Eradication of Acute Myeloid Leukemia with FLT3 Ligand–Targeted miR-150 Nanoparticles

Acute myeloid leukemia (AML) is one of the most common and fatal forms of hematopoietic malignancies (1). With standard chemotherapies, only 30% to 40% of younger (aged <60) and 5% to 15% of older patients with AML survive more than 5 years (1). Thus, it is urgently needed to develop more effective therapies. FMS-like tyrosine kinase 3 (FLT3) encodes a conserved membrane-bound receptor that belongs to the class III receptor tyrosine kinase family (2). Specific binding of its ligand, FLT3-ligand (FLT3L), leads to the activation of multiple downstream signaling pathways (2–5). FLT3 is a critical oncogene in hematopoietic malignancies (5, 6). Internal tandem duplications (ITD) and tyrosine kinase domain (TKD) point mutations of FLT3 are found in more than 30% of AML cases (5–7). FLT3-ITD–carrying AMLs are usually associated with poor prognosis (4, 8–10). Previous studies reported that FLT3 was overexpressed in more than 60% of AML cases tested, relative to normal control hematopoietic cells (8, 11–13). Although a number of FLT3 tyrosine kinase inhibitors (TKI) have been investigated in clinical trials, the long-term therapeutic effects are disappointing, likely due to secondary mutations in FLT3, upregulation of FLT3 expression, and/or activation of FLT3 downstream signaling pathways (5, 7, 14). Thus, decreasing the abundance of FLT3 at the RNA and protein levels would be an attractive strategy to treat AMLs with FLT3 overexpression and/or FLT3-ITD/TKD mutations.

miRNAs are a class of small, noncoding RNAs that play important roles in posttranscriptional gene regulation (15). We recently reported that miR-150 functions as a pivotal tumor suppressor gatekeeper in MLL-rearranged and other subtypes of AML, through targeting FLT3, MYB, HOXA9, and MEIS1 directly or indirectly (16). We showed that MLL fusion proteins and MYC/LIN28 upregulated FLT3 level through inhibiting the maturation of miR-150 (16). These findings strongly suggest a significant clinical potential of restoration of miR-150 expression and function in treating FLT3-overexpressing AML.

However, successful clinical implementation of miRNA-based therapy requires a highly specific and efficient delivery system (17, 18). Viral delivery systems or conventional nonviral systems, such as cholesterol conjugates and cationic liposome packaging, have often shown severe toxicity and induced hypersensitive reactions in vivo (17). Poly(amidoamine) (PAMAM) dendrimers have been shown to be effective nonviral gene delivery vectors (19, 20). Their spherical, chemically well-defined structure, precisely controlled size, multivalency, and capability of forming complexes (dendriplex) with nucleotide oligos make them as a suitable platform for miRNA delivery (21–27).

In this study, we first showed the particular high expression of FLT3 in some subtypes of AML through analyzing a large cohort of AML patients (N = 562). We then developed targeted nanoparticles based on FLT3L-guided dendrimers complexed with miR-150 oligos, which exhibited high selectivity and efficiency in targeting leukemic cells with overexpression of FLT3. Our proof-of-principle preclinical animal model studies demonstrate the potent in vivo therapeutic effect of the FLT3L-guided miR-150 nanoparticles in treating FLT3-overexpressing AML, alone or in combination with other therapeutic agents [e.g., JQ1 (28)].

The patient samples (N = 562) were analyzed by the use of Affymetrix Human Genome U133Plus2.0 GeneChips or Affymetrix Human Genome U133A and B (U133A+B) GeneChips. The RMA5 method (29) was used for data normalization, and the expression values were log(2) transformed. Microarrays and data analyses were conducted as described previously (30, 31). The GEO ID of the entire dataset is GSE37642.

Generation 7 (G7) PAMAM dendrimers were obtained from Sigma-Aldrich. N-Hydroxysuccinimide ester of cyanine5.5 (NHS-Cy5.5) was purchased from Lumiprobe Corporation. Sulfo-SMCC was obtained from Thermo Fisher Scientific. miR-150 and miR-150 mutant RNA oligos (with or without 2′-OMe modification) were purchased from Exiqon Inc. The RNA sequences are miR-150: 5′-UCU CCC AAC CCU UGU ACC AGU G-3′; miR-150 mutant: 5′-UUA UUU UAC CCU UGU ACC AGU G-3′.

Human recombinant FLT3L proteins were ordered from ProSpec-Tany Technogene Ltd., which contain 155 amino acids of the extracellular domain of FLT3L (i.e., the soluble FLT3L form). The synthetic Flt3L peptide containing 74 amino acids with the sequence of SSNFKVKFRELTDHLLKDYPVTVAVNLQDEKHCKALWSLFLAQRWIEQLKTVAGSKMQTLLEDVNTEIHFVTSC was purchased from Pierce Biotechnology, Inc.

G7 PAMAM dendrimers obtained from Sigma-Aldrich were purified and fluorescently labeled using an N-hydroxysuccinimide ester of cyanine5.5 (NHS-Cy5.5; Lumiprobe Corporation), as has been previously reported (32). Then, human soluble FLT3 protein- or mouse Flt3L peptide-conjugated G7 dendrimers were prepared similarly as described previously (32).

The patient samples were obtained at the time of diagnosis with informed consent at the University of Chicago Hospital (Chicago, IL) or other collaborative hospitals and were approved by the Institutional Review Board of the institutes/hospitals. All patients were treated according to the protocols of the corresponding institutes/hospitals. All the AML cell lines used herein were originally obtained from ATCC and maintained at the Chen laboratory.

Total RNA was extracted with the miRNeasy Extraction Kit (Qiagen) and was used as a template for quantitative RT-PCR (qPCR) analysis as described previously (16).

MONOMAC-6 and U937 cells were cultured as described previously (16, 33). Continued testing to authenticate these cell lines was done using qPCR and Western blot analysis to validate the existence or absence of MLL-AF9 when the cell lines were used in this project. Both cell lines were tested for mycoplasma contamination yearly using a PCR Mycoplasma Test Kit (PromoKine) and were proven to be mycoplasma negative.

Bone marrow mononuclear cells from healthy donors or AML patients were grown in RPMI medium 1640 containing 10% FBS, 1% HEPES, and 1% penicillin–streptomycin, supplemented with 10 ng/mL SCF, TPO, Flt3L, IL3, and IL6. Cells were treated with PBS, G7-NH₂-miR-150mut, G7-NH₂-miR-150, G7-Flt3L-miR-150mut, or G7-Flt3L-miR-150 at the indicated doses.

These experiments were conducted as described previously (16, 33) with ApoLive-Glo Multiplex Assay Kit or CellTiter 96 AQ_ueous Non-Radioactive Cell Proliferation Assay Kit (Promega).

Secondary mouse bone marrow transplantation was carried out as described previously (16). After the onset of leukemia [when mice had an engraftment (CD45.1) of more than 20% and/or white blood cell counts higher than 4 × 10⁹/L, usually 10 days posttransplantation], the recipient mice were injected with PBS control, 0.5 mg/kg G7-NH₂-miR-150mut, G7-NH₂-miR-150, G7-Flt3L-miR-150mut, or G7-Flt3L-miR-150 i.v. every other day, until the PBS-treated control group all died of leukemia. For the JQ1 and nanoparticle combination treatment experiment, after the onset of AML, the recipient mice were treated with PBS control or 50 mg/kg JQ1 i.p. alone or together with 0.5 mg/kg G7-NH₂-miR-150 or G7-Flt3L-miR-150 i.v. every other day, until the PBS-treated control group all died of leukemia.

All laboratory mice were maintained and monitored as described previously (16).

The microarray data analysis was conducted by the use of Partek Genomics Suite (Partek Inc). The t test, Kaplan–Meier method, and log-rank test, etc., were performed with WinSTAT (R. Fitch Software), GraphPad Prism version 5.00 (GraphPad Software), and/or Partek Genomics Suite (Partek Inc). P < 0.05 was considered as statistically significant.

Previous studies, based on relatively small numbers of AML patients (merely 200 patients or fewer), have shown that FLT3 is overexpressed in the majority cases of primary AML relative to normal control cells, whereas controversial observations have been reported regarding the association of FLT3 overexpression with specific AML subtypes (11, 13, 34, 35). Here, we analyzed FLT3 expression across various AML subtypes in a microarray-based gene expression dataset of 562 AML patients (GSE37642; Supplementary Table S1; refs. 30, 31). Among various French–American–British classification (FAB) subtypes, the expression level of FLT3 in the FAB M1, M2, and M5 AML subgroups was significantly higher than its average level in the entire set of AML samples (Fig. 1A). Among AML subtypes with different cytogenetic abnormalities, the average expression level of FLT3 in AML with t(11q23)/MLL-rearrangements was significantly higher than that in the whole set of AML samples, whereas the opposite was true in AML with -7/del(7q) or complex karyotypes (Fig. 1B). In AML with distinct molecular abnormalities, the average expression level of FLT3 in AML with FLT3-ITD and/or NPM1 mutations was significantly higher than that in the whole set of AML or AML without the relevant mutations (Fig. 1C). We have also conducted qPCR analysis of expression of both FLT3 and miR-150 in 23 human primary AML samples and 5 normal control samples. As shown in Fig. 1D, FLT3 is expressed at a higher level in 21 (91%) of the 23 AML samples, whereas miR-150 is expressed at a lower level in all the 23 AML samples than its average level in the 5 normal controls.

To selectively deliver miR-150 oligos to FLT3-overexpressing AML cells, we chose poly(amidoamine) (PAMAM) dendrimers (21, 32, 36–38) as the basis of our nanoparticle carriers, and conjugated these nanoparticles with the near-infrared dye cyanine 5.5 (Cy5.5) for monitoring the dynamic distributions of the nanoparticles, and FLT3 ligand (FLT3L) proteins for specific targeting FLT3 on the cell surface. H2B, a nuclear histone protein with similar molecular weight as FLT3L, was conjugated as a negative control (Fig. 2A and Supplementary Fig. S1A–S1C).

We then investigated whether G7-FLT3L dendrimers can efficiently and selectively target AML cells with FLT3 overexpression. After 24-hour treatment of the nanoparticles, the uptake ratios of the G7-FLT3L dendrimers were significantly higher than the G7-H2B control nanoparticles in MONOMAC-6 cells, an AML cell line carrying the t(9;11)/MLL-AF9 [i.e., the most common form of MLL-rearranged AML (39, 40)]. The nanoparticles were taken up in a dose-dependent manner (Fig. 2B and Supplementary Fig. S1D). In addition, the cellular uptake of G7-FLT3L nanoparticles by FLT3-overexpressing AML cells is rapid. Just one hour after the treatment of the nanoparticles, the uptake ratio of G7-FLT3L dendrimers was already 23.6%, significantly higher than that (5.9%) of G7-H2B dendrimers (Fig. 2C and Supplementary Fig. S1E). In contrast, the uptake ratios between G7-FLT3L and G7-H2B dendrimers showed no significant difference in U937 cells [a cell line with very low levels of FLT3 (Fig. 2C and Supplementary Fig. S1E; refs. 16, 41)].

Furthermore, we sought to determine whether the high level of FLT3 expression of target cells is required for the high uptake ratio of G7-FLT3L nanoparticles. As reported previously (16), forced expression of miR-150 significantly reduced endogenous expression of FLT3. Suppression of endogenous FLT3 expression in MONOMAC-6 cells by overexpressing miR-150 resulted in a significant decrease of the uptake ratio of G7-FLT3L nanoparticles, but not that of G7-H2B nanoparticles (Fig. 2D and Supplementary Fig. S1F). Neither G7-FLT3L nor G7-H2B dendrimers showed significant effects on the viability and apoptosis of MONOMAC-6 cells as compared with PBS, indicating low, if any, cytotoxicity of both dendrimer constructs (Fig. 2E).

Collectively, the above results indicate that G7-FLT3L nanoparticles can rapidly, efficiently, and selectively target FLT3-overexpressing AML cells, with minimal nonspecific cellular toxicity.

We then integrated miR-150 oligo with G7-FLT3L dendrimers to form G7-FLT3L-miR-150 dendriplexes (Fig. 3A and Supplementary Fig. S2). G7-FLT3L-miR-150 significantly reduced the viability and increased the apoptosis of MONOMAC-6 cells in a dose-dependent manner, while G7-H2B-miR-150 showed mild effects only at a high concentration (Fig. 3B and C). Neither G7-FLT3L-control nor G7-H2B-control carrying miR-150 mutant control oligos [i.e., control/Ctrl, which has a mutated “seed” sequence (i.e., mutations in the 2nd–6th nucleotides at the 5′ end), as a specific loss-of-function control for the wild-type miR-150] exhibited significant influence (Fig. 3B and C). G7-FLT3L-miR-150 also remarkably inhibited AML cell proliferation even at a dose as low as 10 nmol/L (Fig. 3D), whereas G7-H2B-miR-150 showed no significant inhibitory effects on the cell growth (Fig. 3E).

However, the major problem with the G7-FLT3L-miR-150 nanoparticles is that their inhibitory effect on cell growth is not sustained. Cells treated with even the highest dose of G7-FLT3L-miR-150 tend to grow at a normal speed 4 days posttreatment (Fig. 3D). To increase the stability of miR-150 oligos, we incorporated a 2′-o-methyl (2′-OMe) modification (42) into our miRNA oligos. Indeed, the G7-FLT3L nanoparticles carrying modified miR-150 exhibited a more sustained inhibition on cell growth (Fig. 3F). G7-H2B nanoparticles carrying 2′-OMe modified miR-150 showed a slight inhibitory effect at higher doses (Fig. 3G). Because 2′-OMe modified miR-150 oligos exhibited a more stable and potent effect than unmodified oligos, we used modified oligos in all our following studies.

The increase of intracellular miR-150 level in cells treated with G7-FLT3L-miR-150 resulted in a significant downregulation of miR-150′s direct or indirect target genes, including FLT3, MYB, HOXA9, and MEIS1 (Supplementary Fig. S3; ref. 16), as well as a significant repression on FLT3 level, which, in turn, led to the suppression of the FLT3 signaling pathway (5, 7, 14, 43), featured with reduced activation of ERK, AKT, and STAT5 and downregulation of PIM (Fig. 3H).

We then investigated the potential antileukemia effect of G7-Flt3L-miR-150 nanoparticles in treating AML in vivo. To reduce the cost and the size of the guiding molecule, we used synthetic peptides containing the major functional domains of soluble FLT3L protein (Flt3L peptide) rather than the entire soluble FLT3L (see Supplementary Fig. S4A). The new G7-Flt3L dendrimers exhibited a similar high uptake ratio and similar effects on inhibiting viability and promoting apoptosis (Supplementary Fig. S4B–S4D) as compared with the original G7-FLT3L dendrimers (Fig. 2B and Fig. 3B and C).

To show the in vivo distribution of the nanoparticles, we injected the G7-Flt3L or control nanoparticles into normal mice. Whole-animal imaging shows different distributions of the two nanoparticles: G7-Flt3L was mainly enriched in the bone marrow and the liver, whereas G7-NH₂ showed a more diffuse distribution all over the abdomen (Fig. 4A). Both nanoparticles were taken up at a similar ratio in the bone marrow and the spleen of the injected mice, but more G7-Flt3L⁺ cells are c-Kit⁺ progenitor cells, indicating a preferential recruitment of the G7-Flt3L nanoparticles to leukemic stem cells or progenitors with overexpressed Flt3 (Fig. 4B and C). The level of miR-150 was detected in mouse BM c-Kit⁺ cells from 30 minutes to 48 hours after a single dosage of G7-Flt3L-miR-150. Results showed that the cellular level of miR-150 was maintained at 2.01-fold of the control group 24 hours posttreatment and at 1.32-fold 48 hours posttreatment (Supplementary Fig. S5A).

We further investigated the in vivo therapeutic effects of the miR-150 nanoparticles in a MLL-rearranged leukemic mouse model. G7-Flt3L-miR-150 nanoparticles showed the best therapeutic effect as compared with the control group or the G7-NH₂-miR-150 treated group, whereas miR-150–mutant nanoparticles showed no antileukemia effect (Fig. 5A). Notably, G7-Flt3L-miR-150 treatment almost completely blocked MLL-AF9–induced leukemia in 20% of the mice. The therapeutic effect of G7-Flt3L-miR-150 was also evidenced by the morphologic changes in peripheral blood, bone marrow, spleen, and liver between the treated mice and control AML mice (Fig. 5C). The antileukemia effect of G7-Flt3L-miR-150 nanoparticles is associated with a significant increase in intracellular miR-150 level and decrease in the expression of miR-150′s critical direct targets (e.g., Flt3 and Myb) and indirect targets (e.g., Hoxa9 and Meis1; Supplementary Fig. S5B). Furthermore, neither G7-Flt3L-miR-150 nor G7-NH₂-miR-150 nanoparticles exhibited noticeable effects on normal hematopoiesis (Supplementary Table S2; Supplementary Fig. S6).

JQ1 is a small-molecule inhibitor of BET bromodomains and has been proven to be effective in repressing MLL-AF9/Nras^G12D-induced AML progression in vivo (28, 44). Because MYC is also a downstream target of FLT3 (6, 16), we sought to investigate whether the combination of miR-150 nanoparticles and JQ1 exhibits a stronger therapeutic effect than each alone. Our animal leukemia model studies showed that JQ1 alone and G7-Flt3L-miR-150 nanoparticles alone significantly (P < 0.0001) extended the median survival of MLL-AF9 AML mice from 54 days (control; PBS treated) to 69.5 and 86 days, respectively, whereas their combination substantially extended the median survival to >150 days, suggesting a synergistic effect, or at least an additive effect, between JQ1 and miR-150 nanoparticles in treating FLT3-overexpressing AML (Fig. 5B). Notably, 50% of the leukemic mice treated with this combination survived for more than 200 days, (Fig. 5B) and their pathologic morphologies in PB, bone marrow, spleen, and liver tissues became normal (Fig. 5C). JQ1 and G7-Flt3L-miR-150, each alone or in combination, did not cause any noticeable side effects on normal hematopoiesis in vivo (Supplementary Table S3).

AML cell lines and primary patient samples carrying various cytogenetic or molecular abnormalities were treated with the nanoparticles. While exhibiting very minor inhibitory effect on the viability of normal control mononuclear cells from healthy donors (control), G7-Flt3L-miR-150 nanoparticles exhibited a significant inhibitory effect on the viability of leukemic mononuclear bone marrow blast cells collected from 9 primary AML patients carrying t(11;19)/MLL-ENL, t(6;11)/MLL-AF6, t(4;11)/MLL-AF4, FLT3-ITD, t(4;11)/FLT3-ITD, t(9;11)/MLL-AF9, t(8;21)/AML1-ETO, or inv(16)/CBFB-MYH11 to a degree similar to that of MONOMAC-6 cells, or even better (Fig. 6A–K). In contrast, no inhibitory effect was observed on the viability of U937 cells (Fig. 6L). The sensitivity of the AML cells to G7-Flt3L-miR-150 nanoparticle treatment seems to be related to the expression level of endogenous FLT3 in the cells (Fig. 6M). As expected, only the nanoparticles with both Flt3L-guiding peptides and functional miR-150 oligos can exhibit a potent and selective inhibitory effect on FLT3-overexpressing AML cells. Taken together, these results indicate that G7-Flt3L-miR-150 nanoparticles exhibit a broad antileukemic effect by selectively targeting a variety of AML subtypes with overexpression of FLT3.

Through analysis of a large cohort of AML patients (N = 562), here, we provide a more precise understanding of the patterns of FLT3 expression in AML. Although FLT3 is likely overexpressed in the vast majority of primary AML cases, it is particularly highly expressed in a set of AML subtypes, including FAB M1/M2/M5 AML and AML with t(11q23)/MLL-rearrangements, or FLT3-ITD and/or NPM1 mutations. As AMLs with t(11q23) or FLT3-ITD, accounting for more than 40% of total AML cases, are often associated with poor outcome (4, 8–10, 45, 46), it is urgent to develop targeted therapeutic strategies to treat such FLT3-overexpressing, poor prognosis AMLs.

To selectively target FLT3-overexpressing AMLs, we developed a novel, FLT3L-guided, dendrimer-based nanoparticles for the targeted delivery of miR-150 oligos to FLT3-overexpressing AML cells. Theoretically, our nanoparticle package has the following advantages: (i) it inhibits the growth and proliferation of FLT3-overexpressing AML cells by restoration of the expression and function of a pivotal tumor suppressor miRNA (i.e., miR-150) that posttranscriptionally inhibits the expression of FLT3 (at the RNA level). The direct repression of FLT3 expression will avoid the potential rebound of FLT3 expression induced by other treatments, for example, FLT3 TKI treatment (5, 7, 14, 43); (ii) it employs FLT3L as the guiding molecule to specifically deliver miR-150 oligos to FLT3-overexpressing AML cells and thus achieves selective targeting; and (iii) it uses G7 PAMAM dendrimers as the carriers for miRNA delivery. G7 PAMAM dendrimers have a big surface area, flexible architecture, and multivalent binding capability that facilitate tight cell-surface binding, and the high density of positive charges on their surface also facilitates interaction with the cell membrane and subsequent cellular uptake (21, 36, 37). Indeed, we have shown that our G7-FLT3L-miR-150 nanoparticles have the above unique advantages and exhibit specific targeting and potent antileukemia effects on FLT3-overexpressing AML cells both in vitro and in vivo.

Therefore, our work has made the following conceptual advances and innovative discoveries: (i) our data demonstrate that restoration of miR-150 expression/function holds a great therapeutic potential in treating AML with FLT3 overexpression. The 2′-OMe–modified miR-150 oligos delivered by targeted nanoparticles show a high stability and exhibit long-term and potent inhibitory effects on the proliferation and growth of FLT3-overexpressing AML cells in vitro and on the progression of the leukemia in vivo; (ii) we provide the first successful model, to our knowledge, showing that an engineered ligand protein/peptide can be used as a guiding molecule when conjugated to a G7 PAMAM–miRNA dendriplex for specific targeting of leukemic cells with an abnormally high level of the corresponding cell-surface receptors. We show that both the entire soluble FLT3L proteins (with 155 amino acids) and synthetic Flt3L peptides (with 74 amino acids) work very well as the guiding molecules for specific targeting of FLT3-overexpressing AML cells. Compared with nanoparticles conjugated with control proteins, FLT3L-conjugated nanoparticles showed more than 4 times higher specificity and efficiency in targeting and uptake by FLT3-overexpressing AML cells. Compared with control nanoparticles, Flt3L-conjugated nanoparticles tend to concentrate in the bone marrow rather than distribute pervasively throughout the abdomen following injection, and their accumulation is significantly enhanced in c-Kit⁺ leukemic progenitor cells in both bone marrow and spleen tissues. This suggests that we might be able to extend the application of the nanoparticle system to other AMLs (e.g., AMLs with abnormally high level of c-KIT) by changing the targeting molecule from FLT3L to other molecules (e.g., c-KIT ligand); (iii) we show that G7 PAMAM dendrimer–based conjugates are ideal carriers for miRNA delivery to treat AML. Such carriers do not trigger any noticeable side effects on normal hematopoiesis in vivo; (iv) our proof-of-concept leukemia animal model study demonstrated that the G7-Flt3L-miR-150 nanoparticles were able to substantially inhibit leukemia progression and prolong survival of the treated mice that carried FLT3-overexpressing AML, while exhibiting no obvious side effects on normal hematopoietic system. Furthermore, the combination of G7-Flt3L-miR-150 nanoparticles and JQ1 (a bromodomain inhibitor) exhibited a synergistic or at least an additive effect on inhibition of FLT3-overexpressing AML in vivo compared with administration of each agent alone. Notably, the dosages of both G7-Flt3L-miR-150 nanoparticles and JQ1 used in our animal model studies were at a low level (to minimize the costs of experiments), and thus, a more profound antileukemia effect might be observed if higher dosages are used; and (v) our data demonstrate that specific inhibition of FLT3 expression at the transcript/RNA level by miRNA-based targeted nanoparticles (or other strategies) is a feasible and effective approach to treat FLT3-overexpressing AML.

G7-Flt3L-miR-150 nanoparticles could also be combined with FLT3 TKIs to treat AML carrying FLT3-ITD or TKD mutations, which will inhibit both the expression and enzymatic activity of FLT3 and thereby will likely result in a potent long-term therapeutic effect, as FLT3-TKI–induced rebound of FLT3 level and secondary mutations in FLT3 TKD (5, 7, 14, 43) can be inhibited by G7-Flt3L-miR-150 nanoparticles at the transcript level. With no doubt, our G7-Flt3L-miR-150 nanoparticles can also be combined with other newly developed targeted therapeutic agents (e.g., JQ1), and such combinations may represent promising novel therapeutic strategies with high efficacy and minimal side effects.

In conclusion, our results show that FLT3L-guided nanoparticles complexed with miR-150, a potent tumor-suppressing miRNA, exhibit a high specificity and efficiency in targeting FLT3-overexpressing AML. Our proof-of-principle leukemia animal model studies demonstrate the feasibility and effectiveness of our approach using the targeted miR-150 nanoparticles in treating FLT3-overexpressing AML in vivo. Our studies have made a number of conceptual advances and novel discoveries in the fields of nanoparticle design/application and targeted leukemia/cancer therapy as discussed above and also provided solid basis for the development of new therapeutic strategies for clinical applications in treating FLT3-overexpressing AMLs, the majority of AML cases that are resistant to currently available standard chemotherapies.

J.E. Bradner is the President at Novartis Institute of BioMedical Research. No potential conflicts of interest were disclosed by the other authors.

Conception and design: X. Jiang, Y. Yang, J. Qi, Z. Li, S. Hong, J. Chen

Development of methodology: X. Jiang, J. Bugno, Y. Yang, P. Chen, Z. Li, J.E. Bradner, S. Hong, J. Chen

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): X. Jiang, J. Bugno, C. Hu, Y. Yang, T. Herold, P. Chen, S. Arnovitz, J. Strong, K. Ferchen, B. Ulrich, H. Weng, Y. Wang, R.A. Larson, M.M. Le Beau, S.K. Bohlander, S. Hong, J. Chen

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): X. Jiang, J. Bugno, P. Chen, S. Gurbuxani, H. Huang, S. Hong, J. Chen

Writing, review, and/or revision of the manuscript: X. Jiang, J. Bugno, Y. Yang, T. Herold, H. Huang, S. Li, R.A. Larson, S.K. Bohlander, J. Jin, Z. Li, S. Hong, J. Chen

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): X. Jiang, J. Qi, S. Arnovitz, Y. Wang, S. Li, M.B. Neilly, J. Jin, Z. Li, S. Hong, J. Chen

Study supervision: X. Jiang, S. Hong, J. Chen

Other (providing grants/reagents/tools): J. Chen

Special thanks to our late mentor and colleague Dr. Janet Rowley for her long-term support. The authors thank Drs. Scott Armstrong, Gregory Hannon, and Lin He for providing retroviral constructs. The authors also thank all participants and recruiting centers of the Acute Myeloid Leukemia Cooperative Group (AMLCG) trials.

This work was supported by the NIH R01 grants CA182528, CA178454, and CA127277 (J. Chen), Alex's Lemonade Stand Foundation for Childhood Cancer (J. Chen); Leukemia & Lymphoma Society (LLS) Translational Research Grant (J. Chen), American Cancer Society (ACS) Research Scholar grant (J. Chen), The University of Chicago Committee on Cancer Biology (CCB) Fellowship Program (X. Jiang), LLS Special Fellowship (Z. Li), Gabrielle's Angel Foundation for Cancer Research (J. Chen, Z. Li, X. Jiang, and H. Huang), and the family of Marijanna Kumerich and Leukemia & Blood Cancer New Zealand (S.K. Bohlander).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.