Targeting IL11 Receptor in Leukemia and Lymphoma: A Functional Ligand-Directed Study and Hematopathology Analysis of Patient-Derived Specimens

This article is featured in Highlights of This Issue, p. 2881

Leukemia cells express unique surface receptors that may be leveraged toward targeted delivery of diagnostic and therapeutic agents (1–4). In vivo phage display is one approach that can potentially identify and validate functional ligand-mimics binding to relevant membrane receptors that promote cell internalization within the context of the tumor microenvironment. Our group has pioneered the direct screening of phage display random peptide libraries in cancer patients to enable unbiased discovery of tumor targets (5, 6). In previous work with this platform technology, we isolated a ligand that mimics the IL11 motif (cyclic peptide CGRRAGGSC) and have demonstrated that the IL11 receptor (IL11R) is a tumor target in primary tumors of bone, such as osteosarcoma, and in secondary bone metastases from solid tumors such as prostate cancer (7–10). Based on these findings, we have designed and produced a new ligand-directed agent, bone metastasis targeting peptidomimetic-11 (BMTP-11). BMTP-11 consists of the selected IL11R-targeting motif synthesized in tandem to the sequence _D(KLAKLAK)₂, a peptidomimetic motif that induces cell death via mitochondrial membrane disruption upon cell internalization. The efficacy and toxicology of various ligand-directed versions of _D(KLAKLAK)₂ have been extensively evaluated in preclinical models of human diseases with a vascular component such as cancer, obesity, and retinopathies (7, 10–14).

Given the marked expression of the IL11R in the bone marrow within the context of primary or metastatic solid tumors, along with its absence from normal bone marrow (7, 8, 10), we reasoned that the IL11R might also be a suitable target in human leukemia.

Here, we evaluate the protein expression of the IL11R in a panel of leukemia cell lines and patient-derived bone marrow and peripheral blood samples. Moreover, we assess the effectiveness of the prototype BMTP-11 for inducing cell death in human leukemia cell lines and the clonogenic potential in patient-derived leukemia samples. We also introduce a lead-optimized myristoylated BMTP-11 analogue with an improved antileukemia profile. Together, these data indicate that the IL11R is a relevant molecular target in human leukemia. Given the results presented here, along with extensive toxicology studies and a first-in-human trial in prostate cancer patients, reported in Pasqualini and colleagues (15), the parental BMTP-11 in consort with its derivatives merits attention as targeted drug leads against human leukemia.

A panel of human cell lines was obtained from the Leukemia Cell and Tissue Bank of the Department of Leukemia at The University of Texas MD Anderson Cancer Center (UTMDACC). No authentication was done. The panel (n = 12) included cryopreserved samples of MOLT-4 (T-cell acute lymphoblastic leukemia), CCRF-CEM (T-cell acute lymphoblastic leukemia), HL-60 (acute promyeolocytic leukemia), OCI-AML3 (acute myelogenous leukemia), THP-1 (monocytic acute leukemia), K562 and KBM7 (chronic myelogenous leukemia), SR-786 (anaplastic large T-cell lymphoma), U937 and TUR (monocytic lymphoma), TF-1 (erythroleukemia), and RPMI-8226 (myeloma). Cells were maintained in humidified hypoxia chambers (HeraCell 150; Thermo Electron Corporation) with 5% CO₂ and 5% oxygen at 37°C in RPMI-1640 containing 10% FBS, l-glutamine (0.292 mg/mL), penicillin (100 units/mL), and streptomycin (100 units/mL; complete RPMI-1640).

Primary samples from leukemia patients who had signed written informed consent were obtained from the Leukemia Cell and Tissue Bank of the Department of Leukemia at the UTMDACC. Normal blood and bone marrow samples were commercially obtained (AllCells). Cells were maintained in humidified hypoxia chambers (HeraCell 150; Thermo Electron Corporation) with 5% CO₂ and 5% oxygen at 37°C in StemPro34 SFM (Life Technologies), l-glutamine (0.292 mg/mL), penicillin (100 units/mL), and streptomycin (100 units/mL).

Available Wright–Giemsa-stained peripheral blood and bone marrow aspirate smears, hematoxylin–eosin-stained bone marrow aspirate clot, and trephine biopsy specimens were reviewed. In the bone marrow, the blast percentage was derived from a 500-manual cell differential of all nucleated cells in the aspirate smears. White blood cell counts were produced by a multichannel hematology analyzer (Sysmex XE; Sysmex America Inc.).

BMTP-11 is a synthetic peptidomimetic composed of an IL11R-binding cyclic motif (containing natural L-residues, sequence CGRRAGGSC). A glycinylglycine linker was added to the targeting motif as a spacer to prevent steric hindrance and fused in tandem to the D-residue of cell death–inducing motif _D(KLAKLAK)₂ (8, 10–14). Parental BMTP-11 and all its derivative peptidomimetics were generated by commercial vendors (AnaSpec; Jitsubo; or PolyPeptide as indicated) through solid-phase peptide synthesis assembly process, cyclization by air oxidation, purification by reverse-phase high-performance liquid chromatography, and isolation by lyophilization to our specifications. Final identification of the sequences was carried out by analysis with matrix-assisted laser desorption time-of-flight mass spectrometry. Peptide and peptidomimetic sequences were analyzed and confirmed (Supplementary Table S1).

Exponentially growing leukemia cells were harvested, washed with PBS containing 1% FBS and 2 mmol/L EDTA, and resuspended in ice-cold wash buffer. Mononuclear cells from peripheral blood and bone marrow samples obtained from healthy donors and acute myelogenous leukemia (AML) patients were isolated via Ficoll–Paque PLUS (GE Healthcare) purification. Human Fc Receptor Blocking Reagent (Miltenyi Biotec) was added to each cell suspension (20 μL per 1 × 10⁷ cells) and incubated for 10 minutes on ice. Next, 20 μL of either anti-human IL11Rα-PE–labeled antibody (Clone N-20; Santa Cruz) or PE-conjugated rabbit IgG isotype control (GeneTex) was added per 1 × 10⁶ cells in 100 μL of FACS buffer and incubated at 4⁰C in the dark for 1 hour. To detect the IL11Rα expression, the cells were double-stained with antibodies against the cell type–specific CD markers: anti-human CD3-APC (eBioscience), anti-human CD19-APC-Cy7 (BD Biosciences), anti-human CD14-PE-Cy7 (eBiosciences), anti-human CD33-PE-Cy7 (eBioscience), and anti-human CD34-APC (BD Bioscience) as instructed by the manufacturer. Subsequently, cells were washed with FACS buffer and analyzed using a BD LSR Fortessa flow cytometer (BD Biosciences).

Immunohistochemistry was performed on formalin-fixed, paraffin-embedded sections of bone marrow aspirate clot or biopsy specimens. The sections were deparaffinized, rehydrated, and subjected to heat-induced epitope retrieval process using Diva buffer (Biocare Medical) and a Decloaking Chamber (pressure cooker; Biocare Medical). The slides were incubated with an anti-IL11R mouse monoclonal antibody at a dilution of 1:10 (Clone 4D12; Santa Cruz Biotechnology) or a negative control mouse IgG at room temperature in a dark chamber for 1 hour. Detection of the primary antibody was achieved with the Mach4 Kit (Biocare Medical), used according to the manufacturer's instruction. Chromogen development was performed with DAB+ (DakoCytomation). Slides were counterstained with hematoxylin, dehydrated, mounted, and cover-slipped. A specimen from a prostate carcinoma, metastatic to the bone marrow, was used as a positive control for the immunohistochemical (IHC) staining (8). Two observers (K. Karjalainen and C. Bueso-Ramos), without prior knowledge of the clinical data, evaluated the IHC results. We used ×40 magnification to determine the percentage of tumor cells that exhibited positive immunoreactivity against the IL11R antibody. At least 500 neoplastic cells were counted in each case. The presence of membrane and/or cytoplasmic staining for IL11R in ≥ 5% cells was scored as positive. The intensity of the staining also varied between specimens, and we have indicated the weak-staining pattern as weak positive.

Cytospins were prepared with CD34⁺ cells, air-dried, and fixed with buffered 10% formalin. The cytospins were briefly washed with TBS with 0.1% Tween (which was also used throughout the procedure in washing steps) to remove formalin residue and then subjected to heat-induced epitope retrieval process using Diva buffer (Biocare Medical) and a Decloaking Chamber (pressure cooker; Biocare Medical). The cytospins were incubated with an anti-IL11R mouse monoclonal antibody at a dilution of 1:10 (4D12; Santa Cruz Biotechnology) at room temperature in a dark chamber for 1 hour. Detection of the primary antibody was achieved with the Mach4 Kit (Biocare Medical) used according to the manufacturer's instruction. Chromogen development was performed with DAB+ (DakoCytomation). Slides were counterstained with hematoxylin, air-dried, and cover-slipped.

Human leukemia cells (2 × 10⁴ cells/well) were plated in 96-well dishes containing complete RPMI-1640. Cells were incubated with increasing concentrations of either BMTP-11 or an admixture of CGRRAGGSC plus _D(KLAKLAK)₂ (negative control) at equimolar concentrations overnight at 37°C. Leukemia cell viability was determined through counting viable cells after trypan blue staining by an automated cell counter (TC10 automated cell counter; BioRad) or through measuring of cytoplasmic lactate dehydrogenase (LDH) enzymatic activity with a commercial kit (DHL Cell Viability & Proliferation Assay Kit; AnaSpec). This standard assay enables one-step counting and continuous monitoring of leukemia cell proliferation over time by measuring cytoplasmic LDH activity. Resazurin serves as a sensitive indicator converted to the fluorescent resorufin by cytoplasmic LDH.

Human leukemia cells (2 × 10⁵ cells per well in 1 mL) were plated in 6-well plates containing complete RPMI-1640. The cells were incubated with concentrations indicated of either BMTP-11 or an admixture of CGRRAGGSC plus _D(KLAKLAK)₂ (negative control) overnight at 37°C, or with BMTP-11 structural analogues and negative controls for 4 hours at 37°C. Untreated leukemia cells served as a control to measure background levels of cell death under these conditions. After the incubation, the cells were stained with FITC-conjugated Annexin V antibody and propidium iodide (PI) with a standard cell death detection kit (Sigma). The leukemia cells were subsequently analyzed by flow cytometry as described above.

Clonogenic potential of human hematopoietic progenitor cells was assessed via the human colony-forming cell assay using Complete Methocult methylcellulose-based media (Stemcell Technology). The bone marrow–derived cells were plated at the concentration of 2 × 10⁴ cells per 35 mm dish in the presence of either 100 μmol/L of BMTP-11, 30 μmol/L of BMTP-11A#4, or 100 μmol/L of negative control (_D(KLAKLAK)₂). The cultures were incubated for 16 days after which the colonies were counted manually with an inverted microscope.

All data are reported as the average mean ± SEM. The Student t test or Fisher exact test was used to determine statistical significance as indicated. P values that were considered statistically significant are indicated with asterisks as follows: less than 0.05 (*), less than 0.01 (**), or less than 0.001 (***). We used the following categories for the Fisher exact test. Group A: positive white blood cells (≥ 5% nuclei positive) and negative white blood cells (< 5% nuclei positive) in healthy bone marrow and diseased bone marrow. Group B: positive white blood cells plus endothelial cells (≥ 5% nuclei and vessels positive) and negative white blood cells plus endothelial cells (< 5% nuclei and vessels positive) in healthy bone marrow and diseased bone marrow.

To evaluate IL11R as a membrane target in leukemia, we analyzed the surface expression of IL11R on leukemia cells. Positive anti-IL11R antibody staining was observed on all the human leukemia, myeloma, and lymphoma cell lines tested, namely: MOLT-4, OCI-AML3, K562, KMB7, THP-1, HL-60, CCRF-CEM, TF-1, SR-786, U937, TUR, and RPMI-8226 (Fig. 1A). The cell surface expression of IL11R was subsequently studied via flow cytometry (Fig. 1B; Supplementary Fig. S1). Different hematopoietic cell populations from both healthy (n = 3) and leukemic tissues were analyzed. The overall expression of IL11R was increased in all the AML samples tested compared with the normal bone marrow or peripheral blood tested. At the cell population level, the surface expression was consistently high in CD34⁺ (3/3 cases) and CD33⁺ cell populations (3/3 cases), as well as in CD14⁺ population (2/3 cases). One AML sample also stained positive in CD19⁺ cells (Fig. 1B; Supplementary Fig. S1). Cell surface membrane expression of IL11R was also confirmed via cytospin analysis of CD34⁺ bone marrow cells derived from healthy individuals as well as patients with AML (Fig. 1B; Supplementary Fig. S2).

Next, we examined levels of IL11R in bone marrow samples from a panel (n = 43) of leukemia patients via immunohistochemistry (Table 1). This panel of patient-derived samples was obtained from clinically annotated cases, including AML (n = 33), myelodysplastic syndrome (n = 4), myeloproliferative syndrome (n = 2), and B-cell malignancies (n = 4). The expression of IL11R in bone marrow specimens involved by leukemia (leukemia cells and vessels) was higher than in normal bone marrow (Fisher exact test, P = 0.0448), with over half of the leukemia blasts (23 of 43, 53%) staining positively. Notably, all the leukemic bone marrow samples with evaluable blood vessels (16 of 17, 94%) had positive IL11R vascular staining, whereas there was no detectable expression in the normal vasculature (Fisher exact test, P = 0.0158), regardless of subtype and corresponding blast cell positivity. A majority of cases with leukemic bone marrow (∼99%) had >20% disease involvement, and most of the cases had >40% disease involvement with very few normal bone marrow components such as megakaryocytes. Specifically, 24 AML bone marrow specimens of 33 had >40% disease involvement and many (16 cases) had >60% disease involvement.

However, the bone marrow specimens with myelodysplastic syndrome presented only partial disease involvement (focal involvement). From these six cases, only two (cases 12 and 13) had >10% disease involvement, and only one of them (case 13) stained positive. These data indicate that the IL11R staining is specific and occurs only in certain bone marrow specimens with significant disease involvement. Representative samples for IHC staining of IL11R in normal and diseased bone marrow are shown (Fig. 2). A 1,000-fold original magnification picture of AML subtype 5 (AML-M5) expressing membrane and cytoplasmic IL11R is shown as an example (Supplementary Fig. S2A). Moreover, we further analyzed the IL11R expression by preparing cytospins of isolated CD34⁺ cell populations from healthy bone marrow specimens (n = 2) and bone marrow specimens from patients with AML (n = 2). The expression of IL11R in normal CD34-positive myeloblasts is rare, and the fraction of positive cells in two samples of CD34⁺ blasts purified by immunomagnetic isolation was 1% or less. A distinct membranous IL11R expression was noted in CD34-positive myeloblasts in one of two cases of AML (Supplementary Fig. S2B).

We next investigated whether IL11R would serve as a molecular target for ligand-directed drug delivery of BMTP-11. Upon internalization by target cells via ligand-directed delivery, _D(KLAKLAK)₂ induces cell death by disruption of mitochondrial membranes (8, 10–14). BMTP-11 decreased cell viability at low concentrations (40–100 μmol/L range) in human leukemia cell lines, and in most cases, the maximum inhibitory concentration was reached within the 60 to 80 μmol/L range (Fig. 3). In all cases, a negative control admixture of CGRRAGGSC plus _D(KLAKLAK)₂ had no detectable effect at equimolar concentrations (Fig 3). The cell viability was determined by measuring LDH activity using established methods as described in the Materials and Methods.

To assess the specificity of the BMTP-11, we simultaneously tested the effects of BMTP-11 on purified normal white blood cell populations, including granulocytes, monocytes, and lymphocytes, as well as human MOLT-4 leukemia cells. BMTP-11 had lower activity against normal white blood cells compared with MOLT-4 cells (Fig. 4A). Moreover, no detectable effects of BMTP-11 on mononuclear cells from normal bone marrow were observed (Fig. 4B). We next studied the effect of the IL11R-targeted drug on cell morphology. Again, BMTP-11, but not a negative control admixture of CGRRAGGSC plus _D(KLAKLAK)₂ at equimolar concentrations, affected the morphology of human leukemia cells after an overnight incubation (Fig. 4C). The cells were further analyzed by AnnexinV–FITC/PI staining to assess the extent of cell death induced by BMTP-11 (Fig. 4D). For instance, OCI-AML3 cells treated overnight with BMTP-11 underwent a concentration-dependent cell death induction, and nearly complete leukemia cell death (∼90% of the cells) was observed at 40 μmol/L or higher. In contrast, leukemia cells treated with the negative control admixture exhibited only background levels of cell death (∼5%), similar to untreated control cells.

To further improve the antileukemia activity of our drug prototype through Merrifield synthesis and chemical lead-optimization, we designed and evaluated four BMTP-11 analogues with modified structures. Cell viability of leukemia cell lines incubated with the drug candidates was determined by a LDH activity assay. Three analogue drugs (denominated BMTP-11A#1, BMTP-11A#2, and BMTP-11A#3) with certain L-amino acid residues changed to the cognate D-residue enantiomers (Supplementary Table S1) were produced, but found to be less active than BMTP-11 (Fig. 5A) against leukemia cell lines and were therefore not considered for further studies described here. In contrast, BMTP-11 analogue #4 (BMTP-11A#4), which is myristoylated on its serine residue, had improved antileukemia drug activity against three different cell lines: OCI-AML3, K562, and MOLT-4 (Fig. 5A). Subsequent studies showed that BMTP-11A#4 induced cell death faster than the parental BMTP-11 at equimolar concentrations. For instance, after only 4 hours, BMTP-11A#4 induced cell death in over 90% of the leukemia cells, whereas BMTP-11 induced cell death in fewer than 10% of the cells (Fig. 5B). However, an extended incubation time allowed BMTP-11 to induce cell death as efficiently (Fig 4D). To further analyze the effect of BMTP-11 and BMTP-11#4 on cell viability, we counted the absolute number of viable cells after treatment with drug candidates (Supplementary Fig. S3). These results were similar to the cell viability results obtained by measuring LDH activity (Fig. 3 and Fig. 5).

The clonogenic potential of patient-derived AML cells was assessed via methylcellulose assays, and statistically significant inhibition was obtained at 100 μmol/L concentration for CGRRAGGSC-GG-_D(KLAKLAK)₂ and at 30 μmol/L with the analogue #4 (Fig. 5C), whereas the clonogenic potential of normal bone marrow cells from healthy individuals was not altered at equivalent concentrations (Fig. 5D). The concentrations used for the individual peptides were selected based on their maximal inhibitory concentrations obtained via cell viability and proliferation assays with leukemia cell lines (Fig 5A).

Combinatorial screening of phage display random peptide libraries in vitro, in experimental animal models, and even directly in patients provides tools to identify molecular targets in the context of human disease (1–4, 7–14). Using this technology, we have previously identified IL11R as a functional target in the context of bone metastasis of prostate cancer (7–9) and primary osteosarcoma (10). In the present study, we show that IL11R is expressed at higher levels on the cell surface of human leukemia cell lines as well as in primary bone marrow samples obtained from patients with AML and chronic lymphocytic leukemia (CLL), relative to control bone marrow samples obtained from healthy donors. In addition, we show an increased vascular expression of IL11R in the blood vessels within leukemic bone marrow samples, regardless of disease subtype or percentage of blast positivity, data indicating that IL11R is also expressed on the surface of stromal (i.e., nonmalignant) cells within the tumor microenvironment. We have used an established targeting motif (7–10) and in vitro functional assays to show that the proapoptotic peptidomimetic _D(KLAKLAK)₂ (8, 10–14), when ligand-directed to the IL11R, is internalized by leukemia cells and induces concentration-dependent programmed cell death.

Several previous studies have implicated the IL11/IL11R signaling pathway in the pathogenesis of leukemia (16–18), particularly in AML-M5 blast cells and in B cells from CLL. For instance, AML-M5 cells showed an enhanced clonal proliferation in response to stimulation by IL11 and granulocyte colony-stimulating factor (16). Moreover, the expression level of IL11R is elevated in B-cell CLL compared with peripheral blood lymphocytes from normal donors (17). IL11R transcription has been detected in K562 (CML), Mo7E (megakaryocytes), and TF1 (erythroleukemia) cells (18), suggesting broad expression.

Therefore, we first evaluated the cell surface expression of IL11R in a panel of leukemia cell lines by flow cytometry. An anti-IL11R antibody strongly stained all the human leukemia cell lines tested, indicating broad and robust cell surface receptor expression in established human leukemia cell lines regardless of original disease subtype. We subsequently assessed the IL11R cell surface expression in AML patient–derived samples and in blood and bone marrow samples from healthy donors. The analysis at the cell population level revealed that the IL11R expression was not significant in the normal samples, and was mainly found in the mature and immature myeloid cell populations, as well as in progenitor cell populations, in the samples from AML patients, with the exception of the B-cell population staining positive in one AML patient sample. In addition, the membrane expression of IL11R in the isolated CD34⁺ cell population from AML patient samples was detected in one of two investigated cases via cytospin analysis, whereas the isolated CD34⁺ population from two different normal bone marrow samples did not show such staining. We next studied IL11R protein levels by immunohistochemistry on a large and diverse panel of bone marrow samples derived from leukemia patients. Healthy donors were used as controls. IL11R staining of leukemia blasts was noted in over half of the human leukemia samples of varying subtypes. Most notably, all the leukemic bone marrow samples containing evaluable blood vessels—including AML, myelodysplastic or myeloproliferative syndromes, and CLL, but not normal bone marrow—stained positive for IL11R. These results indicate that the protein expression of this receptor also occurs in the tumor microenvironment and not only in malignant cells.

When we investigated the targeted drug activity of BMTP-11 against human leukemia and lymphoma cell lines, we found that low concentrations (i.e., <100 μmol/L) of BMTP-11 were broadly effective against the cell lines investigated, relative to an admixture of CGRRAGGSC plus _D(KLAKLAK)₂ at equimolar concentration. Notably, we did not observe a significant correlation between IL11R levels and BMTP-11 sensitivity as the IL11R expression was relatively high for all the cell lines tested. In addition, drug sensitivity is often affected by several factors other than the expression level of the target protein. For instance, different cell types could have different levels of p-glycoprotein 1, which would clearly alter drug sensitivity. Similarly, the mutational landscape of a specific cancer cell plays a crucial role in treatment sensitivity, and therapeutic compound responses have been correlated to specific cancer genotypes (19).

To determine specificity, we also compared the effects of BMTP-11 on subpopulations of normal peripheral blood cells versus leukemia cells (serving as positive controls). BMTP-11 had negligible activity against normal granulocytes, monocytes, and lymphocytes compared with index leukemia and lymphoma cell lines. These data suggest that BMTP-11 preferentially targets leukemia cells relative to normal white blood cells. Furthermore, these results are consistent with the finding that mice with targeted genetic disruption of the IL11R did not display defects in hematopoiesis (20).

More than a decade ago, our group used combinatorial screening in a cancer patient to isolate an IL11 like peptide mapping to domain I of IL11 (7). In a subsequent study, extensive structural and functional investigations uncovered a previously unrecognized receptor-binding site for human IL11 (9). Based on these findings, we initiated a pilot drug lead-optimization program for the BMTP-11 in an attempt to find potentially superior analogues. We generated BMTP-11 analogues with modified structures. One such analogue, BMTP-11A#4, a derivative with a Gly¹-Cys⁹ thioether bond and myristoylation on Ser⁸ residue, had improved activity against leukemia cells on side-by-side comparisons with the parental compound BMTP-11 in the cell viability and proliferation assay with established leukemia cell lines. Both BMTP-11 and BMTP-11A#4 reduced the clonogenic potential of patient-derived AML bone marrow cells at their individual effective inhibitory concentrations, whereas these peptides did not have a similar effect on normal bone marrow cells at equivalent concentrations. These data demonstrate that BMTP-11 and its analogue BMTP-11A#4 could potentially be used to therapeutically target leukemic progenitor cells. Myristoyl groups are highly lipophilic, an attribute that may facilitate their incorporation into phospholipid membrane bilayers and cell internalization (21). Thus, it is possible that BMTP-11A#4 has an improved ligand-directed internalization profile. However, confirmation of this working hypothesis must await formal GMP toxicology studies and X-ray crystallography of BMTP-11 and its derivatives binding to the IL11R.

In conclusion, the expression pattern characterization and functional findings reported here establish the presence of targetable IL11R on leukemia cells and tumor-associated vasculature within human bone marrow. To that end, we present evidence that the drug prototype BMTP-11 and its derivatives have activity against leukemia and lymphoma. Given that, in a recent study (15), we performed an extensive formal toxicology study in rodents and primates, and conducted a first-in-human clinical trial in which the BMTP-11 prototype, and perhaps one of its derivatives, may be considered experimental drug candidates for translational applications against hematopoietic malignancies.

W. Arap and R. Pasqualini have ownership interest (including IP licensing, royalty payments, and prior equity interest) in Arrowhead Research Corp., which is subjected to certain limitations and restrictions under university policy. The University of New Mexico manages the terms of these arrangements in accordance with its conflict of interest policies. No potential conflicts of interest were disclosed by the other authors.

Conception and design: K. Karjalainen, D.E. Jaalouk, J.E. Cortes, E. Koivunen, W. Arap, R. Pasqualini

Development of methodology: D.E. Jaalouk, C. Bueso-Ramos, L. Bover, W.H.P. Driessen, E. Koivunen, W. Arap, R. Pasqualini

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): D.E. Jaalouk, C. Bueso-Ramos, L. Bover, A. Kuniyasu, A.J. Zurita, J.E. Cortes, G.A. Calin, W. Arap, R. Pasqualini

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): K. Karjalainen, D.E. Jaalouk, L. Bover, W.H.P. Driessen, A.J. Zurita, S. O'Brien, H.M. Kantarjian, G.A. Calin, E. Koivunen, W. Arap, R. Pasqualini

Writing, review, and/or revision of the manuscript: K. Karjalainen, M. Cardó-Vila, C. Rietz, A.J. Zurita, S. O'Brien, H.M. Kantarjian, J.E. Cortes, E. Koivunen, W. Arap, R. Pasqualini

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): Y. Sun, W.H.P. Driessen, S. O'Brien, W. Arap, R. Pasqualini

Study supervision: E. Koivunen, W. Arap, R. Pasqualini

Other (performed research): K. Karjalainen

Other (design, development, and performing experiments): L. Bover

This work was supported by the Specialized Program in Research Excellence (SPORE) Program in Leukemia at the UTMDACC (to R. Pasqualini and W. Arap) and the Gillson-Longenbaugh Foundation (to R. Pasqualini and W. Arap). D.E. Jaalouk received support from the Kimberly Patterson Fellowship for Leukemia Research.

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.