Diarylpentanoid, a curcumin analog, inhibits malignant meningioma growth in both in vitro and in vivo models

Abstract

BACKGROUND

Malignant meningioma metastasizes systemically, primarily due to its role in epithelial-mesenchymal transition. Although the prognosis is extremely poor, drug development efforts have been limited, because this tumor is categorized as a rare form.

AIM

To examine growth suppressive effect of GO-Y030, a diarylpentanoid curcumin analog, (1E,4E)-1,5-bis [3,5-bis (methoxymethoxy) phenyl] penta-1,4-dien-3-one against the malignant meningioma.

METHODS

The growth suppression of malignant meningioma cells by GO-Y022 and GO-Y030 were examined, using IOMM-Lee and HKBMM cell lines. Male nude mice aged eight weeks, specifically BALB/cSlc-nu/nu mice received a subcutaneous inoculation of IOMM-Lee (10⁷ cells/site) on their back and 30 μg/kg of recombinant hepatocellular growth factor (HGF) was injected into the tumor every three days. After confirmed the growth tumor mass, 500 μL of GO-Y030 diluted with PBS were administrated intraperitoneally daily at doses of 1 mg/kg and 2 mg/kg, respectively.

RESULTS

GO-Y030 exhibits a growth inhibitory effect on malignant meningioma cell lines, IOMM-Lee and HKBMM ranging from 0.8-2.0 μM in vitro. Notably, GO-Y030’s inhibitory effect is about 10 to 16^th times more potent than that of curcumin, which has previously demonstrated potential in combating malignant meningioma. In mouse models, the intraperitoneal administration of GO-Y030 effectively suppresses the growth of malignant meningioma tumors that have been inoculated in the back (P = 0.002). High-performance liquid chromatography analysis has confirmed the distribution of GO-Y030 in the bloodstream and brain tissue. Moreover, GO-Y030 demonstrates the ability to significantly suppress HGF (P < 0.01), nuclear factor kappa B (P < 0.001), and N-cadherin (P < 0.001), all of which contribute to the epithelial-mesenchymal transition.

CONCLUSION

GO-Y030 holds promise as a potent compound for the systemic inhibition of malignant meningioma. GO-Y030 has higher tumor growth inhibitory effect against meningiomas than curcumin, which is known to have antitumor activity through multi-molecular target control resulting in apoptosis induction. GO-Y030 controls at least three molecules of HGF, nuclear factor kappa B, and N-cadherin.

Key Words: Malignant meningioma; Curcumin; Diarylpentanoid curcumin analog; Hepatocellular growth factor; Nuclear factor kappa B; N-cadherin

Core Tip: Malignant meningiomas have a poor prognosis, but drug development is limited due to their rarity. The curcumin analog GO-Y030 showed approximately 10-16 times stronger inhibitory effects than curcumin on IOMM-Lee and HKBMM cell lines, in vitro. Intraperitoneal administration of GO-Y030 also significantly inhibited the growth of malignant meningiomas, IOMM-Lee inoculated in nude mouse models. GO-Y030 significantly inhibited hepatocyte growth factor, nuclear factor kappa B, and N-cadherin, which contribute to epithelial-mesenchymal transition.

INTRODUCTION

Meningioma is a type of brain tumor that originates from arachnoid cap cells and typically displays benign behavior[1]. The estimated incidence of meningioma ranges from 1.8 to 13 per 100000 individuals per year[2]. World Health Organization (WHO) grade I meningiomas are categorized as benign tumors, with a prevalence of 85%-95%[3]. In contrast, WHO grade II and III meningiomas exhibit malignant characteristics, including rapid growth and aggressive invasive and metastatic potential. It leads to a less favorable prognosis, with a 10-year survival rate ranging from 14%-34% for Grade III meningiomas, while Grade I meningiomas have a 10-year survival rate of 80%-90%[3]. Furthermore, the prevalence of Grade III meningioma is relatively low, ranging from 1%-35%[4]. It is worth noting that many pharmaceutical companies do not prioritize the development of specific drugs for malignant meningioma. Consequently, there is a lack of effective treatments for this condition, and existing guidelines only recommend conventional agents such as hydroxyurea[5-8], interferon-alpha[9,10], and somatostatin analogs for its management[11,12]. However, the efficacy of these treatments remains a subject of debate[13]. Under these circumstances, chemotherapeutic drugs for this disease are considered orphan drugs. The molecular mechanisms underlying malignant meningioma are highly complex. Neurofibromatosis type 2 is found in over 50% of meningioma cases[14]. Moreover, additional genetic alterations come into play when a benign meningioma transforms into a malignant form[15]. Currently, various target inhibitors are under investigation. These targets include the mitogen-activated protein kinase (MAPK)/ERK (MEK)[16,17], phosphoinositide 3-kinase/protein kinase B[18]/the mammalian target of rapamycin (mTOR) pathway[16], epidermal growth factor receptor (EGFR)[19], vascular endothelial growth factor receptor (VEGF)[20], c-MET[21], and AXL[21], smoothened (SMO)[22], focal adhesion kinase (FAK)[23], AKT[24], cyclin-dependent kinase[25], histone deacetylase (HDAC)[26], glycogen synthase kinase 3 beta (GSK-3β)[27], and dopamine receptor[28].

In addition, the investigation extends to anti-programmed death 1 (PD-1) inhibitors such as nivolumab[29] and pembrolizumab[30], anti-programmed death-ligand 1 (PD-L1) inhibitors like avelumab[31], as well as radiolabeled somatostatin analog such as 177 Lu-DOTATE[32] and 177 LU-DOTA-JR11[33]. The vulnerability of this disease remains unexplored, and the effectiveness of single target agents has yet to be established. Reports suggest that multi-target agents like sunitinib show some promise in addressing malignant meningiomas at preclinical and clinical stages[28]. However, additional clinical studies have yet to provide conclusive evidence regarding their efficacy.

Curcumin, a component found in the spice plant Curcuma longa, exhibits antiproliferative and proapoptotic effects in human meningiomas[34]. This compound is considered to be a potent lead chemical for malignant meningioma. Additionally, one of its derivatives, a diarylpentanoid (C5) analog named 1,5-bis (4-hydroxy-3-methoxyphenyl)-1,4-pentadiene-3-one (GO-Y022), demonstrates significant permeability across the blood-brain barrier (BBB), with a value of 8.2 ± 1.3 × 10⁻⁶ (cm/s). This permeability is comparable to donepezil, an anti-Alzheimer disease drug, with a permeability of 12.0 × 10⁻⁶ (cm/s). In contrast, the permeability of curcumin is 0.60 ± 0.67 × 10⁻⁶ (cm/s)[35]. We have been involved in the development of C5 analogs like GO-Y022, with the synthesis of over 200 new analogs[36-39]. Among them analogs, (1E,4E)-1,5-bis [3,5-bis (methoxymethoxy) phenyl] penta-1,4-dien-3-one (GO-Y030) demonstrates the most potent anti-tumor potential, which is 30 times higher than curcumin[38]. GO-Y030 exhibits anti-Wnt signaling[40], anti- nuclear factor-kappa B (NF-kB) signaling[41,42], anti-signal transducer and activator of transcription 3 phosphorylation and cancer stem cell growth[43-45], anti-angiogenic potentials[46,47], PD-L1 activity[48,49], anti-regulatory T cell differentiation[50], anti-invasion, anti-metastatic capabilities[51], and pro-apoptotic potential[38]. C5 analogs can suppress cancer cell growth in vivo[52-54]. Furthermore, C5 analogs can improve heart failure[55,56]. As described, C5 analogs have a variety of pharmacological reactions. One of the targets of our C5 analogs is KH-type splicing regulatory protein (KSRP) to which C5 analog can directly bind[57]. KSRP controls the mRNA stability of the genes by initiation of mRNA decay and inhibition of translation, and by enhancing the maturation of microRNAs. KSRP plays a pivotal role in immune cell function and tumor progression[58]. It was revealed that KSRP-binding compound suppresses invasion and distant metastases of cancers[59,60]. The pharmaceutical potential of CO-Y030 surpasses that of curcumin and GO-Y022. It is indicated that the higher the drug's lipophilicity, the better the BBB permeability of the drug. When calculating mLogP, which is a calculated index of lipophilicity, curcumin was + 1.88, while GO-Y030 was +3.92 by ADMEWORKS/predictor (FUJITSU KYUSHU SYSTEMS LIMITED, Fukuoka, Japan). GO-Y030 was predicted to have a more BBB permeability than curcumin. Consequently, we are conducting anti-malignant meningioma activities of GO-Y030 in this study.

MATERIALS AND METHODS

Chemicals

Curcumin was procured from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan). GO-Y022 and GO-Y030 were purchased from Nippon Carbide Industries Co., Inc. (Tokyo, Japan). The purity has been confirmed to be over 95% by HPLC measurement. To prepare stock solutions, these chemicals were dissolved in dimethyl sulfoxide (DMSO) at a concentration of 50 mmol/L. The chemical structures of curcumin, GO-Y022, and GO-Y030 are shown in Figure 1.

Figure 1  Chemical structures of curcumin, GO-Y022 and GO-030.

Cell lines

The malignant meningioma cell line, IOMM-Lee (lot. 1259CRL2542), was purchased from the American Type Culture Collection (VA, United States). Another malignant meningioma cell line, HKBMM (lot. RBRC-RCB0680), was obtained from Riken BRC Cell Engineering Division -CELL BANK- (Ibaragi, Japan). For culturing, IOMM-Lee and HKBMM cells were cultured in DMEM (Thermo Fisher Scientific, Inc. MA, United States) supplemented with 10 g/L fetal bovine serum (FBS, FUJIFILM Wako Pure Chemical Corporation) and HamF12 (FUJIFILM Wako Pure Chemical Corporation) supplemented with 15 g/L FBS, respectively, according to the manufacturer’s protocol. The cells were maintained at 37 °C in a humidified environment with 50 mL/L CO₂ and 950 mL/L air.

Growth inhibition experiment

Cell viability was determined by quantifying the uptake and metabolism of WST-8 (2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium monosodium salt) in accordance with the manufacturer’s instructions (Dojindo Laboratories, Kumamoto, Japan). The half-maximal inhibitory concentration (IC₅₀) was determined as described previously[61].

Animal experiment

Male nude mice aged eight weeks, specifically BALB/cSlc-nu/nu mice, were purchased from Japan SLC, Inc. (Shizuoka, Japan), and they were kept in our animal room for one week. Each mouse (n = 5 in each group) received a subcutaneous inoculation of IOMM-Lee (10⁷ cells/site) on their back because it was easy to observe. Ten days later, once tumors became visible, 30 μg/kg of recombinant hepatocyte growth factor was administered hepatocellular growth factor (HGF), FUJIFILM Wako Pure Chemical Corporation) into the tumor (IT) every three days, following Chen’s protocol[62]. Simultaneously, 500 μL of GO-Y030 diluted with PBS was administrated intraperitoneally (ip) daily at doses of 1 mg/kg and 2 mg/kg, respectively. A 8.4 mg/LDMSO in PBS solution served as the control. No randomization methods were applied. There were no considerable confounders. On the seventh day, the mice were euthanized, and the extracted tumor was fixed in a 10 g/L formalin neutral buffer solution (FUJIFILM Wako Pure Chemical Corporation). The primary outcome of this study was a shrinkage of the tumor volume. Tumor volume was calculated using the following formula: Tumor volume (cm³) = (shortest tumor diameter)² varied within the groups, and to evaluate the antitumor effect, the average size at the start of administration was set as 100% and a relative size was compared. A preliminary experiment was carried out with a minimum number (n = 5) of mice, and the results showed significant differences, so this data was adopted. No specific exclusion and inclusion criterion were provided. All animal experiments adhered to the guidelines set by the Institutional Care and Use Committee at Akita University. This study proposal received approval from the Research Ethics Committee at Akita University as b-1-0440. Research work on animals should be carried out in accordance with the NC3Rs ARRIVE Guidelines. In an animal model in which a tumor was implanted subcutaneously, if it is determined that the animal is suffering intolerably due to tumor growth, etc., from the viewpoint of animal welfare, the mouse will be euthanized, and the efficacy will be judged by a surrogate endpoint such as comparing the tumor burden.

Immunohistochemistry

The cell lines were cultured in 10 cm tissue culture dishes at concentrations equivalent to the IC₅₀ value or at double the IC₅₀ value of GO-Y030. A 1 mg/L DMSO solution was used as the control. After 24 hours, the cells were harvested, collected using iPGell (GenoStaff, Tokyo, Japan), and fixed in a 100 g/L formalin neutral buffer solution, following the manufacturer’s instructions. Paraffin-embedded specimens were prepared for immunohistochemistry (IHC) using the following antibodies: Anti-rabbit polyclonal to HGF (ab216623, Abcam, Cambridge, United Kingdom), anti-rabbit monoclonal N Cadherin (N-Cad) antibody [EPR1791-4] (ab76011, Abcam), anti-rabbit monoclonal NF-κB p65 antibody (D14E12, XP, #8242S, Cell Signaling, MA, United States), and anti-rabbit cleaved Caspase-3 antibody (Asp175, #9661, Cell Signaling, MA, United States). These antibodies were diluted at a 1:200 ratio.

The horseradish peroxidase-conjugated goat anti-rabbit IgG (K4003, Agilent Dako, CA, United States) was used as the secondary antibody. The positivity of each molecule was counted in up to 100 cells at the five different high-power fields, and the percentage was calculated.

For tumor specimens from nude mice, IHC was conducted in a similar manner. In the outcome assessment process, a pathologist was blinded for the information.

Tissue distribution

We determined the blood concentrations and brain distribution of GO-Y030 using the method described previously[54]. We acquired twelve-week-old male C57BL/6 (B6) mice from Japan SLC, Inc, and they were kept in our animal room for one week. GO-Y030 (2 mg/kg) was administrated intraperitoneally, as mentioned above. The mice were sacrificed at 0, 1, 3, and 6 hours after administration (n = 3 in each group). As a control, 0.84% DMSO in PBS was used. A preliminary experiment was carried out with a minimum number (n = 3) of mice, and the results showed significant differences, so this data was adopted. Upon sacrifice, brain tissues and blood samples were collected and analyzed using high-performance liquid chromatography (HPLC), following the previously described protocol[54]. In brief, brain tissue was homogenized in PBS after measuring the tissue sample’s weight. The supernatant from each group of three mice was combined in equal volumes, resulting in a total volume of 450 mL with PBS. This mixture was then subjected to HPLC analysis. The amount of GO-Y030 was adjusted for the tissue’s wet weight and expressed as pg/mg of tissue. Blood was obtained from the orbital sinus after anesthetizing the mice with an intraperitoneal injection of a mixture containing 10 mg of midazolam (SANDOZ, Tokyo, Japan), 0.75 mg of medetomidine (ZENOAQ, Koriyama, Japan), and 12.5 mg of butorphanol (Wako), dissolved in 25 mL of distilled water. Subsequently, serum was collected from the blood, and serum samples from three mice were combined into one. The blood concentration of GO-Y030 was determined using the HPLC method, which utilized a mobile phase consisting of acetonitrile and 5 g/L KH2PO4 (pH 3.5) (70/30, v/v) on a CAPCELL PAK C18 MGII column (5 μm, 4.6 mm × 250 mm, OSAKA SODA, Osaka, Japan) at a flow rate of 0.5 mL/min. R051012 served as the internal standard.

Statistical analysis

Data are presented as mean ± SD. The between-group differences were analyzed using Student’s t-test or Welch's t test with StatMate III version 3.14 (ATMS, Tokyo, Japan). The between-group difference for body weight was analyzed using analysis of variance with Microsoft^® Excel^® for Microsoft 365 MSO Version 2407 Build 16.0.17830.20056 (Microsoft Corp, WA, United States). Statistical analyses were reviewed by SATisSTA (Kyoto, Japan).

RESULTS

C5 curcuminoid, GO-Y030, inhibits malignant meningioma growth in vitro

For the IOMM-Lee cell line, the IC₅₀ value of GO-Y030 was 0.8 μmol/L, while the IC₅₀ value for curcumin and GO-Y022 were 13.3 μmol/L and 4.3 μmol/L, respectively (Figure 2A). Notably, GO-Y030 exhibited 16.6 times and 5.4 times higher inhibitory potential than curcumin and GO-Y022, respectively. In the case of another meningioma cell line, HKBMM, the IC₅₀ value for GO-Y030 was 2.0 μmol/L, whereas the IC₅₀ value for curcumin and GO-Y022 were 20.0 μmol/L and 4.5 μmol/L, respectively (Figure 2B). HKBMM displayed greater resistance to curcuminoid. Nevertheless, GO-Y030 demonstrated 10.0 times and 2.3 times higher inhibitory potential than curcumin and GO-Y022, respectively. To our knowledge, GO-Y030 represents the most potent growth inhibitor among the newly synthesized curcuminoids (data not shown).

Figure 2 Inhibition of meningioma growth by GO-Y030 in vitro. A: The growth inhibition of IOMM-Lee; B: The growth inhibition of HKBMM.

C5 curcuminoid, GO-Y030, inhibits malignant meningioma growth in vivo

We examined the growth-inhibiting impact of GO-Y030 on IOMM-Lee cells inoculated in nude mice, following the procedure outlined in Chen’s protocol (Figure 3A). The tumors in the mice had grown to a size of 0.63 ± 0.20, 0.99 ± 0.31, 0.77 ± 0.08 cm³ in the control, 1 mg/kg (ip), and 2 mg/kg (ip) groups, respectively at the start of administration of the drug. Based on the pilot study, we estimated the amount that could be safely administered to mice to be 2 mg/kg. After 6 days, the volume had increased to 1.12 ± 0.39, 1.38 ± 0.90, 0.76 ± 0.29 cm³ in the control, 1 mg/kg (ip), and 2 mg/kg (ip) groups, respectively. In some mice in the control or 1 mg/kg (ip) groups, subcutaneously implanted tumors invaded the spine and exhibited paralysis of both legs. As results, they were beginning to have difficulty eating and drinking on their own, and it was expected that they would soon become moribund. In the research protocol, when the mice reached this state, from the perspective of animal welfare, instead of observing them until they died of the tumor, they were euthanized and an alternative endpoint was set to compare the tumor volume, and this was adopted.

Figure 3 GO-Y030’s inhibitory effect on meningioma growth in vivo. A: The treatment schedule for the mouse model included intratumor (iT) hepatocellular growth factor administration every 3 days and daily intraperitoneal (ip) administration of GO-Y030; B: Relative tumor volume, with day 1, is set as 100%. GO-Y030 was administered at either 1 mg/kg mouse weight or 2 mg/kg indicated in orange or grey, respectively. The control was indicated in blue. Notably, GO-Y030 significantly suppressed tumor growth. ^aP and ^bP indicate the P values between the control and 1 mg/kg, and between the control and 2 mg/kg, 0.041 and 0.002, respectively. There was no significance between the 1 mg/kg and 2 mg/kg. ^cP = 0.059. The open circle represents the averages of 5 tumors, and the vertical bar depicts the standard deviation. HGF: Hepatocellular growth factor.

By the sixth day, the tumor volume in control mice had increased to 170.0% ± 20.0% of the initial volume (day 0), whereas in mice treated with GO-Y030 (1 mg/kg) and GO-Y030 (2 mg/kg), the tumor volumes were 136.2% ± 24.4% and 97.2% ± 31.3%, respectively (Figure 3B). Notably, intraperitoneal administration of GO-Y030 (2 mg/kg) significantly suppressed tumor growth compared to the control (P = 0.002, Figure 3B). Simultaneously, we monitored the body weights of the mice, and no significant differences were observed. There were no apparent toxic effects in this experiment (Table 1).

Table 1 Body weight of nude mice treated with GO-Y030, mean ± SD.

Body weight (g)	Day 1	Day 3	Day 6	P value
Control	21.98 ± 1.30	24.98 ± 1.10	24.24 ± 1.22	0.55
GO-Y030 (1 mg/kg, ip)	23.7 ± 1.97	24.92 ± 0.92	24.2 ± 0.84
GO-Y030 (2 mg/kg, ip)	22.04 ± 1.52	25.02 ± 1.24	24.42 ± 1.18

C5 curcuminoid, a GO-Y030, suppresses key molecules overexpressed in meningioma cells

We investigated the effect of GO-Y030 on meningioma cells. The IOMM-Lee cells were treated with GO-Y030 at a concentration two times the IC₅₀ value concentration for 24 hours.

The IC₅₀ concentration is the concentration at which proliferation is inhibited by 50%, so to see the cell killing effect, we used a concentration of curcumin derivatives that is twice the IC₅₀. Hematoxylin and eosin (HE) staining indicated that 16.73% ± 14.38% of the cells treated with GO-Y030 exhibited a normal appearance, whereas 96.97% ± 1.81% of the control cells exhibited a normal appearance (P < 0.001, Figure 4A, E and I). Almost all cells exhibited a reduction in their size and condensed nuclei were treated with GO-Y030 (Figure 4E). Previous research by Bukovac reported that the significant role of increased N-Cad expression in tumor invasion and metastasis[63,64]. Consequently, we investigated the expression of N-Cad in IOMM-Lee cells. In untreated cells, N-Cad expression was notably present in 100% ± 0.00% cells (Figure 4B). However, N-Cad expression dimished to 32.15% ± 5.37% in GO-Y030-treated cells that exhibited cell shrinkage and pyknotic nuclei (P < 0.001, Figure 4F and J). On the other hand, in cells where N-Cad expression was retained, cell body and nucleus remained unaffected (Figure 4F). Robert’s finding reported that NF-kB signaling was activated in non-NF2 mutated meningiomas[65]. Consequently, we investigated the expression of NF-kB, specifically p65, in IOMM-Lee cells. In untreated cells, p65 expression was ubiquitously observed in the cytosol (100% ± 0.00%, Figure 4C). Nevertheless, P65 expression disappeared and reduced to 30.17% ± 7.90% in the GO-Y030-treated cells that exhibited cell shrinkage and pyknotic nuclei (P < 0.001, Figure 4G and K). Conversely, in cells where p65 expression was retained, cell shape and nucleus remained unaffected (Figure 4G). Mashayekhi’s study reported a correlation between increased HGF concentration an advanced grade meningioma[66]. We investigated the expression of HGF in IOMM-Lee cells. In untreated cells, HGF expression was prominently localized in the cytosol (100% ± 0.00%, Figure 4D). However, in GO-Y030-treated cells, HGF expression was reduced to 59.42% ± 17.68%, and the cells were characterized by cell shrinkage and pyknotic nuclei (P = 0.01, Figure 4H and L). On the other hand, in cells where HGF expression remained intact, cell shape and nucleus remained unaffected (Figure 4H). Similar results were obtained in an IHC analysis conducted on another meningioma cell line, HKBMM. We also investigated the effect of GO-Y030 on another meningioma cell line, HKBMM. HE stains indicated that 28.84% ± 9.91% of the cells treated with GO-Y030 exhibited a normal appearance, whereas 98.42% ± 2.58% of the control cells exhibited a normal appearance (P < 0.001, Figure 5A, E and I). In untreated cells, N-Cad expression was notably present in 96.19 ± 2.58% cells, whereas N-Cad expression dimished to 18.50 ± 4.74% in GO-Y030-treated cells (P < 0.001, Figure 5B, F and J). In untreated cells, p65 expression was observed in the cytosol of 75.69 ± 3.74% cells, whereas it disappeared and reduced to 34.98 ± 11.49% in the GO-Y030-treated cells (P < 0.001, Figure 5C, G and K). HGF was expressed in 100 ± 0.00% of the untreated cells, whereas in GO-Y030-treated cells, HGF expression was reduced to 33.38 ± 5.13% (P < 0.001, Figure 5D, H and L). These findings suggest that GO-Y030 may suppress the growth of meningioma cells by suppressing N-Cad, NF-kB, and HGF.

Figure 4 Inhibition of molecular targets in meningioma cell lines (IOMM-Lee) with GO-Y030 in vitro. A: Hematoxylin and eosin (HE) stains of IOMM-Lee treated with DMSO alone (control); B: Immunohistochemistry (IHC) of the N-Cadherin (N-Cad) in the control; C: IHC of the hepatocellular growth factor (HGF) in the control; D: IHC of nuclear factor kappa-B p65 (NF-kB) in the control; E: HE stains of IOMM-Lee treated with GO-Y030 at twice concentration of the 50% inhibitory concentration (IC₅₀ × 2); F: IHC of the N-Cad in IOMM-Lee treated with GO-Y030 at IC₅₀ x 2; G: IHC of NF-kB in IOMM-Lee treated with GO-Y030 at IC₅₀ × 2; H: IHC of the HGF in IOMM-Lee treated with GO-Y030 at IC₅₀ × 2; I: The percentage of normal appearance cells in IOMM-Lee treated with GO-Y030 at IC₅₀ × 2. J: The percentage of positive cells for N-Cad in IOMM-Lee treated with GO-Y030 at IC₅₀ × 2; K: The percentage of positive cells for NF-kB in IOMM-Lee treated with GO-Y030 at IC₅₀ × 2; L: The percentage of positive cells for HGF in IOMM-Lee treated with GO-Y030 at IC₅₀ × 2.

Figure 5 Inhibition of molecular targets in meningioma cell lines (HKB-MM) with GO-Y030 in vitro. A: Hematoxylin and eosin (HE) stains of HKBMM treated with DMSO alone (control); B: Immunohistochemistry (IHC) of the N-Cad in the control of HKBMM; C: IHC of the hepatocellular growth factor (HGF) in the control of HKBMM; D: IHC of NF-kB in the control of HKBMM; E: HE stains of HKBMM treated with GO-Y030 at IC₅₀ × 2; F: IHC of the N-Cad in HKBMM treated with GO-Y030 at IC₅₀ × 2; G: IHC of NF-kB in HKBMM treated with GO-Y030 at IC₅₀ × 2; H: IHC of the HGF in HKBMM treated with GO-Y030 at IC₅₀ × 2; I: The percentage of normal appearance cells in HKBMM treated with GO-Y030 at IC₅₀ × 2; J: The percentage of positive cells for N-Cad in HKBMM treated with GO-Y030 at IC₅₀ × 2; K: The percentage of positive cells for NF-Kb in HKBMM treated with GO-Y030 at IC₅₀ × 2; L: The percentage of positive cells for HGF in HKBMM treated with GO-Y030 at IC₅₀ × 2.

Furthermore, we investigated the ability of GO-Y030 to induce apoptosis in meningioma cell lines using an antibody against cleaved Caspase-3. In untreated IOMM-Lee cells, cleaved Caspase-3 expression was observed in 0.1% ± 0.01% cells, whereas it increased to 29.9% ± 3.1% in the GO-Y030-treated cells (P < 0.05, Figure 6A, B and C). In untreated HKBMM cells, cleaved Caspase-3 expression was observed in 5.0% ± 8.8% cells, whereas it increased to 60.0% ± 32.8% in the GO-Y030-treated cells (P < 0.05, Figure 6D, E and F). These findings suggest that GO-Y030 may induce apoptosis of meningioma cells.

Figure 6 Apoptosis induction in meningioma cell lines with GO-Y030 in vitro. A: Cleaved Caspase 3 in the control of IOMM-Lee; B: Cleaved Caspase 3 induction of IOMM-Lee treated with GO-Y030 at twice concentration of the 50% inhibitory concentration (IC₅₀ × 2); C: The percentage of Cleaved Caspase 3 positive cells in IOMM-Lee treated with GO-Y030 at IC₅₀ × 2; D: Cleaved Caspase 3 in the control of HKBMM; E: Cleaved Caspase 3 induction of HKBMM treated with GO-Y030 at IC₅₀ × 2; F: The percentage of Cleaved Caspase 3 positive cells in HKBMM treated with GO-Y030 at IC₅₀ × 2.

C5 curcuminoid, GO-Y030, inhibits N-Cad, NF-kB, and HGF expression in vivo

HE stains indicated distinct differences between the peripheral tumor cells of control mice and GO-Y030 treated mice. In the control group, tumor cells exhibited large nuclei, with some displaying mitotic figures. In contrast, the tumor cells of GO-Y030 treated mice were characterized by smaller, pyknotic nuclei in many cases (Figure 7A and B).

Figure 7 Inhibition of molecular targets by GO-Y030 in vivo. A: HE stains of the control group treated with DMSO alone. The inset indicates a closer magnified view; B: HE stains of IOMM-Lee treated with GO-Y030. The inset indicates a more detailed view. Pyknotic cell nuclei are observed; C: Immunohistochemistry (IHC) of N-Cadherin (N-Cad) in the control group; D: IHC of N-Cad in IOMM-Lee treated with GO-Y030; E: IHC of nuclear factor kappa-B p65 (NF-kB) in the control; F: IHC of the NF-kB of IOMM-Lee treated with GO-Y030; G: IHC of the hepatocellular growth factor (HGF) in the control group; H: IHC of HGF in IOMM-Lee treated with GO-Y030. The red dashed line distinguishes between the cells with intact tumor nuclei that test positive for each molecule, and the cells with pyknotic nuclei that test negative affected with GO-Y030 peripherally from vessels.

Furthermore, we conducted an IHC analysis to assess the expression of N-Cad, NF-kB, and HGF in the tumor specimens. In control mice, N-Cad was abundantly present in tumor cells. However, in GO-Y030 treated mice, N-Cad expression displayed pyknotic nuclei, whereas cells with retained N-Cad expression had larger nuclei (Figure 7C and D). In the tumor cells of control mice, we observed NF-kB p65 in the cytosol. In the tumor cells of GO-Y030-treated mice, where NF-kB p65 expression had disappeared, these cells displayed pyknotic nuclei. On the other hand, cells that retained NF-kB p65 expression exhibited much larger nuclei (Figure 5E and F). In the tumor cells of control mice, HGF was abundantly observed. However, in the tumor cells of GO-Y030-treated mice where HGF expression had disappeared, these cells showed pyknotic nuclei. Conversely, cells that retained HGF expression displayed much larger nuclei (Figure 7G and H). These findings suggest that GO-Y030 may exert a similar inhibitory effect on meningioma growth in vivo by targeting N-Cad, NF-kB, and HGF.

Tissue distribution of GO-Y030

To confirm the blood transfer resulting from intraperitoneal (ip) administration of GO-Y030, we conducted an HPLC analysis. The blood concentration of GO-Y030 was detectable at 18.0 ng/mL, 46.7 ng/mL, and 58.2 ng/mL at 1 hour, 3 hours, and 6 hours after administration, respectively (Table 2). In contrast, the blood concentration resulting from a one-shot intravenous administration of the water-soluble form of GO-Y030 became negative after 6 hours (data not shown). The blood concentration of GO-Y030 following ip administration remained detectable even at 6 hours after injection, which could be attributed to the differing route of administration. Subsequently, we examined the brain distribution of GO-Y030. In brain tissue, GO-Y030 was detected at 0 pg/mg, 26.8 pg/mg, and 23.1 pg/mg at 1 hour, 3 hours, and 6 hours after administration, respectively (Table 2). GO-Y030 was found in brain tissues containing cerebral blood flow. To confirm whether GO-Y030 can migrate to brain tissues and exert its antitumor effect like GO-Y022, it is necessary to establish and verify a brain tumor model using malignant meningioma[35]. It is also very important to examine the anti-invasion and anti-metastatic abilities of curcumin analogs. We will plan it in the near future.

Table 2 In vivo distribution of GO-Y030.

Times after administration (hour)	1	3	6
Blood concentration (ng/mL)	18.0	46.7	58.2
Brain distribution (pg/mg tissue)	0	26.8	23.1

DISCUSSION

Malignant meningioma has shown resistance to many drugs, with only a limited number demonstrating partial response (PR) rates exceeding 10%. Some of these drugs include oral tamoxifen (PR rate: 5%-16.66%), oral mifepristone (1.4%-30.76%), intravenous cyclophosphamide + doxorubicin + vincristine (21%), and oral hydroxyurea (0%-10%)[5,6,7,8]. These outcomes remain inadequate. Ongoing clinical trials for malignant meningioma focus on a range of potential treatments, such as SMO inhibitors like sonidegib[67] or vismodegib[68], RAS/RAF/MEK/MAPK pathway inhibitors like trametinib[69] or selumetinib[70], the PI3K/AKT/mTOR pathway inhibitor capivasertib[24], the CD4/6 inhibitor abemacicilib[25], the FAK inhibitor GSK2256098[23], the EGFR inhibitors like afatinib[19] or brigatinib[71], VEGF receptor 2 inhibitors such as apatinib[20] or carbozantinib[21], the HDAC inhibitor AR-42[26], the GSK-3β inhibitor 9-ING-41[27] and TNF-related apoptosis-inducing ligand ONC206[72]. Many targeted drugs are currently being investigated due to the complex molecular biology of meningioma. Various mutations accumulate, leading benign meningioma to transform into malignant forms. Without identifying an Achilles heel, it could be challenging to control malignant meningioma using a single targeted agent. In this respect, a multi-targeted agent is believed to offer an advantage in managing malignant meningioma. However, until now, apart from sunitinib, no multi-targeted agents have demonstrated effectiveness against meningioma[28]. Sunitinib exhibited a PR rate of only 5.55% and a 6-month progression-free survival rate of 42%[73]. Curcumin is a phytochemical with numerous pharmaceutical properties known to influence several cancer-relating target molecules. However, curcumin’s efficacy is relatively modest, and it suffers from poor bioavailability. Although no clinical trials have been conducted on the effects of curcumin on meningiomas, its potential has been demonstrated at the cellular level and in animal models. GO-Y030 showed 10 to 16^th times stronger growth inhibitory activity against meningioma cell lines. In addition, in an animal model, GO-Y030 significantly inhibited the growth of subcutaneously implanted meningiomas compared to controls when administered intraperitoneally (2 mg/kg, daily) (P = 0.002). GO-Y030 administered intraperitoneally was detected in the blood as early as 1 h after administration, and it was detected in brain tissue at 3 hours after administration, indicating the possibility of crossing the BBB. Furthermore, we have now succeeded in creating a solubilized GO-Y030 compound[38], and we expect to improve its in vivo usefulness, such as by administering it intravenously. In addition, the compound has been shown to have the ability to control multiple targets such as N-Cad, NF-kB, and HGF, making it a promising therapeutic agent for meningiomas.

There have been no reports of clinical trials involving curcumin for treating malignant meningioma. Over 200 C5 curcuminoids have been synthesized[36,40], several of which exhibit superior anti-tumor activities. These curcuminoids function via a variety of modes of actions. In this study, GO-Y030, a C5 curcuminoid, demonstrated almost 10 times higher anti-tumor potential than curcumin. This effectiveness was also validated in an in vivo mouse model with no safety concerns. GO-Y030, administered intraperitoneally, was detectable in the blood and brain tissues. However, due to the unavailability of a suitable mouse model with malignant meningioma in the brain, we were unable to assess its effects on locally advanced brain meningioma. This study confirmed the potential of GO-Y030 in addressing systemic manifestations of malignant meningioma with metastasis.

We focused on three specific molecules: N-Cad, NF-kB, and HGF. These molecules are on meningiomas, its potential has been demonstrated at the cellular level and in animal models. GO-Y030 showed 10 to 16th times stronger growth inhibitory activity against meningioma cell lines. In addition, in an animal model, GO-Y030 significantly inhibited the growth of subcutaneously implanted meningiomas compared to controls when administered intraperitoneally (2 mg/kg, daily) (P = 0.002). GO-Y030 administered intraperitoneally was detected in the blood as early as 1 hour after administration, and it was detected in brain tissue at 3 h after administration, indicating the possibility of crossing the BBB. Furthermore, we have now succeeded in creating a solubilized GO-Y030 compound[38], and we expect to improve its in vivo usefulness, such as by administering it intravenously. In addition, the compound has been shown to have the ability to control multiple targets such as N-Cad, NF-kB, and HGF, making it a promising therapeutic agent for meningiomas.

There have been no reports of clinical trials involving curcumin for treating malignant meningioma. Over 200 C5 curcuminoids have been synthesized[36,40], several of which exhibit superior anti-tumor activities. These curcuminoids function via a variety of modes of actions. In this study, GO-Y030, a C5 curcuminoid, demonstrated almost 10 times higher anti-tumor potential than curcumin. This effectiveness was also validated in an in vivo mouse model with no safety concerns. GO-Y030, administered intraperitoneally, was detectable in the blood and brain tissues. However, due to the unavailability of a suitable mouse model with malignant meningioma in the brain, we were unable to assess its effects on locally advanced brain meningioma. This study confirmed the potential of GO-Y030 in addressing systemic manifestations of malignant meningioma with metastasis.

We focused on three specific molecules: N-Cad, NF-kB, and HGF. These molecules are known to be overexpressed during the progression of meningioma with malignant meningioma. HGF binds to its receptor, c-MET tyrosine kinase, which in turn activates downstream signaling pathways related to epithelial-mesenchymal transformation (EMT)[74]. EMT is a critical process in the early stages of cancer metastasis, and N-Cad plays an important role in it[75]. Although many researchers have attempted to understand the mechanism of action of curcumin, the full picture is still unclear. The same is true for curcumin derivatives. Recently, we have found a protein that the curcumin derivative GO-Y030 directly binds to and regulates. In our next study, we hope to clarify the relationship between the regulation of this protein and HGF, NF-kB, and N-cad. Additionally, the VEGF receptor 2 inhibitor cabozantinib is known to inhibit c-MET. Cabozantinib has already received approval for the treatment of advanced hepatocellular carcinoma[76], making the inhibition of the HGF/c-MET axis a promising strategy. Our study clearly indicated that the C5 curcuminoid GO-Y030 can effectively suppress HGF and N-Cad expression in vivo. Furthermore, the inhibition of HGF/N-Cad was associated with the reduction in meningioma tumor size. This provides strong evidence that targeting the HGF/c-MET axis in the EMT of malignant meningioma holds great promise.

On the other hand, NF-kB signaling was activated in meningiomas. Currently, there are no drugs in development to inhibit NF-kB signaling, except 9-ING-41, which suppresses NF-kB signaling via GSK-3β inhibition. We have been particularly focused on inhibiting NF-kB signaling using the C5 curcuminoid[41,42]. GO-Y030 suppresses NF-kB signaling in malignant meningioma in vivo. It also simultaneously suppresses the HGF/c-MET axis. It is challenging to determine which of these inhibitory effects contribute to tumor shrinkage, as NF-kB signaling and the HGF/c-MET axis are different pathways. Since there are no direct interactions between these pathways, it is possible that the multi-target potential of GO-Y030 may be producing a synergistic or additive effect in this context. GO-Y030 exhibits anti-invasion and anti-metastatic potential[51] and can influence the generation of regulatory T cells, enhancing anti-PD-1 cancer immunotherapy[50]. We plan to explore these potentials in the context of malignant meningioma with GO-Y030 in the near future. C5 curcuminoid is derived from natural sources. Additionally, it is worth mentioning the 2015 Nobel Prize awarded to Campbell and Omura and Tu for their discoveries of avermectins and artemisinin, respectively[77]. These natural products and their analogs have also made significant contributions to cancer pharmacotherapy. Recently, there has been an interest in natural products as potential leads for drug development[78]. However, these results are limited in the pilot study, and it is needed to be further examined by large-scale efficacy and safety preclinical studies. The permeability through the BBB has only been analyzed in silico, and safety studies as well need to be performed in the near future with regard to many parameters other than body weight. For clinical application, efficacy and safety must be examined using multiple animal species, including humans. Curcumin is a compound derived from food, and its derivatives are expected to have low toxicity. In the future, it will be necessary to verify whether combination therapy with other antitumor compounds is possible.

CONCLUSION

GO-Y030 is a potent compound to inhibit malignant meningioma through HGF, NF-kB, and N-Cad inhibition, resulting in apoptosis induction.

ACKNOWLEDGEMENTS

All authors have made substantial contributions, and all authors are in agreement with the manuscript’s content. We thank the Bioscience Educational and Research Support Center, Akita University, for their assistance with histopathological analyses. Special thanks to Maya Chiba and Ikuko Sakaki for their valuable technical support.