Characterization of three new serous epithelial ovarian cancer cell lines.(Research article)

Authors: V�ronique Ouellet (equal contributor) [1]; Magdalena Zietarska (equal contributor) [1]; Lise Portelance [1]; Julie Lafontaine [1]; Jason Madore [1]; Marie-Line Puiffe [1]; Suzanna L Arcand [2]; Zhen Shen [2]; Jos�e H�bert [3,4]; Patricia N Tonin [2,5,6]; Diane M Provencher [1,4,7]; Anne-Marie Mes-Masson (corresponding author) [1,4]

Background

Epithelial ovarian cancer (EOC) is often described as the silent killer or the disease that whispers mainly due to absence of symptoms. This combined with the lack of specific/sensitive markers and/or techniques of screening leads to the diagnosis at late stages of the disease in more than 70% of patients. Unfortunately, the five year survival rate at this point of the disease is less than 30% [1]. Although EOC is not the most prevalent of cancers, it accounts for the highest number of deaths from a gynecologic malignancy.

EOC is a complex disease stratified according to histopathological and morphological criteria. The majority of EOCs are thought to arise from the ovarian surface epithelium (OSE) that is derived from the coelomic epithelium. OSE is composed of multipotent cells that can differentiate and give rise to tumors of different histopathology types [1, 2]. The latter are defined by the International Federation of Gynecology and Obstetrics (FIGO) [3] and represent serous, endometrioid, mucinous, clear cell, de Brenner, mixed and undifferentiated subtypes. Serous type tumors are the most common subtype of EOC identified in more than 50% of cases. EOC tumors are graded according to the degree of differentiation of tumor cells which can vary from well (grade 1), moderately (grade 2) or poorly (grade 3) differentiated cells. Finally, EOC tumors are also classified according to the spread of the disease varying from stage I when tumors are confined to the ovaries to stage IV when distant metastases are observed.

Over the past years several laboratories, including ours [4], have established and characterized cell lines derived from EOC tumors. However, the majority of these EOC cell lines were established from patients ascites [4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29] and only few were derived from solid tumors [4, 12, 30, 31, 32, 33, 34, 35, 36, 37]. Moreover, EOC cell lines have rarely been derived from chemotherapy-naive patients while others were established following viral transformation (SV40 Large T antigen) (such as NMSO cell line) [38, 39] or xenograft passage in immunocompromised mice (such as the HEY, HO-8910PM, and AMOC-2 cell lines) [10, 40, 41]. In addition, few cell lines derived from serous EOC tumors are available even though this subtype represents the most frequently occurring histopathology subtype (such as the TOV-81D, FU-OV-1, and HOC1-7 cell lines) [4, 10, 11, 33, 35].

In this study, we describe three new serous EOC cell lines that were derived in our laboratory from either solid tumors or ascites of two chemotherapy-na�ve patients. This is the first report characterizing cell lines derived from both solid tumor and ascites of the same patient. Moreover, the molecular and growth characteristics of the three cell lines present some unique features thereby providing the research community with new tools in the study of different aspects of serous EOC.

Methods

Sample and Patient data

Tumor samples were collected and banked following surgeries performed within the Division of Gynecologic Oncology at the Centre hospitalier de l'Universit� de Montr�al (H�pital Notre-Dame). The study was approved by the CHUM institutional ethics committee and written consent was obtained from patients prior to sample collection. Stage was determined at the time of surgery. Histopathology and tumor grade were assigned by a pathologist according to the International Federation of Gynecology and Obstetrics (FIGO) criteria [3].

Establishment of the cell lines and culture conditions

All primary cultures and cell lines were cultured in OSE medium (Wisent, Qc, Canada) supplemented with 10% fetal bovine serum (FBS), 2.5 [mu]g/ml amphotericin B and 50 [mu]g/ml gentamicin. Cells were incubated in 5% CO

2 and 5% O2 .

The TOV-1946 cell line was established using the previously described scrape method on tumor tissue from patient 1946 [42, 43]. Briefly, tumor tissue was gently scraped into a 100 mm petri dish containing supplemented OSE medium. TOV-1946 cells were maintained in the same petri dish for the first 40 days and medium was replaced weekly. After 40 days, 80% confluence was attained and TOV-1946 cells were divided into two petri dishes. They were then divided in a proportion of 2:3 once a week for the first 15 passages and 1:2 twice a week thereafter until passage 70. Subsequently, cells were maintained and divided in a proportion of 1:5 twice a week.

The OV-1946 cell line was established from a mass of cells from the ascites of patient 1946. The mass was macro-dissected into small pieces, which were kept in a 100 mm petri dish for 27 days at which point adherent cells reached 80% confluence. Pieces of tissue were then discarded. Cells were divided in a proportion of 2:3 every week for the first 15 passages and then 1:2 twice a week until passage 70. Cells were then maintained and divided in a proportion of 1:5 twice a week.

The TOV-2223 cell line was established from patient 2223 tumor tissue using the collagenase method. Briefly, tumor tissue was macro-dissected onto a 100 mm petri dish containing serum free OSE medium supplemented with 1000 U of collagenase (Sigma-Aldrich, ON, Canada). After 3-4 hours at 37[degrees]C, cells were resuspended into 8 ml of medium and the remaining tumor tissue pieces were discarded. The medium containing cells was then divided into four 60 mm petri dishes and 10% FBS was added. Cells were divided 1:2 once a week until passage 19 and then twice a week until passage 70. Subsequently, cells were maintained and divided in a proportion of 1:3 twice a week.

Antibodies

For immunohistochemistry and western blot analyses, the following antibodies were used: beta actin AC-15 (ab6276 from Abcam inc. MA, USA), p53 (D0-1) (sc-126 from Santa Cruz Biotechnology, CA, USA), anti-c-ErbB2/c-Neu (OP15, Calbiochem, ON, Canada), Keratin 19 Ab-1 (Ms198-P0, Lab Vision Corp., CA, USA), Keratin 7 Ab-2 (MS-1352-P0) and Keratin 8 Ab-4 (MS-997-P0, both from NeoMarker, Medicorp, Qc, Canada).

Immunohistochemistry

Formalin fixed paraffin embedded tumors were sectioned at 4 [mu]m and the slides were stained using the immunoperoxidase method. Briefly, tissue sections were heated at 60[degrees]C for 30 minutes, deparaffinized in toluene and rehydrated in an ethanol gradient. Slides were submerged in boiling citrate buffer (0.01 M citric acid adjusted to pH 6.0) and microwaved for 10 min to unmask antigens. A 3% H

2 O2 treatment was used to eliminate endogenous peroxidase activity. The sections were blocked with a protein blocking serum-free reagent (DakoCytomation Inc., ON, Canada) and incubated with different antibodies for 60 min at room temperature.

The optimal concentration for each primary antibody was determined by serial dilutions. Tissues were incubated with either a secondary biotinylated antibody (DakoCytomation Inc., ON, Canada) or a rabbit anti-goat biotin-conjugated antibody (1:300) (sc-2774, Santa Cruz Biotechnology, CA, USA) for 20 min followed by incubation with a streptavidin-peroxidase complex (DakoCytomation Inc., On, Canada) for 20 min at room temperature. Reaction products were developed using diaminobenzidine containing 0.3% H

2 O2 as a peroxidase substrate. Nuclei were counterstained with hematoxylin and all sections were observed by light microscopy at 400x magnification. Substitution of the primary antibody with phosphate buffered saline served as a negative control.

Growth rate

Growth rates were assessed as previously described [4]. On day 0, 1 x 10

5 cells were seeded onto 60 mm petri dishes. On day 1, 3, 5, 7, 9, 11 and 13 the cells were trypsinized, resuspended in medium and counted using a hemacytometer. Each experiment was performed in triplicate for each harvest and repeated once. Saturation density was defined as the mean maximum number of cells at confluence counted from two independent experiments performed with triplicates and the doubling time was calculated according to the slope of the linear portion of the growth curve.

Anchorage independent growth in soft agarose and three-dimensional culture

Cell lines were assayed for their ability to grow in anchorage independent conditions by culturing 1 x 10

4 cells in agarose (0.33 g/100 ml OSE complete medium for the upper layer and 0.66 g/100 ml OSE complete medium for the base layer) [43]. Cells were cultured in soft agar for three weeks, colonies were photographed and these were used for counting. Two independent experiments performed in duplicate.

Cell lines were tested for their ability to form three-dimensional aggregates or spheroids as previously described [44, 45]. Briefly, 4000 cells were suspended in 15 [mu]l of OSE complete medium. The droplets of medium containing cells are then placed on the cover of non-coated plastic tissue culture plate. The cover is placed on a dish containing 10 ml of PBS to prevent dehydration of the droplets. The ability to form spheroids was assessed after four days.

Low serum growth

Tumor cell growth in low serum conditions was assessed by plating cells in six well plates in OSE medium supplemented of 1% FBS, 2.5 [mu]g/ml amphotericin B and 50 [mu]g/ml gentamicin and cultured for 21 days. The medium was changed every seven days. The experiments were performed in duplicate.

Wound-healing assay

Migration potential was …