JONT

Introduction

Carcinogenesis is a broad process of cellular transformation that stems from variegated genetic alterations, including loss and gainof-function through driver mutations that can ordinarily occur in one single cell [1, 2]. Thereafter it ensues a stepwise sequence of cellular changes from initiation, promotion, and progression that ultimately translate in supraphysiological capabilities for survival, proliferation, dissemination, dormancy, and recurrence [3, 4]. All these processes are also led by an extensive and continuous reediting process of the epigenetic code, rendering a particular cancer landscape with wide plasticity [5]. Immortality, metabolic reprogramming, immune-edition, and autonomy are also hallmarks of malignant transformed cells [4, 6, 7]. Furthermore, epigenetic rearrangements appear to drive cancer cells events of dedifferentiation and trans differentiation, distancing the cells from their original histotype [8, 9]. These events have frequently puzzled pathologists for years given the cancer markers of atypia, pleomorphism, and morphological heterogeneity [10], all of them, resulting from a cellular adaptative response to microenvironmental pressures. These outstanding hallmarks [11, 12], have poised cancer for centuries as the emperor of maladies [13].

In order to asses our hypothesis on the existence of a “pathologic cellular memory” behind the clinical course and phenotype of chronic noncommunicable diseases, we undertook a decade ago the investigation of the concept of diabetic metabolic memory transmissibility, and the subsequent onset of diabetic complications in healthy animals [14]. The inaugural study demonstrated that histopathological traits of diabetic skin and ulcers granulation tissue, could be faithfully recreated in rats through the administration of cell-free filtrates (CFFs) prepared from human fresh diabetic tissue samples [14]. Thereafter, we reproducibly demonstrated that CFFs derived from human pathologic tissue surgical samples, contain “chemical codes” of other forms of non-communicable diseases as arteriosclerosis [15], and cancer [16, 17], and that these signallers could be extracted, passively transferred to otherwise healthy animals, and accordingly recapitulate the histopathological hallmarks of the human donor´s condition [18]. We showed that malignant tumors-derived CFF is able to induce premalignant and malignant changes in different organs, in a relatively narrow temporary window in nude mice [17]. The findings of a subsequent independent and extemporaneous study included the onset of lung, and pancreatic adenocarcinomas, thyroid, renal, and skin carcinomas, an undifferentiated sarcoma, and a variety of preneoplastic changes using different tumor histotypes for the CFF elaboration [16]. These experiments suggested that a malignant transformation code exists, and that this tumorigenic message disrupted cellular proliferation and differentiation programs of normal, mature populations of epithelial and mesenchymal cells in otherwise normal nude mice. More interestingly is the finding that all these changes occurred irrespective to the natural interspecies barriers, suggesting that the cancer genetic and/or epigenetic signallers may be horizontally transferred from eukaryotic-toeukaryotic in mammals’ organisms [16, 17].

Danio rerio, commonly known as zebrafish (ZF), is an animal model first used in research in the early 1980s for the study of developmental biology, and subsequently introduced for cancer research as early as 1982. With the advent of transgenic and mutant models through relative technological simplicity, zebrafish has become a precious biomodel for the study of cancer biology, as for the preclinical evaluation of its pharmacological candidates [1921]. A variety of experimental carcinogenesis approaches spanning from chemicals carcinogens, xenotransplantation of human tumor cells, up to fine and selective genetic manipulation have been successfully implemented in zebrafish [22-25]. No previous study has examined however, the outcomes in ZF resulting from administering an acellular homogenate elaborated from the tissue of a human malignant tumor. We consequently interrogated whether the events of malignant transformation observed in nude mice and possibly due to tumors-derived oncogenic primers, could provoke cancer in an aquatic, non-mammalian specimen. It is shown here that the CFF elaborated from a lethal pleomorphic sarcoma and administered for a few weeks, accounted for a collection of premalignant and malignant changes in the recipient fishes.  

Materials and Methods

Ethics

All the animal procedures were conducted according to local and International Guiding Principles for Biomedical Research. This study was developed under a protocol approved by the Institutional Animal Care and Use Committee of the Center for Genetic Engineering and Biotechnology, and the Center for Investigation and Medicines Development (CIDEM), which is the venue of the ZF aquaculture laboratory. 

Preparation of cell-free filtrates:

The collected sarcoma surgical samples were allowed to thaw and weighed, and approximately 100 mg of wet tissue placed in a 2-mL vial containing 1 mL of normal saline, which was homogenized using a TissueLyser II for 3 min at 30 revolutions/second. The sample was then centrifuged at 10,000 rpm for 10 min at 4°C, sterilized by filtration using 0.2-μm nitrocellulose filters (Sartorius Lab Instruments), aliquoted into sterile Eppendorf vials, and stored at −70°C. As for all the previous studies conducted so far, protein concentration was used as the arbitrary unit of measurement to prepare and administer the inoculums.

CFF biochemical characterization

All the biochemical parameters were determined by spectrophotometric methods using commercial kits. Proinflammatory markers included C-reactive protein (CRP) [C Reactive Protein (PTX1) Human ELISA Kit, Abcam, Cambridge, United Kingdom], interleukin (IL)-1β (IL-1 beta Human ELISA Kit, Abcam, Mass, United States), IL-6 (IL-6 Human ELISA Kit, Abcam), and tumor necrosis factor α (TNFα) (Human TNF alpha ELISA Kit, Abcam). Oxidative stress markers included malondialdehyde (MDA) [Lipid Peroxidation (MDA) Assay Kit, Abcam] and H2O2 (Hydrogen Peroxide Assay Kit, Abcam). Additionally, Sirtuin-1 (SIRT1) levels were determined using the ELISA Kit for SIRT1 (Cloud-Clone Corp., Houston, Texas, United States). In all cases, the manufacturer’s instructions were followed.

Zebrafish husbandry

Wild-type adult ZF were maintained following the established protocols described in previous studies [26]. The aquarium conditions included a recirculating aquatic system with constant oxygenation, controlled environmental conditions, including a 14-hours light/10-hour dark cycle, water temperature of 28°C, and pH maintained at 7.2-7.5. The fishes were fed twice daily with commercial zebrafish food (TetraMin Flakes), and the water quality parameters regularly monitored to ensure optimal conditions for breeding and embryo development. Body weight was registered a day before the study commencement, on a bi-weekly basis, and before termination.

CFF administration to juvenile ZF

Three experimental groups of 20 ZF each were formed. One group received the sarcoma-derived CFF, while the concurrent controls received healthy donor-derived cutaneous tissue CFF, and normal saline solution. Animals received 20µg of protein in 10µL of normal saline solution from each type of tissue source. ZF were about 2-3 months post-fertilization (mpf) during the inoculation period which extended from week 5 to 10. Administrations were done twice a week with a 10µL-Hamilton syringe, through intraperitoneal injection once the animals were sedated by progressive water cooling to 8-10 degrees [27].

Animals’ termination, tissue processing, and histopathologic examination

ZF were euthanized two weeks after the end of the administration period. Gross external changes as spine deviation, brown-reddish skin macules, and abdominal dilation were individually registered. ZF were deeply anesthetized [28] with lidocaine (350mg/L) and immersed in 10% neutral formalin solution, dehydrated, and paraffin embedded. Sections at 4-5 μm thickness were generated and subsequently deparaffinized, rehydrated and stained with Mayer’s hematoxylin and eosin [29]. The slides were captured using a BX53 Olympus microscope coupled to a digital camera and central command unit (Olympus Dp-21). Histological examinations were blindly and independently performed by a skilled ZF board certified experimental pathologist and by a specialist in human anatomical pathology with documented experience in animal models histopathology. Considering that liver cancer is an umbrella term for different types of tumors, previously established histopathological criteria for ZF were consulted for the diagnosis [30, 31].

Results

1-Cells-free filtrate biochemical characterization

An elemental descriptive characterization of the sarcoma-derived CFF is shown in table 1 in reference to healthy donor skin CFF, used as a concurrent control. The biochemical characterization included a marker of lipid peroxidation (MDA), pro-inflammatory reactants (CRP, IL-6, and TNF-α), and an overexpressed growth factor in human soft tissue sarcoma (VEGF). The resulting data indicated that: (1) the accumulation of membrane peroxidated polyunsaturated fatty acids as determined by the MDA content, is almost three times higher in the sarcoma sample than in the healthy donor control.  (2) the sarcoma-derived CRP content is 4.8 superior than the concentration calculated for the healthy donor control tissue.  (3) IL-6 and TNF-α concentrations were 4 and 15 folds (respectively) higher than the values calculated for the healthy control. Finally, tumor homogenate VEGF concentration was 4-fold higher than those found in the healthy control.

Table 1: Biochemical characterization of the cell-free filtrates (CFF).

2-External gross changes in sarcoma-treated ZF.

Juvenile ZF exposed to the sarcoma-derived CFF showed the presence of a growing abdominal white mas, associated to skin reddish brown spots, and spinal cord deviation that somehow limited swimming activity (figure 1).

Figure 1: Clinical appearance of juvenile ZF. (A) Specimen representative of the ZF treated with the sarcoma-derived CFF at 6th week. The red arrow indicates the abdominal dilation due to a growing mass. (B) Representative of the normal appearance of specimens treated with the healthy donor tissue CFF.

As shown in table 2, the sarcoma administration produced a significant fall in body weight at the second week of administration. Although no significant differences were detected as compared to the concurrent controls from that point on, this group showed the lowest weights registered. 

Table 2: Zebrafish body weight.

Data are expressed as media ± SD. Statistical analysis was performed using a two-way ANOVA followed by the Tukey’s multiple comparison test. Different letters indicate statistically significant differences (p≤0.05).

3-Histological findings

Administering the sarcoma-derived material for six weeks resulted in the onset of pre-malignant and malignant changes that involved intestine, liver, and pancreas (table 3).

Table 3: Main histological changes associated to sarcoma CFF administration.

Intestinal stromal-like tumors (figure 2) were identified as a submucosal mass of varied size that produced organs dysmorphia, with a well circumscribed intramural nodular aspect in the muscularis propria, exhibiting dense extracellular matrix, and predominantly made up by spindle cells. The nuclei appeared clearly pleomorphic, atypic, and surrounded by a dense aspect-cytoplasm. Goblet cells and muscularis externa hyperplasia were also observed in different intestinal samples.

Figure 2: Intramural intestinal stroma tumor exhibiting a sarcoma-like aspect with abundant spindle cells. The nuclei appeared intensely basophilic, pleomorphic, atypic, and surrounded by a dense aspect-cytoplasm. H/E, magnification x 40. Scale bar 200 µm.

Liver cells homeostasis was also affected by the tumor-derived signalers. Hepatic premalignant lesions included multiple and scattered parenchymal nodules reminding an adenomatous pattern, containing numerous hyperplastic biliary tubules of varied shape and size (figure 3). Most of these neoformed tubules exhibited luminal papillary neoformations, were lined by cuboidal cholangiocytes arranged as a simple epithelial or as abnormally stratified layers. Within the cholangiocytes layer, dysplastic cells were often observed, with enlarged and hyperchromatic nuclei. These foci of tubules hyperplasia were surrounded by thick collagen bundles and cells of spindle morphology with basophilic nuclei.

Figure 3: Image representing the different histological patterns of intrahepatic nodules with bile ducts hyperplasia. A – shows into the yellow rectangle a nodule with multiple neoformed tubules including a giant one (magnification x 4). B – the yellow rectangle highlights two giant ducts, the left one with a notable papillary formation. On the right, a ductus shows a stratified epithelia with cholangiocytes hyperplasia (magnification x10) . C – the box encircles a giant ductus with multiple papillary neoformations with cuboidal cholangiocytes with hyperchromatic nuclei (magnification x 10). In general terms these represent tubular and papillary proliferation reminiscent of an adenoma pattern. H/E staining. Scale bar 200 µm.

Intrahepatic cholangiocarcinomas (ICC) were detected in fishes exposed to the sarcoma homogenate. These lesions exhibited a tubular pattern with well-to-moderate differentiation, marginal atypia, and embedded within areas of desmoplastic reaction (figure 4). Hepatocytes-originated malignancies consisted in typical hepatocarcinomas (HCC). These tumors were clearly distinguished by loss of hepatocytes cohesiveness (figure 5), perinuclear vacuolation with areas of signet ring cell features, sporadic fibrotic induration with collagen bundles, nuclear atypia, and particularly by the loss of hepatocyte plate structure with tendency toward a nesting pattern of organization. Another interesting finding consisted in a nest of atypical epithelioid cells at the edge of the organ´s parenchyma. These cells appeared rounded, with abundant eosinophilic cytoplasm, and eccentric nuclei, indicating mild to moderate nuclear atypia (not shown).

Figure 4: Images representing an intrahepatic biliary carcinoma with tubular histological pattern. A: the yellow box indicates a papillary neoformation. B: there is a noticeable focus of poorly differentiated cells with hyperchromatic and atypic nuclei encircled to the bottom right. Areas of desmoplastic reaction are observable around the neoformed biliary ducti (magnifications x10 and x 20) H/E staining. Scale bar 200 µm.

Figure 5: Sarcoma-derived CFF induced malignant transformation of hepatocytes with multiple hepatocarcinomas in this group. A: loss of hepatocytes discohesiveness and some degree of cellular disorientation is observable within the circle. B: Perinuclear vacuolation, areas of signet ring-like cells, hyperchromatic and atypic nuclei, and cellular pleomorphism are distinguished. Magnification x 10 and x 40. H/E. Scale bar 200 µm.

Pancreatic changes included: (1) a remarkable Langerhans islets hyperplasia as compared to healthy donor-matched control (figure 6), and (2) the presence of a highly undifferentiated tumor of possible epithelial origin, outside the intestinal wall and adjacent to the exocrine pancreas. This tumor is a typical overcrowded nodule of poorly differentiated, irregular-in-shape cells of epithelioid morphology, and features of high-grade malignancy. The cells are characterized by marked stacking, intense basophilia, pleomorphic nuclei, variable cytoplasm appearance and no inflammatory response (figure 7). None of these lesions were observed in the group of ZF treated with the healthy donors CFF, or in the saline group.

Figure 6: A: Representative of the aspect of a Langerhans islet in ZF treated with CFF derived from healthy donors. B: Representative of hyperplastic response of Langerhans islet in ZF treated with the sarcoma-derived CFF.  Magnification x 10. H/E. Scale bar 200 µm.