Full text links CiteDisplay options Display options Format AbstractPubMedPMID
Abstract
Therapeutic strategies are urgently needed for patients with acute myeloid leukemia (AML). Leukocyte immunoglobulin-like receptor B4 (LILRB4), which suppresses T-cell activation and supports tissue infiltration of AML cells, represents an attractive drug target for anti-AML therapeutics. Here, we report the identification and development of an LILRB4-specific humanized mAb that blocks LILRB4 activation. This mAb, h128-3, showed potent activity in blocking the development of monocytic AML in various models including patient-derived xenograft mice and syngeneic immunocompetent AML mice. MAb h128-3 enhanced the anti-AML efficacy of chemotherapy treatment by stimulating mobilization of leukemia cells. Mechanistic studies revealed four concordant modes of action for the anti-AML activity of h128-3: (i) reversal of T-cell suppression, (ii) inhibition of monocytic AML cell tissue infiltration, (iii) antibody-dependent cellular cytotoxicity, and (iv) antibody-dependent cellular phagocytosis. Therefore, targeting LILRB4 with antibody represents an effective therapeutic strategy for treating monocytic AML.
Disclosure of Potential Conflicts of Interests: The Board of Regents of the University of Texas System has filed patent applications with PCT Application Nos. PCT/US2016/020838, which covers LILRB antibodies and their uses in detecting and treating cancer, and PCT/US2017/044171, which covers the methods for identifying LILRB-blocking antibodies. Authors C.C.Z., M.D., Z.A., N.Z., and X.G. are listed as inventors of PCT/US2016/020838. Authors C.C.Z., Z.A., N.Z., M.D., J.K. and X.G. are listed as inventors of PCT/US2017/044171. Both patent applications have been exclusively licensed to Immune-Onc Therapeutics by the Board of Regents of the University of Texas System. Authors N.Z., C.C.Z., and Z.A. have a sponsored research agreement with Immune-Onc Therapeutics. Authors C.C.Z. and Z.A. are Scientific Advisory Board members with Immune-Onc Therapeutics, who also own equities. Authors T.H. and X.C.L. are employees and own equities of Immune-Onc Therapeutics.
Figures
Fig. 1.. Generation and characterization of LILRB4…
Fig. 1.. Generation and characterization of LILRB4 blocking mAbs.
(A) Single antigen-specific memory B cell… Fig. 1.. Generation and characterization of LILRB4 blocking mAbs. (A) Single antigen-specific memory B cell isolation, culture, and cloning strategy used to generate LILRB4 rabbit mAbs. After isolation and culture of LILRB4-specific memory B cells from immunized rabbits, desired B cells were screened for binding to LILRB4 in ELISA. vH and vL genes were then cloned into rabbit IgG backbones and recombinant mAbs were produced using a transient HEK293F cell expression system. (B) EC50 of 26 LILRB4 rabbit mAbs. An irrelevant rabbit antibody (rIgG) was used as negative control. EC50 ≥ 1.0 nM showed as 1.0 nM. Two independent experiments were performed. (C) Node plot of the epitope bins of 21 LILRB4 rabbit mAbs determined by Octet RED96 using a classic sandwich epitope binning assay. (D) Binding of seven representative rabbit mAbs (from different bins) to ECD (residues 22–259), D1 (first Ig-like domain, residues 27–118), D2 (second Ig-like domain, residues 119–218) and SR (stalk region, residues 219–259) of LILRB4 determined by ELISA. (E) Binding of seven representative rabbit mAbs (from different bins) to LILRB4-expressing THP-1 cells determined by flow cytometry. Two independent experiments were performed. (F) Screening of LILRB4 blocking mAbs in chimeric receptor reporter assay. APOE2 was used as a functional ligand to activate LILRB4 reporter cells. The two LILRB4 blocking antibodies are shown in red. (G and H) Kinetics of 128–3 and 216–1 binding to LILRB4 were assessed using an Octet RED96. (I) Binding ability of 128–3 and 216–1 with LILRB4 determined by ELISA. (J) LILRB4 blocking efficacy of 128–3 and 216–1 was determined by reporter assay. Two independent experiments were performed. (K) Specificities of 128–3 and 216–1 were assessed by ELISA. (L) A combined KABAT/IMGT complementarity determining regions (CDR) graft strategy was used to humanize rabbit mAb 128–3. (M) Binding of humanized 128–3 (h128–3) to LILRB4 was determined by ELISA. (N) Affinity of humanized 128–3 (h128–3) to LILRB4 was determined by Octet RED96. (O) LILRB4 blocking efficacy of h128–3 was determined by reporter assay. Two independent experiments were performed.
Fig. 2.. Decoding of the interaction of…
Fig. 2.. Decoding of the interaction of h128–3 with D1.
(A) Binding of h128–3 to… Fig. 2.. Decoding of the interaction of h128–3 with D1. (A) Binding of h128–3 to D1, D2, and ECD of LILRB4 was determined by ELISA. (B) Overall crystal structure of D1/h128–3-Fab complex. D1 is shown in gray, the antibody h128–3 heavy (H) chain is shown in cyan and its light (L) chain shown in pink. The FG loop of D1 is shown in yellow, the BC loop of D1 is shown in green, and the C’E loop of D1 is shown in blue. (C) Detailed interaction of D1/h128–3-VH. (D) Detailed interaction of D1/h128–3-VL. Residues involved in the hydrogen bond interaction are shown as sticks and labeled. Hydrogen bonds are shown as red dashed lines. (E) The epitope residues in D1 are labeled by black characters. Residues contacted by the h128–3-Fab VH are colored cyan. Residues contacted by the h128–3-Fab VL are colored pink. Residues contacted by both chains are colored red. Binding of h128–3 heavy (F) and light (G) chain mutants to LILRB4 were performed by ELISA. (H) Generation and purification of D1 mutants, which were fused with human IgG1 Fc tag and expressed in HEK293F cells. (I) Binding of h128–3 to D1 mutants was determined by ELISA. (J) Binding of h128–3 with LILRB4 ECD mutants. Three identified critical amino acid residues that were mutated to three different types of amino acid residues. (K) Sequence alignment of h128–3 binding motif (BC loop, C’E loop, and FG loop) in all eleven LILR family members. LILRB4 unique amino acid residues R56, R101, and V104 are marked with green. Total amino acid similarities of D1 in percentages compared with LILRB4 (100%) are shown.
Fig. 3.. h128–3 reverses T-cell suppression in…
Fig. 3.. h128–3 reverses T-cell suppression in vitro and in vivo .
( A ) Representative…
Fig. 3.. h128–3 reverses T-cell suppression in vitro and in vivo. (A) Representative image of T cells in coculture assay. T cells isolated from healthy donors were incubated in the lower chamber of a 96-well transwell plate with irradiated THP-1 cells (E:T of 2:1) in the upper chamber separated by a membrane with 3 μm pores. After coculture with anti-CD3/CD28-coated beads and rhIL2 for 7 days, representative cells were photographed using an inverted microscope (scale bar, 100 μm). Two independent experiments were performed. T cells were stained with anti-CD3 (B), anti-CD4 (C), and anti-CD8 (D) and analyzed by flow cytometry. (E) Quantitative analysis of the cytokines in supernatants of human PBMC and THP-1 cells coculture assay. Human PBMC (1.5 × 106 cells/ml) and THP-1 cell (3 × 105 cells/ml) along with h128–3 or hIgG (20 μg/ml) were cocultured for 48 hours. The supernatants were then harvested and detected using RayBio G-Series human cytokine antibody array 1000 Kit. (F) Study design. C57BL/6 mice were subcutaneously implanted with human LILRB4 forced expressing mouse AML C1498 cells (C1498-hlilrb4) followed by treatment with h128–3, h128–3 along with anti-CD8, control human IgG (hIgG) or hIgG along with anti-CD8. Endpoints were assessed at day 27 after AML cells transplantation. (G) Tumor growth of subcutaneous C1498-hlilrb4-bearing mice (n=5) treated with h128–3 or hIgG. (H) Quantitation of CD8+CD62L+ memory T cells in subcutaneous C1498-hlilrb4-bearing mice treated with h128–3 or hIgG. (I) Tumor growth of subcutaneous C1498-hlilrb4-bearing mice (n=5) treated with h128–3 or hIgG in T cell depletion condition.
Fig. 4.. h128–3 blocks AML cell migration…
Fig. 4.. h128–3 blocks AML cell migration and tissue infiltration.
(A) Comparison of the short-term… Fig. 4.. h128–3 blocks AML cell migration and tissue infiltration. (A) Comparison of the short-term (20 hours) tissue infiltration of wt or lilrb4-KO AML MV4–11 cells in NSG mice (n=4). The numbers of leukemia cells (GFP+) in bone marrow (BM), liver (LV), and spleen (SP) were determined by flow cytometry and normalized to number in peripheral blood (PB). Homing ratio of MV4–11cells in mice treated with hIgG was normalized to 100%. (B) Comparison of transwell migration abilities of THP-1 cells treated with h128–3 or hIgG. Two independent experiments were performed. Short-term homing abilities of CFSE-labeled MV4–11 cells that were injected into NSG mice followed by immediately treated with h128–3 or hIgG at 8 (C) or 20 (D) hours post-injection. (E) Long-term (21 days) tissue infiltration of THP-1 cells in NSG mice after treatment with h128–3 or hIgG. (F) Study design. NSG mice (n=5) were intravenously injected with luciferase expressing THP-1 cells (THP-1-luc) followed by immediate treatment with h128–3 or hIgG. Bioluminescence imaging (G), Survival curve (H) and body weight changes (I) of NSG mice treated with h128–3 or hIgG. (J) NSG mice were injected with 1×106 THP-1-luc cells followed immediately by treatment with 0.01 mg/kg, 0.1 mg/kg, or 1 mg/kg of h128–3 and monitored by bioluminescence imaging. (K) Study design. Primary AML cells from monocytic AML patients were injected into irradiated NSG mice. After 9 weeks of engraftment, mice were sacrificed and tissues (single cell suspension) with engrafted AML cells were then injected into other irradiated NSG mice followed by immediate h128–3 or hIgG treatment. Endpoints were assessed at day 21 after AML cells injection. Human AML cells in bone marrow, liver, spleen, and peripheral blood were analyzed by flow cytometry. (L) Percentages of human AML cells (CD45+LILRB4+) engrafted in indicated organs were analyzed by flow cytometry at day 21 post-transplant. (M) Study design. NSG mice (n=5 or 6) were intravenously injected with THP-1 cells followed by cytarabine (10 mg/kg) and h128–3 or hIgG (10 mg/kg) treatment at indicated time points. Endpoints were assessed at day 21 after THP-1 cells injection. Anti-human CD45 was used to detect human leukemia cells (THP-1 cells) in liver by flow cytometry. (N) Shown are percentages of human cells engrafted in liver at day 21 post-transplantation.
Fig. 5.. h128–3 triggers ADCC and ADCP.
Fig. 5.. h128–3 triggers ADCC and ADCP.
(A) Comparison of the apoptosis of THP-1 cells… Fig. 5.. h128–3 triggers ADCC and ADCP. (A) Comparison of the apoptosis of THP-1 cells induced by h128–3 or hIgG. Two independent experiments were performed. (B) Comparison of the effect on THP-1 cells proliferation treated with h128–3 or hIgG. Two independent experiments were performed. (C) Complement-dependent cytotoxicity (CDC) of h128–3 was assessed in a lactate dehydrogenase (LDH) assay using normal human serum as complement and THP-1 cell as target cell. Two independent experiments were performed. (D) Antibody dependent cellular cytotoxicity (ADCC) triggered by h128–3 or hIgG were assessed by flow cytometry. THP-1-GFP cells were used as target cells and fresh isolated PBMCs from healthy donors used as effector cells in a E:T ratio of 50:1. Two independent experiments were performed. (E-G) Antibody dependent phagocytosis (ADCP) triggered by h128–3 or hIgG was detected in flow cytometry assay using THP-1 cells as target cells and mouse macrophage cell line RAW 264.7 (mouse Macs) or human PBMC derived macrophages (human Macs) as effector cells. After incubation of h128–3 or hIgG together with target and effector cells for 2 hours at 37°C, adherent macrophages were collected and determined by flow cytometry. Phagocytosis percentage was calculated by double positive macrophages/total macrophages. Representative FACS profiles are shown. Three independent experiments were performed. (H) Comparison of the binding of wild-type (h128–3) and N297A mutated h128–3 (h128–3-N297A) to LILRB4 in ELISA. (I) Study design. NSG mice (n=5) were intravenously injected with AML THP-1 cells followed by treated with h128–3 or h128–3-N297A at day 3 after THP-1 cells injection. Endpoints were assessed at day 21 after THP-1 cells injection. Human cells (THP-1 cells) in bone marrow, liver, spleen, and peripheral blood were analyzed by flow cytometry. (J) Percentages of human AML cells engrafted in indicated organs were analyzed by flow cytometry at day 21 post-transplant.
Fig. 6.. Multiple mechanisms of h128–3 contribute…
Fig. 6.. Multiple mechanisms of h128–3 contribute to anti-AML activity.
Fig. 6.. Multiple mechanisms of h128–3 contribute to anti-AML activity. See this image and copyright information in PMC
References
Döhner H, Weisdorf DJ, Bloomfield CD, Acute myeloid leukemia. New England Journal of Medicine 373, 1136–1152 (2015). - PubMed
Siegel RL, Miller KD, Jemal A, Cancer statistics, 2016. CA: a cancer journal for clinicians 66, 7–30 (2016). - PubMed
Meyers J, Yu Y, Kaye JA, Davis KL, Medicare fee-for-service enrollees with primary acute myeloid leukemia: an analysis of treatment patterns, survival, and healthcare resource utilization and costs. Applied health economics and health policy 11, 275 (2013). - PubMed
Curran EK, Godfrey J, Kline J, Mechanisms of Immune Tolerance in Leukemia and Lymphoma. Trends in immunology 38, 513–525 (2017). - PMC - PubMed
Thomas D, Majeti R, Optimizing next-generation AML therapy: activity of mutant IDH2 inhibitor AG-221 in preclinical models. Cancer discovery 7, 459–461 (2017). - PMC - PubMed
