Literature data support evidences that glioblastoma (GBM) patients experience prolonged survival due to sodium valproate (NaVP) treatment. The study assessed the human GBM cell U87 xenograft studied in the chicken embryo chorioallantoic membrane (CAM) model evaluating NaVP effect on tumor. Three groups of tumors (each
Glioblastoma multiforme (GBM) is the most frequent, highly recurrent, and rapidly progressing type of astrocytic brain tumor in adults [
The mechanisms of NaVP without an antiepileptic activity are the known inhibitor of histone deacetylase [
Further studies of GBM markers are needed to know how NaVP regulates tumor growth in experimental models. Polycomb group proteins (PRC1 and PRC2) regulate the chromatin structure and have an important regulatory role in human malignancies and catalyze histone (H2A and H3) modifications. Studies show the role of the PRC2 catalytic component enhancer of the zeste homolog 2 (EZH2) in neoplastic development [
The tumor suppressor gene
The first-line treatment drug Temozolomide for GBM may only increase the survival of patients on average by a few months, and NaVP treatment sensitizes temozolomide-resistant glioma cells [
The main objectives which indicate the novel and original aspects of the current study were to compare the evaluated characteristics of the tumor growth, penetration to the CAM mesenchyme, and correlation of these phenomena with the EZH2 and p53 expression in the tumor as a response to the treatment with different NaVP concentrations in the CAM model.
Fertilized chicken eggs
Commercial human glioblastoma U87 cells line was obtained from the Institute of Neuroscience (Kaunas, Lithuania) and kept in Dulbecco’s modified Eagles medium (DMEM) (Gibco, USA) supplemented with 10% fetal bovine serum (Gibco, USA) and with 100 IU/mL of penicillin and 100
The NaVP-treated and control specimens were collected after 5 days of incubation on the 12 days of embryo development (EDD12), fixed in a buffered 10% formalin solution for 24 h and then paraffin-embedded.
Each embryo of different experimental groups was sacrificed and a CAM was removed, fixed in 10% neutral-buffered formalin, dehydrated, and embedded in paraffin. Serial sections of 3
CAMs with grafted U87 cells were registered in vivo daily under a stereomicroscope (SZX2-RFA16, Japan) equipped with an Olympus DP72 camera for both video recordings and acquiring still images. CAMs were investigated from day 2 (EDD9) after grafting to day 5 (EDD12). Histological slides were investigated under a light microscope Olympus BX40F4 (
Paraffin blocks of CAM with grafted tumors were cut into 3
Serial histological sections of the experimental tumors and CAM were performed to evaluate tumor invasion. According to tumor behavior on the membrane they were distributed into three groups: (1) tumor was formed on the surface of the CAM, chorionic epithelium was not destroyed, and there was no invasion into the mesenchyme; (2) chorionic epithelium was destroyed and cells invaded the mesenchyme, but part of the tumor remained on the membrane surface; (3) tumor cells destroyed the chorionic epithelium, invaded the mesenchyme, and were completely surrounded by the mesenchyme.
Data presented as mean and standard deviation. Data were compared using Tukey’s test applying one-way ANOVA. Statistical package SPSS 20.0 was used. Difference was considered as significant when
The U87 cells tumor xenografts inoculated on egg CAM and photographed daily via the shell window on days 9–12 of embryo development (EDD9–12) are shown in Figure
Tumor xenograft on CAM of nontreated and NaVP-treated U87 cells during days 9–12 of embryo development. Figure
Nontreated
NaVP-treated (4 mM)
NaVP-treated (8 mM)
The CAM with U87 cell tumor at day 12 of embryo development. CE: chorionic epithelium, AE: allantoic epithelium, BV: blood vessels, and T: tumor. (a) Control membrane without tumor. (b) Membrane with tumor of U87 cells without treatment. Arrowheads show the growth of chicken blood vessels into the tumor, and a dotted arrow shows the chorionic epithelium destroyed by tumor cells. (c) Membrane with tumor developed from U87 cells treated with 4 mM NaVP. The dotted arrow shows the destroyed chorionic epithelium and partial tumor invasion into the mesenchyme. (d) Membrane with a tumor developed from U87 cells without treatment. The tumor is completely encapsulated into the CAM mesenchyme. (e) Membrane with tumor developed from cells treated with 4 mM of NaVP, with the intact chorionic epithelium; the tumor is not vascularized and located on the surface of the CAM. (f) Tumor developed from cells treated with 8 mM of NaVP, located on the surface of the membrane, not vascularized. Scale bar: 200
Images of H–E-stained slides of control CAM, the CAM with nontreated U87 cell tumors, and CAM with tumors of U87 cells treated with 4 mM and 8 mM NaVP are presented in Figure
Of the tumors developed from nontreated U87 cells, 50% destroyed the chorionic epithelium, invaded the chorioallantoic membrane mesenchyme, and had been completely formed in the mesenchyme; 40% of the tumors destroyed the chorionic epithelium and invaded the mesenchyme but partly remained on the CAM surface, and only 10% of tumors were formed on the surface of the CAM without invading the mesenchyme. When U87 cells were treated with 4 mM of NaVP, 60% of tumors were formed on the surface of the CAM and 30% destroyed the chorionic epithelium and invaded the mesenchyme, but part of the tumor remained on the surface of the CAM. In this group, 10% of tumors were observed only in the mesenchymal layer of the CAM. When U87 cells were treated with 8 mM of NaVP, 90% of the tumors were formed on the surface of the CAM without invading the mesenchymal layer, the chorionic epithelium was also not destroyed, and only 10% of cases showed a complete invasion into the mesenchyme. Compared with nontreated tumors, the incidence of tumor formation on the CAM surface without invading the mesenchyme was significantly higher when U87 cells were treated with 4 mM and 8 mM of NaVP (10%, 40%, and 90%, resp.;
Frequency of U87 cell tumor groups with invasion into the CAM mesenchyme and localized only on the CAM. The investigated groups were U87 cell tumors nontreated with NaVP (
In tumors formed by the U87 cells without treatment a high expression of the EZH2 protein was found, and positively stained cells were distributed in all tumor (Figure
Expression of EZH2 and p53 proteins in tumors formed by U87 cells treated with 4 mM and 8 mM of NaVP and by nontreated cells. (a) EZH2 expression in nontreated tumors, (b) EZH2 expression in tumors treated with 4 mM of NaVP, (c) EZH2 protein expression in tumors treated with 8 mM of NaVP. (d) p53 protein expression in tumors formed by nontreated U87 cells, (e) p53 expression in tumors treated with 4 mM of NaVP, and (f) p53 protein expression in tumors treated with 8 mM of NaVP. (g) EZH2-positive cells and (h) p53-positive cell tumors developed from U87 cells in tumor groups without treatment and from cells treated with 4 mM and 8 mM of NaVP, respectively. Arrows indicate EZH2- and p53-positive cells.
In tumors formed by nontreated U87 cells, a high expression of the mutant p53 protein was found (Figure
GBM is the most lethal form of cancer with a median survival of up to 12 months [
In vivo experimental studies demonstrate that the CAM model allows a successful testing of investigational medicinal products. The CAM model has been used to implant several malignant cell lines to investigate tumor growth and the metastatic process, angiogenic potential, identifying therapeutic targets, and evaluating antitumor drugs, as these transplanted cells on CAM keep producing human antigens [
The study biomicroscopy follow-up of control-nontreated and NaVP-treated U87 cell tumors on CAM shows a progressive growth and vascularization. The majority of tumors which developed from U87 cells without NaVP treatment during tumor growth destroyed the chorionic epithelium and invaded the mesenchyme. Biomicroscopy showed that the nontreated xenograft of U87 cells on CAM invaginated into the underlying mesenchyme starting from day 2 after inoculation. The study shows that increasing the NaVP concentration diminished the U87 cell capability to destroy the chorionic epithelium and significantly increased the number of tumors formed only on the surface of the CAM without invading the mesenchyme. The incidence of tumor formation on CAM without invasion into the mesenchyme was significantly higher when U87 cells have been treated with NaVP, and this effect significantly increased with the NaVP concentration. NaVP is a histone-deacetylase (HDCE) inhibitor and specifically inhibits HDAC classes I and IIa [
The U87 tumors on CAM in relationship with tumor progression concomitantly induced blood vessel angiogenesis, and a clearly expressed spoked-wheel pattern was observed on days 4 and 5 after grafting in nontreated tumors. The treatment of U87 cells with 8 mM of NaVP did not show clear dynamics of tumor growth during 5 days of development at EDD9–12; at the same time, the blood vessel angiogenesis failed. Other researchers reported that the progression of the U78 tumor was related to angiogenesis which indicates a glioma development relationship with growth factor receptors [
The studied GBM U87 cell tumor samples showed a strong nuclear staining of EZH2 in U87 cell tumors without NaVP treatment, and a high expression of the EZH2 protein in the tumor cell nuclei was found. It is known that increased EZH2 expression correlates with glioma grade and its recurrence, suggesting that EZH2 could be a marker of glioma aggressiveness and correlate with a decreased GBM patient survival, and the EZH2 protein was found to be strongly expressed in U87 cell lines. EZH2 is present only in dividing cells [
In our study, the expression of EZH2 protein was observed also in the chorionic and allantoic epithelium and CAM mesenchyme in nontreated with NaVP tumors. This positive staining may be associated with the fact that the EZH2 protein participates in embryo development: the EZH2 knockdown was shown to be embryo-lethal in mice [
We found that the number of EZH2-positive cells was significantly lower in tumors treated with 4 mM and 8 mM of NaVP as compared with the cell count in nontreated tumors, and the NaVP effect was higher as its concentration increased to 8 mM. It was reported that the inhibition of EZH2 may be a potential therapeutic strategy to target GBM proliferation, migration, and angiogenesis as the inhibition of EZH2 in vitro by pre-miR-101, EZH2 siRNA, or small molecule DZNep attenuated GBM cell growth, migration/invasion, and GBM-induced endothelial tubule formation in a U87-Fluc-mCherry GBM xenograft mouse imaging model resulted in a reduced tumor growth and migration/invasion. A significant correlation between the expression of 28 out of 279 genes associated with cell migration and EZH2 expression was observed [
In the nontreated U87 cell tumors, the EZH2 expression in cells has been very strongly expressed in tumor areas which invaded across the chorionic membrane to the mesenchyme and were surrounded by the thickened mesenchyme. In this respect, of significance could be the results of other studies indicating that tumor infiltrating front, molecular heterogeneity in GBM can improve the rationale of potential molecular targets. The key genes involved in gliomas tumor cell proliferation, invasion, migration, response to immune system, and stemness markers are highly enriched in the peritumoral brain zone, and these genes probably contribute to the resistance of cells to standard therapy, resulting in a tumor recurrence [
The study shows that, in tumors formed by the U87 cells without NaVP treatment, a high expression of the p53 protein was found. The p53 is a tumor suppressor gene implicated in the genesis of malignancies. The overexpression of the p53 protein is often used as a surrogate indicator of mutations in the p53 gene [
The p53 abnormalities are common in the progression from a low-grade lesion to a high-grade lesion of GBM in patients [
Our study revealed a high expression of p53 in U87 cell tumors which were not treated with NaVP, and this expression was accompanied with pronounced angiogenesis in tumors in the CAM model. The p53 distribution in the tumor was not related to tumor periphery or vascularization areas. Other investigators also found that the pattern of p53 expression was not related to a particular region, such as the infiltrating edges or vascularization areas in GBM of patients [
In our U78 tumors on CAM treated with NaVP, the number of p53-positive cells significantly diminished, and only a significantly decreased number of p53-positive cells in tumors treated with 8 mM of NaVP were observed. By others, the mechanistic insight shows that p53 directly interacts with the antiapoptotic proteins bcl-xL and bcl-2, and mutant p53 proteins could be related to this binding [
The study results demonstrate that the CAM model allows a successful testing of anticancer drugs designed to interfere with the p53 and EZH2 molecular pathways important for glioma progression. The experimental findings of the study indicate that NaVP has a function in blocking the proliferation, migration, and angiogenesis of human U87 glioma cells in the CAM tumor model, thereby supporting the NaVP potential in glioblastoma therapy. However, the U87 cell model system used in the study has limitations due to the known high cell line mutability and variability, but the study results are important for further studies to evaluate the NaVP effect on the p53 and EZH2 expression of different (pediatric and adult) cell and brain tumor stem cell lines as well as the primary GMB tumor cell cultures.
The authors declare no conflicts of interests.
The present study was supported by the Lithuanian University of Health Science (Grant no. V–1238).