Rational Targeting of Cdc42 Overcomes Drug Resistance of Multiple Myeloma

Multiple myeloma (MM) drug resistance highlights a need for alternative therapeutic strategies. In this study, we show that CASIN, a selective inhibitor of cell division cycle 42 (Cdc42) GTPase, inhibited proliferation and survival of melphalan/bortezomib-resistant MM cells more profoundly than that of the sensitive cells. Furthermore, CASIN was more potent than melphalan/bortezomib in inhibiting melphalan/bortezomib-resistant cells. In addition, CASIN sensitized melphalan/bortezomib-resistant cells to this drug combination. Mechanistically, Cdc42 activity was higher in melphalan/bortezomib-resistant cells than that in the sensitive cells. CASIN inhibited mono-ubiquitination of Fanconi anemia (FA) complementation group D2 (FANCD2) of the FA DNA damage repair pathway in melphalan-resistant but not melphalan-sensitive cells, thereby sensitizing melphalan-resistant cells to DNA damage. CASIN suppressed epidermal growth factor receptor (EGFR), signal transducer and activator of transcription 3 (STAT3), and extracellular signal-regulated kinase (ERK) activities to a larger extent in bortezomib-resistant than in melphalan-sensitive cells. Reconstitution of ERK activity partially protected CASIN-treated bortezomib-resistant cells from death, suggesting that CASIN-induced killing is attributable to suppression of ERK. Importantly, CASIN extended the lifespan of mouse xenografts of bortezomib-resistant cells and caused apoptosis of myeloma cells from bortezomib-resistant MM patients. Finally, CASIN had negligible side effects on peripheral blood mononuclear cells (PBMC) from healthy human subjects and normal B cells. Our data provide a proof of concept demonstration that rational targeting of Cdc42 represents a promising approach to overcome MM drug resistance.


INTRODUCTION
Multiple myeloma (MM) is an incurable blood cancer of plasma cells and is the second most prevalent hematological malignancy (1,2). Melphalan, a DNA cross-linking agent, was the most widely used drug to treat MM patients in the 1960s. However, despite initial responses, most patients develop acquired resistance to melphalan (3). Bortezomib, the first proteasome inhibitor, was introduced into clinical practice for the treatment of relapsed MM (4). Although bortezomib shows an impressive anti-myeloma activity with an overall response rate of 43% (5), patient resistance to this drug has developed (6)(7)(8). A second generation of proteasome inhibitors as well as proteasome inhibitors in combination with immunomodulatory drugs (IMIDs), monoclonal antibodies (MoAbs), or histone deacetylase (HDAC) inhibitors have proven effective in overcoming bortezomib resistance (9). Nonetheless, novel therapeutics are desired to further conquer primary or acquired drug resistance of MM.
Numerous signaling pathways are involved in drug resistance of MM. For instance, increased activity of extracellular signal-regulated kinase 1 and 2 (ERK1/2) of the mitogenactivated protein kinase (MAPK) pathway plays a critical role in thioredoxin-mediated bortezomib resistance (10). MAPK kinase (MEK)/ERK and Janus kinase (JAK)/signal transducer and activator transcription 3 (STAT3) pathways have been reported to mediate CDC28 protein kinase regulatory subunit 1B (CKS1B)-induced drug resistance in aggressive CKS1Boverexpressing MM (11). Epidermal growth factor receptor (EGFR)/JAK1/STAT3 signaling is associated with sensitivity of MM cells to proteasome inhibitors (12). In addition, enhanced expression of the Fanconi Anemia (FA) DNA damage repair pathway and elevated mono-ubiquitination of the associated FA complementation group D2 (FANCD2) are critical mechanisms of melphalan resistance (13,14). Despite these advances in knowledge, the mechanisms underlying MM drug resistance have not been fully elucidated.
The Rho family of small GTPases are a class of molecular switches that cycle between GTP-bound active and GDPbound inactive forms (15)(16)(17). Our previous study showed that B cell-specific deletion of cell division cycle 42 (Cdc42) of Rho GTPases reduces plasma cells (18). Therefore, we question whether Cdc42 could be targeted to benefit patients with MM. In this study, we found that a selective Cdc42 inhibitor, CASIN (19), significantly inhibited proliferation and survival of melphalan-and bortezomib-sensitive and -resistant MM cells. Importantly, CASIN had a stronger effect on melphalan/bortezomib-resistant cells than it did on -sensitive cells. Furthermore, CASIN sensitized melphalan/bortezomibresistant MM cells to melphalan/bortezomib. Concomitantly, CASIN extended the lifespan of mouse xenografts of bortezomibresistant MM cells and caused apoptosis of primary myeloma cells from bortezomib-resistant MM patients.
Bortezomib (Cat# S1013) and BVD523 (Cat# 869886-67-9) were obtained from Selleckchem. CASIN was obtained from Cayman Chemical Co. (Cat# 425399-05-9). For the in vitro experiments, CASIN was dissolved in DMSO to make the stock solution, followed by diluting it with the culture medium to a series of the testing solutions. For the in vivo experiments, CASIN was dissolved in cyclodextran. Melphalan was purchased from Sigma-Aldrich (Cat# 148-82-3). The protease inhibitor cocktail tablets were obtained from Roche Diagnostics GmbH (Ref# 11836170001). The phosphatase inhibitor cocktail was purchased from Goldbio (Cat# GB-450).

Flow Cytometry
For cell apoptosis assay, cells were incubated with Annexin V for 15 min, according to the manufacture's protocol (Invitrogen). For DNA damage assay, cells were fixed, permeabilized, and then incubated with anti-γ H2A histone family member X-positive (γH2AX, eBioscience; Cat# 50-9865-42) for 30 min. The cells were then analyzed by flow cytometry.

Human Samples
Peripheral blood mononuclear cells (PBMC) from healthy human subjects were collected after obtaining informed consent according to the protocol approved by the review board of Cincinnati Children's Hospital Medical Center. The cells were cultured in RPMI1640 medium containing 10% FBS.
PBMC from bortezomib-resistant MM patients were collected by Dr. Stephen Medlin, after obtaining informed consent according to the protocol approved by the review board of University of Cincinnati Medical Center. MM patients with other cancer(s) were excluded. CD138 + plasma cells were isolated using magnetic-activated cell sorting. The cells were cultured in RPMI1640 medium containing 10% of FBS and 2 ng/mL recombinant IL-6 (R&D Systems) (23).

Xenograft Mouse Model
NSG mice (8-10-week-old) were obtained from Cincinnati Children's hospital. The mice were conditioned by administering busulfan (3 mg/kg) followed by intrafemoral injections of 1 × 10 6 V10R cells per mouse the next day. Two days after cell injection, the mice were randomly allocated to the following four groups and treated as indicated: vehicle [150 µL/mouse, intraperitoneally (ip)]; CASIN (20 mg/kg, 2 times/day, ip); bortezomib (0.2 mg/kg, 2 times a week, ip); or CASIN plus bortezomib (CASIN, 20 mg/kg, 2 times/day). The mice were monitored daily and euthanized when they developed signs of reduced mobility including paralysis, hunched posture, respiratory distress, or a combination of these signs according to the institutional ethical guidelines. No blinding was performed. All the mice were housed under specific pathogen-free conditions in the animal facility at Cincinnati Children's Hospital Research Foundation. Mice were anesthetized when necessary, using ketamine [80-100 mg/kg, intramuscularly (im)], acepromazine (4-6 mg/kg, im), and atropine (0.1 mg/kg, im). Anesthesia was maintained using ketamine (30 mg/kg, im) as needed. During the experiments, mice were isolated in microisolator cages and cared for in the Laboratory Animal Resource Center by a trained technician and two veterinarians. Animals were checked daily by qualified personnel in the laboratory. The method of euthanasia was CO 2 -induced. The animal study was in compliance with the Cincinnati Children's Hospital Medical Center Animal Care and Use Committee protocols and the National Institutes of Health (NIH) guide for the care and use of Laboratory animals (NIH Publications No. 8023, revised 1978).

Statistics
For cell line culture experiments, the data represent means ± standard deviation (SD, error bars) of technical replicates (triplicates) from three independent experiments. The results of the MM patient cell culture experiments represent means ± SD of biological replicates (n = 4). A two-sided t-test was used to determine differences between two experimental groups. For animal experiments, the data represent means ± SD of biological replicates. Each experimental group (Vehicle, CASIN, bortezomib, CASIN plus bortezomib) consisted of 10 mice to ensure a power of 85%, effect size of 1.0, 30% SD, and P < 0.05. A log-rank (Mantel-Cox) test was used to determine differences between two experimental groups. For all of the experiments, P < 0.05 or P < 0.01 was considered statistically significant.

Cdc42 Is Essential for MM Cell Proliferation and Survival
To determine whether Cdc42 has a role in MM cell proliferation and survival, we used Cdc42 shRNA to knockdown Cdc42 in S and LR5 cells ( Figure 1A). We found that Cdc42 knockdown significantly inhibited cell proliferation ( Figure 1B) and caused cell apoptosis ( Figure 1C) in both S and LR5 cells, suggesting that Cdc42 plays an important role in MM cell proliferation and survival.

Half-Maximal Growth Inhibition Concentration of CASIN
Considering the role of Cdc42 in MM cell proliferation and survival, we postulated that targeting of Cdc42 might benefit MM patients. To test this hypothesis, we examined the effects of a Cdc42 inhibitor, CASIN (19). We first determined the halfmaximal growth inhibition concentration (GI 50 ) of CASIN in a number of MM cell lines and found that it ranged from 2.56 to 4.87 µM (Figures 2A-G). CASIN was used at 5 µM, unless otherwise noted, in all subsequent in vitro studies.

CASIN Preferentially Suppresses Melphalan-Resistant MM Cells
We used S cells to confirm that CASIN selectively inhibited Cdc42 activity without affecting that of Rac1, another Rho GTPase family member closely related to Cdc42 (Figure 3A). We found that CASIN suppressed the proliferation of S and LR5 cells to a similar extent (Figure 3B), with a GI50 of 5.454 µM for LR5  Figure S1A). CASIN also caused apoptosis of both S and LR5 cells ( Figure 3C). Notably, CASIN was more potent in inducing apoptosis of LR5 cells than that of S cells (10and 2.5-fold increase, respectively, compared to vehicle group, Figure 3C). While CASIN was less effective than melphalan in killing S cells, it was more effective than melphalan in killing LR5 cells ( Figure 3C). Moreover, CASIN sensitized both S and LR5 cells to melphalan-induced cell death. Intriguingly, the sensitizing effect of CASIN was more profound in LR5 than in S cells ( Figure 3C). Our data indicate that CASIN could be particularly effective in counteracting melphalan resistance in MM.

Mechanisms of Action of CASIN in Suppressing Melphalan-Resistant MM Cells
Next, we wanted to understand the mechanisms underlying the dominant effect of CASIN on melphalan-resistant cells over their sensitive counterparts, and we detected an increase in Cdc42 protein expression and activity in LR5 cells (Figures 1A,  3D). Furthermore, we wondered why CASIN was more effective in sensitizing melphalan-resistant cells to melphalan-induced cell death. It has been shown that melphalan resistance is attributable to increased mono-ubiquitination of FANCD2 of the FA DNA damage repair pathway and its mediated DNA repair (13). Consistently, we found that the level of melphalan-induced FANCD2 mono-ubiquitination was higher and DNA damage was lower in LR5 cells than that in S cells, as evidenced by decreased frequency of γH2AX + cells (Figures 3E,F). Interestingly, CASIN abolished melphalaninduced FANCD2 mono-ubiquitination in LR5 but not S cells ( Figure 3E). Concomitantly, CASIN increased melphalaninduced DNA damage in LR5 but not S cells, as evidenced by increased frequency of γH2AX + cells ( Figure 3F). Our data indicate that CASIN sensitizes melphalan-resistant cells to melphalan-induced cell death by blocking FANCD2-mediated DNA damage repair.

CASIN Preferentially Suppresses Bortezomib-Resistant MM Cells
We then determined whether CASIN could ameliorate bortezomib resistance. CASIN (5 µM) inhibited the proliferation ( Figure 4A) and survival (Figure 4B) of bortezomib-resistant V10R cells to a larger extent than -sensitive WT cells, with a GI50 of 4.191 µM for V10R cells (Supplemental Figure S1B). While CASIN was less effective than bortezomib (10 nM) in suppressing WT cells, it was substantially more potent in inhibiting V10R cells (Figures 4A,B). Treatment of V10R cells with CASIN at a lower concentration (3 µM) increased cell sensitivity to bortezomib (7.5 and 10 nM, Supplemental Figures S2A,B).

Mechanism of Action of CASIN in Suppressing Bortezomib-Resistant MM Cells
Similar to melphalan-resistant MM cells, bortezomib-resistant V10R cells showed elevated Cdc42 activity ( Figure 4C). Thus, increased Cdc42 activity seemed to contribute to bortezomib resistance and the superior efficiency of CASIN in suppressing bortezomib-resistant MM cells. Recent studies have shown that targeting of PI-3 kinase/AKT signaling pathway enhances the sensitivity of MM cells to bortezomib and overcomes bortezomib resistance (25,26). We confirmed that the activity of AKT was markedly elevated in bortezomib-resistant V10R cells compared to that in the sensitive WT cells (Figure 4D). However, CASIN did not affect the activity of AKT in V10R cells (Figure 4D), suggesting that PI-3 kinase/AKT pathway does not contribute to the effects of CASIN on bortezomibresistant MM cells. EGFR/STAT3 signaling pathway is also involved in the sensitivity of MM cells to bortezomib (12). Consistently, both p-EGFR and p-STAT3 were elevated in V10R cells (Figures 4E,F). The increased activities of EGFR and STAT3 were suppressed by CASIN treatment, suggesting that the dominant effects of CASIN on bortezomib-resistant MM cell proliferation and survival are attributable to its suppression of EGFR/STAT3 signaling pathway. In addition, ERK signaling is associated with chemo-resistance of MM (10,12). We found that the activity of ERK, which was increased in V10R cells (Figure 4G), was attenuated by CASIN ( Figure 4G). Treatment of V10R cells with an ERK inhibitor, BVD523, mimicked the effect of CASIN on inhibition of V10R cell survival ( Figure 4H). Importantly, restoration of ERK activity in CASIN-treated V10R cells by viral transduction of MEK1 (Figure 4I) partially inhibited the apoptosis of V10R cells (Figure 4J), demonstrating that inhibition of ERK pathway critically contributed to the inhibitory effect of CASIN on survival of bortezomib-resistant cells. Collectively, these results imply that the mechanism of action of CASIN on bortezomib-resistant MM cells involved inhibition of EGFR/STAT3 and ERK signaling pathways.

CASIN Prolongs Lifespan of Mouse Xenografts of Bortezomib-Resistant MM Cells and Causes Apoptosis of Bortezomib-Resistant MM Patient Cells
Considering the potency of CASIN in bortezomib-resistant MM cells in vitro, we examined its efficacy in mouse xenografts of bortezomib-resistant cells. As shown in Figure 5A, while bortezomib treatment only slightly extended the lifespan of V10R cell-bearing mice, the CASIN-and combined CASIN and bortezomib-treated V10R cell-bearing mice exhibited a  significantly prolonged lifespan. Of note, the lifespan of the cotreated group was comparable to that of the CASIN group, suggesting that the effect of the co-treatment was mostly mediated by CASIN, similar to the in vitro effect of cotreatment (Figures 4A,B). Perhaps, a lower dose of CASIN would sensitize V10R-xenografted mice to bortezomib, as seen in vitro (Supplemental Figure S2). Our data indicate that CASIN may benefit bortezomib-resistant MM patients. To substantiate this notion, we tested the effect of CASIN on human CD138 + primary myeloma cells from bortezomib-resistant MM patients. As shown in Figure 5B, CASIN treatment increased apoptosis of MM patient cells by more than 2-fold. Taken together, these results suggest that Cdc42 could serve as a potential therapeutic target to overcome drug resistance of MM.

CASIN Has Negligible Side Effects
Finally, we determined the potential side effects of CASIN. We found that 7 µM of CASIN inhibited the viability of both PBMC from healthy human subjects and MM cells (Figure 6A). While CASIN at 5 µM used for all the in vitro experiments significantly reduced the viability of MM cells, it did not inhibit PBMC ( Figure 6A). CASIN at 5 µM also had no effect on the survival of normal B cells (Figure 6B), and only modestly suppressed their proliferation ( Figure 6C). Furthermore, we did not note significant weight loss in the above mouse xenografts treated with CASIN (data not shown). Our data suggest that the side effects of CASIN would be limited.

DISCUSSION
MM remains an incurable disease and nearly all patients eventually develop resistance to currently available drugs (27). Thus, novel therapeutics that conquer drug resistance of MM are urgently needed. In this study, we provide evidence that CASIN, a Cdc42 inhibitor, of which we own intellectual property (19), represents a promising experimental drug to circumvent melphalan/bortezomib resistance of MM. We suggest that CASIN overcomes melphalan resistance by decreasing FANCD2 mono-ubiquitination and, subsequently, increasing DNA damage. Considering that bortezomib also inhibits FANCD2 mono-ubiquitination (14), it would be interesting to examine whether bortezomib augments CASIN activity in melphalan-resistant MM cells. On the other hand, CASIN may overcome bortezomib resistance through inhibiting the activities of EGFR, STAT3, and ERK, but not AKT. We have provided evidence to support that CASIN inhibits bortezomib-resistant cell survival partially through inactivating ERK. Whether inactivation of EGFR and STAT3 critically contribute to CASIN-mediated suppression of bortezomibresistant cell survival needs further investigation. Overexpression of the β5 proteasomal subunit, the primary target of bortezomib, and c-Maf have been shown to regulate bortezomib resistance (28). Whether CASIN suppresses the β5 proteasomal subunit and c-Maf expression remains to be determined. In addition, both melphalan and bortezomib resistance can be conferred by upregulation of multidrug resistance (MDR) proteins (e.g., P-glycoprotein) and glycolytic signaling (28). Thus, examining whether CASIN dampens MDR protein expression and glycolytic signaling is warranted.
An interesting finding of our study is that Cdc42 activity is increased in both melphalan-and bortezomib-resistant MM cells, suggesting that enhanced Cdc42 activity contributes to both melphalan and bortezomib resistance. Thus, our observations that inhibition of Cdc42 activity by CASIN inhibits FANCD2 mono-ubiquitination in melphalan-resistant cells and EGFR, STAT3, and ERK activities in bortezomibresistant cells suggest that Cdc42 acts upstream of FANCD2 to confer melphalan resistance and upstream of EGFR, STAT3, and ERK to confer bortezomib resistance. An increase in Cdc42 activity in melphalan/bortezomib-resistant cells also indicates that enhanced Cdc42 activity may be used as a biomarker to identify melphalan/bortezomib-resistant MM patients who are most likely to benefit from CASIN treatment.
CASIN acts by binding to inactive GDP-bound Cdc42 to inhibit its cycling to active GTP-bound Cdc42 and is highly specific to Cdc42 (19). To confirm its specificity, we show that 5 µM CASIN does not affect the activity of the closely related Rac1. We believe that CASIN has great applicability for treating drug-resistant MM considering the following points: (1) Cdc42 activity is enhanced in melphalan/bortezomib-resistant MM cells, suggesting that CASIN would likely cause an oncogene addiction-like effect; (2) CASIN only marginally affects healthy human PBMC and human B cells; (3) CASIN has no effect on thymocyte development and T cell homeostasis in mice (29); and (4) CASIN does not cause weight loss and systemic inflammation in steady-state mice and in mice bearing bortezomib-resistant MM cells (data not shown). Thus, a therapeutic window is presumably achievable in clinical treatment of drug-resistant MM patients with CASIN.
In conclusion, our study provides a proof of concept that rational targeting of Cdc42 may be utilized to treat drug-resistant MM and, therefore, warrants clinical evaluation of CASIN in drug-resistant MM patients.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this manuscript will be made available by the authors, without undue reservation, to any qualified researcher.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Cincinnati Children's Hospital Medical Center. The patients/participants provided their written informed consent to participate in this study. The animal study was reviewed and approved by Cincinnati Children's Hospital Medical Center.

AUTHOR CONTRIBUTIONS
PN designed and performed the research, analyzed the data, and wrote the paper. JC and YL performed the research and analyzed the data. KK and MZ performed the research. LZ contributed vital new reagents or analytical tools. YZ designed the research and analyzed the data. FG designed the research, analyzed the data, contributed vital new reagents or analytical tools, and wrote the paper.

FUNDING
This work was supported in part by grants from the National Institutes of Health (R01GM108661 and R21CA198358 to FG) and the Center for Clinical and Translational Science and Training (Pilot Translational Research Grant to FG, under Award Number UL1TR001425, the National Center for Advancing Translational Sciences of the National Institutes of Health).