Thermal Bioprinting Causes Ample Alterations of Expression of LUCAT1, IL6, CCL26, and NRN1L Genes and Massive Phosphorylation of Critical Oncogenic Drug Resistance Pathways in Breast Cancer Cells

Abstract

Bioprinting technology merges engineering and biological fields and together, they possess a great translational potential, which can tremendously impact the future of regenerative medicine and drug discovery. However, the molecular effects elicited by thermal inkjet bioprinting in breast cancer cells remains elusive. Previous studies have suggested that bioprinting can be used to model tissues for drug discovery and pharmacology. We report viability, apoptosis, phosphorylation, and RNA sequence analysis of bioprinted MCF7 breast cancer cells at separate timepoints post-bioprinting. An Annexin A5-FITC apoptosis stain was used in combination with flow cytometry at 2 and 24 h post-bioprinting. Antibody arrays using a Human phospho-MAPK array kit was performed 24 h post-bioprinting. RNA sequence analysis was conducted in samples collected at 2, 7, and 24 h post-bioprinting. The post-bioprinting cell viability averages were 77 and 76% at 24 h and 48 h, with 31 and 64% apoptotic cells at 2 and 24 h after bioprinting. A total of 21 kinases were phosphorylated in the bioprinted cells and 9 were phosphorylated in the manually seeded controls. The RNA seq analysis in the bioprinted cells identified a total of 12,235 genes, of which 9.7% were significantly differentially expressed. Using a ±2-fold change as the cutoff, 266 upregulated and 206 downregulated genes were observed in the bioprinted cells, with the following 5 genes uniquely expressed NRN1L, LUCAT1, IL6, CCL26, and LOC401585. This suggests that thermal inkjet bioprinting is stimulating large scale gene alterations that could potentially be utilized for drug discovery. Moreover, bioprinting activates key pathways implicated in drug resistance, cell motility, proliferation, survival, and differentiation.

Keywords: bioprinting; drug discovery; kinase phosphorylation; molecular properties; tumor model.

Figure 1

Percentage of Apoptotic MCF7 breast cancer cells post-bioprinting. The annexin A5-FITC kit was used for this analysis. Cells were collected post-bioprinting at 3 time frames, 2, 24 h, and 7 days Cells were BP in a modified HP thermal inkjet printer, onto a petri dish with EMEM, and the cells were incubated immediately. Graph (A) depicts the total percentage of apoptotic cell population is expressed as the sum of early and late apoptosis percentage (F2+F4), green bars. Dark bars depict necrotic cells, those are the cells stained with PI. Each bar represents the average of triplicates. Controls used were live cells stained, unstained, and dead cells (not shown). Error bars represent the standard deviations. (B) Dot plot graphs depict results at 2 h post-bioprinting. Graph (C) displays results at 24 h post-bioprinting. Graph (D) are the results from non-printed (manually seeded) MCF7s cells. P-values are from a two-tailed student t-test for independent tests (shown over the bars in A). 10,000 events were obtained per sample. The Kaluza Analysis v.1.3 software (Beckman Coulter) was used to extract the data and graphs.

Figure 2

(A) Bioprinted MCF7 cells 24 h post-BP. Image showing in all three channels. (B) Manually Seeded MCF7 fixed at 24 h post seeding. (C) Bioprinted MCF7 Cells and (D) manually seeded MCF7 cells, single channel image, stained with Neu (sc-33684) primary antibody and goat IgG anti-mouse secondary conjugated with Alexa 568 channel. The same parameter settings were used for both cell samples. Fluorescent Intensity measurements were significantly different between the two samples, cytosol and nucleus were measured separately. In (E) Mean intensity measurements for the cytosol and the nucleus of the BP cells were 28.9 (1.6) and 24.3 (1.7), respectively. The mean intensity measurements for the cytosol and nucleus of MS cells were 87.0 (7.4) and 183.6 (9.8), respectively (BP Cyto = BP Cytosol, MS Cyto = Manually Seeded Cytosol, MS Nuc = Manually Seeded Nucleus, BP Nuc = BP Nucleus).

Figure 3

Activation of cellular kinases by thermal inkjet bioprinting. Chemiluminescent images in iBright FL1000 of a Proteome Profiler Human Phospho-MAPK Array (Catalog # ARY002B). (A) Membranes of manually seeded and BP MCF7 breast cancer cells. Membrane arrangement: LEFT = manually seeded (MS) MCF7 cells, RIGHT = BP MCF7. Signal for each kinase is represented by a pair of duplicate spots; three reference pairs are shown in three upper/lower corners. (B) Histogram profiles for selected analytes were generated by quantifying the mean spot pixel density exposure in the iBright FL1000 and Invitrogen™ iBright™ Analysis Software v.3.0. Kinases that show increased levels of phosphorylation are identified. Mean pixel density for the analytes is shown in the bar graphs. Twenty one kinases appeared phosphorylated in the BP samples, whereas 10 kinases showed in the manually seeded. Of the MS cells 6 kinases were strongly phosphorylated by >1.6-fold as compared to the BP cells.

Figure 4

Network of analytes phosphorylated in BP MCF7 breast cancer cells. Functions selected in this network were regulator of apoptosis (green, 8), response to stress (red, 17), intracellular signal transduction (yellow, 21), and signal regulators (blue, 16). This network depicts functional interactions among BP BC predisposed genes. In this network of phosphorylated sites, there are significantly more interactions than expected (p ≤ 0.001).

Figure 5

(A) Schematic summary of the intracellular pathways activated by bioprinting MCF7 cells. Kinases in green ovals not included in the phosphorylated kinases but they represent critical kinases, pink ovals represent overlapping kinases, present in both, manually seeded, and bioprinted cells. Blue ovals represent kinases phosphorylated only in BP cells. Red connector lines ending in a perpendicular line mean the originating kinase blocking signaling activity or deactivation. Representation of the critical kinases and their pathways in the bioprinted cells. (B) Network with links classified based on curated evidence in STRING. Biological functions selected are p53 binding process (red, 4), regulation of phosphorylation (blue, 14), stress activated protein kinase signaling cascade (pink, 7), regulation of cellular response to heat (dark green, 4), and cellular response to stress (light green, 16). These nodes have a significant number of interactions, as expected due to phosphorylated targets were extracted. The interaction score was set at 0.7 with k-mean clustering set at 5, thus only links that have a high confidence probability are displayed (P < 0.001). A total of 24 targets were observed from the phosphorylated targets in Bioprinted cells whereas for manually seeded cells, 6 phosphorylated sites were observed in this network. Despite the complexity of this network, we observed that MAPK1, TP53, CREB1, MAPK3, MAPK8, AKT1, HSPB1, AKT2, and MAP2K3 proteins display more than 10 protein-protein interactions.