Small molecule kinase inhibitor LRRK2-IN-1 demonstrates potent activity against colorectal and pancreatic cancer through inhibition of doublecortin-like kinase 1

Abstract

Background: Doublecortin-like kinase 1 (DCLK1) is emerging as a tumor specific stem cell marker in colorectal and pancreatic cancer. Previous in vitro and in vivo studies have demonstrated the therapeutic effects of inhibiting DCLK1 with small interfering RNA (siRNA) as well as genetically targeting the DCLK1+ cell for deletion. However, the effects of inhibiting DCLK1 kinase activity have not been studied directly. Therefore, we assessed the effects of inhibiting DCLK1 kinase activity using the novel small molecule kinase inhibitor, LRRK2-IN-1, which demonstrates significant affinity for DCLK1.

Results: Here we report that LRRK2-IN-1 demonstrates potent anti-cancer activity including inhibition of cancer cell proliferation, migration, and invasion as well as induction of apoptosis and cell cycle arrest. Additionally we found that it regulates stemness, epithelial-mesenchymal transition, and oncogenic targets on the molecular level. Moreover, we show that LRRK2-IN-1 suppresses DCLK1 kinase activity and downstream DCLK1 effector c-MYC, and demonstrate that DCLK1 kinase activity is a significant factor in resistance to LRRK2-IN-1.

Conclusions: Given DCLK1's tumor stem cell marker status, a strong understanding of its biological role and interactions in gastrointestinal tumors may lead to discoveries that improve patient outcomes. The results of this study suggest that small molecule inhibitors of DCLK1 kinase should be further investigated as they may hold promise as anti-tumor stem cell drugs.

Figure 1

LRRK2-IN-1 inhibits DCLK1 kinase activity. An in vitro kinase assay was performed using Purified active DCLK1 kinase (0.25 μg) with 2.5 μg of autocamtide II substrate, 1 μM ATP, and either DMSO, 0.6, 2.5, 5, 10, or 50 nM LRRK2-IN-1 (A). Using relative luminescent units (RLU) data, a sigmoidal-dose response curve was plotted in GraphPad Prism 6.0 (adj. R² = 0.952) revealing an IC₅₀ value of 2.61 nM (B). AsPC-1 cells were treated with LRRK2-IN-1 at varying concentrations for 48 h. Following treatment cells were lysed, protein was isolated and quantified by BCA assay, and immunoblotting was performed with α-phospho-DCLK1. The ratio of phospho-DCLK1 to total DCLK1 (Figure  4B; 48 h) was determined and demonstrated decreased phosphorylation of DCLK1 (p < 0.05) following treatment (C). Schematic demonstrating the shared protein kinase domain between DCLK1 isoforms referenced in Uniprot [Swiss-Prot: O15075] (D). Three dimensional view of LRRK2-IN-1 binding site in DCLK-long-β revealing predicted interactions with residues of the hinge region, catalytic loop (C-loop), activation loop (A-loop), αC-helix (αC), and the highly conserved lysine (Lys112) of the kinase catalytic domain suggesting that LRRK2-IN-1 competes with ATP for the DCLK1 kinase binding pocket. Dashed lines mark the hydrogen bond formed with the conserved aspartate (“D” of the “DFG” motif) of the activation loop (E).

Figure 2

LRRK2-IN-1 elicits anticancer activity in vitro. HCT116 and AsPC-1 cells were seeded into 96-well plates at 10⁴ cells per well and allowed to attach overnight at 37°C. LRRK2-IN-1 was added in triplicate to the wells at concentrations of 0 (DMSO), 0.3, 0.6, 1.25, 2.5, 5, 10, and 20 μM and incubated at 37°C. After 48 h an MTT proliferation assay was performed and revealed significant inhibition of cell proliferation starting at 0.62 μM (A-B). A live/dead assay in AsPC-1 cells was performed 24 h post treatment to confirm LRRK2-IN-1’s cytotoxic effect. Additionally, a luminescent assay revealed that this effect co-ocurred with Caspase 3/7 activation, both p < 0.0001 by ANOVA (C). AsPC-1 cells were seeded into 6-well plates allowed to reach confluence, scratched, and treated with DMSO, 0.5, 5, or 10 μM LRRK2-IN-1 and incubated at 37°C. The wound area was imaged at baseline, 12, 24, 48, and 72 h and quantified using ImageJ (D-E). mRNA expression analysis of AsPC-1 cells treated with DMSO or LRRK2-IN-1 at concentrations of 0.5 μM or 5.0 μM for 8 h revealed a significant downregulation (p < 0.01) of apoptosis regulators BCL2, BCL2L1 (BCL-XL), and MCL1 mRNA expression (F). Western blots showed decreased BCL2 and Phosphohistone H3 at both 24 and 48 h post LRRK2-IN-1 treatment (G).

Figure 3

LRRK2-IN-1 induces G1 and G2 arrest in vitro. Immunofluorescence of AsPC-1 cells treated with LRRK2-IN-1 demonstrating a decrease in mitotic cells at 48 h as determined by phosphohistone H3 (PHH3) positive cells (A). Cell cycle analysis of AsPC-1 cells 48 h post LRRK2-IN-1 demonstrating maximum G1 arrest at 5.0 μM and G2/M arrest at 20 μM regardless of ploidy level. Yellow peaks denote aneuploid populations and arrows identify these populations in the density maps (B). Cell cycle analysis of AsPC-1 cells 24 and 48 h post LRRK2-IN-1 treatment demonstrating G1 arrest at low doses and potent G2/M arrest at higher doses (C).

Figure 4

LRRK2-IN-1 downregulates DCLK1 expression and downstream target c-Myc. LRRK2-in-1 downregulates DCLK1 mRNA expression in a dose and time-dependent manner in AsPC-1 cells (A). LRRK2-IN-1 significantly downregulates DCLK1 protein expression at 24 and 48 h and downregulates protein expression of DCLK1 downstream target c-MYC (B-C). NCBI GEO data demonstrating inhibition of MAPK7/ERK5 with 10 μM U0126 resulting in downregulation of DCLK1 gene expression in SW480 cells (D). Treatment of AsPC-1 cells with 10 μM U0126 downregulates DCLK1 as well as shared downstream target c-MYC’s mRNA and protein levels. However, whereas combined treatment with this dose of U0126 (10 min pretreatment) and 5 μM LRRK2-IN-1 results in synergistic downregulation of c-MYC, DCLK1 is unaffected or slightly upregulated suggesting that U0126 and LRRK2-IN-1 may regulate DCLK1 mRNA and protein expression by the same mechanism of action (E-G).

Figure 5

LRRK2-IN-1 induces anti-oncogenic molecular changes and LRRK2-IN-1 induced cell death depends on DCLK1 kinase activity. AsPC-1 cells treated with LRRK2-IN-1 show decreased expression of stem (A), oncogenic (B), and EMT-related gene expression (C) (Vehicle = white; 0.5 μM = gray; 5.0 μM = striped bars. p < 0.05). Consistent with the changes seen in EMT-related gene expression, LRRK2-IN-1 significantly decreases invasion in AsPC-1 cells at 5 μM (D). DCLK1 isoform 1 is the most highly expressed referenced variant in colon cancer (E – top panel). Lentiviral overexpression of this isoform and its kinase dead form demonstrates approximately equal levels of DCLK1 transcript as determined by real-time RT-PCR of infected HCT116 cells (E – bottom panel). HCT116(vector), HCT116-DCLK1, and HCT116-DCLK1-KD cell lines were treated with LRRK2-IN-1 at various concentrations. After 48 h an MTT proliferation assay was performed. Compared to vector control cells, HCT116-DCLK1-KD cells demonstrated no change from 0.63 – 2.5 μM and decreased resistance at 5 and 10 μM, while HCT116-DCLK1 cells demonstrated significantly increased resistance to LRRK2-IN-1 cytotoxicity at 0.63 and 1.25 μM concentrations leading to a shift in relative cytotoxicity at low doses (F). AsPC-1(vector), AsPC-1-DCLK1, and AsPC-1-DCLK1-KD cells were treated with vehicle, 0.5, or 5 μM of LRRK2-IN-1 and allowed to form colonies for 9 days. Colonies were stained with crystal violet and plates were divided into grids and counted. AsPC-1-DCLK1 cells demonstrated significant resistance at 0.5 μM compared to vector control cells, while AsPC-1-DCLK1-KD cells demonstrated a significant decrease in resistance compared to vector control cells (G).

Figure 6

LRRK2-IN-1 is effective in vivo. AsPC-1 cells (0.5×10⁶) were injected subcutaneously into the flanks of athymic nude mice (n = 4) and allowed to grow until the tumor reached an average volume of 100 mm³. LRRK2-IN-1 was solubilized in 20% Captisol® and xenografts were injected peritumorally with either 20% Captisol (vehicle) or LRRK2-IN-1 (100 mg/kg) for a total of 12 injections. LRRK2-IN-1 treatment resulted in a significant decrease in tumor volume as a function of time (A) and reduced final excised tumor volume (p = 0.09; B-C) and tumor weight (p = 0.14; D). The + symbol denotes the mean.