New therapeutic target for pediatric anaplastic ependymoma control: study of anti-tumor activity by a Kunitz-type molecule, Amblyomin-X

EPNs comprise a heterogeneous group of neuroepithelial tumors, accounting for about 10% of all intracranial tumors in children and up to 30% of brain tumors in those younger than 3 years. Actually, the pattern therapy for low-grade EPNs includes complete surgical resection followed by radiation therapy. Total surgical excision is often not possible due to tumor location. The aim of this study was to evaluate, for the first time, the anti-tumor activity of Amblyomin-X in 4 primary cultures derived from pediatric anaplastic posterior fossa EPN, Group A (anaplastic, WHO grade III) and one primary culture of a high grade neuroepithelial tumor with MN1 alteration, which was initially misdiagnosed as EPN: i) by in vitro assays: comparisons of temozolomide and cisplatin; ii) by intracranial xenograft model. Amblyomin-X was able to induce cell death in EPN cells in a more significant percentage compared to cisplatin. The cytotoxic effects of Amblyomin-X were not detected on hFSCs used as control, as opposed to cisplatin-treatment, which promoted a substantial effect in the hAFSCs viability. TEM analysis showed ultrastructural alterations related to the process of cell death: mitochondrial degeneration, autophagosomes and aggregate-like structures. MRI and histopathological analyzes demonstrated significant tumor mass regression. Our results suggest that Amblyomin-X has a selective effect on tumor cells by inducing apoptotic cell death and may be a therapeutic option for Group AEPNs.


Results
EPN cells (n = 5) were previously characterized by pathological and molecular assay and one case was rediagnosed as high-grade neuroepithelial tumor with MN1 alteration; in the sequence, research was continued with the investigation by following approach: Amblyomin-X induces cytotoxicity in epN cells. Our first stage was to estimate Amblyomin-X effects in vitro, by comparing its viability's effect in the EPNs primary cell culture and hAFSCs culture as normal cells control. Both EPNs and hAFSCs cells had a time-dependent and Amblyomin-X concentration-dependent (Fig. 1B,D).
Morphology study of EPN cells treated 24 h with Amblyomin-X shown minor changes and cell adhesion damage (Fig. 1B). Cells treated in forty-eight hours were certified as cell contraction, adhesion damage, inter-cell elongations loss and dispersed cell aggregates formation (Fig. 1D.q,r). EPN control cell (treated with vehicle in the same experimental conditions), exhibited a homogeneous culture distribution on the plate, exhibited a fusiform arrangement and multidirectional bundles set ( These results corroborated by cells viability assays, which demonstrate that after 24 h (Fig. 1A) and 48 h (Fig. 1C) of treatment with Amblyomin-X (10 µM and 20 µM) the EPNs cells display approximately 80% and 55% of viability, respectively. Cytotoxic effects of Amblyomin-X observed in EPN cells were more significant than observed after incubation with temozolomide for 48 h, 90% of viability.
Amblyomin-X cytotoxic effects were reproduced in five samples used in study: four EPN cells lines and CNS high grade neuroepithelial tumor with MN1 alteration, in which we choose to keep in the study considering the www.nature.com/scientificreports www.nature.com/scientificreports/ important effect of Amblyomin-X on high grade tumor cells by induce apoptotic cell death. It is worth mentioning that similar results were observed when we used cells derived from a neuroepithelial tumor that can be easily misdiagnosed when the molecular and epigenetic characterization is not performed.
Cell apoptosis analysis by Flow Cytometry. Induction of apoptosis after treatment with Amblyomin-X (10 and 20 µM) and cisplatin (20 µM) was evaluated in EPNs and hFSCc for 24 and 48 h by FITC Annexin V/7AAD.
Our results show that Amblyomin-X induced cell death in EPN cells and, more significantly, were not observed cytotoxic effects in hFSCs ( Fig. 2A), in which were used as normal control.
However, the treatment with cisplatin (20 µM) affected hFSCs viability considerably. After 24 h and 48 h of treatment with cisplatin (20 µM) the cells display approximately 90% and 70% of viability, respectively (Fig. 1A,C); and the percentages of cell death obtained after 48 h of treatment were the follow: 15,2% early apoptosis, 16,2% late apoptosis or necrosis and 5,64% death ( Fig. 2A). Untreated EPN cells analysis presented the following cytological description, spindle morphology and oval nuclei (Fig. 3a). The cell nucleus contains heterochromatic bundles and nucleoli (Fig. 3a). EPN cells exhibited significantly the presence of mitochondria with evident crests (Fig. 3h) and rough endoplasmic reticulum organized in tube-like structures known as cisternae (Fig. 3o).

Histological analysis of epN's-brain after in vivo
Amblyomin-X treatment. Histological, histochemical for Prussian Blue (EPN cells labeled with MION-Rh) and immunohistochemical staining, of rat brains (Fig. 5), demonstrated that EPN that has receive only vehicle exhibited high cellularity located in the third ventricle ( Fig. 5b-d), invading periventricular areas (Fig. 5a,e-h) and proliferating through surrounding regions (Fig. 5g,h). Immunohistochemical analysis for the expression of human GFAP confirmed the glial origin of tumor ( Fig. 5f), Ki67 detection shown high number of cycling cells in tumor tissue (Fig. 5g,h) and high expression of vascular endothelial growth factor (VEGF) in tumor.
Analysis of rats that received Amblyomin-X (1.0 mg/kg/day by intra-peritoneal injection) treatment (

Discussion
EPNs are one of the most severe neoplasms in Pediatric Oncology because it presents an important pluripotentiality profile, which justifies the lack of efficient chemotherapy agents in the treatment. The long-term clinical outcome of pediatric intracranial EPN is poor with a high rate of recurrence. Therefore, effective therapeutic agents are urgently needed.
We evaluated, for the first time, the potential antitumor therapeutic application of Amblyomin-X in EPN primary culture. www.nature.com/scientificreports www.nature.com/scientificreports/ Amblyomin-X is a recombinant protein identified on cDNA library from salivary glands of Amblyomma cajennense tick, which have shown antitumor, antiangiogenic and anticoagulant properties 18 .
Herein, Amblyomin-X treatment was able to induce cell death on EPN cells, but, excitingly, it did not decrease the viability of normal cells, preserving their original morphological characteristics of fibroblast-like cell type. On the other hand, the morphological and viability analysis of hAFSCs under temozolomide or cisplatin treatment revealed significant changes induced by cytotoxic effects, such as, dispersed cell aggregates formation and inter-cell elongations loss.
Concerning selective potential of Amblyomin-X by tumor cells, our results corroborated previous data, which demonstrate that Amblyomin-X acts on tumor cells with low toxicity in normal cells 17 . In addition, this great advance was confirmed for EPN's, where the compared treatment with other current drugs preconized its tumor was superior [19][20][21][22] . For example, Phi et al., 23 have been used, as cells control, neural stem cells (NSCs) derived from the neocortex and immortalized astrocytes of the treatment with STAT3 inhibitors, which exhibit low viability compared to our results, i.e. 90-95% of hAFSCs viability.
Rudà et al., 22 described temozolomide as recover treatment for recurrent intracranial EPNs; however, complete response was achieved in only 5% of the patients and with median progression-free survival of 9.69 months. Platinum-based regimens is another treatment option, yet limited, in EPNs 21 .
In our studies, morphological analysis of EPNs cells after 24 h of Amblyomin-X treatment displayed minor changes or cell adhesion damage. Cells treated in forty-eight hours were certified as cell adhesion damage, dispersed cell aggregates formation and inter-cell elongations loss, compared to EPN control cell (vehicle). Amblyomin-X was able to reduce EPN cell viability in a time and dose-dependent manner. These results indicates that Amblyomin-X is more effective against EPNs than temozolomide or cisplatin.
In addition, EPNs cells morphological aspects changes, the phosphatidylserine exposition and DNA degradation were evaluated by FITC Annexin V/7AAD assay, after treatment with Amblyomin-X. We found that Amblyomin-X was able to induce apoptosis selectively in the EPNs cells, in accordance with the findings studies in renal cell carcinoma (RCC) 24,25 .
MTT assay results and flow cytometry corroborate with ultrastructural investigation, which demonstrate various alterations involved on cell death process, i.e: pyknotic-like nucleus, degenerated aspect of numerous mitochondria, rough endoplasmic reticulum with disarranged cisternae, autophagosomes or autophagy corpuscle, late endosomes, aggregate-like structures and aggresomes. Another studies using Amblyomin-X treatment also revealed, by TEM, spherical-type aggresome formation and alteration in rough endoplasmic of the tumor cell lines 14,24 .
Preclinical evidence 25,26 has shown that aggresomes and the autophagy pathway have a crucial compensatory role in the protein-clearance mechanisms that eradicate potentially toxic proteins to promote resistance to proteasome inhibitors and, hence, tumor survival.
Amblyomin-X was able to recognizing renal tumor mass in animal bearing and remained for 3 times more than healthy animal, which has excreted via renal system 25 . Additionally, in healthy animal were not observed cytotoxic effects when used the concentration 100 fold more in acute dose-Amblyomin-X treatment during preclinical assays 25 . Several previous data have evaluated the Amblyomin-X action on process of tumor cell death with modulations related to the cell cycle arrest and inhibition of the ubiquitin-proteasome system, disturbing its catalytic activity and leading to cell death via apoptosis [13][14][15][16][17]26 . Ours hypothesis could be that the same mechanism occurs in EPN cells. The present data concerning EPN's-Amblyomin-X treated, have demonstrated series of intracellular events linked to cytotoxic effects leading to tumor cell death by induction aggresomes formation, as a possible non-exclusive ubiquitin pathway 27,28 . The power anticancer activity of Amblyomin-X on EPN cells compelled us to examinate this compound effectiveness in vivo, applying the established xenograft models 7 .
The intracranial xenograft model built from pediatric EPNs has showed an important approach to preclinical drug discovery 7 . Fluorescence, X-Ray and MRI detected tumor growth progression.
After 21 days of the daily treatment of EPN xenograft model with Amblyomin-X (1 mg/kg), by intra-peritoneal injection, we observed a significant tumor mass regression compared to the control (vehicle), in which was investigated by fluorescence, X-Ray, MRI and histological analysis. www.nature.com/scientificreports www.nature.com/scientificreports/ MRI assay was extremely important to control tumor growth; T 2 -weighted images showed hypointense zones in third ventricle, with subsequent ventricular dilatation, in which were more evident in the T 2 * -weighted images in the same area. These results revealed tumor mass regression on the side of histopathological test, which confirmed reduction of dilatation of the third ventricle with decreased tumor cells proliferation in the same area.
These results are in accordance with the in vitro assays, were Amblyomin-X treatment have showed cytotoxic activity on EPN cells, induce cell death by apoptosis and a slight or no activity. Concluding, our results showed that Amblyomin-X acts selectively on EPN cells whereas it as a safety-potential anti-EPNs drug candidate.
Certainly, it would be of great importance the DNA methylation array study in molecular subtypes of EPN in both cell lines and tissue samples in vivo, building thus, a comparative profile among these subtypes and treatment with Amblyomin-X. Future trials will be developed, so we could lead to more precise prognostic assessments.

Materials and Methods
Five human anaplastic EPNs (WHO grade III), located in the PF from 1-10 year-old patients, were submitted to resection at Pediatric Oncology Institute, Grupo de Apoio ao Adolescente e à Criança com Câncer (GRAACC), Department of Neurology and Neurosurgery, Federal University of São Paulo(UNIFESP), São Paulo, Brazil. All parent and/or legal guardians signed an informed consent for the study and all methods involving human samples were performed in accordance with the relevant guidelines and regulations (Ethical Committee in Research of Federal University of São Paulo_UNIFESP).
Recently, EPNs were classified according to the DNA methylation profile 6 , such classification showed to be increasingly dependent on molecular biology findings 29 . We highlight the WHO revised 2016 classification of CNS tumors 4 , which incorporates, for the first time, genetic information in addition to morphology for the classification of many tumor entities 30,31 .
As follows, both the pathologic anatomy and molecular diagnostics of EPNs were described in the Table 1. We performed molecular diagnostics through the methylation profile using Illumina DNA methylation arrays according Capper et al. 32  Collection Amniotic Fluid (AF) and culture of hAFsCs. We used hAFSCs as control samples, since this cell type is one of our ongoing scientific research projects. hAFSCs have been classified as a novel type of broadly multipotent/pluripotent stem cells sharing characteristics of both embryonic and adult stem cells 33 .
All pregnant women signed an informed consent for the study and all methods involving human samples were performed in accordance with the relevant guidelines and regulations (Ethical Committee in Research of Federal University of São Paulo_UNIFESP). Fetal AF (40 mL each) was obtained from six pregnant women with fetuses undergoing repair of Myelomeningocele (MMC), with gestational age of 26 weeks.
Amblyomin-X. De Souza et al. 25 described Amblyomin-X, a 15 kDa protein, obtained in a recombinant form. www.nature.com/scientificreports www.nature.com/scientificreports/ Cytotoxicity assays. Cell viability. The 3-(4,5-dimethylthiazol-2-yl)−2,5-diphenyltetrazolium bromide (MTT) assay was applied to assess the cell viability on primary cultures of EPNs (1 × 10 4 cells/well) and hAFSCs (2 × 10 4 cells/well) in the plate of 96 wells,according 12  transmission electron Microscopy (teM). EPNs primary culture establishment. was performed using ACLAR ® film. In the sequence, the cells were treated with Amblyomin-X [1 µM] for 48 h. EPN cells adhered on ACLAR ® film were fixed in 1% glutaraldehyde and processed according to Pavon et al. 7 . Semi and ultrathin sections were obtained with the aid of a Porter Blum ultramicrotome. The ultrathin sections (70 nm) were placed on copper grid, which were photographed under a TEM (Philips CM100). 10 3 EPNs cells were plated in 24-well plates and incubated overnight, (for approximately 18 hours at 37 °C, 5% CO 2 ) in DEMEM-LG medium, with 40 μg Fe/mL MION-Rh. After incubation, the culture medium solution was removed and the cells were washed twice with PBS (1×) to remove extracellular MION-Rh. The labeled cells were treated with 0.25% Tryple Express, and then harvested and manually counted using 0.4% Trypan Blue (Gibco/Invitrogen Corporation) and fixed with 4% paraformaldehyde. Next, fluorescence analysis were done using diamidino-2-phenylindole (DAPI, Sigma-Aldrich) to label the cell nuclei and an Rh-B filter (530 nm and 550 nm) to detect the MION-Rh, according Pavon et al. 7 . The Intracellular detection of MION-Rh in labeled EPNs cells were performed using a fluorescence microscope (IX51 Olympus, Tokyo, Japan). After skin incision on the dorsal region of skullcap, bone cap trepanation was made using a dental drill. According to Swanson's Stereotaxic Atlas guidelines (1992), the implantation position was determined following coordinates: 0.8 mm anteroposterior, 1.4 mm laterolateral, and a depth of 3.8 mm, as described before 28 . Hamilton syringe was used to implant 10 4 EPN cells in 10 µL of culture medium into the third ventricle. Tumor development was monitored over 45 days, according Pavon et al. 7 .

Animal ethics statement.
In vivo image of tumor progression tracking. Tumor development were monitored using an in vivo imaging device, Bruker model MS FX PRO (Bruker, Ettlingen, GE). Throughout image acquisition, animals were placed in dorsal recumbence and remained anesthetized with inhaled 2% isoflurane in oxygen at 2 L/min. Initially, the skull images were acquired by X-ray. The fluorescence of the labeled cells was evaluated using the excitation (540 nm) and emission (585 nm) of MION-Rh. The images were acquired and evaluated using multiplex location software. The images were acquired and evaluated using multiplex-located software, according Pavon et al. 7 .
Histological analysis of tumor tissues. After image acquisition, the animals were anesthetized and transcardially perfused with a buffered saline solution and 4% paraformaldehyde (PFA). The brains were removed and stored in PFA for 24 hours; they were then cryoprotected in a 40% sucrose solution for 48 hours. Subsequently, 40μm thickness coronal sections were cut using a cryostat (Leica) and stained using standard procedures for hematoxylin-eosin, Prussian Blue staining and for immunohistochemical staining was used human primary antibodies included the following: glial fibrillary acidic protein (GFAP) (1:200), Ki-67 (1:20) and VEGFR-1 (1:150) (Epitomics, Inc.), according Pavon et al. 7 .