Dual PAK4-NAMPT Inhibition Impacts Growth and Survival, and Increases Sensitivity to DNA-Damaging Agents in Waldenström Macroglobulinemia

Waldenström macroglobulinemia (WM) is characterized by lymphoplasmacytic cell infiltration of bone marrow (BM) and secretion of a serum monoclonal immunoglobulin M (IgM) protein. Recent genomic studies have contributed to a better understanding of the molecular mechanisms supporting WM pathogenesis and progression (1–3). These findings have been rapidly translated from laboratory to clinical setting, with the identification of novel therapeutic approaches and the initiation of several clinical trials for the treatment of patients with relapsed/refractory WM (4, 5). However, despite this substantial progress in treating patients with WM, only a few patients achieve complete remission whereas others develop drug resistance. Therefore, there is a need for identifying novel targets as well as for innovative therapeutic regimens to effectively treat patients with WM.

p21-activated kinase 4 (PAK4) is a member of PAK family of serine/threonine kinases with a role in cytoskeleton reorganization, cell proliferation and survival, and drug resistance (6–8). We have recently reported the oncogenic potential of PAK4 in multiple myeloma (MM) cells and provided evidence of the significant impact of the PAK4 inhibitor KPT-9274 on MM cell growth and viability, representing a potential novel therapeutic intervention in this malignancy (9). KPT-9274 is an orally administered small molecule, currently in clinical trial for the treatment of patients with advanced solid malignancies or non-Hodgkin's lymphoma (NCT02702492). It allosterically binds to, destabilizes and causes the degradation of PAK4. In addition to which, we have also reported that KPT-9274 depletes the synthesis of nicotinamide adenine dinucleotide (NAD) by blocking the activity of nicotinamide phosphoribosyl transferase (NAMPT), the rate-limiting enzyme in the NAD biosynthesis salvage pathway.

NAD is an essential metabolite required for sustaining energy production (TCA cycle) and regulating various cellular processes. Cancer cells are characterized by higher NAD⁺ turnover than normal cells due to the increased energy required for their cell proliferation and metabolism, as well as regulation of transcription, chromatin dynamics, and DNA repair-processes (10). As NAD⁺ is rapidly consumed and converted to nicotinamide, NAMPT plays a crucial role in replenishing the intracellular NAD⁺ pool. Aberrant activation of NAMPT has been reported in a number of solid and hematologic malignancies, including leukemia, MM, and importantly WM (11, 12).

Here, for the first time we describe a role for PAK4 in WM cell viability and report a significant impact of dual PAK4-NAMPT inhibition via KPT-9274 on WM cell growth and survival. At the molecular level, we identified the DNA damage and repair pathway to be significantly impacted by KPT-9274 leading to synergistic activity in combination with DNA-damaging agents in vitro and in vivo.

The WM cell lines (BCWM-1, MWCL-1, and RPCIWM-1) were cultured in RPMI containing 10% FBS (GIBCO, 10437028), 2 mmol/L l-glutamine, 100 U/mL penicillin, and 100 mg/mL streptomycin (GIBCO, 15140122). Peripheral blood mononuclear cells (PBMCs) were isolated using Ficoll-Hypaque density gradient sedimentation. Bone marrow stromal cells (BMSCs) were isolated from BM aspirate of patients with WM and were cultured as previously described (13). Primary WM cells were separated from BM samples by antibody-mediated positive selection using anti-CD19⁺ magnetic-activated cell separation microbeads (Miltenyi Biotech, 130-050-301). A purity of 95% CD19⁺ cells was obtained. KPT-9274 was provided by Karyopharm Therapeutics Inc.; melphalan and bendamustine were purchased from Selleckchem (Selleck Chemicals, S1212).

Cell viability was analyzed by CellTiter-Glo (CTG; Promega, G7572). Apoptosis was evaluated by flow cytometric analysis following FITC Annexin-V (BD Biosciences, 556419), PE–Annexin-V (Biolegend, 640947), DAPI (BD Biosciences, 564907), and PI (BD Biosciences, 51-6621E) staining, according to the manufacturer's instructions. Caspase activities were evaluated by specific Caspase-Glo assays (Promega).

NAMPT activity was analyzed by NAD/NADH Assay Kit (Colorimetric; Abcam, ab65348) according to the manufacturer's instructions.

Normal and tumor tissue specimen sections of formalin-fixed, paraffin-embedded BM biopsies were prepared and processed for immunohistochemistry to detect PAK4 and p-PAK4 (Ser474) protein expression by using specific antibodies (LifeSpan BioSciences, lsc-287254, and Santa Cruz Biotechnology, sc135774).

Western blotting (WB) was performed to evaluate the expression levels of total protein and phospho-specific isoforms using following antibodies: PAK4 (LifeSpan BioSciences, lsc-287254); FANCD2 (Santa Cruz Biotechnology, sc-20022), RAD51 (Santa Cruz Biotechnology, sc-398587), RAD51C (Santa Cruz Biotechnology, sc-56214), p-PAK4 (Ser 474; Santa Cruz Biotechnology, sc135774); NAMPT (Cell Signaling Technology, 86634), total and cleaved form of Caspase3 (Cell Signaling Technology, 9662), Caspase7 (Cell Signaling Technology, 9692), Caspase8 (Cell Signaling Technology, 9746), Caspase9 (Cell Signal, 9502s), PARP (Cell Signaling Technology, 9542s), Bcl-xl (Cell Signaling Technology, 2764s), BIM (Cell Signaling Technology, 2819), γ-H2AX (Ser139; Cell Signaling Technology, 9718s), p-CHK1 (Ser 345; Cell Signaling Technology, 2341), p-BAD (Ser112; Cell Signaling Technology, 5284s), p-P53(Ser15; Cell Signaling Technology, 9284s). Glyceraldehyde-3-phosphate dehydrgenase (GAPDH; Cell Signaling Technology, 2118), Histone3 (Cell Signaling Technology, 4499), Vinculin (Santa Cruz Biotechnology, sc25336), and α-Tubulin (Santa Cruz Biotechnology, sc8035) were used as loading control.

Human GIPZ PAK4 shRNA vectors were purchased from Dharmacon (Lafayette). Scramble: non-silencing Lentiviral shRNA control (Catalog #RHS4346), shRNA: 396# V3LHS_646396, 682# V2LHS_197682, 934# V3LHS_643934, 937# V3LHS_643937.

Human DNA Damage Signaling Pathway RT² Profiler PCR Array (PAHS-029Z, Qiagen) was used to evaluate the impact of KPT-9274 on mRNA levels of 84 related genes. Relative expression was calculated using the comparative δ,δ(C_t) method, using untreated cells as control. Fold-change [2⁁(−δ,δ C_t)] is the normalized gene expression [2⁁(−δ C_t)] in the treated cells divided the normalized gene expression [2⁁(−δ C_t)] in the control cells. Fold-change values greater than 1 indicate a positive or an upregulation; fold-change values less than 1 indicate a negative or downregulation.

In vitro HR assay was carried out using FRET-based DNA strand exchange assay (14). Briefly, the BCWM1 cells were treated with different combinations of drugs. After 24 hours, cells were harvested and lysed in chaps buffer. Cell lysate (20,000 cells) was incubated for 45 minutes with homologous ssDNA (Oligo 25; 48-mer; denoted as “Homologous DNA”) to form the nucleoprotein filament. Then, fluorescently labeled dsDNA (Oligo 25-FLU and 26-BHQ1) was added to the filament to initiate DNA strand exchange. After 30 minutes of incubation, fluorescence intensity was measured.

CB17-SCID mice were purchased from Charles River Laboratories. All animal studies were conducted according to protocols approved by the Animal Ethics Committee of the Dana-Farber Cancer Institute. Mice were irradiated (200 cGy) and then inoculated subcutaneously in the right flank with 10 × 10⁶ BCMW-1 cells in 100-μL RPMI-1640. Following detection of tumor (∼3 weeks after the injection), mice were treated intraperitoneally with: KPT-9274 (100 mg/kg body weight) daily for 5 consecutive d/wk for 3 weeks; or bendamustine (25 mg/kg) 1 single dose/wk for 2 consecutive weeks; or the combination with the same dosing regimen used for the individual agents. The control group received the vehicle at the same schedule as the combination group. Tumor volume was evaluated in 2 dimensions by caliper measurements performed approximately twice a week using the following formula: V = 0.5a × b², where “a” and “b” are the long and short diameter of the tumor, respectively.

Mice tumor sections were subjected to immunohistochemical (IHC) staining for hematoxylin and eosin (H&E) and caspase-3 activation to detect apoptotic cell death.

All values are expressed as mean plus or minus the standard deviation (SD). The statistical significance of differences between treatments was analyzed using the Student t test using GraphPad Prism analysis software; differences were considered significant when P was less than or equal to 0.05. Drug interactions were assessed by CalcuSyn 2.0 software (Biosoft), which is based on the Chou–Talalay method. Combination index (CI) = 1, indicates additive effect; CI < 1 indicates synergism; CI > 1 indicates antagonism.

We have analyzed and observed high expression of both PAK4 and phosphorylated (p)-PAK4 Ser474 in primary cells from patients with WM and WM cell lines, with significantly lower expression in normal B lymphocytes (CD19⁺ cells) and CD19⁻ cells isolated from four normal donors (Fig. 1A and B and Supplementary Fig. S1A). Undetectable levels of PAK4 were also previously observed in BM specimens from healthy donors (9). Moreover, expression of PAK4 has been confirmed by RNA-seq in a large cohort of patients with WM (Supplementary Fig. S1B). Genetic depletion of PAK4 via stable lentiviral knockdown decreased tumor cell survival proportional to the reduction in PAK4 levels in the BCWM1 cell line (Fig. 1C); this effect was accompanied by induction of apoptosis (Fig. 1D).

We next evaluated the effect of pharmacological inhibition of PAK4 using KPT-9274, a dual PAK4-NAMPT inhibitor. We observed a significant dose- and time-dependent decrease of WM cell viability (Fig. 2A); and suppression of WM-BMSC interaction-mediated growth of WM cells (Supplementary Fig. S2A) after treatment with KPT-9274, whereas no effect was observed on BMSCs (data not shown). Moreover, PAK4-related biological processes such as migration, invasion and adhesion to BMSCs were negatively affected by the inhibitor (Supplementary Fig. S2B and data not shown). The inhibition of cell viability observed after treatment with KPT-9274 was associated with a dose- and time-dependent induction of apoptosis (Fig. 2B) via activation of effector caspases, mainly through the intrinsic apoptotic pathway and PARP cleavage (Fig. 2C and D). Moreover, increased BIM levels and BCL-2 antagonist of cell death (BAD) activation, via inhibition of serine 112 phosphorylation, as well as decreased Bcl-xL levels were observed in BCWM1 cells treated with the inhibitor compared with control cells (Fig. 2D).

KPT-9274 allosterically binds to and destabilizes PAK4 causing its degradation. In addition, KPT-9274 also inhibits the activity of nicotinamide phosphorybosyltransferase (NAMPT), the rate limiting enzyme for NAD⁺ production from nicotinamide in mammalian cells. We confirmed that treatment with KPT-9274 significantly decreases PAK4 expression in WM cells, and while not affecting NAMPT protein levels, has a profound effect on cellular NAMPT activity (Fig. 2E and F). Moreover, NAMPT expression was not affected by PAK4 knockdown in BCWM1 cells (Supplementary Fig. S1C); however, we observed a slight (but not significant) reduction of NAMPT activity in PAK4 depleted cells as compared with control cells (data not shown).

The effect of KPT-9274 treatment on NAMPT inhibition can be rescued by repletion of NAD⁺ through biosynthesis from nicotinamide (NAM); consistently, we found that exogenous NAM rescued KPT-9274-induced WM cell death in a dose-dependent manner (Fig. 2G), further confirming the importance of NAD⁺ depletion in the antitumor effect of KPT-9274 on WM cells.

Aside NAMPT, alternative NAD⁺ sources available to tumors have been reported, including nicotinate phosphorybosyltransferase (NAPRT), which mediates NAD⁺ production from nicotinic acid (NA). We have evaluated the expression of NAPRT1 in BCWM1 cells and observed no significant impact after treatment with KPT-9274 (Supplementary Fig. S2C); importantly, KPT-9274–mediated inhibition of WM cell growth was largely abolished by supplementation with NA (Supplementary Fig. S2D).

Primary WM cells were obtained from BM aspirates of patients with WM following CD19-positive selection. Treatment with KPT-9274 significantly inhibited cell viability and triggered apoptosis in primary WM cells from both newly diagnosed as well as relapsed patients, in a time- and dose-dependent manner (Fig. 3A and B and Supplementary Fig. S2E). Sensitivity to KPT-9274 was also observed in primary cells resistant to the BTK inhibitor Ibrutinib (Fig. 3C), and in the context of the patient-derived BM microenvironment (data not shown). Finally, we investigated the anti-WM effect of KPT-9274 in vivo in a murine xenograft model of human WM using BCWM1 in SCID mice. Following detection of tumors, mice were treated with either 100 mg/kg KPT-9274 or vehicle orally 5 d/wk for 3 weeks. As shown in Fig. 3D, treatment with KPT-9274, compared with vehicle alone, significantly inhibited WM cell tumor growth. No related toxicity was observed in mice, as determined by daily evaluation of activity and overall body weight change during the course of treatment (data not shown). Histologic examination of tumors retrieved from BCWM1-bearing mice confirmed significant tumor cell apoptosis (caspase-3 staining; Fig. 3E). Moreover, WB analysis of protein lysates from retrieved tumor cells confirmed significant induction of caspase-7 as well as inhibition of p-BAD and decreased PAK4 expression (Fig. 3F).

Our reanalysis of transcriptomic data in MM cells following KPT-9274 treatment (Gene Expression Omnibus accession number GSE93745) revealed a significant enrichment of DNA damage response and repair (DDR) genes in KPT-9274–treated cells compared with untreated cells (Supplementary Fig. S3A). We have therefore used a targeted human PCR array to elucidate the impact of KPT-9274 on the transcriptional regulation of DNA damage and repair genes in WM cells. We observed repression of DNA damage and repair genes in WM cell lines treated with KPT-9274 compared with control cells (Fig. 4A). Specifically, we observed significant inhibition of Fanconi Anemia (FA)/BRCA pathway-related genes (FANCD2 and BRCA1 among others) as well as DDR genes such as RAD51. These observations were also confirmed by qPCR (Supplementary Fig. S3B) and at the post-transcriptional level by WB analysis, as observed for RAD51 and other components of the homologous recombination (HR) repair machinery (Fig. 4B). This effect was accompanied by inefficient HR-mediated repair activity (Fig. 4C), along with induction of DNA damage (assessed by γH2AX) and p53 activation in both KPT-9274-treated (Fig. 4D). These observations correlate with the described activity of NAD on DNA damage and repair (10, 15, 16). Consistent with these findings, we also observed that nicotinamide and nicotinic acid supplementation partly rescued the effect of KPT-9274 on DNA damage and repair (Fig. 4C–E and Supplementary Fig. S3C–S3D).

Fanconi anemia and DDR pathways have been implicated in resistance to DNA-alkylating agents (17–20). Because bendamustine is currently one of the main drugs used in patients with WM (21–23), we have explored the potential of KPT-9274 to impact sensitivity of WM cells to bendamustine. WM cells were therefore simultaneously treated with different concentration of KPT-9274 and bendamustine. The combination treatment resulted in significant time- (data not shown) and dose-dependent inhibition of cell viability (Fig. 5A); isobologram and combination index analysis revealed strong synergism of the combination as compare to single agents, with a combination index (CI)<1.0 at all tested doses (Fig. 5A and Supplementary Fig. S4A). The synergistic inhibition of WM viability was accompanied by significant induction of apoptotic cell death (Fig. 5B) and activation of caspase-3/7 (Fig. 5C) in combined versus single-agent therapy; and the effect was partially rescued by NA supplementation (Supplementary Fig. S4B). Importantly, neither single agent nor the combination triggered death of healthy donor PBMCs, suggesting a favorable therapeutic index (Supplementary Fig. S4C). Interestingly, in PAK4-depleted WM cells, treatment with bendamustine significantly increased the rate of early and late apoptosis compared with scrambled control cells (Supplementary Fig. S4D). Importantly, a potent synergistic anti-WM activity of KPT-9274 in combination with bendamustine was observed in primary WM patient cells, with a CI <1.0 with all tested doses (Fig. 5D). Finally, DNA damage markers were significantly affected by the combination regimen when compared with either drug alone (Fig. 5E); the effect was partially rescued by NA supplementation (Supplementary Fig. S4E). A similar synergism was observed in WM cells treated with KPT-9274 in combination with the DNA-damaging agent melphalan (Supplementary Fig. S5).

Having shown that combined KPT-9274 plus bendamustine induced synergistic apoptosis in WM cells in vitro, we next examined the in vivo efficacy of the combination in a human xenograft mouse model using BCMW.1 cells injected subcutaneously in SCID mice. After tumor development, mice were treated with vehicle, KPT-9274 (100 mg/kg, daily for 5 days a week, administered orally), bendamustine (25 mg/kg, 1 injection intraperitoneally/week for 2 weeks) as single agents or in combination. As seen in Fig. 6A and B, the combination of bendamustine with KPT-9274 induced a significant reduction in tumor growth compared with mice receiving vehicle or either single agents. The combination did not have increased toxicity (data not shown). Furthermore, WB analysis of cell lysates from retrieved tumors confirmed the effect of the combination on the DDR pathway (Fig. 6C).

We have recently shown a significant biological and functional role of PAK4, and the therapeutic value of a dual PAK4-NAMPT inhibitor (KPT-9274) in MM. On the basis of the promising findings in MM and other malignancies, including PDAC, Renal Cell Carcinoma (RCC), triple-negative breast cancer, and B-cell acute lymphoblastic leukemia (B-ALL; refs. 9, 24–28), we here characterize and report a significant antitumor effect of KPT-9274 in WM, including ibrutinib-resistant primary patients WM cells. In this study, we also confirmed the dual inhibition properties of KPT-9274, via its impact on both PAK4 protein expression and cellular NAD+ levels in WM cells. Co-inhibition of these targets via KPT-9274 led to synergistic anti-tumor effects through energy depletion, inhibition of proliferation, and ultimately apoptosis in WM cells. Rescue experiments with nicotinamide and nicotinic acid confirmed the importance of NAD⁺ depletion for the growth inhibitory effect exerted by KPT-9274 on WM cells

Recent studies have implicated activity of PAK family members as well as NAD⁺ in DNA damage and repair (16, 29–31). We have observed a significant induction of DNA damage along with inhibition of DNA repair in WM cells after treatment with KPT-9274. Interestingly, induction of DNA damage was also observed in PAK4 knockdown cells (data not shown). The intrinsic apoptotic pathway is a major mechanism of cell death in response to DNA damage, and the p53 tumor suppressor is the primary regulator of apoptosis in response to damage (32, 33). We have observed a significant increase in the phosphorylation of Chk1, leading to the activation of p53 in cells treated with KPT-9274. At the same time, KPT-9274 significantly affected expression of genes involved in the FA/BRCA as well as HR-mediated repair sustaining the induction of DNA damage. In line with the hypothesis that defective NAD⁺ synthesis is responsible for the increased DNA damage observed upon KPT-9274 treatment in WM cells, we found that exogenously added NAD⁺ reduced DNA damage in WM-treated cells, and partly rescued the inhibition of HR.

FA/BRCA and DDR pathways have been shown to affect sensitivity to alkylating agents (17–20), and our mechanistic studies indeed demonstrated that dual PAK4 and NAMPT inhibition by KPT-9274 in WM cells sensitizes them to the activity of alkylating agents, such as melphalan or bendamustine. The combination treatment resulted indeed in a synergistic time- and dose-dependent inhibition of cell viability in cell lines and primary WM cells, while sparing the viability of healthy PBMC. We also observed a synergistic induction of apoptosis and caspase activation following treatment with combined versus single-agent therapies with greater induction of DNA damage. The anti-WM activity of this combination was finally confirmed in vivo in mice xenografted with BCWM1. We believe these results are highly significant, considering that bendamustine is an approved agent for treatment of NHL and WM (34, 21–23). Importantly, we show here for the first time the impact of targeting DDR and HR pathways in WM.

In summary, we report significant antiproliferative activity of dual PAK4-NAMPT inhibition in WM cells, with induction of apoptosis, DNA-damage response and FA/BRCA pathway disruption and a strong synergistic activity in combination with DNA damaging agents such as bendamustine; suggesting KPT-9274 as a novel therapeutic strategy in WM as monotherapy or in combination with alkylating agents.

S. Oliva is a consultant/advisory board member for Celgene, Jansenn, Amgen, Adaptive Biotechnologies. W. Senapedis holds ownership interest (including patents) in Karyopharm Therapeutics. E. Baloglu is an employee of and holds ownership interest (including patents) in Keryopharm Therapeutics. No potential conflicts of interest were disclosed by the other authors.

Conception and design: N.C. Munshi, M. Fulciniti

Development of methodology: N. Li, M.A. Lopez, M. Linares, S. Oliva, M.A. Shammas

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): J. Martinez-Lopez, L. Xu, Y. Xu

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): N. Li, M. Linares, S. Kumar, J. Martinez-Lopez, T. Perini, Z. Hunter, S.P. Treon, N.C. Munshi, M. Fulciniti

Writing, review, and/or revision of the manuscript: N. Li, W. Senapedis, E. Baloglu, K.C. Anderson, S.P. Treon, N.C. Munshi, M. Fulciniti

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): N. Li, M.A. Lopez, K.C. Anderson, M. Fulciniti

Study supervision: N.C. Munshi, M. Fulciniti

Other (drug discovery and target identification of dual inhibitor): E. Baloglu, W. Senapedis

Other (assisted with DNA repair related assays): S. Kumar, M.A. Shammas

This work was supported by NIH grants PO1-155258 (to M. Fulciniti and N.C. Munshi) and P50-100707 (to N.C. Munshi); Department of Veterans Affairs Merit Review Award 1 I01BX001584-01 (to N.C. Munshi).

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.