The VR23 Antitumor Compound Also Shows Strong Anti-Inflammatory Effects in a Human Rheumatoid Arthritis Cell Model and Acute Lung Inflammation in Mice

Well-regulated inflammatory response is essential for the control of microbial infections and clearing dead cells and other debris. However, its dysregulation may lead to general inflammatory conditions, such as acute lung inflammation (ALI), pneumonia, and rheumatoid arthritis (RA). It is also increasingly clear that chronic inflammation often results in permanent genomic alterations, eventually leading to the initiation, promotion, and malignant conversion of tumors (1–4). Many different inflammatory mediators (including proinflammatory cytokines, growth factors, and free radicals) are involved in these processes (1, 4).

Currently, nonsteroidal anti-inflammatory drugs and glucocorticoids are most often used to treat general inflammatory conditions (5). Unfortunately, however, these drugs often cause toxic gastrointestinal side effects, such as ulceration and hemorrhage (5). In addition, nonsteroidal anti-inflammatory drugs also increase the risk of vascular and coronary problems, heart failure, and hypertension (6, 7). Glucocorticoids are a class of corticosteroids that upregulate anti-inflammatory proteins and downregulate multiple inflammatory genes (8). Although dexamethasone (DEX), one of the most widely used glucocorticoids, is very effective in reducing inflammation, it can also lead to serious toxic side effects if used over a prolonged period (9, 10). The common side effects include musculoskeletal weakness, endocrine abnormality, pancreatitis, increase of blood pressure, and problems in the cardiovascular system and CNS, underscoring the importance of developing effective new drugs with minimal side effects.

Autoimmune diseases are difficult to control and, unfortunately, the incident rates are on the rise (11, 12). One of the most common autoimmune diseases is RA, which results in inflammatory arthritis of the joints and, in some cases, bone and cartilage destruction (13). Like in many other autoimmune conditions, a high level of proinflammatory cytokines is usually involved in the initiation and progress of RA (12). RA is currently treated with disease-modifying antirheumatic drugs (DMARDs), such as methotrexate, hydroxychloroquine (HCQ), glucocorticoids, and biological agents (14). However, monotherapy with these agents typically is ineffective because of the development of drug resistance (15). The combination of conventional DMARDs with biological agents is effective for 50% of RA patients, allowing them to achieve clinical remission. However, for the remaining 50% of patients, there is insufficient disease reduction (15). Therefore, RA still requires the development of new drugs with novel modes of actions.

We previously reported the synthesis and characterization of 7-chloro-4-(4-[2,4-dinitrophenylsulfonyl]piperazin-1-yl)quinoline 7-chloro (VR23), which possesses effective antitumor activity (16, 17). The major molecular target of VR23 is β2 of the 20S proteasome catalytic subunit. It was suggested previously that proteasome inhibitors may possess an anti-inflammatory property because they can theoretically downregulate the NF-κB pathway by inhibiting Ikβ degradation (18–20). However, this assumption remains controversial as others have shown proinflammatory activity of certain proteasome inhibitors (21, 22).

Chloroquine (CQ) and HCQ possess anti-inflammatory properties (23). In addition, CQ/HCQ can decrease reactivity against autoantigenic peptides by interfering Ag processing, thus effectively controlling autoimmune disorders such as RA and systemic lupus erythematosus (SLE) (24, 25). As a quinoline-sulfonyl hybrid compound, VR23 contains the main scaffold of CQ. Therefore, we hypothesized that VR23 may also possess anti-inflammatory and antiautoimmune activities. We report in this article that VR23 does indeed possess these properties, just much stronger than CQ/HCQ.

VR23 was synthesized by Dalton Pharma Services (Toronto, ON, Canada) following the protocol described previously (17), and its stock solution (20 mM) was prepared in DMSO, as previously described (16). Bortezomib (BTZ), carfilzomib (CFZ), and Cell Counting Kit–8 (Enzo Life Sciences) were purchased through Cedarlane (Burlington, ON, Canada). The following items were purchased from Sigma-Aldrich (Oakville, ON, Canada): DEX (>97%), HCQ sulfate (>98%), and LPS from Escherichia coli O111:B4. The following items were purchased from Thermo Fisher Scientific (Waltham, MA): TNF-α, IL-1β, RPMI 1640, high-glucose DMEM, FBS, and antibiotic and antimycotic solutions. The single-analyte ELISA kits were purchased from Qiagen (Montreal, QC, Canada).

THP-1 and SW982 cells, purchased from the American Type Culture Collection, were cultured in RPMI 1640 (the former) or high-glucose DMEM (the latter) supplemented with 100 μg/ml streptomycin, 100 U/ml penicillin, and 10% (v/v) FBS. The THP-1 and SW-982 cell lines were used at passages <10. Primary human fibroblast-like synoviocytes (HFLS) from three healthy donors (HFLS-N; catalog number 408-05a; lot numbers 2941, 3377, and 3378) and three donors with RA (HFLS-RA; catalog number 408RA-05a; lot numbers 2956, 3077, and 3025) were purchased from Cell Applications (San Diego, CA) through Cedarlane. The cells were cultured in synoviocyte growth medium from the same supplier. The cells in each vial were isolated from a single donor and provided at passage 1, and all experiments were conducted at passages 3–8, as described previously (26).

The sulforhodamine B assay (27) was used to determine the drug concentration that reduces cell viability by 50% (IC₅₀). Cells were cultured in 96-well plates, with 5000 cells in 100 μl of culture medium per well. The rest of the protocol was described previously (28, 29).

Cells were cultured in 11 columns of 96-well plates, with ∼5000 cells in 100 μl of culture medium per well, excluding one column for a medium-only control. The plates were incubated overnight in a humidified atmosphere at 37°C with 5% CO₂ and 95% air. Following incubation, the plates were treated with test compounds diluted in media at a total volume of 100 μl. Each column was treated with a different condition, and each treatment consisted of eight repeats to ensure accuracy of the results. The plates were then incubated for 6 h in a cell-culture incubator, and then 10 μl of Cell Counting Kit–8 reagent was added to each well, followed by the incubation of an additional 2 h to allow color formation. The absorbance was measured using an automatic plate reader (Synergy H4 Hybrid Multi-Mode Microplate Reader; BioTek) at 450-nm wavelength. Finally, cell viability was determined using the following equation: cell viability (%) = (Abs_trtmt − Abs_media)/(Abs_UT − Abs_media) × 100, where Abs_trtmt is the absorbance value of cells with the various treatment conditions, Abs_media is the absorbance of the wells containing media only, and Abs_UT is the absorbance of the wells with untreated cells.

THP-1 cells were plated at the density of 500,000 cells/ml in 1-ml volume, and the SW982 and HFLS cells were at 50,000 cells/ml in 0.5 ml for cytokine analysis. Single-Analyte ELISArray human kits for each of the individual cytokines, TNF-α, IL-1β, IL-6, and IL-8 (all from Qiagen), were used following recommendation by the supplier. A mouse TNF-α ELISA Kit (Thermo Fisher Scientific) was used for the analysis of cytokine secretion in the mouse bronchoalveolar lung fluid.

Cell migration assays were performed using Corning Invasion chambers with Matrigel matrix (Thermo Fisher Scientific). The lower chamber contained compounds (e.g., VR23 or HCQ) and chemoattractant (e.g., 10% FBS). The upper chamber was loaded with 25,000 cells in FBS-free medium. The analysis of the trans-well inserts was performed as described previously (30). After 24 h of incubation, the trans-well insert was removed, and then the remaining culture medium and cells on the upper chamber were also removed from the top of the membrane with cotton swab. The trans-well insert was placed back into the well containing 1 ml of 70% ethanol for 10 min to fix the cells. The residual ethanol was carefully removed with cotton swab. Cells on the trans-well insert membrane were air dried for 15 min at room temperature. To stain cells, the trans-well insert containing cells was placed in 1 ml of 0.2% (w/v) crystal violet for 10 min. The excessive crystal violet was carefully removed from the membrane with cotton swab. Cells on the membrane were air dried at room temperature prior to analysis by the Olympus IX73 inverted microscope. Cell images (10×) were captured at five different fields. The average number of cells that migrated through the membrane toward the chemoattractant was determined. Adherent cells attach to the membrane on the lower chamber side upon migration, and suspension cells drop into the lower chamber.

Female BALB/c mice strain 028 (6–8 wk old), purchased from Charles River Laboratories (Senneville, QC, Canada), were fed regularly with laboratory chow and water upon arrival at the Laurentian University Animal Care Facility (Sudbury, ON, Canada) where this work was carried out.

To determine an optimal dose of LPS, three mice per group were treated intranasally for 24 h under the following three different conditions: 1) no LPS, 2) 0.25 mg of LPS per kg of body weight, and 3) 0.50 mg of LPS per kg of body weight. After the animals were sacrificed, the bronchoalveolar lavage fluid (BALF) of each mouse was collected and analyzed for LPS-induced inflammatory response. The results indicated that 0.40 mg/kg (i.e., between 0.25 and 0.5 mg/kg) is the optimal concentration.

Experimental design involving the use of VR23 and the control was based on acute lung injury instillation described by Wei and Huang (31). Briefly, mice were divided into five groups, with three to five mice in each, as follows: no acute lung injury (i.e., no treatment), sham treatment, LPS only, LPS + DEX, and LPS + VR23. VR23 (30 mg/kg) and DEX (4 mg/kg) were i.p. injected 1 h prior to LPS administration. The mice were given LPS (0.4 mg/kg) in PBS intranasally to induce acute lung injury. The sham-treated mice were given PBS plus DMSO (i.p.; because VR23 was dissolved in DMSO). The mice in the no-acute-lung-injury group were not given any drugs or stimulation. The mice were sacrificed by isoflurane inhalation at 24 h postinduction of acute lung injury with LPS. BALF was collected, and cells infiltrated into the bronchoalveolar fluid were counted. The fluid was also used to measure the amount of TNF-α by ELISA. The lungs were removed to determine the levels of myeloperoxidase (MPO) and histopathological analysis. The animal study protocol was reviewed and approved by the Animal Care Committee at Laurentian University.

The accumulation of neutrophils in the lung tissue was analyzed using a colorimetric MPO assay kit (Sigma-Aldrich), which measures the amount of MPO activity. After 24 h of lung injury induction, the upper and lower lobes of the right lung were snap frozen and kept at −80°C until analysis. Immediately prior to experiments, the lungs were weighed, thawed, and homogenized using cold MPO assay buffer (included in the kit) with a sonic dismembrator. The MPO activity was then determined as per the protocol provided by the supplier (Sigma-Aldrich).

Twenty-four hours after the lung injury induced with LPS, lung tissues were embedded in paraffin blocks, followed by slicing them into 10-μm sections using a microtome. Each slice was stained with H&E for examination of cell morphology and tissue type. For microscopy, at least 10 fields were analyzed per sample using an Olympus IX73 microscope (20× magnification).

Each experiment was repeated in three biological replicates unless otherwise indicated. The mean values of these results were used for statistical analysis and expressed as mean ± SE. Comparison between experimental groups was made by p value determination using one-way ANOVA. The p value <0.05 is considered to be statistically significant. The Dunnett post hoc test was performed when necessary to determine the significance between the treatment groups and controls. Analyses were performed using GraphPad Prism software, version 7.0e (San Diego, CA).

To determine the optimal concentrations of LPS and TNF-α in stimulating monocyte cells, we measured the level of TGF-β or IL-8 induction in the presence of various concentrations of LPS or TNF-α (Supplemental Fig. 1A, 1B). The resultant data determined by ELISA indicated that the optimal doses of LPS and TNF-α are 5 μg/ml and 10 ng/ml, respectively. Thus, throughout this work, we used these concentrations unless stated otherwise. Similarly, we determined optimal time poststimulation. As shown in Supplemental Fig. 1C–G, the levels of secreted cytokines IL-1β, TNF-α, and IL-8 reached a plateau at 6 h after stimulation. One exception was the level of IL-6 induced by LPS, which never reached a peak by 24 h poststimulation, the last time point used in this experiment. Nevertheless, the 6 h poststimulation time point could still be used to measure the relative amounts of secreted IL-6.

As VR23 contains the main scaffold of the anti-inflammatory CQ, we examined whether VR23 can also downregulate inflammation. To standardize the doses for different compounds used for comparison, we first determined the values of growth inhibition by 50% (IC₅₀) for SW982 and THP-1 cells in response to VR23, CQ, HCQ, BTZ, CFZ, or DEX (Table I). We then determined changes in the levels of four major proinflammatory cytokines after THP-1 cells were stimulated with either LPS or TNF-α for 6 h. As shown in Fig. 1A and 1B, VR23 at 3 μM downregulates IL-1β and IL-6 secretion to near the background level. At the same concentration, VR23 also substantially (∼50%) downregulates the level of TNF-α (Fig. 1C). VR23 downregulates IL-8 by 19 and 76% at 3 and 6 μM, respectively, in LPS-stimulated THP cells (Fig. 1D). Interestingly, under our experimental conditions, VR23 more effectively downregulates IL-8 in TNF-α–stimulated THP-1 cells (60% at 3 μM and almost completely at 6 μM) (Fig. 1E).

In contrast, the other two proteasome inhibitors, BTZ and CFZ, do not downregulate the levels of IL-1β, IL-6, TNF-α, or IL-8 at indicated concentrations. Thus, proteasome inhibitors do not necessarily have the same effects on the regulation of proinflammatory cytokines. Compared with VR23, the “parental” CQ is much less effective on the level of IL-1β and TNF-α at 25 μM, although it does downregulate IL-6 in a similar level rendered by 3 μM VR23, indicating that VR23 is clearly superior to CQ in modulating proinflammatory cytokines. At the IC₅₀ concentration, DEX also effectively downregulates IL-1β, TNF-α, and IL-6. However, VR23 is generally more effective than DEX (Fig. 1).

Having established that VR23 has strong anti-inflammatory activity in monocytes, the next question was whether VR23 can also regulate chronic inflammation, a major factor involved in the initiation and progress of autoimmune disease, such as in RA patients. To address this question, a study was carried out with SW982, a cell line derived from human synovial sarcoma. The SW982 cell line is widely used as a model for studying RA, as it possesses similar immunological properties as primary synovial cells (32). Because IL-6 is strongly implicated in the progress of RA (33), we measured IL-6 as its surrogate. We found that VR23 effectively downregulates IL-6 by 6 h posttreatment in SW982 cells. HCQ also shows a trend of downregulation but with no clear significance (Fig. 2A). Note that HCQ was included in this experiment as it is often used to treat RA and SLE (34–36). In addition, both HCQ and VR23 contain a 7-chloro-4-aminoquinoline group found in many anti-inflammatory molecules (37). Our findings thus indicate that VR23 can be effective and potentially safe for the treatment of RA patients.

We next examined the effects of VR23 on primary human synovial cells. Our model in this study was HFLS isolated from the synovial fluid in joint capsules of healthy donors (HFLS-N) or RA patients (HFLS-RA). It is known that synoviocytes in RA patients produce harmful molecules that contribute to joint destruction and the exacerbation of the disease. For these experiments, we first determined the dose of compounds that are effective without killing cells at 6 h poststimulation (Supplemental Fig. 2). For both HFLS-N and HFLS-RA, up to 10 μM of VR23 did not affect cell viability either used alone or in combination with IL-1β (Supplemental Fig. 2D, 2E). We then determined changes in the level of the IL-6 proinflammatory cytokine after HFLS-N or HFLS-RA cells were stimulated with IL-1β or LPS for 6 h. We found that both VR23 (5 μM) and HCQ (15 μM) downregulate IL-6 by ∼30% in comparison with the IL-1β–stimulated sample in the HFLS-N cells (Fig. 2B). The picture is quite different in HFLS-RA cells: VR23 (5 μM) downregulated IL-6 by 63% in comparison with the IL-1β–stimulated sample, whereas HCQ (30 μM) did so only by 45% (Fig. 2C). These data thus suggest that VR23 can be more effective in modulating IL-6 production in RA patients than healthy persons. When LPS is a stimulant, 2.5–5.0 μM VR23 effectively downregulates IL-6 to the background level, whereas HCQ was ineffective (Fig. 2D, 2E). Therefore, HCQ is much less effective in downregulating IL-6 than VR23 on HFLS-RA cells stimulated by either IL-1β or LPS. These findings are consistent with data from the study carried out with SW982 (Fig. 2A) and strengthen the idea that VR23 could be an effective drug against RA.

The migration of human fibroblast-like synovial cells plays a key role in RA progression as they eventually invade and destruct the cartilage and bone (38, 39). Because we previously found that VR23 prevents tumor cells spreading to adjacent tissues (16), we examined whether VR23 also inhibits synovial cell migration. As shown in Fig. 3A and 3B, 5 μM VR23 downregulates cell migration to the background level when 10% FBS was used as chemoattractant. We also examined the effect of VR23 on the level of MCP-1, secreted in both HFLS-N and HFLS-RA cells. As shown in Fig. 3C and 3D, VR23 at 5 μM downregulates MCP-1 secretion to the background level in both HFLS-N and HFLS-RA. In contrast, HCQ is effective on neither HFLS-N nor HFLS-RA. We also carried out a scratch-wound healing assay using an IncuCyte S3 Live Cell Analysis System (Supplemental Fig. 3). The scratch wound inflicted to the SW982 cell population was completely healed by 24 h postscratch in the presence of 25 μM CQ. In contrast, even by 48 h postscratch, the wound was not completely healed in the presence of 5 μM VR23. Thus, these data indicate that VR23 is much better in preventing synoviocyte migration than CQ/HCQ. Together, our data again indicate that VR23 may be an effective treatment option against RA and possibly other autoimmune conditions.

Because our data from in vitro studies showed that VR23 can substantially reduce or almost completely prevent the secretion of proinflammatory cytokines in THP-1 and SW982 cells, we examined its anti-inflammatory effects using a mouse model. Toward this goal, we first determined an optimal LPS dose to induce TNF-α in the mouse bronchoalveolar. We found that 0.4 mg of LPS per kilogram of body weight is an optimal dose when the subjected animals’ lungs were analyzed at 24 h post-LPS (Supplemental Fig. 4).

Acute lung injury is characterized by increased permeability of the capillary barrier, which is accompanied by an increase in lung edema and an influx of neutrophils into the bronchoalveolar space (40). The influx of neutrophils caused by lung injury induces the release of proinflammatory cytokines, such as IL-1β, TNF-α, and IL-8 (40). To determine the anti-inflammatory effects of VR23, the level of TNF-α in the BALF was measured 24 h after the animal was treated with 0.4 mg of LPS. We found that VR23 effectively reduces the level of TNF-α (Fig. 4A). Furthermore, VR23 at 30 mg per kilogram of body weight completely inhibited cell invasion into the bronchoalveolar space (Fig. 4B). In both cases, VR23 and DEX showed similar efficacy.

MPO is a marker distinctly correlated with inflammation. Neutrophils that influx into the bronchoalveolar space because of lung injury caused by acute inflammation will degranulate and release the proinflammatory enzyme MPO (41). Therefore, the level of MPO can be an excellent inflammation marker in lung (41). Data in Fig. 4C show that VR23 is a strong inhibitor of MPO.

LPS-induced lung injury can lead to lung tissue inflammation due to the influx of immune cells into the alveolar space. To gain insight into this aspect, we examined the effects of VR23 on lung tissue sections stained with H&E. The resultant histopathological data show that alveolar inflammation caused by LPS is largely prevented when animals are treated with 30 mg/kg VR23 (Fig. 4D). DEX treatment also showed a similar result (Fig. 4D). This suggests that the anti-inflammatory activities of VR23 observed in the cell culture system can be translated into a preclinical model.

We previously showed that VR23 possesses strong antitumor activity in vitro and in animal models (16, 17). VR23 is especially notable because it is not only nontoxic and preferentially kills cancer over noncancer cells but also shows strong synergy when combined with other therapeutic agents, such as paclitaxel (16). In this article, we report that VR23 also possesses a strong anti-inflammatory property, which is consistent with the notion that the 7-chloro-4-aminoquinoline group renders an anti-inflammatory property (37, 42). The anti-inflammatory effect of VR23 is not specific to certain cell lines only because it can reduce the response in a number of different model systems, including monocyte cell lines, primary synoviocytes, and the bronchoalveolar system in mice.

In our cell culture models, we found that VR23 substantially downregulates the levels of IL-1β, TNF-α, IL-6, and IL-8, all of which are associated with inflammation (33, 43–46). IL-6 is especially sensitive to VR23, suggesting that the IL-6 pathway may be a major target of VR23. Under conditions of acute stimulation, IL-6 can function as an anti-inflammatory cytokine. However, overwhelmingly stimulated, IL-6 can be proinflammatory, leading to the development of chronic inflammation, autoimmune diseases, and cancer (47). Data from our experiment with high doses of stimulants are likely translatable to the chronic inflammation caused by proinflammatory IL-6.

All of the four cytokines mentioned above are key inflammatory mediators that can induce damage even under acute inflammatory conditions and are intimately associated with the initiation of autoimmune conditions when unregulated for a prolonged period. The ability of VR23 to downregulate all of these proinflammatory cytokines demonstrates its substantial potential for the treatment of both acute and chronic inflammatory conditions.

Although VR23 effectively downregulates four proinflammatory cytokines examined (i.e., TNF-α, IL-1β, IL-6, and IL-8), the BTZ and CFZ proteasome inhibitors do not show the same effects on the monocytic cells. It was previously suggested that proteasome inhibitors may downregulate inflammation through the modulation of the NF-κB pathway (18–20). BTZ, in particular, was previously suggested to ameliorate inflammation if administered prior to inflammatory induction. For example, Han et al. (48) demonstrated that BTZ (25 or 50 nM) could reduce inflammation when administered prior to inflammation induction. Similarly, Chen et al. (20) demonstrated that BTZ at a dose of 25 nM downregulated the inflammatory response in endotoxin-induced uveitis. Contrarily to these reports, Hideshima et al. (21) demonstrated that BTZ induces inflammatory response in multiple myeloma cells. At least in part, the progression of multiple myeloma by BTZ could be due to its ability to induce proinflammatory macrophages (22). A study by Tilahun et al. (49) also demonstrated the detrimental effects of BTZ in acute systemic inflammatory conditions, such as sepsis. Our data shown in this article clearly demonstrate that BTZ and CFZ do not notably downregulate or upregulate inflammatory cytokines at IC₅₀ doses. The discrepancies discussed above may be stemmed from different doses and/or different cells used for each experiment. In any event, it is highly likely that the anti-inflammatory effect shown by VR23 is not a general property shared by proteasome inhibitors.

The antimalarial HCQ is one of five nonbiologic DMARDs. HCQ and its parental drug CQ were initially documented for their uses to treat autoimmune disorders, such as RA and SLE (50), mainly because of their reported ability in downregulating the levels of TNF-α, IL-6, and IL-1β in cells stimulated with endotoxin (51). Today, HCQ is still used as a DMARD for RA because of its low-toxicity profile, although it only has moderate clinical effects (52). In the treatment of early RA, the use of nonbiologic DMARD monotherapy is the recommended treatment by the American College of Rheumatology guidelines (52). If the patient has still high disease activity, combinations of methotrexate with HCQ or other DMARDs are recommended as initial therapy (52). However, if the patient has poor prognostics with high disease activity, biologics such as anti-TNF agents are recommended (52). Thus, DMARDs, including HCQ, remain essential treatment options for patients with RA, even if more-effective (and expensive) biological agents are available (52).

In our study, we found that CQ is effective only in downregulating IL-6 in monocytes and not effective on TNF-α, IL-6, and IL-8. This discrepancy may be due to the differences in CQ concentrations used. For example, Karres et al. (51) used 100–200 μM (which is 5- to 10-fold of IC₅₀ values), whereas we used more clinically relevant doses (15 or 30 μM). Our data thus indicate that VR23 would be more appropriate than CQ/HCQ for the treatment of inflammatory and autoimmune conditions. In particular, VR23 has the potential to be a highly effective agent for the treatment of RA, as it not only can effectively and preferentially downregulate the level of IL-6 in HFLS-RA over HFLS-N but also can inhibit synovial cell migration (Figs. 2, 3). It should be noted that the dysregulation of IL-6 plays a large role in the development and progression of RA (and often other autoimmune conditions) (53). IL-6 plays a critical role in the progression of RA; however, it is not the only factor involved in the manifestation and progression of RA. RA is a complex disease, involving the dysregulation of many cytokines and factors. Nevertheless, the ability of VR23 to specifically downregulate IL-6 in addition to other proinflammatory cytokines makes it an excellent candidate as a novel DMARD for RA treatments.

Our animal-based study clearly demonstrates that VR23 can be very effective for the treatment of ALI caused by LPS (and, potentially, by bacterial infection) (Fig. 4). When compared for efficacy alone, VR23 is similar to DEX, a glucocorticoid widely used at clinics. However, the prolonged use of corticosteroid hormone can lead to serious side effects on musculoskeletal function, endocrine regulation, cardiovascular system, and CNS (9, 10). Considering VR23 does not show any ill effects in preclinical models (16), it would be better than DEX as a treatment agent against inflammation and autoimmune conditions.

Together with our previous reports (16, 17), data presented in this article suggest that VR23 is not only an effective and safe antitumor agent but may also have substantial potential for the treatment of inflammation and autoimmune conditions, especially RA. Thus, VR23 can be a uniquely effective drug for cancer treatment as well as for the prevention of cancer development and progression caused by prolonged inflammation.

We are grateful to Elizabeth Acosta-Ramirez for demonstrating the in vivo BALF collection and to Vandana Srivastava for initiation of the SW982 cell line work.

The authors have no financial conflicts of interest.