMVHVVENPECPTGPD

[0088] The HN3 sequence involved (SEQ ID NO.8):

[0089] MQVQLVQSGGGLVQPGGSLRLSCAASYFDFDSYEMSWVRQAPGKGLEWIGSIYHSGSTYYNPSLKSRVTISRDNSKNTLYLQMNTLRAEDTATYYCARVNMDRFDYWGQGTLVTVSSS

[0090] (2) Verification of the targeting ability of ITGB1-mcIL12+HN3+EV (or HN3+EV for short): To verify the targeting ability of EV, HepG2 cells with high expression of human GPC3 protein were used for immunofluorescence verification, and HepG2 cells, Hepa1-6, 293T, and modified Hepa1-6-hGPC3 and 293T-hGPC3 cells with high expression of human GPC3 protein were used for cell binding verification.

[0091] Immunofluorescence: Cells were pre-seeded in 24-well plates placed on slides. The next day, the supernatant was removed, and the cells were washed once with PBS. Fixed with paraformaldehyde for 10 min, followed by three 5-min washes with PBS. Blocked with 1 ml of 5% BSA / PBS buffer at room temperature for 2 h, followed by three 5-min washes with PBS. WT EVs were added at a cell:EV ratio of 1:30000, and the cells were incubated at 25℃±5℃ for 2 h. The cells were washed three times with PBS for 5 min each. 5E+10 cells of ITGB1-HEK293 F EV or ITGB1-mscIL12+HN3+EV were added to 1 ml of PKH67 dye buffer and 6 μl of dye, and incubated at 25℃±5℃ in the dark for 30 min. Free dye was removed using Core 700. 2E+10 cells were added to a slide (1 ml volume), and incubated at 4℃ for 4 h. The cells were washed five times with PBST for 5 min each. After staining with DAPI for 3 minutes, wash 3 times, mount the slide, and take a picture after the slide dries.

[0092] (3) Cell binding: After digestion with EDTA, cells were centrifuged at 1000 rpm for 5 min and resuspended in culture medium. 400 μl of WT EV was added at a concentration of 4E6 cells / 1.5 ml, with a EV storage concentration of 3.88E11 p / ml. The cells were shaken at 4°C to mix and incubated for 1 h. Pre-incubation of the ITGB1-mscIL12+HN3+EV+hGPC3 proteome was performed at 4°C for 30 min. The ITGB1-mscIL12+HN3+EV storage concentration was 2.15E+11. 367 μl of this EV was added to 4000 μl of PBS to prepare a concentration of 1.8E10 p / ml. The ITGB1-mscIL12+HN3+EV+hGPC3 group consisted of 1400 μl EV plus 100 μl of human GPC3 protein at a storage concentration of 100 μg / ml; the ITGB1-mscIL12+HN3+EV+Fc-hIgG1 group consisted of 1400 μl EV plus 100 μl of human Fc-IgG1 protein at a storage concentration of 100 μg / ml. Pre-blocked WT EV cells were aliquoted into 100 μl tubes, approximately 2E+5 cells / tube, and 150 μl of EV was added to each group. Cells were incubated at 4°C for 1 h. The cells were washed twice with 1 ml of 2% FBS / PBS each time. The supernatant was discarded, and the cells were resuspended in 200 μl of PBS. 30 μl of cells were mixed with 30 μl of NanoLuc substrate for NanoLuc enzyme activity assay. ITGB1-hscIL12+EV in vitro stimulation of human PBMCs to express IFNγ: 200,000 human PBMCs were added to 96-well plates, along with 1 μg / ml antiCD3 antibody, 1 μg / ml antiCD28 antibody, and different concentrations of control EV (without hscIL12 loading), hscIL12+EV, or hIL12 protein. After 4 days, the cell supernatant was collected and the concentration of human IFNγ in the supernatant was measured using an ELISA kit (Shanghai Enzyme Linked Laboratory, EK-180-96).

[0093] (4) ITGB1-mscIL12+HN3+EV (or mscIL12+EV for short) in vitro stimulation of mouse spleen cells to express IFNγ: 200,000 mouse spleen cells were added to a 96-well plate, and 1 ng / ml antiCD3 antibody and different concentrations of control EV (without mscIL12) or mscIL12+EV were added. After 4 days, the cell supernatant was collected and the concentration of mouse IFNγ in the supernatant was detected by ELISA kit (Shanghai Enzyme Linker, EK-280-96).

[0094] (5) Targeting of ITGB1-mscIL12+HN3+Deg EV (or HN3+EV for short) in Hepa1-6-hGPC3 subcutaneous tumor model mice: Establishment of Hepa1-6-hGPC3 subcutaneous tumor model mice: 5×106 cells / 100μl were seeded subcutaneously in the axilla of C57 mice. When the tumors grew to 500-800mm3, the mice were divided into two groups according to tumor size. The concentration of ITGB1-mscIL12+HN3+Deg EV was adjusted to 2.5E+11 cells / ml, and 200μl was injected via the tail vein. After 30 minutes, the tumors of the tumor-bearing mice were harvested, and the number of EV particles in the tumors was detected.

[0095] (6) Antitumor effect of ITGB1-mscIL12+HN3+Deg EV in Hepa1-6-hGPC3 subcutaneous tumor model mice: On the 4th and 7th days after tumor cell inoculation, the EV group was administered 5E+10 cells / mouse, and rmIL12 was administered 0.7ng / mouse and 3μg / mouse via tail vein. The tumor volume of mice was measured and weighed every two days during the period. At the end of the experiment, mouse tumors were taken for pathological sectioning.

[0096] 2. Results:

[0097] like Figure 10-15 As shown in Table 8-13, the cell-level targeting validation of ITGB1-mscIL12+HN3+EV (or abbreviated as HN3+EV) was performed as follows: Immunofluorescence detection: In the figure, blue represents the cell nucleus and green represents PKH67-labeled EV. The results showed that EV loaded with HN3 bound to GPC3 on the surface of HepG2 cells, forming a green ring around the membrane, while the control EV (without HN3) bound to the cells very weakly.

[0098] Cell binding assay: After pretreatment with hGPC3 protein, the HN3 site on the EV loaded with HN3 binds to free hGPC3. Therefore, the binding of EV to cells expressing hGPC3 is significantly lower than that of control EV, while there is no significant difference in cells not expressing hGPC3. This indicates that HN3+EV can specifically bind to hGPC3 on cells.

[0099] hscIL12+EV stimulates IFNγ secretion in human PBMCs: Compared with control EVs (unloaded hscIL12, cont. EV), hscIL12-loaded EVs can activate human PBMCs and stimulate IFNγ secretion. The activation effect shows a clear dose-dependent effect, and the activation effect on PBMCs gradually increases with the increase of EV particle number. 6 The activation effect of 10 ng / mL (2.4 ng) hscIL12+EV was stronger than that of 10 ng / mL (2.4 ng) hscIL12 protein;

[0100] mscIL12+EV stimulates mouse spleen cells to secrete IFNγ: Compared with the control EV (ContEV) without mscIL12 loading, mouse spleen cells loaded with mscIL12 can be stimulated to secrete IFNγ regardless of whether they are pre-activated with anti-CD3 antibody.

[0101] Targeting of ITGB1-mscIL12+HN3+DegEV (HN3+EV for short) in Hepa1-6-hGPC3 subcutaneous tumor model mice: The results showed that HN3+EV had significantly higher particle number concentration per gram of tumor and organ index (particle number concentration per gram of tumor / particle number concentration in plasma) in the mouse Hepa1-6-hGPC3 subcutaneous tumor model mice than unloaded HN3+EV, indicating that loading HN3 can significantly enhance the targeting ability of EV;

[0102] The tumor-suppressing effect of ITGB1-mscIL12+HN3+DegEV in Hepa1-6-hGPC3 subcutaneous tumor model mice: The results showed that ITGB1-mscIL12+HN3+DegEV could significantly slow down the tumor growth rate and inhibit tumor growth ( Figure 14 The tumor growth curve showed that the tumor-suppressing effect was better than that of the 3μg IL12 protein group. HE staining results showed that the rmIL12 3μg group and the ITGB1-mscIL12+HN3+DegEV group had severe tumor liquefaction, while the rmIL12 70ng group and the ITGB1-Deg EV group only showed necrosis in the central area due to lack of nutrition and oxygen. Figure 15 (Tumor HE staining).

[0103]

[0104] Table 8. Data on the binding of ITGB1-mscIL12+HN3+EV to different cells

[0105]

[0106]

[0107] Table 9. Data on IFNγ secretion by human PBMCs stimulated by hscIL12+EV

[0108]

[0109] Table 10. Data on IFNγ secretion from mouse spleen cells stimulated by mscIL12+EV

[0110] ITGB1-mscIL12+Deg EV ITGB1-mscIL12+HN3+Deg EV 895000 2410000 877000 2630000 2200000 2340000 1650000 2190000 732000 2200000 1110000 2240000

[0111] Table 11. EV particle number concentration data per gram of tumor tissue

[0112] ITGB1-mscIL12+Deg EV ITGB1-mscIL12+HN3+Deg EV 0.05 0.24 0.06 0.26 0.11 0.19 0.07 0.21 0.07 0.12 0.07 0.11

[0113] Table 12. Ratio of particle number concentration in tumor tissue per gram to EV particle number concentration in plasma

[0114]

[0115]

[0116] Table 13. Tumor growth data

Claims

1. A method for preparing extracellular vesicles (EVs) with reduced nonspecific binding / endocytosis, characterized in that: The specific steps include: (1) Knock out ITGB1 expression at the HEK293F cell level to achieve ITGB1 knockout at the EV level; (2) Sorting and validation of the ITGB1 knockout HEK293F cell line; (3) Loading EVs into the ITGB1 knockout HEK293F cell line with NanoLuc; (4) Verify at the cellular level the changes in binding / endocytosis function of EVs derived from HEK293F cells with ITGB1 knockout; (5) In vivo verification of the effect of ITGB1 knockout on the tissue distribution of EV in major organs; The amino acid sequence of ITGB1 is shown in SEQ ID NO.

7.

2. The preparation method according to claim 1, characterized in that: In step (1), the ITGB1 CRISPR / Cas9 / gRNA plasmid was transiently transfected into HEK293F suspension cells using PEI transfection reagent. After 72 hours, the cells were co-incubated with APC anti-human ITGB1 antibody, and APC negative cell populations were sorted by flow cytometry. The sorted cell populations were then expanded and cultured. This cell population is the ITGB1 knockout HEK293F cell line.

3. The preparation method according to claim 2, characterized in that: In step (2), the Western Blot method is used to identify changes in ITGB1 expression on ITGB1-EV.

4. The preparation method according to claim 3, characterized in that: In step (3), NanoLuc was constructed using VB220306-1137jmq as a vector to construct the pFc-TMD(NPTN-TMD)-ICD(EWI-F-ICD)-NanoLuc plasmid, which was co-transfected into cells with the Sleeping Beauty transposase expression vector pCMV-(CAT)T7-SB100X. After 72 h, Zeocin antibiotic with a final concentration of 100 μg / ml was added and pressurized until all cells in the unstable transfection group died. The cells were expanded and cultured, and the cell supernatant was collected by centrifugation at 5000 rpm for 30 min. EVs were purified from the supernatant, and a standard curve of EV particle number concentration-NanoLuc enzyme activity was established to verify the NanoLuc activity of EVs. EVs were treated with Triton X-100 and Proteinase K to verify the loading position of NanoLuc in EVs.

5. The preparation method according to claim 4, characterized in that: In step (4), recipient cells are seeded into a 24-well plate at a density of 0.5E+5 cells per well. After treatment, 50 μl of NanoLuc detection substrate is added to detect the NanoLuc activity of the sample.

6. The preparation method according to claim 5, characterized in that: In step (5), EV was injected via the tail vein at a dose of 5E+10 mice. Peripheral whole blood was collected at 5, 15, 30, 60, 120, and 240 min to obtain plasma, liver, spleen, lung, and kidney tissue homogenates. The NanoLuc enzyme activity in the plasma and tissue homogenates was detected. A standard curve was established using the EV particle number concentration-NanoLuc enzyme activity, and the EV particle number concentration in the tissue was calculated back.

7. The preparation method according to claim 1, characterized in that: It also includes step (6), which involves combined deglycosylation of ITGB1 knockout cell-derived EVs.

8. The preparation method according to claim 7, characterized in that: In step (6), for every 2E+10 EVs from ITGB1 knockout cells, 10 μl of PNGase F and 100 μl of 10×Reaction Buffer were added, and the system was supplemented with PBS to 1 ml. The EV concentration was 2E+10 p / ml. The cells were incubated at 37℃ for 1 h, and CD63 and GAPDH were detected by Western Blot.

9. The preparation method according to claim 8, characterized in that: To verify changes in EV binding / endocytosis with receptor cells after deglycosylation at the cellular level, NanoLuc was added as a substrate and the NanoLuc activity of the sample was measured.

10. The preparation method according to claim 8, characterized in that: To verify the effect of deglycosylation on the tissue distribution of EVs in major organs in vivo, EVs were injected via the tail vein at a dose of 5E+10 mice. Peripheral whole blood was collected at 5, 15, 30, 60, 120, and 240 min to obtain plasma, liver, spleen, lung, and kidney tissue homogenates. The NanoLuc enzyme activity in plasma and tissue homogenates was detected. A standard curve was established using EV particle number concentration versus NanoLuc enzyme activity, and the EV particle number concentration in tissues was calculated.

11. The preparation method according to claim 1, characterized in that: It also includes step (7), loading antibodies onto EVs to give them a certain targeting function.

12. The preparation method according to claim 11, characterized in that: In step (7), the pIgkappa-HA-HN3-TMD(PDGFRβ) plasmid is constructed using pHEK293_Ultra_Expression_I as a vector. The plasmid is stably infected or transfected into ITGB1 knockout HEK293F cells by constructing a lentivirus pLenti-HN3-Fc-ITGB1_TMD-ITGB1_ICD-PGK-Blasticidin S-WPRE or a transfection system.

13. The preparation method according to claim 12 further includes step (8), which enables the immune system to be activated by simultaneously loading scIL12 onto the EV.

14. The preparation method according to claim 13, characterized in that: In step (8), hscIL12 and mscIL12 were used as vectors to construct the pIL12-Fc-TMD(NPTN-TMD)-ICD(EWI-F-ICD)-NanoLuc plasmid, which was co-transfected into cells with the Sleeping Beauty transposase expression vector pCMV-(CAT)T7-SB100X. HN3 lentivirus and IL12 Sleeping Beauty plasmid were co-infected / transfected into ITGB1 knockout HEK293F cells. The cells were further screened and the cell supernatant was collected by centrifugation and purified to obtain ITGB1-scIL12+HN3+ EV. After further deglycosylation, ITGB1-scIL12+HN3+Deg EV was obtained.

15. The preparation method according to claim 14, characterized in that: In step (8), to verify the targeting of EVs, immunofluorescence verification was performed using HepG2 cells that highly expressed human GPC3 protein, and cell binding verification was performed using HepG2 cells, Hepa1-6, HEK293T, and modified Hepa1-6-hGPC3 and HEK293T-hGPC3 cells that highly expressed human GPC3 protein.

16. The preparation method according to claim 14, characterized in that: Human PBMCs were treated with CD28 antibody, CD3 antibody, and different concentrations of control EV. The expression of IFNγ in human PBMCs stimulated by ITGB1-hscIL12+EV in vitro was detected by ELISA.

17. The preparation method according to claim 14, characterized in that: Mouse spleen cells were treated with CD3 antibody and different concentrations of control EV, and the expression of IFNγ in mouse spleen cells stimulated in vitro by ITGB1-scIL12+HN3+EV was detected by ELISA.

18. The preparation method according to claim 14, characterized in that: A mouse model of Hepa1-6-hGPC3 subcutaneous tumor was established. ITGB1-mscIL12+HN3+Deg EV was injected via the tail vein to verify the targeting effect of ITGB1-mscIL12+HN3+Deg EV in the mouse model of Hepa1-6-hGPC3 subcutaneous tumor.

19. The preparation method according to claim 18, characterized in that: On the 4th and 7th day after tumor cell inoculation, the drug was administered. The tumor volume of the mice was measured and weighed every two days. At the end of the experiment, the tumors were taken for pathological sections to verify the tumor-suppressing effect of ITGB1-mscIL12+HN3+Deg EV in Hepa1-6-hGPC3 subcutaneous tumor model mice.

20. An extracellular vesicle with reduced nonspecific binding / endocytosis prepared by the method of any one of claims 1-19.

21. The use of the extracellular vesicles as described in claim 20 in the preparation of a medicament.

22. The application according to claim 21, characterized in that: The drug is a treatment for hepatocellular carcinoma that highly expresses GPC3.

23. A drug, characterized in that, Its raw materials include the extracellular vesicles as described in claim 20.

Images