DYRK1A suppression restrains Mcl-1 expression and sensitizes NSCLC cells to Bcl-2 inhibitors

Objective: Mcl-1 overexpression confers acquired resistance to Bcl-2 inhibitors in non-small cell lung cancer (NSCLC), but no direct Mcl-1 inhibitor is currently available for clinical use. Thus, novel therapeutic strategies are urgently needed to target Mcl-1 and sensitize the anti-NSCLC activity of Bcl-2 inhibitors. Methods: Cell proliferation was measured using sulforhodamine B and colony formation assays, and apoptosis was detected with Annexin V-FITC staining. Gene expression was manipulated using siRNAs and plasmids. Real-time PCR and Western blot were used to measure mRNA and protein levels. Immunoprecipitation and immunofluorescence were used to analyze co-localization of dual specificity tyrosine-phosphorylation-regulated kinase 1A (DYRK1A) and Mcl-1. Results: Suppression of DYRK1A resulted in reduced Mcl-1 expression in NSCLC cells, whereas overexpression of DYRK1A significantly increased Mcl-1 expression. Suppression of DYRK1A did not alter Mcl-1 mRNA levels, but did result in an accelerated degradation of Mcl-1 protein in NSCLC cells. Furthermore, DYRK1A mediated proteasome-dependent degradation of Mcl-1 in NSCLC cells, and DYRK1A co-localized with Mcl-1 in NSCLC cells and was co-expressed with Mcl-1 in tumor samples from lung cancer patients, suggesting that Mcl-1 may be a novel DYRK1A substrate. We showed that combined therapy with harmine and Bcl-2 antagonists significantly inhibited cell proliferation and induced apoptosis in NSCLC cell lines as well as primary NSCLC cells. Conclusions: Mcl-1 is a novel DYRK1A substrate, and the role of DYRK1A in promoting Mcl-1 stability makes it an attractive target for decreasing Bcl-2 inhibitor resistance.


Introduction
The B-cell lymphoma-2 (Bcl-2) family of proteins are characterized by four homology domains, BH1-BH4, and are subdivided into two major groups. The first of these is the anti-apoptotic subgroup, which includes Bcl-2 and Mcl-1. The second is the proapoptotic subgroup, which includes Bax-like molecules (such as Bax and Bak) that contain the BH1-3 domains, as well as BH3-only proteins (such as Bid and Bad) 1 . When BH3-only proteins are activated, their amphipathic α-helical BH3 domain inserts into a hydrophobic groove on Bcl-2 anti-apoptotic proteins, subsequently initiating apoptosis 2 . Small molecule Bcl-2 inhibitors are promising for the treatment of hematological malignancies in addition to other cancers 3 . ABT-737 (Navitoclax), the first BH3 mimetic, has shown promising effects in clinical trials for patients with relapsed chronic lymphocytic leukemia (CLL) 4 . Consequently,  (Venetoclax) became the first BH3 mimetic approved for cancer treatment by the Food and Drug Administration (FDA) on April 11, 2016 5 . ABT-199 is an oral Bcl-2 specific inhibitor and is approved to treat CLL patients with the 17p deletion who have received at least one prior therapy 6,7 . Because Bcl-2 inhibitors bind to Mcl-1 with low affinity, Mcl-1 overexpression confers acquired resistance to Bcl-2 inhibitors in multiple cancer types, including non-small cell lung cancer (NSCLC) and acute myeloid leukemia 8,9 . NSCLC is the most common malignancy worldwide and accounts for ~85% of all lung cancers, with small cell lung cancer comprising ∼15% 10 . In addition, Mcl-1 is a novel biomarker of poor prognoses for NSCLC patients 11 . However, there is no direct Mcl-1 inhibitor currently available in clinical use. Thus, novel therapeutic strategies are urgently needed to target Mcl-1 and sensitize the anti-cancer activities of Bcl-2 inhibitors for effective treatment of NSCLC.
DYRK1A is located on the critical region of Down syndrome (DS) chromosome 21, and its overexpression in DS patients contributes to cognitive impairments 12 . In addition, DYRK1A is thought to be involved in neurodegenerative diseases including Alzheimer's disease, Parkinson's disease, and Huntington's disease 13 . DYRK1A is also overexpressed in multiple human malignancies, including hematological and brain cancers, where it contributes to tumor growth by manipulating cell cycle progression [14][15][16] . Thus, its role in multiple human diseases makes DYRK1A an attractive target for therapeutic drugs 17 . Harmine, a β-carboline alkaloid, is a selective ATP-competitive inhibitor of DYRK1A, which acts by binding to the ATP-binding pocket of DYRK1A 18 . Inhibition of DYRK1A with harmine reverses resistance to multiple anti-cancer drugs including mitoxantrone, camptothecin, and AZD9291 19,20 .
DYRK1A is a serine/threonine kinase that regulates diverse pathways involved in cell differentiation and apoptosis, gene transcription and splicing, and the cell cycle 21 . DYRK1A phosphorylates NFATc1, p27, EGFR, and c-myc at serine or threonine residues 22 . Here, we showed that Mcl-1 was a novel DYRK1A substrate, and DYRK1A suppression sensitized NSCLC cells to Bcl-2 inhibitors.

The sulforhodamine B (SRB) assay
Cell viability was measured using the SRB assay. Cells (5 × 10 3 ) were seeded on 96-well plates and treated with compounds for 72 h. The cells were fixed with 10% (w/v) trichloroacetic acid overnight, washed 5 times with deionized water, and dried at 60 °C. After the plates were entirely dry, cells were stained using 100 μL 0.4% SRB solution for 30 min, washed using 1% (v/v) acetic acid, and dried at 60 °C. Protein-bound dye was dissolved with a Tris-based solution and cell viability was measured at 540 nm using a multi-scan spectrum.

Colony formation assay
NSCLC cells were cultured on 6-well plates (1,000 cells per well) overnight, incubated with the compounds for 14 days, fixed with 4% paraformaldehyde for 15 min, and stained using Giemsa solution for another 15 min. Colonies were photographed using ChemiDoc XPS (Bio-Rad, Hercules, CA, USA).

Apoptosis assay
Cells were collected, washed with phosphate-buffered saline (PBS), stained using the Annexin V-FITC Apoptosis Kit (BD Biosciences, Franklin Lakes, NJ, USA), and detected using flow cytometry. Briefly, 1 × 10 5 cells were harvested and resuspend in 200 μL of Annexin V binding buffer containing 5 μL of Annexin V-FITC for 15 min at room temp in the dark. And 5 μL of 0.5 mg/mL propidium iodide solution was added before flow cytometer detection.

Real-time reverse transcription PCR
Total cellular RNA was extracted with TRIzol (Invitrogen, Carlsbad, CA, USA), precipitated with isopropyl alcohol, and rinsed with 70% ethanol. Single stranded cDNA was generated using oligo (dT) priming, followed by SYBR Green-based real-time reverse transcription PCR (Bio-Rad Laboratories, Hercules, CA, USA). After the primers were received, they were centrifuged and then diluted to 10 μM as stock solutions. SYBR Green qPCR Master Mix Kit was used to establish a 10 μL reaction system [cDNA: 1.0 μL, primer-forward: 1.0 μL, primer-reverse: 1.0 μL, ddH20: 1.

Plasmid transfection
NSCLC cells were cultured on 6-well plates overnight, and transfection was performed the next day using jetPRIME (Polyplus, Illkirck, France) according to the manufacturer's instructions. Briefly, 1 μg plasmid was diluted into 100 μL jet-PRIME ® buffer. Then, 2 μL jetPRIME ® was added, vortexed, and incubated for 10 min at room temperature. Transfection mix (100 μL/well) was added to the cells.

Western blot analysis
Cells were lysed using RIPA Buffer (Thermo Fisher Scientific, Waltham, MA, USA). Total protein (20 μg) was subjected to 10% SDS-PAGE and transferred to a polyvinylidene difluoride membrane. The membranes were blocked and then incubated with primary and secondary antibodies. Protein expression was analyzed using an enhanced chemiluminescence (ECL, Abbkine, Wuhan, China) system.

Immunoprecipitation
Cell lysates were centrifuged at 10,000 × g for 30 min at 4 °C, incubated with primary antibody using a slow rotation for 4 h, and then incubated with protein G magnetic beads for 1 h at 4 °C. The beads were washed a minimum of five times, mixed with loading buffer, heated to 95 °C for 5 min, followed by Western blot analysis.

Immunofluorescence
Cells were seeded into 96-well plates and cultured overnight. The next day, the cells were fixed with 4% paraformaldehyde for 30 min at room temperature, washed with PBS, and permeabilized with 200 μL of 0.3% Triton X-100 in PBS for 30 min. Cells were washed again with PBS, blocked with 1% bovine serum albumin in PBS for 30 min, incubated with primary antibodies at 4 °C overnight, washed three times with PBS, and incubated with fluorescent dye-conjugated secondary antibodies for 1 h in the dark. Nuclei were stained with 4′,6-diamidino-2-phenylindole for 5 min in the dark. Cells were visualized and photographed using a fluorescence microscope.

Statistical analysis
The results are expressed as the mean ± SD. The data presented were obtained at least three times. The synergistic effects of harmine plus ABT-199/ABT-737 were quantitatively determined by calculating the combination index (CI) values using Calcusyn software. A CI value < 0.9 indicated synergism; 0.9-1.1, additive; > 1.1, antagonism. A two-tailed Student's t-test was used to compare the difference between two groups. A value of P < 0.05 was accepted as statistically significant, and the levels of significance were indicated as follows: * P < 0.05, ** P < 0.01, and *** P < 0.001.

DYRK1A upregulates Mcl-1 expression in NSCLC cells
DYRK1A expression was knocked down in NSCLC cell lines using siRNA, and DYRK1A knockdown resulted in decreased Mcl-1 expression in NSCLC cells ( Figure 1A). In contrast, overexpression of DYRK1A in NSCLC cells significantly increased Mcl-1expression ( Figure 1B). Expression of other Bcl-2 family members, such as Bcl-2 and Bcl-xL, was not altered with DYRK1A knockdown or overexpression in NSCLC cells (Figure 1A and 1B). Treatment of NSCLC cells with harmine, a DYRK1A inhibitor, resulted in a dose-and time-dependent inhibition of Mcl-1 expression (Figure 1C and 1D). However, Bcl-2 and Bcl-xl expression was not changed when DYRK1A was inhibited with harmine in NSCLC cells (Figure 1C and 1D). These data suggested that DYRK1A upregulated Mcl-1 expression in NSCLC cells.

DYRK1A promotes the stability of Mcl-1 in NSCLC cells
To investigate whether DYRK1A regulates Mcl-1 protein expression at the transcriptional level, Mcl-1 mRNA expression was measured in NSCLC cells after treatment with siDYRK1A or harmine. After treatment with siRNA, depletion of DYRK1A did not inhibit Mcl-1 mRNA expression ( Figure 2D). Furthermore, treatment of cells with the proteasome inhibitor, MG132, prevented the degradation of Mcl-1 that was induced by DYRK1A depletion or inhibition (Figure 2E and 2F). These data demonstrated that DYRK1A promoted stability of Mcl-1 in NSCLC cells.

DYRK1A directly interacts with Mcl-1 in NSCLC cells and is co-expressed with Mcl-1 in tumor samples from lung cancer patients
DYRK1A regulates phosphorylation of many substrates and is involved in controlling a variety of molecular processes 17 . Phosphorylation of Mcl-1 at Thr-163 increases protein stability and is responsible for chemoresistance in NSCLC cells 23 . We found that treatment of cells with harmine significantly reduced Mcl-1 phosphorylation (Thr-163; Figure 1D). Thus, we hypothesized that DYRK1A interacted with Mcl-1 to regulate its phosphorylation. Indeed, our data suggested that DYRK1A directly interacted with Mcl-1 in A549 cells, and co-localization of DYRK1A with Mcl-1 was observed in A549 cells using a dual immunofluorescent assay (Figure 3A and 3B). Next, co-expression of DYRK1A and Mcl-1 was analyzed in NSCLC patient samples. We found that DYRK1A was co-expressed with Mcl-1 in tumor samples from lung cancer patients ( Figure 3C) 24 . Mcl-1 overexpression confers acquired resistance to Bcl-2 inhibitors, and therefore we hypothesized that DYRK1A overexpression would also hinder the anti-cancer effects of Bcl-2 inhibitors. Overexpression of DYRK1A by transfecting DYRK1A plasmid decreased the anti-proliferative effects of Bcl-2 inhibitors in both A549 and NCI-H1299 cells ( Figure 3E). In contrast, suppression of DYRK1A by transfecting with DYRK1A siRNA enhanced the anti-proliferative effects of Bcl-2 inhibitors in NSCLC cells ( Figure 3F). These data suggested that overexpression of DYRK1A and Mcl-1 promoted acquired resistance to Bcl-2 inhibitors in NSCLC cells.

DYRK1A inhibition sensitizes NSCLC cells to Bcl-2 inhibitors
Next, we hypothesized that DYRK1A inhibition would sensitize NSCLC cells to Bcl-2 inhibitors via suppression of Mcl-1. As expected, we observed synergistic cytotoxic effects when the DYRK1A inhibitor, harmine, was combined with either ABT-737 or ABT-199 in NSCLC cells, with mean CI NCI-H460 values below 0.9 ( Figure 4A). In contrast, treatment of cells with a combination of harmine and Bcl-2 antagonists significantly suppressed NCI-H460 cell colony formation compared to treatment with harmine alone or Bcl-2 antagonists alone ( Figure 4B and 4C, combination therapy vs. monotherapy, P < 0.001). Thus, combined treatment with harmine and Bcl-2 antagonists dramatically reduced NSCLC cell proliferation.
To determine whether harmine could enhance Bcl-2 antagonist-mediated apoptosis, we compared the effect of the Bcl-2 antagonist treatment with or without harmine on NSCLC cells. Harmine and Bcl-2 antagonists each elicited apoptosis as single agents, but a more profound effect was observed using combined treatment in NSCLC cells (Figure 5A and 5B, combination therapy vs. monotherapy, P < 0.05). Furthermore, PARP and caspase-3 showed greater activation with combined treatment compared to single treatment ( Figure 5C). Treatment of NSCLC cells with Bcl-2 antagonist alone resulted in increased expression of Mcl-1, whereas Mcl-1 expression was markedly reduced following combined treatment with harmine and either ABT-737 or ABT-199. These data demonstrated that harmine enhanced Bcl-2 antagonist-mediated apoptosis in NSCLC cells by reducing Mcl-1 expression.

DYRK1A inhibition sensitizes primary NSCLC cells to Bcl-2 inhibitors by reducing Mcl-1 expression
Finally, to evaluate the clinical potential of combined harmine and Bcl-2 antagonist therapy, we established four primary NSCLC cancer cell lines from lung cancer patients. Combined treatment of primary NSCLC cell lines with harmine and Bcl-2 antagonist showed synergistic, antiproliferative effects with the CI values below 0.9 ( Figure 6A). Furthermore, combined treatment with harmine and Bcl-2 antagonist significantly suppressed primary NSCLC cell colony formation when compared to treatment with single agents (Figure 6B, combination therapy vs. monotherapy, P < 0.05). Although suppression of DYRK1A with siRNA had no effect on the level of Mcl-1 mRNA (Figure 6C), suppression of DYRK1A expression with siRNA or harmine reduced Mcl-1 protein expression in primary NSCLC cells (Figure 6D and 6E). In addition, harmine treatment enhanced Bcl-2 inhibitor-induced apoptosis in primary NSCLC cells by suppressing Mcl-1 expression (Figure 6F  and 6G). Together, these data demonstrated that DYRK1A suppression enhanced the anti-NSCLC activity of Bcl-2 antagonists by reducing Mcl-1 expression, both in NSCLC cell lines and primary NSCLC cells.

Discussion
Mcl-1 is an anti-apoptotic protein that plays a critical role in the development of resistance to anti-cancer drugs, including epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs) and Bcl-2 inhibitors 25 . Although direct Mcl-1 inhibitors are currently in preclinical and clinical development, none has been approved by the Food and Drug Administration. (C) NSCLC cells were incubated with harmine and/or Bcl-2 inhibitor at the indicated concentrations for 48 h, and then Western blot was used to detect the expression of the indicated proteins. Significant results are presented as * P < 0.05, ** P < 0.01, or *** P < 0.001.  [30][31][32] . We demonstrated that DYRK1A suppression by siRNA or harmine decreased the phosphorylation of Mcl-1 at Ser159/Thr163. DYRK1A suppression also accelerated Mcl-1 degradation, indicating that DYRK1A mediated proteasome-dependent degradation of Mcl-1. Furthermore, DYRK1A likely directly bound and phosphorylated Mcl-1, thereby enhancing Mcl-1 protein stability and contributing to the development of acquired resistance to Bcl-2 inhibitors. In this regard, our findings suggested that targeting of DYRK1A may serve as a novel approach to reduce Mcl-1 protein stability in NSCLC cells and thereby prevent Bcl-2 inhibitor resistance. However, further studies are required to identify the binding site between DYRK1A and Mcl-1.
NSCLC patients harboring EGFR-activating mutations derive greater benefit from EGFR-TKI therapy than from chemotherapy, either in first-line or subsequent lines of treatment 33 . However, the overall EGFR mutation rate is only 16.7% in NSCLC patients and most NSCLC patients harboring EGFR wild-type have poor responses to EGFR-TKI therapy. Thus, platinum-based chemotherapy is still the standard therapeutic regimen for these patients 34 . For these reasons, there is an urgent need to develop novel targeted therapeutic strategies for NSCLC patients with wild-type EGFR. Our previous work has demonstrated that DYRK1A positively regulates EGFR/Met via regulating the expression and activation of STAT3, and that suppression of DYRK1A enhances the anti-cancer activity of EGFR-TKIs in wild-type EGFR NSCLC cells via inhibiting the STAT3 signaling pathway 20 . Furthermore, inhibition of DYRK1A suppresses angiogenesis and tumor growth via the p53 signaling pathway in EGFR wild-type NSCLC cells 35 . These findings suggested that targeting of DYRK1A could be a novel, effective strategy for treatment of NSCLC patients with EGFR wild-type. In the present study, we demonstrated that combined treatment with a DYRK1A inhibitor and a Bcl-2 inhibitor produced a synergistic, anti-proliferative effect in EGFR wild-type NSCLC cell lines (NCI-H1299, A549, and NCI-H460) as well as in primary NSCLC cells (derived from three EGFR wild-type NSCLC patients and one patient with an EGFR exon 20 insertion mutation resistant to EGFR-TKI treatment; Table 1) 36 . Thus, this combined treatment regimen provides an attractive therapeutic strategy for the treatment of wild-type EGFR NSCLC patients.

Conclusions
In summary, we showed that either genetic or pharmacological inhibition of DYRK1A selectively perturbed the stability of Mcl-1. Furthermore, inhibition of DYRK1A enhanced the anti-NSCLC activity of Bcl-2 inhibitors by inhibiting Mcl-1 expression. Our data identified Mcl-1 as a novel substrate of DYRK1A and highlighted DYRK1A as a potential therapeutic target for conquering Bcl-2 inhibitor resistance in NSCLC (Figure 7).