Morphometric analysis of a triple negative breast cancer cell line in hydrogel and monolayer culture environments

Cell culture and experimental samples

HCC70, a cell line derived from primary ductal carcinoma of an African-American female donor, was cultured in parallel in monolayer and in hydrogel scaffolds. HCC70 cells were cultured using biosafety level 2 (BSL2) aseptic technique without antibiotics. All cell culture reagents were either purchased sterile or filter-sterilized with 0.2 µm membrane filters. HCC70 cells (ATCC^® CRL2315^™, Lot #58033362; ATCC, Manassas, VA, USA) were cultured as per the manufacturer’s instructions (ATCC, Manassas, VA, USA). Stock ampules (termed Passage 1) were prepared from the vendor’s vial when cells were received on dry ice in a frozen vial as previously described (Jogalekar, Cooper & Serrano, 2017). For each experiment, one Passage 1 stock ampule was seeded into two T25 flasks. Cells were cultured in RPMI-1640 media (ATCC, Lot #62285353; Manassas, VA, USA) and 10% fetal bovine serum (FBS; Lot #1025354; Gibco, Carlsbad, CA, USA) in an incubator maintained at 37 ^∘C and 5% CO₂. Once the cells reached 80% confluence, (∼48 h), they were trypsinized, titered, and seeded at a density of 4.5 ×10⁴ cells per cm² in two, 4-well slides. Monolayer and hydrogel cultures were established on each slide with two chambers designated for each condition. One slide was designated for light microscopy (VWR, Cat #62407-294; Radnor, PA, USA) and the other was processed for TEM (Cat #62407-330; Nalge Nunc International, Penfield, NY, USA).

Matrix preparation

Geltrex™ LDEV-Free Reduced Growth Factor Basement Membrane Matrix (Cat #A1413202; Invitrogen, Carlsbad, CA, USA) was allowed to thaw at 4 °C. The matrix was triturated gently with a pipette to prevent the formation of bubbles, prior to adding 100 µl of matrix per cm² to the slides under sterile conditions. The Geltrex ^TM was allowed to solidify in 37 °C incubator for 30 min, before the cells were seeded on top of the matrix.

Phase contrast microscopy

Prior to fixation for confocal microscopy, live cultures were imaged using 20 × objective of an inverted Nikon TE-2000 microscope equipped with a 2.8MP CoolSNAP MYO CCD camera (Photometrics) and Metavue image capture software (Molecular Devices).

Sample preparation for confocal microscopy

Sample processing was carried out as described previously (Jogalekar, Cooper & Serrano, 2017). Live cultures were rinsed with 1 × Phosphate buffered saline, pH 7.4 (PBS, P3813; Sigma Aldrich, St. Louis, MO, USA) for 2 min and fixed for 10 min using 4% paraformaldehyde (PFA; CAS #30525-89-4; Electron Microscopy Sciences, Hatfield, PA, USA). The cells were washed twice with 1 × PBS for 2 min each and treated with 0.2% Triton X-100 (Cat #T8787; Sigma Aldrich, St. Louis, MO, USA) for 20 min. The cells were rinsed briefly twice with 1 × PBS, probed with 165 nM Alexa Fluor^® 488 phalloidin dye (Cat #A12379; Invitrogen, Carlsbad, CA, USA) in 1% bovine serum albumin (BSA; Cat #A2153; Sigma Aldrich, St. Louis, MO, USA), for 20 min in the dark. The cells were washed briefly twice with 1 × PBS and counterstained with 2 µg/ml Hoechst 33342 (Cat #H3570; Invitrogen, Carlsbad, CA, USA) in 1 × PBS, for 10 min in the dark. Equal amounts of 1% BSA in 1 × PBS was added to the auto-fluorescence controls. The cells were briefly rinsed twice with 1 × PBS and re-fixed in 4% PFA, before exposing them to SlowFade^® AntiFade kit (Cat #S2828; Invitrogen, Carlsbad, CA, USA). Coverslips were placed on samples and adhered to the slide with nail polish prior to imaging.

Confocal microscopy

The digitized images of monolayer and 3D samples were captured using a TCS SP5 II confocal microscope system (Leica Microsystems Inc., Buffalo Grove, IL, USA) using a 40X objective (NA–0.70) and a pinhole of 1 Airy Unit. Alexa Fluor^® 488 phalloidin was excited using the 65 mW Argon laser line and images were captured with Alexa 488 emission settings. Hoechst 33342 was excited using a 50 mW 405 diode UV laser line and images were captured with DAPI emission settings. TIF image stacks were used for confocal projection images.

Sample processing for TEM

Samples were processed for TEM as described previously (Jogalekar, Cooper & Serrano, 2017). The cells were fixed overnight at 4 ^∘C (∼16 hr) with a fixative containing 2.5% glutaraldehyde (CAS #111-30-8; Electron Microscopy Sciences, Hatfield, PA, USA) and 0.1 M cacodylate buffer (#11650; Electron Microscopy Sciences, Hatfield, PA, USA) followed by washing twice with 0.1 M cacodylate buffer, 10 min each. The post-fixation was completed with 2% osmium tetroxide (CAS #20816-12-0; Electron Microscopy Sciences, Hatfield, PA, USA) / 0.1 M cacodylate buffer for 2 hr. The cells were rinsed in double distilled water (ddH₂O) for 15 min and subjected to dehydration in a series of ethyl alcohols -50%, 70% and 95% (twice, 10 min each) and 100% (thrice, 10 min each). Cells were incubated in the transitional solvent containing a mixture of acetone: ethanol (2:3) thrice for 10 min each. Infiltration was carried out with 1:1 and 1:3 proportion of absolute ethanol: EMbed 812 resin (#14120; Electron Microscopy Sciences, Hatfield, PA, USA) for 30 min each, before incubating in pure resin for 15 min. Fresh resin was added to slides before they were cured for 48 hr in a 60 °C oven. The chambers were detached from the slide and immersed in liquid nitrogen to separate the resin blocks from the slide. The blocks were double-embedded in fresh resin and cured for 48 hr in a 60 °C oven before preparing ultra thin sections (∼70 nm) with a UC6 ultra-microtome (Leica Microsystems Inc., Buffalo Grove, IL, USA). Sections collected on Formvar carbon grids (Cat #FF100-Cu; Electron Microscopy Sciences, Hatfield, PA, USA) were stained with 2% uranyl acetate solution (#22400; Electron Microscopy Sciences, Hatfield, PA, USA) for 10 min, then rinsed with ddH₂O, dried, and stored in grid boxes.

Transmission electron microscopy

A transmission electron microscope, H-7650 (Hitachi High-Technologies, Pleasanton, CA, USA) and a CCD camera (AMT Corp., Woburn, MA, USA) were used to capture digital images of 2, 944 × 2, 944 pixel frame size acquired with an 80 kV accelerating voltage.

Image analysis for TEM

Three grids were analyzed from each resin block and one image was captured for each grid, yielding a total of six images per chamber. The number of mitochondria, autophagic vacuoles and intercellular junctions was determined from all images at the same magnification (15, 000 ×, 80 kV) using the ‘count’ tool in Adobe Photoshop CS6 (Adobe Systems Inc., San Jose, CA, USA). Criteria for identifying the structures of mitochondria, autophagic vacuoles and intercellular junctions were defined before the scoring was undertaken (Fig. 1). The characteristic double-membrane enclosed structures containing cristae or folds were scored as mitochondria (m, Fig. 1A). Several types of intercellular junctions (ij, Figs. 1C,1D) were found to be present in HCC70 cells. They were identified by the presence of electron-dense areas connecting the membranes. The structures enclosed by double membranes and containing unwanted or damaged cellular organelles, and partially digested material were considered as autophagic vacuoles (av, Fig. 1B) (Tu et al., 2011; Hale et al., 2013; Galluzzi & Kroemer, 2015).

Numbers for mitochondria, autophagic vacuoles and intercellular junctions were normalized to tissue area values that were measured using ImageJ software (Schneider, Rasband & Eliceiri, 2012). Briefly, the tissue area (µm²) of a TEM image was determined using the free-hand drawing tool. The counts for the number of structures were then divided by the tissue area of an image to calculate structures per unit area. Values were reported as number of structures per 100 µm² of tissue area (File S1).

Statistical analysis

The statistical analysis was carried out using Microsoft Excel 2011 for Macintosh (File S1). The counts obtained from six images of each block were grouped and means and totals were computed. A one-way Analysis of Variance (ANOVA) test was used to determine the statistical significance of the data collected from TEM images of monolayer and hydrogel cultures.

Figure preparation

Phase contrast, confocal and TEM images were captured in TIFF format and saved at 300 dpi. Minimal image processing was carried out using ‘levels’ sliding bar, before labeling and composing figures using Adobe Photoshop CS6 (Adobe Systems Inc., San Jose, CA, USA). Microsoft Excel 2011 software for Macintosh was used for the preparation of graphs.

Responsible conduct

The human breast cancer cell line HCC70 (ATCC^® CRL2315™; Lot #58033362; RRID: CVCL_1270) was obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA) and cultured in the laboratory according to the vendor’s protocol. The use of the HCC70 cell line in a Biosafety Level 2 tissue culture facility was approved by the New Mexico State University Institutional Biosafety Committee (#1401SE2F0103). HCC70 is a commercial, publicly available cell line established using a mammary gland tissue from primary ductal carcinoma of a female donor, age 49 years, as listed on the vendor’s website (ATCC, Manassas, VA, USA). Consequently, HCC70 meets the exemption criteria for review by the Institutional Review Board under the Code of Federal Regulations 45 CFR 46.101(b)(4) (“Human Specimens, Cell Lines or Data — Research Involving Human Subjects,” 2016). The vendor has confirmed that the cells are of epithelial origin and are negative for three surface receptors: ER, PR and HER2.

Phase contrast and confocal microscopy of monolayer and hydrogel cultures

The general morphology of HCC70 cells was examined using phase contrast microscopy (Fig. 2). The monolayer and matrix cultures demonstrated strikingly different morphologies. Monolayer cells formed a single layer on the rigid substrate, spread horizontally, and appeared to contact neighboring cells (Fig. 2A). In contrast, HCC70 cells cultured with the basement membrane matrix Geltrex™ formed compact aggregates and spheroidal structures (Fig. 2B).

The morphology of monolayer and matrix cultures was further analyzed using laser confocal microscopy to study the spatial arrangement of cells and localize actin filaments and DNA. The two groups of cells were stained using Alexa Fluor^® 488 Phalloidin dye specific for F-actin and Hoechst 33342, a DNA-specific probe (Fig. 3). Confocal images of HCC70 cells in monolayer cultures indicated that cells were flattened and adhered to the rigid glass substratum, while cells cultured in the hydrogel formed well-rounded, multilayered structures. We noted that inter-nuclear distances between monolayer cells (Figs. 3A–3C) were higher than those in matrix-organized cells (Figs. 3D–3F).

Ultra-structural analysis of monolayer cultures

TEM images of HCC70 cells cultured in a monolayer environment showed that cells adhered to adjacent cells with the help of intercellular junctions (Figs. 4A–4C, *). A few mitochondria (Figs. 4A–4D, m) and autophagic vacuoles (Figs. 4A, 4C, av) were observed inside the cells. The mitochondria had a variety of shapes such as round, oval, and cylindrical. The sum of individual counts for mitochondria, autophagic vacuoles and intercellular junctions is shown in Table 1 together with the sampled area.

Ultra-structural analysis of hydrogel cultures

TEM images of hydrogel cultures of HCC70 cells showed that cells formed multiple intercellular junctions with neighboring cells (Figs. 5A, 5B, *). Cells in hydrogels comprised autophagic vacuoles (Figs. 5A, 5C, av), as well as mitochondria with a variety of shapes ranging from round to oval to cylindrical (Figs. 5A–5D, m). The sampled area, the sum of individual counts for mitochondria, autophagic vacuoles and intercellular junctions are shown in Table 1.

ANOVA analysis of ultrastructure data from the two growth conditions showed that HCC70 cells cultured in hydrogel comprised a higher number of mitochondria (Pvalue < 0.001), autophagic vacuoles (Pvalue < 0.01) and intercellular junctions (Pvalue < 0.001) than those present in monolayer cells (Figs. 6B–6D). The tissue area used for quantitative analysis was equivalent (Pvalue > 0.05) for both culture conditions (Fig. 6A).