Gene interference vector, adeno-associated virus and construction method thereof, and application of composition of gene interference vector and adeno-associated virus in treatment of acute myelogenous leukemia

Abstract

The invention discloses a gene interference vector as well as a construction method and application thereof in treating acute myelogenous leukemia. The pNM6-miGBA prepared by the invention can respond to the activity of NF-kappa B in acute myelogenous leukemia cells to the greatest extent, accurately and efficiently silence GBA, inhibit the enzymatic activity of GBA, accumulate GlcCer, induce lysosomal storage disorder and trigger multiform death, mainly ferroptosis, of the acute myelogenous leukemia cells, but has no significant toxicity to normal cells. The invention is expected to provide a new technology and a new reagent for the treatment of acute myelogenous leukemia.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of cancer gene therapy biotechnology, and particularly relates to a gene interference vector, an adeno-associated virus and a construction method thereof, and application of a composition thereof in treating acute myeloid leukemia. BACKGROUND

[0002] Acute myeloid leukemia (AML) is a highly heterogeneous malignant tumor derived from bone marrow hematopoietic stem / progenitor cells, and is mainly characterized by acute onset, high relapse rate and treatment resistance. Although targeted therapy and immunotherapy have made progress in some subtypes, traditional induction chemotherapy is often difficult to cure in elderly patients and cases containing drug-resistant leukemia stem cells (LSCs). At the molecular level, the NF-κB signaling pathway is abnormally activated in AML cells, which participates in the regulation of cell proliferation, anti-apoptosis, antioxidant defense and stress response. Persistent NF-κB activity (p-p65 / p50) helps the survival and chemotherapy resistance of LSCs. Therefore, inhibiting the NF-κB-related antioxidant network and enhancing intracellular oxidative stress has become one of the important strategies to overcome drug resistance and eliminate LSCs. However, direct targeting of the NF-κB signal still faces great challenges in drug development, because some normal cells need moderate NF-κB activation to maintain normal physiological functions, and NF-κB inhibitors have broad spectrum, which may injure normal cells and cause serious side effects. Therefore, how to take advantage of the significant difference in NF-κB activity between AML cells and normal cells to achieve efficient treatment of AML and minimize damage to normal cells is a major problem to be solved in the development of AML therapy based on NF-κB.

[0003] Previous studies have shown that lysosomal dysfunction in AML cells is closely related to lipid metabolism reprogramming, and also confers stronger antioxidant capacity and drug resistance to tumor cells. By precisely targeting the regulation of lysosomal membrane permeability and function, inducing lysosome-dependent cell death, disrupting redox balance and inhibiting the survival of leukemia stem cells become a very potential treatment strategy. When the lysosomal membrane permeability increases or the integrity is damaged, a large amount of toxic contents such as free iron in the lysosome can escape to the cytoplasm, triggering multiple lethal pathways, including ferroptosis, autophagy, apoptosis, necrosis, pyroptosis, lysosome-dependent cell death, etc. GBA (Glucocerebrosidase) is responsible for hydrolyzing glucosylceramide (GlcCer) into ceramide (Cer) and glucose (Glu), and is a key enzyme for maintaining lysosomal lipid metabolism and membrane composition homeostasis. Inactivation of GBA will lead to GlcCer accumulation and induce lysosomal storage disorders and functional impairment, changing lipid homeostasis and membrane permeability. Therefore, the activity of GBA can directly affect lysosomal function and trigger multiple forms of cell death, and is a potential target for the treatment of malignant tumors.

[0004] The interaction between NF-κB signaling pathway and lysosomal function will provide new treatment ideas. On the one hand, the sustained activation of NF-κB signal can regulate the expression of lysosome-related genes and affect cell survival ability; on the other hand, the release of free iron and other contents from lysosomal membrane rupture can further activate the stress response and exacerbate the positive feedback of NF-κB signaling pathway, forming a cyclic regulation. NF-κB is abnormally activated in AML and is related to the occurrence, progression and chemotherapy resistance of leukemia. Although there are significant differences in NF-κB activity between AML and normal cells, current NF-κB inhibitors lack selectivity, and direct inhibition of NF-κB will damage normal cells, so it is difficult to be used for clinical treatment of AML. Abnormal lysosomal function in AML cells confers drug resistance and antioxidant properties. GBA is a key enzyme in lysosomes that catalyzes the hydrolysis of GlcCer to generate Cer and glucose, and its inhibition will lead to GlcCer accumulation in lysosomes and lysosomal storage disorders, causing cell death. Lysosomes contain a large amount of contents, and once the lysosomal membrane is damaged or the permeability is changed, the contents such as free iron and hydrolytic enzymes will be released into the cytoplasm, causing cell stress and multiple forms of cell death, including apoptosis, necrosis, autophagy, ferroptosis, pyroptosis, etc. Inactivation of GBA will lead to GlcCer accumulation and induce lysosomal storage disorders and functional impairment, changing lipid homeostasis and membrane permeability. Therefore, the activity of GBA can directly affect lysosomal function and trigger multiple pathways of cell death, and is an important target for the synergistic treatment of tumors and nanomedicines, especially IONPs.

[0005] Since the concept of ferroptosis was proposed in 2012, its potential value in cancer treatment has attracted widespread attention. Studies have shown that ferroptosis can kill tumor stem cells and overcome tumor drug resistance, so this form of cell death is considered as the lifeline of refractory tumors. Ferroptosis is a form of cell death that depends on iron and is driven by lipid peroxidation (lipROS), and the immune system may prevent tumor occurrence partly through ferroptosis. Ferroptosis can spread in waves between cells and show strong killing effect on adjacent cells. Therefore, the anti-tumor effect of ferroptosis has been rapidly and widely explored in different cancers. Iron oxide nanoparticles (IONPs) as ferroptosis inducers to induce iron overload in cells are widely used in tumor treatment and imaging. Feraheme (ferumoxytol, FeNPs) is an FDA-approved intravenous iron supplement used clinically to treat iron deficiency anemia in patients with chronic kidney disease, and its core is composed of iron oxide (Fe3O4 / γ-Fe2O3), mainly localized in lysosomes. In addition, FeNPs can kill AML cells with low expression of ferroportin (FPN) through Fenton reaction, while also having a protective effect on hematopoietic stem cells. However, single FeNPs, even at a high dose (5 mg / kg), have not shown ideal killing effect on AML cells. Through efficient use of Fenton reaction catalyzed by FeNPs to achieve significant killing of AML cells, it will become an important breakthrough in the clinical application of FeNPs.

[0006] Therefore, on the one hand, how to effectively combine NF-κB with AML to achieve precise intervention on AML through specific targeting mechanism while protecting normal tissues from damage is an important challenge that needs to be solved in the current AML treatment field. In-depth research and clinical application of this strategy will play a crucial role in overcoming treatment bottlenecks, improving efficacy and improving patient prognosis. On the other hand, taking advantage of the abnormal activation of NF-κB in AML as a molecular background to regulate the expression of effector genes to promote lysosome rupture or inhibit GBA to enhance the Fenton reaction catalyzed by FeNPs may become an important strategy to enhance killing of AML (especially LSCs), overcome drug resistance and promote clinical transformation of FeNPs, but the selectivity and safety of normal cells need to be balanced. SUMMARY

[0007] Invention purposes: In view of the problem of poor prognosis of AML and accompanied by NF-κB activation and lysosomal dysfunction, the purpose of the present application is to provide a gene interference vector for AML treatment, which can express miGBA in response to highly activated NF-κB in AML cells, specifically silence GBA and inhibit its activity, cause GlcCer accumulation, and ultimately cause lysosomal storage disease, induce AML cells to die in multiple forms dominated by ferroptosis. The vector is an AML treatment vector with AML selectivity and no significant toxicity to normal cells.

[0008] In view of the problem that existing AML gene therapy and intravenous iron supplements are difficult to significantly induce AML cell ferroptosis, the second purpose of the present application is to provide a composition based on recombinant adeno-associated virus and intravenous iron supplement for AML and its application. The recombinant adeno-associated virus and intravenous iron supplement based on the present application can significantly induce AML cell ferroptosis. The composition kills AML cells by cooperating with the two materials of recombinant adeno-associated virus and intravenous iron supplement, thereby achieving the treatment of AML.

[0009] Technical scheme: In order to solve the above technical problems, the first aspect of the present application provides a gene interference vector, which comprises a vector skeleton, wherein an AML-specific promoter NM gene fragment and a miGBA gene fragment are sequentially included in the vector skeleton, the AML-specific promoter NM gene fragment comprises a plurality of NF-κB Motif gene fragments, a spacer gene fragment and a minimal promoter gene fragment, the miGBA gene fragment sequentially comprises a 5' end 20-60 bp base fragment of miR30, an shGBA and a 3' end 20-60 bp base fragment of miR30, and the shGBA is a gene fragment of shRNA based on miR30 structure and targeting GBA mRNA.

[0010] GBA is a key enzyme in lysosome that catalyzes the hydrolysis of glucosylceramide (GlcCer) to generate ceramide (Cer) and glucose, and its inhibition directly leads to lysosomal storage disease. miGBA can target the coding region of GBA, form a complex with RISC, effectively interfere with GBA mRNA, and inhibit the expression and enzyme activity of GBA.

[0011] Among them, the vector skeleton comprises a pDMP-miR vector or a pAAV-MCS, preferably, the plurality of NF-κB Motif gene fragments comprise 1-10 NF-κB Motif DNA fragments, and the sequence of the NF-κB Motif DNA fragment is 5'-GGACTTTCC-3', and the Motif has high affinity with NF-κB.

[0012] The interval fragment is 10-20 base pairs between the NF-κB Motif DNA fragments, and the sequence of the interval fragment is 5'-GGGAATTTCCGG-3' as preferred.

[0013] The sequence of the 5' end of the miR30 is CTGGAGGCTTGCTGAAGGCTGTA, and the sequence of the 3' end of the miR30 is CAGGACACAAGGCCTGTTACTAGCACTCACATGGAACAAATGGCC.

[0014] The minimum promoter sequence is 5'-TAGAGGGTATATAATGGAAGCTCGACTTCCA-3', which is located downstream of the Motif and is used to start the expression of the downstream microRNA.

[0015] The sequence of the GBA mRNA is shown in SEQ ID NO. 1-SEQ ID NO. 6, and the sequence of the GBA mRNA is shown in SEQ ID NO. 1 or SEQ ID NO. 5 as preferred.

[0016] The sequence of the shGBA is shown in SEQ ID NO. 8-SEQ ID NO. 13, and the sequence of the shGBA is shown in SEQ ID NO. 8 or SEQ ID NO. 12 as preferred.

[0017] The vector further comprises a reporter gene, and the reporter gene includes but is not limited to zsGreen gene as preferred.

[0018] The application further provides a construction method of the gene interference vector, which comprises the following steps: obtaining the NF-κB Motif gene fragment, the minimum promoter gene fragment and the miGBA gene fragment, and inserting them into a vector skeleton to obtain the gene interference vector.

[0019] The application further provides a cell, which is obtained by transfecting a host cell with the gene interference vector.

[0020] The host cell comprises AML cells or normal cells, preferably, the AML cells comprise one or more of THP-1, U937, KG1a, HL60, WEHI-3, NB4, SKNO01, ML-2, EL9611 or C1498, and the normal cells comprise one or more of MCF12a, BEAS-2B, HMEC-1, NIH-3T3, L929, GES-1, MRC-5, AML-12, HL-1 or C2C12.

[0021] The application also provides the use of the gene interference vector or the cell in the preparation of a drug for treating acute myeloid leukemia, preferably, the gene interference vector can promote the expression of miGBA in AML cells in response to NF-κB through a positive feedback mechanism; preferably, the gene interference vector can express miGBA in response to the activity of NF-κB in acute myeloid leukemia cells, induce lysosomal storage disease, and trigger multiple forms of death dominated by ferroptosis in acute myeloid leukemia cells.

[0022] The NM fragment of the interference vector constructed in the application contains different numbers of motif sequences of NF-κB, and the optimal number of motifs is screened by using zsGreen as a reporter gene in ten AML cells and ten normal cells, 6 motifs can maximally respond to the activity of NF-κB in AML cells and do not affect normal cells. In the application, three different human or mouse miGBAs are designed, and the expression amount of miGBA in AML cells and normal cells, the influence of GBA mRNA, protein and enzyme activity are demonstrated. Then, the human or mouse miGBA with the best silencing effect on GBA is used for killing experiments of AML cells and normal cells, lysosome morphology, metabolite and function detection, cell death form rescue experiments, and cell ferroptosis characterization experiments, so as to demonstrate the killing ability and mechanism of pNM6-miGBA on AML cells and biological safety.

[0023] The second aspect of the present application provides a recombinant adeno-associated virus, which is obtained by transfecting cells with the gene interference vector, the packaging plasmid and the helper plasmid, wherein the gene interference vector comprises a vector skeleton, and the vector skeleton comprises an AML-specific promoter, an NM gene fragment and a miGBA gene fragment in sequence, the AML-specific promoter comprises a plurality of NF-κB Motif gene fragments, a spacer gene fragment and a minimal promoter gene fragment, the miGBA gene fragment comprises a 5' end 20-60 bp base fragment of miR30, an shGBA and a 3' end 20-60 bp base fragment of miR30 in sequence, the shGBA is a gene fragment of shRNA based on the structure of miR30 and can be targeted to GBA mRNA, and the sequence of the shGBA is shown in SEQ ID NO. 8-SEQ ID NO. 13.

[0024] Preferably, the vector skeleton comprises a pDMP-miR vector or a pAAV-MCS, and the plurality of NF-κB Motif gene fragments comprise 1-10 NF-κB Motif DNA fragments, and the sequence of the NF-κB Motif DNA fragment is 5'-GGACTTTCC-3', and the Motif has high affinity with NF-κB.

[0025] Preferably, the number of NF-κB Motif sequences in the gene interference vector is 6 (NM6), the NM6 can maximize the expression of downstream effect genes or microRNAs in AML cells in response to NF-κB activation without affecting normal cells, and the functional sequence of the NM6 controlling the expression of miGBA can respond to the activity of NF-κB in AML cells and specifically express miGBA, but not in normal cells.

[0026] Preferably, the miGBA is connected by the 5' end and the 3' end of miR30 and shGBA, can be targeted to the coding region of GBA, the target sequence of human GBA is 5'-TATGTTCAGCATTGCTGTA-3', the target sequence of mouse GBA is 5'-TGCGCTCAATCCTTGCTTT-3', the miGBA can express microRNA, form RISC complex in cells, target and cut lysosomal enzyme GBA gene, cause the protein level of GBA gene to be reduced, damage the function of lysosome, induce lysosomal storage disorder, and release free iron and other contents.

[0027] The sequence containing the 5' end of miR30 is CTGGAGGCTTGCTGAAGGCTGTA, and the sequence containing the 3' end of miR30 is CAGGACACAAGGCCTGTTACTAGCACTCACATGGAACAAATGGCC.

[0028] The interval fragment is 10-20 base pairs between the NF-κB Motif DNA fragments, and the sequence of the interval fragment is 5'-GGGAATTTCCGG-3' as preferred.

[0029] The minimum promoter sequence is 5'-TAGAGGGTATATAATGGAAGCTCGACTTCCA-3'.

[0030] The recombinant adeno-associated virus includes any one of various serotypes of adeno-associated viruses AAV1-AAV12, and the adeno-associated virus serotype is AAV2 as preferred.

[0031] The application also includes a construction method of the recombinant adeno-associated virus, which is obtained by transfecting cells with the gene interference vector, a packaging plasmid and a helper plasmid, the packaging plasmid is pAAV-RC, the helper plasmid is pHelper, and the cells are HEK-293T cells as preferred.

[0032] The application also includes a composition containing the recombinant adeno-associated virus, and an intravenous iron supplement.

[0033] The intravenous iron supplement is an oxidized iron nanoparticle capable of entering cells to produce •OH through a Fenton reaction to induce cell ferroptosis, and the oxidized iron nanoparticle is Feraheme (FeNPs) as preferred, which has a core composed of oxidized iron (Fe3O4 / γ-Fe2O3) and is clinically approved for treating iron deficiency anemia in patients with chronic kidney disease.

[0034] Further, the FeNPs concentration is crucial for the vector and virus to express miGBA. First, the FeNPs content cannot be too high, and a high concentration of FeNPs will produce a large amount of •OH, causing damage to normal cells. Second, the •OH content released by FeNPs cannot induce the activation of NF-κB in normal cells, otherwise the activated NF-κB will start the expression of miGBA, which is disastrous to normal cells. The FeNPs concentration is 10 μg / ml as preferred.

[0035] The application also includes the recombinant adeno-associated virus and the composition in the preparation of a drug for treating acute myeloid leukemia.

[0036] The combination of the recombinant adeno-associated virus and intravenous iron supplement can induce a significant increase in acute myeloid leukemia cell •OH, activate NF-κB through a positive feedback mechanism to promote miGBA expression, cause lysosomal storage disease in cells, and aggravate ferroptosis.

[0037] The application includes the use of the recombinant adeno-associated virus alone or in combination with intravenous iron supplement, and the combination includes one-time mixture administration or administration of intravenous iron supplement after virus injection. The recombinant adeno-associated virus can be administered in vivo alone or in combination with intravenous iron supplement, and both have a significant therapeutic effect on acute myeloid leukemia mice and greatly prolong the survival of mice. The recombinant adeno-associated virus and intravenous iron supplement can be administered in a one-time mixture or the virus can be injected first and then the intravenous iron supplement is administered, and the interval can be 1 to 7 days, all of which have a significant therapeutic effect.

[0038] As a preferred embodiment, the adeno-associated virus is injected in vivo first, and the intravenous iron supplement is injected after an interval of 2 days, which is an AML treatment scheme.

[0039] In the present application, the FeNPs and adeno-associated virus gene vector composition can clear most AML cells by inducing ferroptosis and lysosomal storage disease in vitro, but have no effect on normal cells. In vivo, the FeNPs and adeno-associated virus can significantly reduce the infiltration of AML in the liver, spleen, lung, and bone marrow; greatly prolong the survival of AML mice, and have no significant toxicity to normal combination and blood biochemistry.

[0040] Advantages: Compared with the prior art, the present application has the following advantages:

[0041] (1) The present application constructs a pNM series promoter containing NF-κB Motif and finally determines that the vector containing 6 motifs is the preferred vector. The vector can maximize the expression of the effect element (zsGreen / miGBA) in AML cells, but has no effect on normal cells. Normal cells can only significantly express miGBA under the stimulation of TNFα (NF-κB activator), indicating that pNM6-miGBA can utilize the activity of NF-κB to achieve selective expression in AML, greatly reducing the toxic side effects on normal cells.

[0042] (2) The pNM6-miGBA constructed in the application can effectively and specifically inhibit GBA in AML cells, leading to significant accumulation of GlcCer and decrease of Cer and Glu, accompanied by up-regulation of lysosome markers LAMP1 / LAMP2, increase of lysosome volume, decrease of LysoSensor fluorescence and lysosome proton transport activity, proving that GBA silencing can induce lysosomal storage disease and reduce lysosome function in AML cells, while there is no such effect in normal cells with low activation of NF-κB. These results show that pNM6-miGBA can selectively induce GBA inactivation, lysosomal storage disease and impaired function in AML, and has good biological safety.

[0043] (3) In the application, pNM6-miGBA can significantly enhance the killing effect on AML cells over time, and has a strong anti-AML effect. Cell death rescue experiments show that pNM6-miGBA treated AML cells undergo pyroptosis, apoptosis inhibitor, necrosis inhibitor, autophagy inhibitor, and ferroptosis and other forms of death, among which ferroptosis is the most significant. After induction of pNM6-miGBA, the contents of lipROS, Fe 2+ in AML cells significantly increase, and the activity of GPX4, the key antioxidant enzyme of ferroptosis, is inhibited. More importantly, miGBA has no effect on the mitochondrial and lysosomal morphology of normal cells, and the activity of lipROS, Fe 2+ , GPX4, but the ferroptosis positive drug Erastin has a significant ferroptosis effect, lacking AML specificity. This result fully demonstrates the specificity and treatment safety of pNM6-miGBA for AML.

[0044] (4) In the application, pNM6-miGBA causes lysosomal storage disease in AML cells, releasing a large amount of contents, which further activates NF-κB, positively feedback amplifies the mechanism of action of pNM6-miGBA, and makes AML die significantly. In general, the pNM6-miGBA provided in the application is a new type of AML gene therapy tool, which has killing effect on AML and safety to normal cells, makes up for the shortcomings of existing NF-κB inhibitors and ferroptosis drugs, and has potential application value as an AML targeted therapy strategy.

[0045] (5) The recombinant adeno-associated virus of the application has high targeting and selectivity, miGBA expression is driven by AML-specific promoter NM6 containing multiple NF-κB response elements, and miGBA transcription is induced only in the context of highly activated NF-κB in AML cells, so as to achieve selective attack on AML cells at the molecular level and avoid non-specific effects on normal tissue cells as much as possible. NM6 sets the response threshold, taking into account targeting and safety, so that miGBA expression has AML specificity. The recombinant adeno-associated virus of the application has a highly efficient and specific gene silencing tool, and the use of miR30 modified shGBA construction makes the expressed microRNA more easily processed and assembled into the RISC complex, which can efficiently target the GBA coding region and reduce protein expression and enzyme activity. This design reduces the off-target risk and toxic side effects that may be caused by traditional microRNAs or broad-spectrum GBA inhibitors, and the delivery of adeno-associated virus in vivo provides sustained gene inhibition effect.

[0046] (6) The composition of the application has an innovative mechanism for synergistically inducing iron death, and gene interference and nano-chemical effects are mutually amplified. miGBA targets and inhibits lysosomal enzyme GBA, leading to lysosomal storage disorders, causing FeNPs to accumulate in the lysosome due to slow metabolism, continuously generating a large amount of •OH through Fenton reaction, inducing NF-κB to further activate, forming a positive feedback loop, amplifying the lethal effect of miGBA and FeNPs, and significantly inducing AML cells to undergo iron death. Compared with single drugs or simple gene interference, the composition of the application has complementary mechanism and amplification effect. The iron oxide nanomaterials and adeno-associated viruses described in the application are beneficial to clinical transformation, and the application preferably uses Feraheme, an iron oxide nanomaterial that has been used in the treatment of iron deficiency anemia in the clinic, and its safety and pharmacokinetic characteristics have been supported by clinical data, significantly reducing the regulatory and safety barriers from animal experiments to human trials. As a mature gene delivery system, AAV vectors also have precedents of being approved or entering clinical trials, which helps to evaluate the feasibility of the overall therapy.

[0047] (7) The composition of the application has good in vivo safety, by precisely controlling the dose of FeNPs and calibrating the sensitivity of NM6 response, it is ensured that AML cell iron death is effectively induced while miGBA is not mistakenly expressed in normal cells due to the triggering of NF-κB, and excessive •OH is also avoided from directly causing oxidative damage to normal cells. Experimental data show that there is no significant toxicity to normal cells and blood biochemical indicators at the recommended dose. The administration method of the composition is flexible and can be individually adjusted, and the application allows the adeno-associated virus and intravenous iron supplement to be mixed and administered at one time, and also supports the strategy of separate-time administration of injecting the virus first and then injecting the intravenous iron supplement. The preferred scheme is to first administer the AAV vector, then to infuse FeNPs after two days, in order to obtain the optimal therapeutic effect.

[0048] (8) The in vitro and in vivo anti-AML experiments of the present application show that the combination of the adeno-associated virus and FeNPs composition can efficiently clear most AML cells by inducing ferroptosis and lysosomal storage disorder with little effect on normal hematopoietic cells. In fact, drug-resistant AML cells are susceptible to ferroptosis, and lysosomal storage disorder is irreversible damage, and this composition greatly reduces the risk of drug resistance and recurrence. The results of the in vivo AML mouse model show that the combination therapy significantly reduces the infiltration of leukemia in the liver, spleen, lung and bone marrow and greatly prolongs the survival of animals. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 It is a schematic diagram of pNM6-miGBA plasmid map and gene structure composition of pNM6-miGBA plasmid;

[0050] Figure 2 It is a comparison of the expression amount of p-p65 in different cells and a comparison of the fluorescence results of cells transfected with different plasmids; wherein a is the expression amount of p-p65 in ten different AML cells (THP-1, U937, KG1a, HL60, WEHI-3, NB4, SKNO01, ML-2, EL9611, C1498) and normal cells (MCF12a, BEAS-2B, HMEC-1, NIH-3T3, L929, GES-1, MRC-5, AML-12, HL-1, C2C12) (n=3); b is the fluorescence map of two AML cells (KG1a, C1498) and two normal cells (MCF12a, HL-1) transfected with pNM-zsGreen containing different amounts of NF-κB motif 48h; c is the quantitative analysis of fluorescence intensity (n=3).

[0051] Figure 3Fig. 1. The principle of construction and expression of pNM6-miGBA plasmid; a, the principle of construction of pNM6-miGBA plasmid, which is driven by NF-κB (pol II) specificity; b, the expression of various plasmids pNM6-NT, pNM6-mihGBA-1, pNM6-mihGBA-2, pNM6-mihGBA-3, pNM6-mimGBA-1, pNM6-mimGBA-2, pNM6-mimGBA-3 at different times (24-72 h) in KG1a and C1498 (n=3); c, the expression of various plasmids pNM6-NT, pNM6-mihGBA-1, pNM6-mihGBA-2, pNM6-mihGBA-3, pNM6-mimGBA-1, pNM6-mimGBA-2, pNM6-mimGBA-3 at different times (24-72 h) in MCF12a and HL-1 with or without TNFɑ (10 ng / mL) induction for 1 h (n=3); d, AO&EB fluorescence staining of AML cells (KG1a, C1498) and normal cells (MCF12a, HL-1) after pNM6-miGBA transfection for 72 h; the rest of the time period is shown in the supplementary Figures 7-9 .

[0052] Figure 4 Fig. 2. The expression results of various plasmids; a, the quantitative detection of GBA mRNA in KG1a and C1498 cells after transfection of various plasmids pNM6-NT, pNM6-mihGBA-1, pNM6-mihGBA-2, pNM6-mihGBA-3, pNM6-mimGBA-1, pNM6-mimGBA-2, pNM6-mimGBA-3 for 48 h (n=3); b, the quantitative analysis of GBA mRNA in MCF12a and HL-1 cells after transfection of various plasmids pNM6-NT, pNM6-mihGBA-1, pNM6-mihGBA-2, pNM6-mihGBA-3, pNM6-mimGBA-1, pNM6-mimGBA-2, pNM6-mimGBA-3 for 48 h with or without TNFɑ (10 ng / mL) induction for 1 h (n=3).

[0053] Figure 5Figure 6 shows the expression results of various plasmids, wherein a is the quantification of protein expression level and enzyme activity of GBA in KG1a cells after transfection of various plasmids (pNM6-NT, pNM6-mihGBA-1, pNM6-mihGBA-2, pNM6-mihGBA-3) for 48 h (n=3); b is the quantification of protein expression level and enzyme activity of GBA in C1498 cells after transfection of various plasmids (pNM6-NT, pNM6-mimGBA-1, pNM6-mimGBA-2, pNM6-mimGBA-3) for 48 h (n=3); c is the quantification of protein expression level and enzyme activity of GBA in MCF12a cells after transfection of various plasmids (pNM6-NT, pNM6-mihGBA-1, pNM6-mihGBA-2, pNM6-mihGBA-3) for 48 h (n=3); d is the quantification of protein expression level and enzyme activity of GBA in HL-1 cells after transfection of various plasmids (pNM6-NT, pNM6-mimGBA-1, pNM6-mimGBA-2, pNM6-mimGBA-3) for 48 h (n=3); e is the quantification of protein expression level and enzyme activity of GBA in MCF12a cells after transfection of various plasmids (pNM6-NT, pNM6-mihGBA-1, pNM6-mihGBA-2, pNM6-mihGBA-3) for 48 h after induction of TNFɑ (10 ng / mL) for 1 h (n=3); f is the quantification of protein expression level and enzyme activity of GBA in HL-1 cells after transfection of various plasmids (pNM6-NT, pNM6-mimGBA-1, pNM6-mimGBA-2, pNM6-mimGBA-3) for 48 h after induction of TNFɑ (10 ng / mL) for 1 h (n=3). The data are presented in the form of mean ± standard deviation (SD).

[0054] Figure 6 Figure 7 shows the schematic diagram and result diagram of the best plasmid, wherein a is the schematic diagram of pNM6-miGBA plasmid killing AML cells but having no effect on normal cells; b is the expression analysis of GBA substrates GlcCer and downstream metabolites Cer, Glu under the intervention of PBS, pNM6-NT, pNM6-miGBA (n=3); c is the p-p65 content detection in AML cells (KG1a, C1498) and normal cells (MCF12a, HL-1) under the intervention of PBS, pNM6-NT, pNM6-miGBA (n=3).

[0055] Figure 7Figures of fluorescence staining and cell viability analysis of each transfected cell; wherein a is the AO&EB fluorescence staining of KG1a treated with PBS, NT, miGBA plasmid for 24 h, 48 h, 72 h; b is the quantitative data of cell number of KG1a under various treatments (n=3); c is the cell viability analysis of KG1a under various treatments (n=3); d is the AO&EB fluorescence staining of C1498 treated with PBS, NT, miGBA plasmid for 24 h, 48 h, 72 h; e is the quantitative data of cell number of C1498 under various treatments (n=3); f is the cell viability analysis of C1498 under various treatments (n=3).

[0056] Figure 8 Figures of fluorescence staining and cell viability analysis of each transfected cell; wherein a is the AO&EB fluorescence staining of MCF12a treated with PBS, NT, miGBA plasmid for 24 h, 48 h, 72 h; b is the quantitative data of cell number of MCF12a under various treatments (n=3); c is the cell viability analysis of MCF12a under various treatments (n=3); d is the AO&EB fluorescence staining of HL-1 treated with PBS, NT, miGBA plasmid for 24 h, 48 h, 72 h; e is the quantitative data of cell number of HL-1 under various treatments (n=3); f is the cell viability analysis of HL-1 under various treatments (n=3).

[0057] Figure 9 Figures of fluorescence staining and cell viability analysis of each cell after TNFɑ induction; wherein a is the AO&EB fluorescence staining of MCF12a treated with PBS, NT, miGBA plasmid for 24 h, 48 h, 72 h after TNFɑ (10 ng / mL) induction for 1 h; b is the quantitative data of cell number of MCF12a under various treatments (n=3); c is the cell viability analysis of MCF12a under various treatments (n=3); d is the AO&EB fluorescence staining of HL-1 treated with PBS, NT, miGBA plasmid for 24 h, 48 h, 72 h after TNFɑ (10 ng / mL) induction for 1 h; e is the quantitative data of cell number of HL-1 under various treatments (n=3); f is the cell viability analysis of HL-1 under various treatments (n=3).

[0058] Figure 10Figures for mRNA quantification results analysis and proton detection analysis results of each transfected cell; Wherein a is the mRNA quantification results of lysosome markers lamp1 and lamp2 in AML cells (KG1a, C1498) and normal cells (MCF12a, HL-1) after PBS, NT, miGBA plasmid treatment for 72 h (n=3); b is the Lysosensor and hoechst fluorescence intensity ratio quantification data (n=30); c is the proton transport capacity detection data (n=30). Statistical significance is tested by two-tailed, unpaired Student's t-test.

[0059] Figure 11 Figures for each transfected cell different cell death inhibitor rescue experiment results, MDA content determination, fluorescence quantification, Fe 2+ content determination and GPX4 enzyme activity determination; Wherein a is the rescue experiment of different cell death inhibitors incubated for 48 h after pNM6-miGBA overnight transfection of KG1a and C1498 (n=3); Pyroptosis inhibitor (Ac, 20 μM), apoptosis inhibitor (ZVAD, 50 μM), necrosis inhibitor (Nec1s, 10 μM), autophagy inhibitor (BA-1, 1 nM), ferroptosis inhibitor (Lipro-1, 50 nM); b is the MDA content determination (n=3); c is the transmission electron microscopy graph of AML cells (KG1a, C1498) and normal cells (MCF12a, HL-1) treated by NT, miGBA and erastin (10 μM), the red arrow points to the mitochondria, and the green arrow points to the lysosome; d is the Lipid ROS fluorescence quantification graph (n=3); e is the Fe 2+ content determination data (n=3); f is the GPX4 enzyme activity determination (n=3).

[0060] Figure 12 Figures for the activity change and p-p65 expression level results of FeNPs and pNM6-miGBA treated cells; a is the cell activity change of KG1a, C1498, MCF12a, HL-1 after different FeNPs (Feraheme) concentration (0-200 μg / mL) treatment for 72 h and b is the p-p65 expression level data (n=3); c is the cell activity change of KG1a, C1498, MCF12a, HL-1 after different FeNPs (Feraheme) concentration (0-200 μg / mL) and pNM6-miGBA co-treatment for 72 h (n=3);

[0061] Figure 13Figure 2. a, Schematic diagram of pNM6-miGBA plasmid combined with FeNPs treatment results; b, Schematic diagram of pNM6-miGBA plasmid combined with FeNPs killing AML cells but no effect on normal cells; c, TEM images of AML cells (KG1a, C1498) and normal cells (MCF12a, HL-1) treated with NT+FeNPs, miGBA+FeNPs, miGBA and FeNPs (10 μg / mL), green arrow points to swollen lysosome, yellow arrow points to normal lysosome; d, BODIP 581 / 591 C11 fluorescence staining merge images of AML cells (KG1a, C1498) and normal cells (MCF12a, HL-1) treated with NT+FeNPs, miGBA+FeNPs, miGBA for 72 h, non-oxdized and oxdized fluorescence images are shown in Supplementary Figure 2a; e, Lipid ROS fluorescence quantification (n=3); f, •OH fluorescence quantification (n=3); g, Fe2+ content quantification (n=3); h, GPX4 enzyme activity assay (n=3). Figure 8 a, FeNPs final concentration is 10 μg / mL; d, Lipid ROS fluorescence quantification (n=3); e, •OH fluorescence quantification (n=3); f, Fe2+ content quantification (n=3); g, GPX4 enzyme activity assay (n=3). 2+ a, FeNPs final concentration is 10 μg / mL; d, Lipid ROS fluorescence quantification (n=3); e, •OH fluorescence quantification (n=3); f, Fe2+ content quantification (n=3); g, GPX4 enzyme activity assay (n=3).

[0062] Figure 14 Figure 3. a, BODIP 581 / 591 C11 fluorescence staining images of AML cells (KG1a, C1498) and normal cells (MCF12a, HL-1) treated with NT+FeNPs, miGBA+FeNPs, miGBA for 72 h, non-oxdized and oxdized fluorescence images are shown in Supplementary Figure 3a; b, Cell viability data of AML cells (KG1a, C1498) and normal cells (MCF12a, HL-1) after various treatments (PBS, NT, FeNPs, NT+FeNPs, miGBA, miGBA+FeNPs) for 72 h (n=3); c, p-p65 expression levels of AML cells (KG1a, C1498) and normal cells (MCF12a, HL-1) after various treatments for 72 h (n=3).

[0063] Figure 15Figure for the results of KG1a and C1498 after transfection by pNM6-miGBA and incubation with FeNPs; wherein a is the DEGs (differentially expressed genes) Wayne chart of KG1a and C1498 after transfection by miGBA (KG1a-miGBA, C1498-miGBA) and co-incubation with FeNPs (10 μg / mL) for 48 h; b is the volcano plot of DEGs; c is the enrichment results of KEGG pathway analysis of DEGs after KG1a-miGBA and FeNPs (10 μg / mL) incubation treatment; d is the enrichment results of KEGG pathway analysis of DEGs of KG1a-miGBA group.

[0064] Figure 16 Figure for the enrichment results of KEGG pathway analysis of DEGs of C1498-miGBA and FeNPs with or without; a is the enrichment results of KEGG pathway analysis of DEGs after C1498-miGBA and FeNPs (10 μg / mL) incubation treatment; b is the enrichment results of KEGG pathway analysis of DEGs of C1498-miGBA group;

[0065] Figure 17 Figure for the KEGG pathway and GO enrichment results of pNM6-miGBA and FeNPs synergistic treatment; wherein a is the KEGG enrichment analysis circle diagram of DEGs of KG1a-miGBA+FeNPs, KG1a-miGBA, C1498-miGBA+FeNPs, C1498-miGBA four groups; b is the GO enrichment analysis of DEGs of KG1a-miGBA+FeNPs, KG1a-miGBA, C1498-miGBA+FeNPs, C1498-miGBA four groups;

[0066] Figure 18 Figure for the results of AML mouse animal experiment; wherein a is the schematic diagram of AML mouse animal experiment treatment; b is the body weight change data (n=10); c is the survival curve (n=10); d is the weight of spleen, viscera, and lung (n=10); e is the H&E staining diagram of main tissues (heart, liver, spleen, lung, kidney); f is the fluorescence diagram of C1498-GFP in bone marrow (BM); g is the proportion of C1498-GFP positive cells in bone marrow (n=6).

[0067] Figure 19Figures for the results of co-administration of miGBA and FeNPs to AML cells; wherein a is the content data of miGBA, GBA mRNA, GBA protein and GBA enzyme activity in C1498 (n=6); b is the content of GBA substrate GlcCer, downstream metabolite Cer, Glu and the quantitative detection of p-p65 (n=6); c is the content determination of Lamp1 and Lamp2 mRNA, lysosensor fluorescence intensity and lysosome proton transport rate determination (n=6); d is the content determination of MDA, •OH fluorescence, Fe 2+ content and GPX4 enzyme activity determination data (n=6); e is the blood metabolism and distribution in main tissues of FeNPs, f is the blood metabolism and distribution in main tissues of rNM6-miGBA (12h, 24h, 48h) (n=6).

[0068] Figure 20 Figures for the results of mouse safety experiments; wherein a is the schematic diagram of mouse safety experiments, which is divided into 6 groups: Health (healthy mouse group), NT (blank control group), FeNPs group, NT+FeNPs group, miGBA group, miGBA+FeNPs group; b is the result figure of mouse body weight change (n=6); c is the result figure of rAAV DNA abundance in each tissue (n=6); d is the iron content in each tissue (n=6); e is the H&E staining diagram of main tissues (heart, liver, spleen, lung, kidney).

[0069] Figure 21 Figures for mouse blood routine and biochemical indicators (n=3); WBC: white blood cells, RBC: red blood cells, PLT: platelets, HGB: hemoglobin, ALT: alanine aminotransferase, AST: aspartate aminotransferase, ALP: alkaline phosphatase, BUN: blood urea nitrogen, Cr: creatinine; UA: uric acid. Data is presented in the form of mean ± standard deviation (SD). DETAILED DESCRIPTION

[0070] 1. Construction of each vector in the application

[0071] In view of the two characteristics of NF-κB transcriptional activity and high activation in AML cells, the present application designs a series of pNM vectors responsive to NF-κB. The DMP sequence in the pDMP-miR vector is modified, which is derived from the previous article (Gao J, Luo T, Wang J. Gene interfered-ferroptosis therapy for cancers. Nat Commun. 2021 Sep 7; 12(1): 5311) of the inventor, and the DMP sequence is modified into a series of vectors containing 1 to 10 NF-κB motifs by seamless cloning, the NF-κB motif nucleotide sequence is 5'-GGACTTTCC-3', the spacer nucleotide sequence is 5'-GGGAATTTCCGG-3', and the two NF-κB motifs are connected by a spacer. Specifically, the nucleotide sequences containing 1 to 10 NF-κB motifs and the spacer therebetween (for example: NF-κB motif+spacer, NF-κB motif+spacer+NF-κB motif, NF-κB motif+spacer+NF-κB motif+spacer, NF-κB motif+spacer+NF-κB motif+spacer+NF-κB motif, and so on to 10 NF-κB motifs) are synthesized by the Sangon Company, the pDMP-miR vector is double digested by Mlu I and EcoRI restriction endonucleases, and then the synthesized nucleotide sequence is connected to the digested vector by seamless cloning, and the series of vectors containing 1 to 10 NF-κB motifs are obtained by cloning and sequencing identification. Then, the pAAV-MCS (Stratagene Company) is double digested by Mlu I and Afe I, and the target fragment containing the NF-κB motif-minimal promoter-miR30 sequence is amplified from the series of vectors containing 1 to 10 NF-κB motifs as a template and inserted into the digested vector to obtain a series of NF-κB responsive pNM1-10 vectors. The amplification primers of the target fragment are F: 5'-TGCGGCCGCACACGT-3' and R: 5'-GCTATCTCGAGTGCGGCC-3'. In addition, the present application also constructs a pNM-zsGreen vector for screening the pNM vector that can be used for AML specific treatment.Specifically, the zsGreen sequence is amplified from a pHIV-zsGreen plasmid (Addgene) using primers F: 5'-GGATCCATGGCCCAGTCC-3' and R: 5'-AGCGCTTTAGGGCAAGGCGG-3'. Then, the amplified zsGreen sequence is ligated to the pNM1~10 series vectors digested with EcoRI and Afe I by means of seamless cloning, and the pNM-zsGreen series vectors containing 1 to 10 NF-κB motifs are obtained by cloning and sequencing.

[0072] Meanwhile, the pNM6-miR30-shGBA vector (the best number of 6 motif vectors are screened) is constructed by combining the miR30 system, and is named pNM6-miGBA (a) in the Figure 3 The miR30 gene interference technology is to insert shRNA into the miR30 skeleton of the NF-κB responsive pNM1~10 series vectors by using the structural characteristics of miR30, so as to silence the expression of the target gene in a specific tissue or cell by means of a tissue-specific promoter. The reason for not directly expressing shGBA is that shRNA is mainly regulated by pol III (without cell specificity), and NF-κB belongs to pol II, which can express miGBA (guaranteeing cell specificity) (a) in the Figure 3 Finally, the miR30-shGBA is combined with NM6, and three different miR30-shGBA for human and mouse sources are designed, which are named mihGBA-1, 2, 3 and mimGBA-1, 2, 3, respectively.

[0073] The specific steps are: using BLOCK-iT™ RNAi Designer tool (https: / / rnaidesigner.thermofisher.com / rnaiexpress / ) to design three groups of human or mouse shRNA targeting GBA CDS region, the specific target points are shown in Table 1, the sequence of shGBA designed based on the target points is shown in Table 2, and the oligonucleotide sequence for synthesizing shGBA is shown in Table 3. The forward and reverse strands of the oligonucleotide synthesized in Table 3 (synthesized by Sangon Company) are denatured and annealed (95 ℃, 10 min; then cooled to room temperature) to obtain double-stranded DNA (dsDNA). Then, T4 DNA ligase and BsmBI endonuclease are used to perform a Golden Gate reaction, and the reaction system is composed of 10 units of BsmBI, 600 units of T4 DNA ligase, 1×T4 DNA ligase buffer, 1 nM dsDNA and 50 ng pNM6 vector (10 μL). The Golden Gate reaction is performed on a PCR instrument: 37 ℃, 5 min and 16 ℃, 10 min for 10 cycles, 37 ℃, 30 min and 80 ℃, 5 min. The pNM6-mihGBA-1, 2, 3 and pNM6-mimGBA-1, 2, 3 vectors are obtained respectively. The vector pNM6-mihGBA-1 / pNM6-mimGBA-2 with the highest silencing efficiency on human / mouse cell GBA is named pNM6-miGBA. As a negative control vector, we copy the NT fragment in plasmid pcDNA6.2-GW / EmGFP-miR-Neg (Thermo Fisher Scientific Company) into the pNM6 vector and name it pNM6-NT. The plasmid map of pNM6-miGBA is shown in ™ 6.2-GW / EmGFP-miR-Neg (Thermo Fisher Scientific Company) into the pNM6 vector and name it pNM6-NT. The plasmid map of pNM6-miGBA is shown in Figure 1 The gene structure of pNM6-mihGBA-1 is shown in Figure 1 As can be seen from B, the structures of the pNM6-mihGBA2, 3 and pNM6-mimGBA-1, 2, 3 vectors of the application only need to replace the bases in the boxes with the sequences SEQ ID NO. 9-SEQ ID NO. 13 in Table 2, respectively. All plasmids are verified correct by DNA sequencing.

[0074] Table 1 miGBA target points

[0075] Table 2 Sequence of shGBA designed according to miGBA target points

[0076] Table 3 Oligonucleotide sequences used to construct miRNA expression plasmids

[0077] 2. Cell culture

[0078] The AML cells and normal cells used in the present application are: KG1a (human acute myeloid leukemia cells), WEHI-3 (mouse myeloid leukemia cells), SKNO01 (human acute myeloid leukemia cells), C1498 (mouse acute myeloid leukemia cells) purchased from American Type Culture Collection (ATCC). THP-1 (human monocytic leukemia cells), U937 (human monocytic leukemia cells), HL60 (human acute promyelocytic leukemia cells), MCF12a (human normal mammary epithelial cells), BEAS-2B (human normal lung epithelial cells), HMEC-1 (human microvascular endothelial cells), NIH-3T3 (mouse embryonic fibroblasts), L929 (mouse fibroblasts), GES-1 (human gastric mucosa epithelial cells), MRC-5 (human embryonic lung cells), AML-12 (mouse normal hepatocytes), HL-1 (mouse cardiomyocytes), HEK-293T (human embryonic kidney cells) and C2C12 (mouse myoblasts) purchased from Cell Resource Center of Shanghai Life Science Research Institute, Chinese Academy of Sciences. C1498-GFP cells were purchased from Shanghai Sunny Biotech Co., Ltd., EL9611 (mouse erytholeukemia cells) were derived from Nanjing Hospital of Traditional Chinese Medicine. NB4 (human acute promyelocytic leukemia cells) and ML-2 (human acute myelomonocytic leukemia cells) were purchased from German Collection of Microorganisms and Cell Cultures (DSMZ). All cell lines have been identified and mycoplasma tested.

[0079] KG1a, WEHI-3, SKNO01, C1498, THP-1, U937, HL60, NB4, ML-2, EL9611 were cultured in RPMI-1640 (Gibco); MCF12a, NIH-3T3, L929, GES-1, MRC-5, C2C12, HEK-293 were cultured in DMEM (Gibco); HL-1 was cultured in Klebs' medium (Sigma, 51800C) with additional 0.1 mM norepinephrine (Sigma, 489350) and 2 mM L-glutamine (Sigma, G8540). HMEC-1 was cultured in MCDB131 (Gibco, 10372019) with additional 10 ng / ml hEGF (Sigma, GF144). AML-12 was cultured in DMEM / F12 with additional 1% ITS liquid media supplement (Sigma, I3146) and 40 ng / mL dexamethasone (Sigma, 265005). All cells were cultured in 10% fetal bovine serum (Gibco, 10091148) with 100 U / mL penicillin (Gibco) and 100 μg / mL streptomycin (Gibco) in a humidified incubator at 37 °C with 5% CO2.

[0080] 3. Cell transfection and acridine orange & ethidium bromide (AO & EB) staining experiment

[0081] The various cells (1 x 105cells / well) in logarithmic growth phase cultured in step 2 were seeded in 6-well plates and cultured for 24 h. Then, the cells were transfected with 2 μg of the plasmid DNA using Lipofectamine 2000 (Invitrogen, 11668019) according to the manufacturer's instructions. After 6 h, the medium was replaced with fresh medium. After 48 h, the cells were collected and stained with AO & EB (Sigma, E7522) for 15 min. The cells were observed under a fluorescence microscope (Olympus, BX51) and photographed. 5C1498 and HL-1 cells were transfected with mouse-targeted miGBA vectors for 24 h, 48 h, and 72 h, and KG1a and MCF12a cells were transfected with human-targeted miGBA vectors for 24 h, 48 h, and 72 h according to the species of cells. At the corresponding time nodes, all cells were stained with AO & EB (Sangon, E607308). According to the kit instructions, for adherent cells, the culture medium was removed and replaced with PBS buffer, followed by the addition of 10 μL of AO and 10 μL of EB per 180 μL of PBS buffer. The cells were incubated at room temperature in the dark for 5 min. For suspended cells, centrifugation was performed at 1,000 rpm for 5 min, and the culture medium was then removed. The subsequent steps were the same as for adherent cells. Live cells were uniformly green, and dead cells were orange. The cells were imaged under a fluorescence microscope to observe live and dead cells, and data statistical analysis was performed using Image J software.

[0082] 4. CCK-8 experiment

[0083] The reagents used in the CCK-8 experiment were TNFɑ, Bafilomycin A1 (BA1), ZVAD-FMK (ZVAD), Necrostatin-1s (Nec1s), Liproxstatin-1 (lipro-1), and Ac-DMLD-CMK (Ac), which were purchased from MedChemExpress.

[0084] Cell viability was determined and analyzed using a Cell Counting Kit-8 (CCK-8, Beyotime). Cells (5×10 3) were seeded into 96-well plates and incubated overnight at 37 °C in a 5% CO2 incubator. Then the plasmids to be transfected (200 ng / well) were transfected into cells using Lipofectamine 2000. The induction of TNF-a (final concentration of 10 ng / mL) was performed 1 h before transfection. After overnight transfection, FeNPs (final concentration of 10 pg / mL) were added according to experimental requirements, and incubated for 24 h, 48 h, 72 h. FeNPs were derived from AMAG Pharmaceuticals company; in the experiment of the effect of various inhibitors on cell viability, the overnight transfected cells were incubated with inhibitors for 48 h; the inhibitors used included pyroptosis inhibitor (Ac, 20 pM, MCE, HY-P10939), apoptosis inhibitor (ZVAD, 50 pM, MCE, HY-16658B), necrosis inhibitor (Nec1s, 10 pM, MCE, HY-14622A), autophagy inhibitor (BA-1, 1 nM, MCE, HY-100558), ferroptosis inhibitor (Lipro-1, 50 nM, MCE, HY-12726). TNFɑ, BafilomycinA1 (BA1), ZVAD-FMK (ZVAD), Necrostatin-1s (Nec1s), Liproxstatin-1 (lipro-1), Ac-DMLD-CMK (Ac) were purchased from MedChemExpress. As needed, in the experiment of the effect of FeNPs concentration on cell viability, cells were incubated with culture medium containing different concentrations of FeNPs (0-200 pg / mL) for 72 h. After incubation, CCK-8 reagent (10 pL / well) was added to the cells and incubated for 1 h, and finally the absorbance of the solution was measured at 450 nm using a microplate reader.

[0085] 5. GBA enzyme activity detection

[0086] According to the method of the GBA enzyme activity detection kit (Abeam, ab273339), the cell pellet (1 x 10 7 ) was homogenized in lysis buffer, placed on ice for 10 min, and centrifuged at 12,000 x g, 4 °C for 10 min to obtain the supernatant as the sample. The prepared sample and GBA substrate containing a fluorescent group were added to a 96-well plate, mixed well, and incubated at 37 °C for 30 min. Finally, the fluorescence data (Ex / Em = 360 / 445 nm) were read on a microplate reader.

[0087] 6. GPX4 enzyme activity detection

[0088] According to GPX4 enzyme activity test kit (Sigma, 353919), 1,000 x g, 4°C centrifugation for 10 min to collect cells. The centrifuged cells were lysed on ice with homogenate buffer (50 nM Tris-Hcl, pH 7.5; 5 mM EDTA, 1 mM DTT). Then, 10,000 x g, 4°C centrifugation for 15 min. 20 µL of centrifuged supernatant was added to the wells of a 96 plate in turn with 50 µL Diluted Assay Buffer, 50 µL of Co-Substrate Mixture, 50 µL NADPH. 20 µL of hydrogen peroxide was added to all reaction wells, carefully shake the 96-well plate for 5 s, and finally use the enzyme label meter to read the absorbance at 340 nm wavelength per minute, at least 5 time points were obtained to calculate the GPX4 enzyme activity.

[0089] 7. GlcCer, Cer, Glu assay

[0090] Firstly, a known amount of stable isotope internal standard [Glucose- 13 C6 (MCE, 110187-42-3), d7-Cer (d18: 1-d7 / 18: 0) (Sigma, 860677P), d5-GlcCer (18: 1) (Sigma, 860673P)] was added to the cell precipitate, and then the Folch extraction system (methanol: chloroform: water = 2: 1: 0.8, vortex mixing for 10 min, 1,6000 x g centrifugation for 10 min, 4°C) was used. The organic phase and the aqueous phase were recovered respectively and each was divided into two parts; the organic phase was blown dry with nitrogen and then redissolved with methanol. Cer and GluCer were quantified with d7-Cer and d5-GlcCer as internal standards in the positive ion MRM mode by triple quadrupole mass spectrometry (Waters Xevo TQ-XS). The aqueous phase was blown dry with nitrogen and then resuspended with acetonitrile: water 1: 1. Glucose- 13 C6 (Sigma, 389374) was used as an internal standard to quantify glucose. The whole process was completed after adding the internal standard to correct the recovery and matrix effect, and a matrix matching standard curve was established in the matrix. The recovery rate, matrix effect, linear range, recovery correction precision and stability were verified according to the Matuszewski method to ensure the reliable differentiation and repeatable quantification of GlcCer, Cer and Glu.

[0091] 8. Fe 2+ detection

[0092] Fe 2+The kit (Boxbio, AKIC004M) was used for quantitative analysis. Fe 2+ Under acidic conditions, it can form a blue complex with tripyridyl triazine (3,3',5,5'-Tetramethylbenzidine, TPTZ), and the product has a characteristic absorption peak at 593 nm. The content of Fe 2+ was quantitatively detected by the change in absorbance. The specific operation includes sample pretreatment (cell lysis and tissue homogenate) on ice, addition of color reagent for color reaction, and use of a microplate reader to read the absorbance at 593 nm.

[0093] 9、ELISA detection

[0094] We used ELISA kits to quantitatively detect protein expression, including p-p65 (abcam, ab176647), GBA (Aviva, mouse OKEH03366; human OKCD08164). The main steps include: prepare the reagent according to the instructions, and place it at room temperature. Add 50 µL of the sample to be tested or control to each well, then add 50 µL of antibody mixture, incubate at room temperature for 1 h, then discard the supernatant and wash 3 times with Wash Buffer. Add 100 µL TMB substrate, avoid light color development for 15 min; then add 100 µL of stop solution to terminate the reaction, immediately read the OD value at 450 nm.

[0095] 10、MDA determination

[0096] Collect 2×10 6 cells, wash the cells in ice-cold PBS. Resuspend the cells in 300 µL lysis buffer, and add 3 µL BHT stock solution (BHT can prevent further peroxidation of the sample during processing), then homogenize and sonicate. After processing, centrifuge at 13,000×g for 10 min, collect the supernatant to obtain the sample to be tested. According to the instructions (Abeam, ab118970), configure the MDA standard solution and Developer Mix, mix 600 µL of Developer Mix with 200 µL of standard or sample to be tested, incubate at 95°C for 60 min, then place on ice for 10 min. Take 200 µL of the above reaction solution and add it to the wells of a 96-well plate, immediately read the OD value at 532 nm.

[0097] 11、ROS detection

[0098] Use lipid peroxidation kit BODIPY ®581 / 591 C11 (Invitrogen) to detect lipid peroxidation in cells and imaged by 590 nm and 510 nm emission filter and 40x objective in fluorescence microscope. Finally the lipid peroxidation of cells was quantified by analyzing and calculating the fluorescence intensity and corresponding ratio in 590 and 510 channel by quantitative software ImageJ software. Hydroxyl radical (•OH) detection and quantification kit (Biorab, HR8841) was used to detect and quantify •OH in cells. This product is a hydroxyl radical detection kit using •OH specific green fluorescent probe O27. O27 can freely penetrate the cell membrane into the cell and be oxidized by •OH in the cell to produce green fluorescent product, and the fluorescence data is read by a microplate reader (Ex / Em = 488 / 525 nm).

[0099] 12. miGBA assay

[0100] Each cell (1 x 105) was lysed using 1 mL of TRIzol (Sigma) 7), 200 μΐ of chloroform was added to extract the RNA water phase, and the RNA was precipitated by 1 mL of isopropanol, washed with 200 μΐ of 75% ethanol, dissolved in 100 μΐ of ddH2O, and measured for total RNA concentration by Nanodrop. The whole process was required to be free of RNase to ensure that the RNA was not degraded, and the total RNA of the individual cells was obtained. The miRNA 1st Strand cDNA Synthesis Kit was used to synthesize the first strand of miRNA cDNA (Vazyme, MR101). First, the genomic DNA was removed by gDNA Wiper Mix, and the reaction system was 2 μΐ of 5x gDNA Wiper Mix, 1 μg of total RNA, and ddH2O was added to 10 μΐ. The reaction program was as follows: 42°C for 2 min. Then, the stem-loop primer (Table 3) and reverse transcriptase were used to synthesize the first strand of cDNA, and the reaction system was 10 μΐ of genomic DNA removed, 1 μΐ of Stem-loop primer (2 μΜ), 2 μΐ of 10x RT Mix, 2 μΐ of HiScript II Enzyme Mix, and 5 μΐ of ddH2O. The reaction program was as follows: 25°C for 5 min; 50°C for 15 min; 85°C for 5 min. The synthesized product was used for quantitative detection of miRNA, and the primer was shown in Table 4. The reaction system was 0.4 μΐ of forward primer, 0.4 μΐ of Universal miGBA-R, 10 μΐ of 2x Taq Pro Universal SYBR qPCR MasterMix, and 9.2 μΐ of cDNA. The reaction program was as follows: 95°C for 15 min; 35 cycles of 95°C for 5 s, 60°C for 30 s were performed.

[0101] Table 4 miRNA detection primer

[0102] 13、TEM imaging

[0103] AML cells and normal cells were transfected with pNM6-NT and pNM6-miGBA overnight, and positive drug groups were treated with Erastin (final concentration 10 µM, 8 h). After drug treatment, cells were fixed with 2.5% glutaraldehyde in cacodylate buffer (0.1 M, pH 7.4) at 4°C overnight. Then, cells were further fixed with a solution containing 2% osmium tetroxide, 0.1 M cacodylate and 1.5% potassium ferrocyanide at room temperature for 1 h. Subsequently, cells were dehydrated in a series of graded ethanol, and the dehydrated cells were embedded with epoxy resin. The embedded cells were cut into ultrathin sections, stained with uranyl acetate and lead citrate, and finally observed and photographed under a transmission electron microscope.

[0104] 14. LysoSensor staining

[0105] LysoSensor (Invitrogen) was used to monitor the degree of lysosomal acidification. After the treatment of cells with pNM6-NT, pNM6-miGBA, LysoSensor (final concentration 1 µM) was added and incubated for 2 h. Then, fresh medium was replaced, and fluorescence microscopy was used for observation, and Image J software was used for data statistical analysis.

[0106] 15. Lysosomal proton transport activity assay

[0107] AML cells, normal cells and the cells transfected with pNM6-NT and pNM6-miGBA were placed in DMEM containing 2 mg / ml FITC-dextran in an ice bath for 5 min, then transferred to a 37°C incubator for 30 min, and replaced with DMEM for continuous culture for 30 min to transport FITC-dextran to lysosomes. Lysosomes were collected and resuspended in detection buffer (125 mM KCl, 1 mM EDTA, 20 mM HEPES, pH 7.5) and equilibrated in an ice bath for 1 h, then mixed with a final concentration of 5 µM Concanavalin A (MCE, HY-P2149) and DMSO, respectively, and incubated at 37°C for 10 min, and the FITC fluorescence was recorded using a microplate reader. Then, a final concentration of 5 mM Mg-ATP (Shanghai Xinyu Biological Technology Co., Ltd., XY90005PZ) was added, and the initial slope of fluorescence quenching was measured to evaluate the lysosomal proton transport activity.

[0108] 16. Quantitative PCR (Q-PCR)

[0109] Cells were lysed using 1 mL of TRIzol (Sigma) (1×10 7), 200 μL chloroform was added to extract the RNA water phase, and the RNA was precipitated by 1 mL of isopropanol, washed with 200 μL of 75% ethanol, and dissolved in 100 μL of ddH2O. The total RNA concentration was measured by Nanodrop. The whole process needed to be kept free of RNase to ensure that the RNA was not degraded. The HiScript II 1stStrand cDNA Synthesis Kit (Vazyme) was used for reverse transcription to synthesize cDNA. Subsequently, the fluorescent quantitative PCR amplification was performed by the HiScript II One Step RT-PCR Kit (Vazyme). The reaction system was as follows: 2 μL of forward primer, 2 μL of reverse primer, 25 μL of One Step Mix, 2.5 μL of One Step Enzyme Mix, 8.5 μL of cDNA, and 10 μL of ddH2O. The reaction program was as follows: 94°C for 3 min; 35 cycles of 94°C for 30 s, 60°C for 30 s. The mRNA transcription level (RQ = 2 –ΔΔCt ) was corrected by taking GADPH as the internal reference. The primer information is shown in Table 5. All experiments were repeated three times for verification.

[0110] Table 5 Primer sequences for qPCR

[0111] 17. Preparation and titration of viruses

[0112] 5 x 10 6 HEK-293T cells were inoculated into a 75 cm 2The cells were cultured in DMEM medium overnight in a culture flask, and then transfected with three plasmids, including two helper plasmids pAAV-RC (Zoman, ZK735) and pHelper (Zoman, ZK736) and one pAAV plasmid (pNM6-NT or pNM6-miGBA). After 72 h of continuous culture, the cells and culture solution were collected in a 50 mL centrifuge tube and frozen overnight at -80°C. Then the frozen cells and culture solution were thawed by incubation in a 37°C water bath for 2 h. The whole freezing and thawing process was repeated three times. After the freezing and thawing, pure chloroform (1:10 volume) was added, and the 50 mL centrifuge tube was placed in a 37°C water bath for 1 h of vigorous shaking. After the shaking, NaCl was added to a final concentration of 1 M, and the centrifuge tube was centrifuged at 15,000 rpm for 15 min at 4°C. Then the supernatant was removed, PEG8000 was added to a final concentration of 10% (w / v), and the solution was shaken until it dissolved, and then centrifuged at 15,000 rpm for 15 min at 4°C. The supernatant was discarded, the precipitate was dissolved with PBS, and DNase and RNase were added to a final concentration of 1 μg / mL, and incubated at room temperature for 30 min. Finally, the incubated reaction solution was extracted once with chloroform (1:1 volume), and after the layers were separated, the upper aqueous phase was collected into a new tube to obtain the purified virus, and the obtained serotype AAV2 virus was named rNM6-NT and rNM6-miGBA, respectively.

[0113] 18. RNA-seq

[0114] The present application includes 6 groups of RNA-seq samples (blank groups C1498 and KG1a, C1498 and KG1a co-treated with FeNPs and pNM6-miGBA and only pNM6-miGBA transfected for 72 h), and each group is biologically repeated three times. RNA is extracted by TRIzol, and degradation and contamination are detected by 1% agarose gel electrophoresis. The purity is evaluated by Nanodrop OD260 / OD280, the concentration is quantified by Qubit, and the integrity and insert size are determined by Agilent 2100. After quality control, Illumina platform 150 bp paired-end sequencing is used, and clean reads are aligned to the reference genome using HISAT2.

[0115] 19. GO and KEGG analysis

[0116] The expression levels of mRNA of each gene in the sample were analyzed using HTSeq software, and DESeq R software package was used to screen the differentially expressed genes (DEGs, P<0.05). The DEGs with FDR<0.05 and absolute value of gene expression fold change (|fold change|)≥2 were further subjected to GO and KEGG analysis. DAVID was used for GO function enrichment analysis of DEGs, and signal pathways were mapped in KEGG database to complete annotation.

[0117] 20. Treatment of acute myeloid leukemia mice

[0118] C57BL / 6 (male, 8 weeks old) healthy mice were injected with 5 x 10 6 C1498 cells / mouse to construct an AML mouse model, and treatment was started 10 days after administration.

[0119] The mouse experiment was divided into three batches. The first batch of 12 groups (n=10): healthy mouse group, PBS injection group, rNM6-NT group, FeNPs group, rNM6-NT+FeNPs (0) group, rNM6-miGBA group, rNM6-miGBA+FeNPs (0) group, rNM6-miGBA+FeNPs (1) group, rNM6-miGBA+FeNPs (2) group, rNM6-miGBA+FeNPs (3) group, rNM6-miGBA+FeNPs (5) group, rNM6-miGBA+FeNPs (7) group; the interval days between rNM6-miGBA and FeNPs administration are indicated in the brackets. The administration dose of rNM6-miGBA was 1 x 10 11 mg / kg mouse. The body weight and survival curve of the mice were recorded.

[0120] The second batch consisted of four groups (n=6): healthy mouse group, rNM6-NT+FeNPs group, rNM6-miGBA group, rNM6-miGBA+FeNPs group. The interval days between rNM6-miGBA and FeNPs administration was 2 days, and the administration dose of rNM6-miGBA was 1 x 10 11 mg / kg mouse. On day 20 after modeling of AML mice, the mice were euthanized, and the main organs (heart, liver, spleen, lung, kidney) of the mice were weighed and observed by H&E staining.

[0121] The third batch of mice, C57BL / 6 (male, 8 weeks old) healthy mice were injected with 5 x 10 6C1498-GFP cells / mice, a total of four groups (n=6) were divided: healthy mice group, NT+FeNPs group, rNM6-miGBA group, rNM6-miGBA+FeNPs group, the interval days of rNM6-miGBA and FeNPs administration was 2 days, the administration dose of rNM6-miGBA was 1 x 10 11 / mice, the administration dose of FeNPs was 1 mg / kg mice.

[0122] 21. Biosafety analysis

[0123] We divided healthy mice C57BL / 6 (male, 8 weeks old) into 6 groups (n=6), including healthy mice group, rNM6-NT group, FeNPs group, rNM6-NT+FeNPs group, rNM6-miGBA group, rNM6-miGBA+FeNPs group. The administration dose of rNM6-miGBA was 1 x 10 11 / mice, the administration dose of FeNPs was 1 mg / kg. The interval days of rNM6-miGBA and FeNPs administration was 2 days. The body weight, blood routine, liver and kidney biochemical indicators, rAAV abundance, iron content and pathological conditions (H&E) of main tissues of mice were detected. At the same time, the metabolism of rNM6-miGBA and FeNPs in mice in blood (0 min, 1 min, 5 min, 10 min, 30 min, 1 h, 2 h, 4 h, 8 h, 12 h, 24 h) and main tissues (12 h, 24 h, 48 h) was detected.

[0124] 22. Data analysis

[0125] All data were expressed as mean ± standard deviation (SD) and analyzed statistically and plotted using GraphPad Prism 8.0 software. First, the normality of data was evaluated by Shapiro-Wilk test, and the homogeneity of variance between groups was evaluated by Levene test. If the data meet the normal distribution and homogeneity of variance, two-tailed, unpaired Student's t test is used for significance analysis; if the data do not meet these conditions, non-parametric test (Mann-Whitney U test) is used. The statistical significance level is set as *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

[0126] 23. Experimental results

[0127] 23.1 Experimental results of viral vectors

[0128] The NF-κB-responsive pNM1~10 series vectors constructed above contain NF-κB-high affinity motifs, minimal promoters, and downstream effector elements (genes or RNAi). Figure 1 The activation of the p65 / p50 heterodimer is a key process in the NF-κB activation pathway, with p65 phosphorylation (p-p65) playing a decisive role in NF-κB transcriptional activity. Therefore, we first examined the activity of p-p65 in 10 AML cell lines and 10 normal cell lines. The results showed that p-p65 was highly activated in all 10 AML cell lines, with relatively high expression levels in THP-1, U937, and KG1a, and the lowest in C1498. Figure 2 (a) In normal cells, p-p65 is almost not expressed ( Figure 2 (a) in the middle.

[0129] Although NF-κB is almost inactive in normal cells, to ensure the in vivo safety of the pNM vector (no response to normal cell substances) and the maximum expression of effector elements in AML cells, we integrated 1-10 designed NF-κB motifs into the pNM vector and used zsGreen as a reporter gene for detection. Figure 2 (b) Data showed that the fluorescence intensity of zsGreen significantly increased in AML cells (KG1a and C1498) with increasing NF-κB motif numbers. However, in normal cells (MCF12a and HL-1), weak green fluorescence was observed in MCF12a when the number of NF-κB motifs increased to 7. Figure 2 (b and c in the original text). Although the fluorescence intensity of pNM7-zsGreen in MCF12a was very weak and not observed in HL-1, considering in vivo safety, we used pNM6 as the subsequent therapeutic carrier ( Figure 3 (a) in the middle.

[0130] The results of cell transfection with pNM6-NT, pNM6-mihGBA-1,2,3, and pNM6-mimGBA-1,2,3 vectors constructed in this invention showed that their expression in AML cells (KG1a, C1498) gradually increased over time (24~72h). Figure 3 (b) effectively silences GBA mRNA ( Figure 4 a) Protein level ( Figure 5 (a and b in the text) significantly inhibited GBA enzyme activity ( Figure 5 (a and b in the text); but it has no effect on normal cells (MCF12a, HL-1). Figure 3 c inFigure 4 b and d in Figure 5 (c and d in the original text). Furthermore, under the induction of TNFα (a classic NF-κB activator), normal cells, like AML cells, will also produce miGBA (…). Figure 3 c), silencing GBA mRNA ( Figure 4 b) Protein level ( Figure 5 e and f in the formula significantly inhibit GBA enzyme activity ( Figure 5 (e and f in the text). This indicates that pNM6-miGBA has NF-κB specificity. We used mihGBA-1, 2, 3, and mihGBA-1 and mimGBA-2, which have the strongest silencing efficiency among mihGBA-1, 2, 3, for subsequent studies ( Figure 4 and Figure 5 ), and named together as miGBA.

[0131] To investigate the cytotoxicity of miGBA against AML cells, we transfected miGBA into AML cells and normal cells. The results showed that miGBA could time-dependently enhance the killing rate of AML cells but had no effect on normal cells. Figure 3 d and Figure 7 and Figure 8 Furthermore, normal cells can also be killed by miGBA under the induction of TNFα, indicating that miGBA is an AML-specific killer based on NF-κB. Figure 3 d and Figure 9 ).

[0132] To investigate the mechanism of miGBA's killing effect on AML cells, we first quantitatively analyzed the GBA substrate GlcCer and its downstream metabolites Cerebrolysin (Cer) and glucose (Glu). The results showed that GBA inhibition led to the accumulation of GlcCer in AML cells, while Cerebrolysin (Cer) and glucose (Glu) were significantly reduced, but normal cells were unaffected. Figure 6 (a and b in the original text). GlcCer is an important component of the lysosomal membrane, and its excessive accumulation can lead to lysosomal storage impairment. The reduction in Glu is not as significant as that in Cer because GBA-mediated GlcCer metabolism is not the main pathway for Glu production. The significant increase in the expression levels of lysosomal markers lamp1 and lamp2 also indicates that miGBA induces lysosomal storage impairment in AML cells. Figure 10 (a) Furthermore, by detecting the fluorescence intensity of the lysosensor probe and the proton transport activity of lysosomes, we found that lysosomal storage impairment further leads to a decrease in lysosomal activity in AML cells (a). Figure 10 (b and c in the text). However, miGBA has no effect on the lysosomal function of normal cells ( Figure 10). We detected the expression of p-p65 in miGBA transfected cells, and the results showed that miGBA could stimulate the further activation of p-p65 in AML cells ( Figure 6 ). This is because lysosomes are accumulated, leading to increased membrane permeability, and a large amount of released contents activate NF-κB. The activation of p-p65 forms a positive feedback loop, further promoting the expression of miGBA, ultimately leading to the death of AML cells due to lysosomal accumulation disorder ( Figure 6 ). While in normal cells, almost no p-p65 activation cannot start this cycle ( Figure 7 ). Figure 6 Figure 8 ).

[0133] To explore the specific mechanism of miGBA-induced AML cell death, we first rescued the killing caused by miGBA with various cell death inhibitors. The results showed that pyroptosis inhibitor (Ac), apoptosis inhibitor (ZVAD), necrosis inhibitor (Nec1s), autophagy inhibitor (BA-1), and ferroptosis inhibitor (Lipro-1) could all rescue miGBA-induced AML cell death to some extent, but Lipro-1 was the most significant ( Figure 11 a). This is because lysosomes are the cellular iron pool, and their accumulation disorder will release iron ions to induce ferroptosis in cells. To confirm this guess, we first detected the content of the ferroptosis marker MDA, and the data showed that miGBA, similar to the ferroptosis positive drug Erastin, could stimulate AML cells to significantly up-regulate MDA, but could be reversed by Lipro-1 ( Figure 11 b). Transmission electron microscopy (TEM) observation directly confirmed that miGBA could induce AML cells to have large lysosome volume and accumulation disorder; mitochondria showed shrinkage, high membrane density, and thickened cristae ( Figure 11 c). It is worth noting that the effect of miGBA on mitochondria is similar to that of Erastin, but Erastin does not cause lysosomal accumulation ( Figure 11 c). This implies that the accumulation disorder of lysosomes caused by miGBA is the key to inducing ferroptosis in AML cells. We further detected the lipid peroxidation level (lipROS), Fe 2+ and GPX4 enzyme activity of AML cells, and the results showed that after induction by miGBA, the content of lipROS, Fe 2+ in AML cells increased significantly, and the key antioxidant enzyme GPX4 activity of ferroptosis was inhibited ( Figure 11 d, Figure 11 e, and Figure 11 ​f). More importantly, miGBA had no effect on normal cell mitochondrial, lysosome morphology, and lipROS, Fe 2+ GPX4 activity, but Erastin had a significant ferroptosis effect, lacking AML specificity ( Figure 11 d, Figure 11 e and and Figure 12 f). This result fully demonstrated the specificity and safety of miGBA for AML treatment.

[0134] 23.2, Experimental results of vectors and their packaging viruses

[0135] 23.2.1, pNM6-miGBA synergistically induces severe ferroptosis in AML cells with Feraheme (FeNPs)

[0136] In view of the high activation of NF-κB in AML cells, we designed a pNM series of vectors responsive to NF-κB. The vector contains a highly affine Motif of NF-κB, a minimum promoter and a microRNA downstream. Although NF-κB is almost not activated in normal cells, in order to ensure the in vivo safety of pNM vectors (no response to normal cells) and the maximum expression of the effector element in AML cells, we designed 1-10 NF-κB motifs integrated into the pNM vector, and detected the expression of miGBA. The data showed that with the increase of the number of NF-κB motifs, miGBA in AML cells (KG1a and C1498) increased significantly. However, in normal cells (MCF12a and HL-1), when the number of NF-κB motifs increased to 7, miGBA had a lower expression level. Although the expression of miGBA in normal cells by pNM7-miGBA was weak, considering the in vivo safety, we used pNM6-miGBA as the subsequent therapeutic vector, which can be used for packaging of adeno-associated virus.

[0137] Excess Fe 2+ can occur Fenton reaction in cells, producing a large number of toxic hydroxyl radicals (•OH), further inducing the activation of NF-κB, i.e. the phosphorylation of p65 (p-p65), which positively feedback amplifies the therapeutic effect of miGBA. In addition, Fe 2+ can also cause lipid peroxidation in cells and induce ferroptosis. Therefore, the present application introduces the clinical intravenous iron supplement Feraheme (FeNPs). As a kind of nanomedicine, its core is composed of iron oxide (Fe3O4 / γ-Fe2O3), which contains a large amount of Fe 2+, outer layer wrapped by carboxylic dextran, with a diameter of about 17-31 nm. Studies have reported that FeNPs are mainly located in lysosomes, and can selectively kill AML cells with low ferroportin (FPN) through iron overload and Fenton reaction, while also having a protective function for hematopoietic stem cells (HSCs). The localization and biological function of FeNPs are highly consistent with the killing mechanism of miGBA on AML. In order to expand the ferroptosis effect of miGBA on AML, we investigated the possibility of FeNPs and pNM6-miGBA synergistic therapy for AML. Data showed that FeNPs concentrations of 0-60 μg / mL had good biological safety for AML cells and normal cells (a in Figure 12 , but FeNPs exceeding 30 μg / mL would induce the activation of p-p65 (b in Figure 12 ). Importantly, ultra-low dose of FeNPs (10 μg / mL) could kill most AML cells when synergistically treated with pNM6-miGBA, but had no effect on normal cells (c in Figure 13 ). In fact, the concentration of FeNPs is crucial for pNM6-miGBA. First, the content of FeNPs cannot be too high, as high concentration of FeNPs would produce a large amount of •OH, causing damage to normal cells. Second, the content of •OH released by FeNPs cannot induce the increase of p-p65 level in normal cells, otherwise the activation of p-p65 would start pNM6-miGBA, which is disastrous for normal cells. Considering safety and therapeutic effectiveness, we used ultra-low dose of FeNPs (10 μg / ml) as the experimental dose in the subsequent experiments.

[0138] We speculate that after induction by miGBA, lysosomal storage disorders occur in AML cells, which leads to the accumulation of FeNPs in lysosomes, continuous production of •OH, stimulation of p-p65 activation, and formation of a positive feedback loop, ultimately leading to the death of AML cells due to lysosomal damage and ferroptosis (a in Figure 13 . To verify this hypothesis, we directly observed AML cells and normal cells treated with miGBA and FeNPs synergistically by TEM, and the results showed that both single miGBA and miGBA synergistically treated with FeNPs would cause lysosomal storage disorders in AML cells, but had no effect on normal cells (b in Figure 13 . It is worth noting that a large amount of FeNPs accumulated in the lysosomes of the pNM6-miGBA+FeNPs group, but FeNPs alone had no effect on all cells (c in Figure 13(b) This suggests the biocompatibility of low-dose FeNPs with lysosomes. We characterized ferroptosis in AML cells and normal cells and found that the pNM6-miGBA+FeNPs group showed the most significant increases in lipROS and •OH, which were much higher than those in the pNM6-miGBA group (b). Figure 13 FeNPs alone do not affect intracellular lipROS and •OH because cells can metabolize low doses of FeNPs (ce). Figure 14 ce in Figure 13 (a) In addition, we also measured intracellular Fe. 2+ The results showed that the pNM6-miGBA+FeNPs group could significantly increase Fe in AML cells. 2+ The level improved the single miGBA response to Fe 2+ Insufficient induction defects ( Figure 13 (f in the text). Similarly, the pNM6-miGBA+FeNPs group significantly inhibited the enzyme activity of GPX4 in AML cells, but had no effect on normal cells. Figure 14 Finally, we examined the synergistic killing effect of pNM6-miGBA and FeNPs on AML cells. The results showed that the pNM6-miGBA+FeNPs group was able to clear the vast majority of AML cells, significantly improving the single efficacy of pNM6-miGBA and FeNPs, and exhibiting good biocompatibility with normal cells. Figure 14 (b) This is due to the positive feedback regulation that continuously activates p-p65 after treatment with pNM6-miGBA and FeNPs. Figure 15 (c) In conclusion, the synergistic treatment of pNM6-miGBA and FeNPs is devastating in its effectiveness against AML.

[0139] To explore the molecular mechanism of synergistic treatment with pNM6-miGBA and FeNPs, we performed RNA-seq analysis on AML cells (C1498 and KG1a). Venn diagrams and volcano plots showed that GBA was significantly knocked down in both C1498 and KG1a cells after treatment with pNM6-miGBA or pNM6-miGBA and FeNPs. Figure 15 (a and b in the text) confirmed the effectiveness of miGBA in silencing GBA mRNA. Furthermore, the ferroptosis marker gene ferritin FTL was significantly upregulated in all treatment groups ( Figure 15 (a and b in the text) suggests that both miGBA and miGBA combined with FeNPs significantly affected iron metabolism in AML. KEGG pathway analysis showed that ferroptosis was most pronounced in KG1a cells after treatment with pNM6-miGBA and FeNPs. Figure 15c). Single pNM6-miGBA treatment is more likely to trigger neurological disorders such as Parkinson's and Alzheimer's ( ). Figure 16 The d in the text is consistent with the physiological function of GBA. In C1498 cells, both single treatment with pNM6-miGBA and synergistic treatment with pNM6-miGBA and FeNPs significantly induced ferroptosis and the Fatty acid metabolism pathway, indicating that iron metabolism and lipid metabolism are disordered in AML cells. Figure 17 KEGG pathway and GO analysis of all Top20 showed that synergistic treatment with pNM6-miGBA and FeNPs induced the disaggregation of AML cellular components; pNM6-miGBA was more inclined to respond to oxidative stress ( Figure 18 The RNA-seq results fully demonstrate the effectiveness of miGBA in interfering with GBA mRNA, and also support the fact that the synergistic treatment of pNM6-miGBA with pNM6-miGBA and FeNPs is ferroptosis as the killing mechanism of AML.

[0140] 23.2.1.2. In vivo anti-AML effect of synergistic administration of rNM6-miGBA and FeNPs

[0141] Inspired by the synergistic effect of miGBA and FeNPs on AML cell killing, we packaged pNM6-miGBA into adeno-associated virus rNM6-miGBA (serotype 2) for in vivo treatment of AML in mice. The core reason for using adeno-associated virus and FeNPs in synergistic administration is that the safety of both has been extensively validated in clinical practice, which will help advance the clinical application of this treatment. We treated patients by intravenous injection of rNM6-miGBA and FeNPs, with the dose of rNM6-miGBA being 1×10⁻⁶. 11 The FeNPs dosage was 1 mg / kg, which is 5 times lower than the clinical dose, greatly improving in vivo safety. Figure 18 (a) Since miGBA requires 72 hours to achieve a significant inhibitory effect on GBA activity in cell experiments, while FeNPs typically have a half-life of around 12 hours in the blood, there is a significant difference in their onset time. Therefore, to explore the appropriate dosing interval and determine the optimal dosing time for FeNPs, we set up dosing groups with intervals of 0, 1, 2, 3, 5, and 7 days. Day 0 means simultaneous administration of a mixture of rNM6-miGBA and FeNPs. Data showed that both the rNM6-miGBA+FeNPs group and the rNM6-miGBA group were able to restore mouse body weight and significantly prolong mouse survival (…). Figure 18(b and c in the text). Experimental results showed that the optimal dosing interval between rNM6-miGBA and FeNPs was 2 and 3 days, enabling 60% of mice to survive for more than 150 days. Figure 18 c). Mouse body weight and survival curve data fully demonstrate that the synergistic administration of rNM6-miGBA and FeNPs has significant anti-AML properties and good safety in vivo. Figure 18 (b and c in the original text). Considering the therapeutic effect, we chose to administer rNM6-miGBA and FeNPs 2 days apart for subsequent animal experiments. To further investigate the in vivo therapeutic mechanism of synergistic administration of rNM6-miGBA and FeNPs, we examined the infiltration of leukemia cells in major mouse tissues. The results showed that synergistic administration of rNM6-miGBA and FeNPs significantly reduced the AML cell load in the spleen, liver, and lungs of mice, and the weight of these tissues was close to that of the healthy mouse group. The effect of miGBA was second best, while FeNPs alone had no effect. Figure 18 (d in the text). The H&E staining results further support this finding. Figure 18 (e). In addition, we introduced C1498-GFP stable cells to track AML cell infiltration in mouse bone marrow. The results showed that synergistic administration of rNM6-miGBA and FeNPs significantly reduced the number of AML cells in the bone marrow, with rNM6-miGBA showing the next best effect, while FeNPs alone had no effect. Figure 19 (f and g in the text). These results fully demonstrate the potent therapeutic effect and biocompatibility of synergistic administration of rNM6-miGBA and FeNPs in vivo for AML.

[0142] To investigate the in vivo therapeutic mechanism of AML synergistic administration of rNM6-miGBA and FeNPs, we enriched bone marrow AML cells from each group and detected GBA expression. The results showed that both the rNM6-miGBA+FeNPs group and the rNM6-miGBA group induced significant miGBA expression in AML cells, downregulated GBA mRNA and protein levels, and inhibited GBA enzyme activity. Figure 19 (a) These data strongly support the in vivo interference function of rNM6-miGBA against GBA. Simultaneously, we also examined the expression levels of GlcCer and its metabolites Cere and Glu. The results were similar to those in the cell experiments: GlcCer significantly accumulated in both the rNM6-miGBA+FeNPs group and the rNM6-miGBA group, while Cere and Glu levels significantly decreased. Figure 19(b) indicates that lysosomes are impaired in their storage. We further investigated lysosomal function, and the results showed that the rNM6-miGBA+FeNPs group had significantly increased lysosomal markers lamp1 and lamp2, significantly inhibited lysenseor fluorescence intensity and proton transport rate, with miGBA showing the next best effect, and FeNPs alone having no effect. Figure 19 The c) indicates that AML lysosomes suffer from storage disorders and functional impairment. Furthermore, we also investigated the iron death markers MDA, •OH, and Fe. 2+ The content and GPX4 enzyme activity were measured, and the results showed that the rNM6-miGBA+FeNPs group could significantly increase MDA, •OH and Fe in AML cells in vivo. 2+ The content of [specific ingredient] significantly inhibited GPX4 enzyme activity, with miGBA showing the next best effect, while FeNPs alone were ineffective. Figure 20 (d) This result indicates that the core mechanism of synergistic administration of rNM6-miGBA and FeNPs in vivo anti-AML is ferroptosis. In summary, synergistic administration of rNM6-miGBA and FeNPs significantly enhances the in vivo anti-AML effect of rNM6-miGBA and FeNPs alone, and synergistic administration can induce lysosomal storage impairment and ferroptosis in AML cells in vivo.

[0143] 2.3.3 In vivo safety of synergistic administration of rNM6-miGBA and FeNPs

[0144] The in vivo safety of rNM6-miGBA and FeNPs is a key indicator of concern in anti-AML treatment. Figure 20 (a) We investigated the in vivo safety of rNM6-miGBA, FeNPs, and their synergistic therapy. Body weight data confirmed that these drugs had no effect on mouse body weight. Figure 20 (b) Furthermore, we also examined the viral abundance and iron content in major tissues (heart, liver, spleen, lung, kidney, and bone marrow) after administration of rNM6-miGBA and FeNPs. The results showed that rNM6-miGBA and FeNPs were distributed in all tissues, with the highest accumulation levels observed in the liver and spleen, respectively. Figure 20 (c and d in the text), which is also an important reason why the synergistic treatment of rNM6-miGBA and FeNP can clear leukemia cell infiltration in the liver, spleen, and lungs. H&E staining, blood routine and biochemical indicators further show that neither rNM6-miGBA nor FeNPs, whether used alone or in combination, has toxicity to major tissues, blood, liver and kidney function. Figure 21 e and Figure 19). Finally, we measured the metabolism of rNM6-miGBA and FeNPs in blood and tissues in vivo, and the results showed that the half-life of FeNPs in blood was 8.18 h (e in Figure 19 , and that of rNM6-miGBA was 11.83 h (f in Figure 19 ). In tissues, rNM6-miGBA and FeNPs reached the peak of accumulation in liver at 24 h (e in Figure 19 and f in ​ ). At 48 h, FeNPs and miGBA had about 15% and 18% accumulation in bone marrow, respectively. In summary, whether it was rNM6-miGBA, FeNPs or the combination of the two, all had good in vivo safety.

[0145] In summary, the present application develops a composition of a gene interference vector, a recombinant adeno-associated virus and an intravenous iron supplement for acute myeloid leukemia treatment, which has strong AML killing property and in vivo safety, and has clinical conversion potential.

Claims

1. A gene interference vector, characterized in that, The gene interference vector includes a vector backbone, in which an AML-specific promoter NM gene fragment and a miGBA gene fragment are sequentially included. The AML-specific promoter NM gene fragment contains multiple NF-κB Motif gene fragments, spacer gene fragments, and a minimal promoter gene fragment. The miGBA gene fragment sequentially contains a 20-60 bp fragment from the 5' end of miR30, a shGBA fragment, and a 20-60 bp fragment from the 3' end of miR30. The shGBA is a gene fragment of shRNA based on the miR30 structure that can target GBA mRNA.

2. The gene interference vector according to claim 1, characterized in that, The vector backbone includes a pDMP-miR vector or pAAV-MCS. Preferably, the plurality of NF-κB Motif gene fragments contain 1 to 10 NF-κB Motif DNA fragments, and the sequence of the NF-κB Motif DNA fragments is 5'-GGACTTTCC-3'.

3. The gene interference vector according to claim 1 or 2, characterized in that, The spacer fragment is 10-20 bp located between NF-κB M Otif DNA fragments. Preferably, the sequence of the spacer fragment is 5'-GGGAATTTCCGG-3'.

4. The gene interference vector according to claim 1 or 2, characterized in that, The minimum promoter sequence is 5'-TAGAGGGTATATAATGGAAGCTCGACTTCCA-3'.

5. The gene interference vector according to claim 1 or 2, characterized in that, The sequence of the GBA mRNA is shown in SEQ ID NO.1 to SEQ ID NO.

6. Preferably, the sequence of the GBA mRNA is shown in SEQ ID NO.1 or SEQ ID NO.

5.

6. The gene interference vector according to claim 1 or 2, characterized in that, The sequence of the shGBA is shown in SEQ ID NO.8~SEQ ID NO.

13. Preferably, the sequence of the shGBA is shown in SEQ ID NO.8 or SEQ ID NO.

12.

7. The method for constructing the gene interference vector according to any one of claims 1 to 6, characterized in that, Includes the following steps: The NF-κB Motif gene fragment, the minimal promoter gene fragment, and the miGBA gene fragment are obtained and inserted into the vector backbone.

8. A cell, characterized in that, The cells are obtained by transfecting host cells with the gene interference vector according to any one of claims 1 to 7.

9. The cell according to claim 8, characterized in that, The host cells include AML cells or normal cells. Preferably, the AML cells include one or more of THP-1, U937, KG1a, HL60, WEHI-3, NB4, SKNO01, ML-2, EL9611, or C1498, and the normal cells include one or more of MCF12a, BEAS-2B, HMEC-1, NIH-3T3, L929, GES-1, MRC-5, AML-12, HL-1, or C2C12.

10. The use of the gene interference vector according to any one of claims 1 to 7 or the cell according to any one of claims 8 to 9 in the preparation of a drug for treating acute myeloid leukemia, wherein, preferably, the gene interference vector can continuously respond to NF-κB in AML cells and promote miGBA expression through a positive feedback mechanism; preferably, the gene interference vector can respond to NF-κB activity in acute myeloid leukemia cells, express miGBA, induce lysosomal storage disorder, and trigger multiform cell death, mainly ferroptosis, in acute myeloid leukemia cells.

Images