Bortezomib Congeners Induce Apoptosis of Hepatocellular Carcinoma via CIP2A Inhibition

CIP2A is an oncoprotein that upregulates p-Akt and promotes cancer cell proliferation and survival. The proteasome inhibitor bortezomib has been shown to reduce CIP2A and lead to cell apoptosis. Here; we modified the functional group of bortezomib to generate a series of novel compounds and conducted a structure–activity relationship (SAR) study. The results showed that compound 1 was able to repress CIP2A expression and cell apoptosis in the same manner as bortezomib, but with less potency in inhibition of proteasome activity. This finding provides a new direction for the design of CIP2A inhibitors.


Introduction
Cancerous inhibitor of protein phosphatase 2A (CIP2A) is a human oncoprotein which is overexpressed in many cancers, including hepatocellular carcinoma (HCC), lung, colon, prostate, ovarian, cervical, breast, head and neck cancers and leukemia [1]. Overexpression of CIP2A in transformed human cells promotes anchorage-independent cell growth and in vivo tumor formation [2]. A mechanistic study showed that CIP2A inhibits protein phosphatase activity which reduces PP2A-mediated dephosphorylation of c-myc or Akt. This inhibition prevents c-myc degradation, maintains Akt activity and further promotes cell progression [3][4][5][6].
A natural product, rabdocoetsin B, has been reported to inhibit CIP2A level and reduce cancer cell proliferation [7]. Although rabdocoetsin B shows CIP2A inhibition activity, the pure compound is not easy to obtain and purify. In addition, designing a synthetic route for rabdocoetsin B is complicated by the chiral centers in its structure. Moreover, this natural product may induce off-target effects and activation or inactivation of other enzyme activity at high concentrations.
Botezomib (Velcade ® , Millennium Pharmaceuticals, Inc., Cambridge, MA, USA, Figure 1) is a proteasome inhibitor used clinically for relapsed multiple myeloma and mantle cell lymphoma [8]. Studies have shown that bortezomib is able to reduce the phosphorylation level of Akt in HCC [4]. Further study showed that bortezomib reduced p-Akt through activating PP2A phosphatase activity and downregulating CIP2A expression in HCC cells [9,10]. To date, no correlation has been found between proteasome inhibition and inhibition of the CIP2A/PP2A/Akt pathway by bortezomib. In addition, there are no reports showing the relationship between CIP2A expression and proteasome activity induced by bortezomib derivatives that inhibit CIP2A expression.
In light of the ability of bortezomib to reduce CIP2A level, here we synthesized some novel bortezomib derivatives that retain the ability to reduce cell survival, but with less reduction of proteasome activity than bortezomib. We provide data showing that these novel bortezomib derivatives mediate apoptosis correlated with downregulation of CIP2A.

Chemistry
Bortezomib is a modified dipeptide that can be written as Pyz-Phe-boroLeu, where the N-terminus is capped by pyrazinoic acid and the carboxylic acid of leucine is replaced with a boronic acid moiety. The boronic acid plays a crucial role in inhibiting 20S activity. For example, the boron atom acts as an electrophile to covalently interact with the nucleophilic oxygen of Thr10. In addition, Thr1 of the proteasome generates a hydrogen bridge with the boronate hydroxyl group further stabilizing the whole complex. To elucidate the relationship between proteasome activity and downregulation of CIP2A, we used a chemical approach to reduce the interaction between the boronic acid of bortezomib and Thr1 of the proteasome by adding a bulky group to the boronic acid. The proteasome activity of the resulting bortezomib derivative (1, Figure 1) was tested by ELISA. To explore the structure-activity relationship of the downregulation of CIP2A, we replaced the boronic acid of bortezomib with various functional groups yielding compounds 11-14. Moreover, we replaced the pyrazinoic ring with benzene and methyl groups and used it as a platform to carry out structural modifications, which generated compounds 15-20. These bortezomib derivatives were synthesized according to a general procedure which is described in Scheme 1. The inhibition of CIP2A and the proteasome activity of these compounds were tested by ELISA and western blot. The original synthesis of bortezomib started with the pinanediol ester of leucine boronic acid; however, we started with a carboxylic acid reaction with 1,1′-carbonyldiimidazole that was further coupled with N,O-Bis(trimethylsilyl)acetamide. The intermediate product was incorporated with a boronic ester to generate compound 4. To investigate the CIP2A inhibition ability of boronic acid and boronic ester, we hydrolyzed the boronic esters with the sequence of potassium hydrogen difluoride and silica gel to obtain boronic acid compounds in high yield.

Development of a Bortezomib Derivative with a Bulky Pinanediol Group
We synthesized a bortezomib derivative 1 with a bulky pinanediol group connected to the boron atom which reduces its hydrogen acceptor and donor ability. Theoretically, compound 1 should have less binding affinity than bortezomib to the proteasome Thr 10 and Thr1 residues and less cell toxicity. We then explored the ability of bortezomib and compound 1 to inhibit proteasome activity in Sk-Hep1 cells. Bortezomib was able to inhibit 75% of the proteasome activity of untreated Sk-Hep1 cells at 100 nM ( Figure 2). In contrast, compound 1-treated cells showed only 40% inhibitory activity as compared to the vehicle control. Figure 2. Effects of bortezomib and its derivatives on proteasome inhibition. Sk-hep1 cells were exposed to 100 nM bortezomib and compounds 1, 16, 17, 19, and 20 for 5 h before measurement of proteasome activity. Columns, mean (n = 3); bars, SD.

Structure-Activity Relationship between Boronic Acid and the Pyrazinoic Ring, and Cell Growth Inhibition
The boronic acid functional group of bortezomib was replaced with ester and hydroxylmethylene groups, generating compounds 11 and 12 ( Table 1). None of these derivatives with hydrogen donor and acceptor groups showed greater cell toxicity than bortozomib and compound 1. The isobutyl group is also an important factor in cell growth inhibition. For example, compounds 13 and 14 which have no isobutyl group have no growth inhibition effect. Next, we replaced the pyrazinoic ring with a benzene ring and a methyl group to generate compounds 16 and 17 ( Table 2). The benzene ring showed the same growth inhibitory effect as the pyrazinoic ring whereas the methyl group showed less cytotoxicity. We also generated compound 15 by removing the benzyl group at the R2 position, thus dramatically reducing the cytotoxicity. We further hydrolyzed the pinanediol group of compounds 15, 16, and 17 at the boron position resulting in compounds 18, 19 and 20, respectively (Table 3). Compounds 18 and 19 showed similar effect to compounds 15-17 in inducing cell death.

Target Validation of Bortezomib Derivatives
The bortezomib derivative with a boron-pinanediol component showed the same potency against cancer cell growth as bortezomib. However, some of the compounds showed less potent proteasome inhibition activity. Therefore, we hypothesized that the bortezomib derivatives may enhance cell death by downregulatation of CIP2A. The bortezomib derivatives were tested for inhibition of CIP2A expression by western blot. As shown in Figure 3, boron-pinanediol (compounds 1 and 16) and boronic acid compounds (bortezomib and compound 19) showed a significant repression of CIP2A expression at 125 nM. Compounds 17 and 20 were able to reduce CIP2A at 500 nM. The result indicated that CIP2A plays an important role in regulating cell survival. At the same time, we also examined proteasome regulation proteins including P27, P53, cyclin E and E2F-1 with drug treatment (Figure 3).

Figure 3.
Effects of bortezomib and derivatives on protein levels of CIP2A in Sk-hep1 cells. Cells were exposed to bortezomib and derivatives at the indicated doses for 24 h. Cell lysates were assayed by Western blot. Immunoblots were scanned by a UVP BioSpectrum AC image system and quantitated using VisionWork LS software to determine the ratio of the level of CIP2A to actin. Columns, mean (n = 3); bars, SD. The cell survival was not correlated with the expression of these proteins, suggesting that these proteins were not the major factors in cell survival. We also applied flow cytometry assays to confirm that the decreased level of CIP2A induced by these derivatives correlated with cell toxicity in Sk-hep1 cells ( Figure 4). In other words, these derivatives induced cell death in part through inhibition of CIP2A.
Moreover, these compounds were tested in Huh7 and Hep3B with MTT assay and the trend of potency in Huh7 and Hep3B is similar to Sh-hep-1 in IC50 (Table 4). In addition, we also tested the downstream signaling pathway. Expression levels of p-Akt, a downstream target of CIP2A, were assessed in bortezomib, and compounds 1, 16, 17, 19, and 20 at 500 nM. As shown in Figure 5, compounds 1, 16, and 19 with CIP2A inhibitory activity, were able to reduce p-Akt levels, and compounds 17, and 20 had moderate effects on either protein. The activation of caspases and the cleavage of PARP were examined for these agents with western blotting assays. We further explored whether CIP2A plays an important role in regulating cell survival. Genetic overexpression of CIP2A showed a rescue effect on bortezomib-induced cell toxicity. Knockdown of CIP2A increase cell toxicity with the treatment of bortezomib, indicating that CIP2A is one of the major targets of bortezomib ( Figure 6).    Figure 5. Western blot analysis of Akt (Akt1), p-Akt (Ser473), caspases and PARP levels. Sk-hep1 cells were exposed to bortezomib and derivatives at 500 nM in DMEM with 5% FBS for 24 h. Cell lysates were prepared for Akt (Akt1), p-Akt (Ser473), caspase-8, caspase-9, caspase-3, and PARP. CF, cleaved form (activated form). In summary, these data suggest that boron-pinanediol derivative induced-biological activity was due to the inhibition of CIP2A and triggered the apoptotic signals.

Discussion
Recent data suggest that CIP2A is an oncoprotein which activates p-Akt and its downstream cell survival and proliferation signals. This study confirms that bortezomib and its derivatives induce cell death through CIP2A inhibition. These findings have several implications. First, the results confirm that cell death induced by bortezomib and its derivatives does not totally depend on proteasome inhibition. Although bortezomib has been approved by the FDA as a proteasome inhibitor in multiple myeloma and mantle cell lymphoma, its effect on solid tumors was not significant. Our findings suggest that CIP2A is involved in this event. Second, our data suggest that CIP2A is a potential drug target for anticancer agents as the drug sensitivity induced by bortezomib and its derivatives correlated with CIP2A expression. Third, we also found that downregulation of CIP2A by bortezomib and its derivatives results in downregulation of p-Akt, suggesting that CIP2A also influences the interaction between PP2A and its p-Akt substrate. Since CIP2A is also over expressed in other cancers, we plan to explore the CIP2A ablative effect of these compounds before the preclinical study.

Cell Culture
The Sk-Hep1 cell line was obtained from American Type Culture Collection (Manassas, VA, USA). Cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 100 units/mL, penicillin G, 100 mg/mL streptomycin sulfate and 25 mg/mL amphotericin B in a 37 °C humidified incubator and an atmosphere of 5% CO 2 in air.

Proteasome Activity Assay
A 20S Proteasome Activity Assay kit (Chemicon, Billerica, MA, USA) was used to determine the inhibition of proteasome in drug-treated Sk-Hep1 cells. Cell lysates were prepared, and the fluorogenic peptide substrate, Suc-Leu-Leu-Val-Tyr-AMC was used according to the manufacturer's instructions. In brief, control or compound-treated cells were broken in a lysis buffer without protease inhibitors. Total cell lysate (50 μg) was incubated with 20 μmol/L of fluorogenic substrate Suc-Leu-Leu-Val-Tyr-AMC at 45 °C in 100 μL of assay buffer. Free AMC liberated by the substrate hydrolysis was quantified for 90 min at 1-min intervals on a microtiter plate fluorometrer (excitation, 355 nm; emission, 460 nm). The percentage of proteasome activity values (% control) were derived by dividing the slope obtained in the presence of bortezomib or compounds by the slope obtained in its absence ×100 [12].

Apoptosis Analysis
Apoptotic cells were measured by flow cytometry (sub-G1). After treatment with compounds, cells were trypsinized, collected by centrifugation and resuspended in PBS. After centrifugation, the cells were washed in PBS and resuspended in potassium iodide (PI) staining solution. Specimens were incubated in the dark for 30 min at 37 °C and then analyzed with an EPICS Profile II flow cytometer (Coulter Corp., Hialeah, FL, USA). All experiments were performed in triplicate.

Conclusions
We synthesized a series of bortezomib derivatives that showed CIP2A inhibition activity, and the inhibition of CIP2A-Akt signaling pathway correlated with the cytotoxicity of these derivatives in HCC cells. Several compounds were found equally as potent as bortezomib in inhibition of CIP2A and cell growth. A mechanistic study showed that compound 1 also reduced expression of p-Akt after repressing the CIP2A signaling cascade. In addition, compound 1 provides a useful pharmacological tool to the study therapeutic relevance of HCC with unregulated CIP2A expression and drug resistance. These compounds will be tested in other cancers. Moreover, a series of new derivatives provide structure activity relationship which provide a direction for future lead optimization of CIP2A inhibitors. Therefore, these data suggest the potential value of anti-CIP2A agents in the treatment of cancers. Testing of compound 1 in an in vivo HCC model and pre-clincal studies are currently being pursued.