052020V1

1

Comparing Drug Response in 2D Cultures

and 3D Bioprinted Tumoroids Priyanka Koti, MEng, Shubhankar Nath, PhD, Josefin Blell, MSc,

Christen Boyer, PhD, and Itedale Namro Redwan, PhD CELLINK LLC, Boston, MA, USA

Abstract Three-dimensional bioprinting is gaining significant attention in cancer research, where there is an urgent need for predictive and representative tumor models. This study investigates the use of 2D cell cultures and 3D bioprinted tumor models for evaluating drug efficacy in aggressive forms of breast and pancreatic cancer. 2D and 3D tumor models were treated with cisplatin and gefitinib and compared for changes in cell morphology and cytotoxicity. Cisplatin and gefitinib present nonoverlapping mechanisms of action that interfere with DNA repair mechanisms and epidermal growth factor receptor (EGFR) signaling, respectively. Our findings validate bioprinted tumoroids as robust models for evaluating drug efficacy and show that 3D models allow for relevant cell morphologies and migratory patterns, as well as unique responses to anticancer drugs that differ from traditional 2D cell culture systems .

Introduction In cancer biology, the tumor microenvironment (TME) is a critical battle zone between tumor cells and the immune system. The TME is a bundle of extracellular matrix (ECM), immune cells, signaling molecules, blood vessels and fibroblasts that encapsulates tumors and affects cancer progression. Components of the TME interact with each other by secreting small signaling molecules that influence all aspects of tumor behavior, including cell proliferation, invasion, metastasis and resistance to anticancer therapies (Bremnes, 2011). Reconstructing the TME is therefore critical for anticancer studies, but a major setback has been the inability to develop a predictive 3D tumor model for high-throughput drug assessment. 3D tumor models should recapitulate cell-cell interactions within the tumor stroma and overcome the limitations of 2D cell culture systems. Here, 3D bioprinting offers a promising solution for predicting in vivo outcomes, modeling the TME and evaluating drug response.

Metastasis and chemoresistance threaten the survival of cancer patients. One therapeutic modality that has shown promise in the field of cancer management is chemotherapy, which uses small anticancer molecules to attack specific growth pathways and kill cancer cells. Among such molecules are cisplatin (CIS) and gefitinib (GEF), FDA-approved anticancer drugs that target DNA and EGFR pathways, respectively. In short, CIS causes apoptosis by inhibiting cell division and mRNA production, and GEF interferes with upregulated EGFR-signaling in cancer cells. Interestingly, while CIS and GEF have both been used to treat lethal forms of pancreatic and breast cancer, they have also been linked to false-negative or false-positive predictions in vitro, suggesting that they affect cells differently in 2D and 3D (Reynolds, 2017). To further address this discrepancy, we compared the effects of CIS and GEF on 2D monolayers and 3D bioprinted tumoroids using 2 breast cancer (MCF7, MDA MB 231) and 2 pancreatic cancer (BxPC3, Panc-1) cell lines.

© CELLINK AB 2020

052020V1

2

Materials and Methods

Cell preparation

Two pancreatic cancer cell lines (BxPC3, Panc-1) and 2 breast cancer cell lines (MCF7, MDA MB 231) were purchased from the American Type Culture Collection (ATCC). All cell lines were cultured according to the supplier’s protocols and passaged every 3 to 4 days. BxPC3 cells were grown in RPMI medium with L-glutamine (Corning, Ref #10-040-CV). Panc-1 cells were grown in DMEM with 4.5 g/L glucose, L-glutamine and sodium pyruvate (Corning, Ref #10-013-CV). MCF7 cells were cultured in EMEM (BD, Ref #670086); and MDA MB 231 cells were grown in bicarbonate-free Leibovitz L-15 medium with L-glutamine (Corning, Ref #10-045-CV). All media were supplemented with 10% FBS (Gibco, Cat #16000044) and 1% penicillin streptomycin (Gibco, Ref #1509-70-063).

Bioink preparation and bioprinting

Three mg/mL Coll 1 (CELLINK, Ref #IK4000002001) and 5% GelMA (CELLINK, Ref #IK3051020303) were prepared for bioprinting according to CELLINK protocols. A total of 3 mL Coll 1 or GelMA was mixed (10:1) with 5 x 106 cells/100 µL media and loaded into clear and amber cartridges (CELLINK, Ref #CSO010311502), respectively, for droplet printing at ~3 kPa. A temperature-controlled printhead (TCPH, SKU #000000020346) set to 8°C and a pneumatic printhead were used to bioprint Coll 1 and GelMA droplets, respectively, on an 8°C printbed. Each bioink was printed in an untreated 96-well plate (Thermo Fisher Scientific, Cat #267427) using the Droplet Print function on a BIO X (CELLINK, SKU #000000022222). After printing, Coll 1 droplets were thermally crosslinked for 20 minutes at 37°C, and GelMA droplets were UV-crosslinked at 365 nm for 6 seconds. One hundred µL media was added to each well and refreshed every 2 to 3 days.

2D monolayer culture

For 2D comparison, each cell line was seeded in treated 96-well plates (Thermo Fisher Scientific, Cat #167425). Cell seeding density was optimized for each cell type to reach 90% confluency after 48 hours of culture. Panc-1 cells were seeded at 1.2 x 104 cells/well, BxPC3 cells were seeded at 1.7 x 104 cells/well, MCF7 cells were seeded at 2.0 x 104 cells/well and MDA MB 231 cells were seeded at 2.0 x 104 cells/well.

Drug treatment and analysis

Bioprinted tumoroids and 2D monolayers were treated for 96 hours and 48 hours, respectively, with varying concentrations of gefitinib (LC Laboratories, #G-4408) or cisplatin (Cayman Chemical Company). An MTS Assay (Sigma-Aldrich) and LIVE/DEAD staining kit (Invitrogen) were used to assess cell viability for 2D and 3D conditions. All assays were performed according to the manufacturers’ instructions. Images were acquired using an EVOS Auto 2 Fluorescent Microscope (Thermo Fisher Scientific).

Statistical analysis

Statistical analysis was performed using GraphPad Prism 8.2.1. All data are expressed as mean ± SEM of two experiments carried out in sextuplets. Differences in treated conditions were analyzed using a one-way analysis of variance (ANOVA) and considered significant for p values < 0.05.

© CELLINK AB 2020

052020V1

3

Figure 1. Phase contrast and LIVE/DEAD images of bioprinted tumoroids on Day 11. A Calcein-PI assay was optimized for 3D culture and used to illustrate cell death at high concentrations of cisplatin or gefitinib. The benefits of this assay show the robust effects of

antitumor drugs on all 4 cell lines and depict variations in cell morphology for each cell type and ECM. Scale bar = 1000 µm or 650 µm. Green = live, red = dead.

© CELLINK AB 2020

052020V1

4

Results and Discussion Spheroid formation and cell morphology in bioprinted tumoroids

Tumors adapt different morphologies based on cell type and culture conditions (Nath, 2016). After 7 days of culture in GelMA and Coll 1, cancer cells had aggregated to form spheroids with various morphologies. As shown in Figure 1, MDA MB 231 cells formed concentric stellate networks, MCF7 cells formed round spheroids, BxPC3 cells formed grape-like ellipsoids and Panc-1 cells formed mass spheroids. The use of GelMA and Coll 1 as tumor scaffolds also affected spheroid formation due to differences in porosity, stiffness and composition. Interestingly, cancer cells grown in 2D cultures lacked the described morphologies, possibly because of their lack of ECM to support cell-cell interactions, tight junctions and nutrient- and oxygen-gradients (data not shown).

Hypoxia in 3D models

Hypoxia is another variable in drug response that is unique to 3D models and in vivo tissue. The Warburg Effect describes hypoxia as a survival mode for cancer cells in which they switch from producing oxygen and ATP to upregulating EGFR and AKT signaling for increased proliferation. This switch increases toxicity, acidity and waste buildup in 3D models, thereby creating a 3-ring hypoxic gradient. Hypoxic gradients are illustrated in Figure 1, in which cells toward the center of spheroids appear dead (red) and cells around the edges are viable (green). The outermost ring is a layer of proliferating cells, the middle ring is a layer of viable cells, and the innermost ring is a core of necrotic cells due to waste buildup and oxygen deficiencies (Nath, 2016).

Efficacy of cisplatin in 2D and 3D

Low to high doses of CIS were added to 2D monolayers and 3D bioprinted tumoroids on Day 2 and Day 7, respectively. Treatment lasted for 48 hours for 2D monolayers and 96 hours for 3D bioprinted tumoroids. An MTS assay revealed dose-dependent cytotoxicity in 2D monolayers for all cell lines, as well as 3D breast cancer tumoroids (Figure 2A). Interestingly, BxPC3 and Panc-1 cell lines displayed a higher IC50 in 3D than in 2D. In other words, both pancreatic cancer cell lines were largely unaffected by CIS in 3D bioprinted tumoroids. Here, one explanation is that pancreatic cancer cells displayed resistance to increasing CIS concentrations (Wang, 2016; Kelland, 2007; Sangster-Guity, 2011). In response to drug treatment, pancreatic cancer cells may have induced their survival pathways, upregulating senescence, DNA damage response signaling and translesion DNA synthesis (Gomes, 2019).

Efficacy of gefitinib in 2D and 3D

EGFR oncoproteins are often overexpressed in breast cancer and pancreatic cancer cell lines. Drug inhibition of the EGFR pathway can therefore lead to cell cycle arrest, senescence or apoptosis (Jacobi, 2017). As shown in Figure 2B, gefitinib significantly decreased cell viability in 3D and 2D. The IC50 for 3D Coll 1 and GelMA was lower than the IC50 for 2D cultures for all cell types, indicating that GEF caused more death in 3D bioprinted tumoroids than in 2D cultures.

Limitations of 2D Cell Culture Systems

Two-dimensional cell culture systems fail to mimic intrinsic properties of in vivo tumors, including the natural barriers, hypoxic gradients and tight cell-cell junctions that slow down drug diffusion. Moreover, they lack the tissue-specific environments and ECM that support 3D growth and the upregulation of oncoproteins (Reynolds, 2017). Another look at Figure 2A shows that pancreatic cancer cells in 3D are more resistant to CIS than the same cells grown in 2D monolayers. Here it is clear that a 2D study alone would be a misleading and inaccurate prediction for in vivo pancreatic cancer treatment.

© CELLINK AB 2020

052020V1

5

Figure 2. Dose-response curves of breast cancer cells and pancreatic cancer cells treated with cisplatin (A) or gefitinib (B). Cell viability, relative to untreated controls, was determined using a metabolic MTS assay. Data are read as mean ± SEM for 6 replicates.

Conclusions

The use of CELLINK GelMA and Coll 1 as scaffolds for tumoroids provided a stable TME for spheroidformation and drug diffusion.

Different kill-curve patterns for constructs made with GelMA and Coll 1 suggest that the ECM playsa critical role in drug response. Future studies to determine which scaffolds are appropriate forspecific tumor models are warranted.

Our findings show dose-dependent and cell-specific responses to cisplatin and gefitinib treatment in2D and 3D tumor models. Breast cancer and pancreatic cancer cell lines were more sensitive to GEFtreatment in 3D than in 2D. Similarly, breast cancer cell lines were more sensitive to CIS treatment in

© CELLINK AB 2020

6

3D than in 2D, but pancreatic cell lines displayed the opposite effect, suggesting an elevated level of drug resistance in 3D models.

Bioprinted tumoroids are relevant 3D models for drug screening that can be used to reduce instancesof false-negative and false-positive predictions. Future studies can scale-up tumoroid production witha BIO X for high-throughput drug testing.

References

1. Bremnes, R. M., Dønnem, T., Al-Saad, S., et al. The role of tumor stroma in cancer progression and prognosis: Emphasis oncarcinoma-associated fibroblasts and non-small cell lung cancer. Journal of Thoracic Oncology. 2011; 6(1): 209-211. DOI:10.1097/JTO.0b013e3181f8a1bd.

2. Gomes, L. R., Rocha, C. R. R., Martins, D. J., et al. ATR mediates cisplatin resistance in 3D-cultured breast cancer cells viatranslesion DNA synthesis modulation. Cell Death and Disease. 2019; 10(6): 459. DOI: 10.1038/s41419-019-1689-8.

3. Jacobi, N., Seeboeck, R., Hofmann, E., et al. Organotypic three-dimensional cancer cell cultures mirror drug responses in vivo:Lessons learned from the inhibition of EGFR signaling. Oncotarget. 2017; 8(64): 107423-107440. DOI: 10.18632/oncotarget.22475.

4. Kelland, L. The resurgence of platinum-based cancer chemotherapy. Nature Reviews Cancer. 2007; 7: 573–584. DOI:10.1038/nrc2167.

5. Nath, S., and Devi, G. R. Three-dimensional culture systems in cancer research: Focus on tumor spheroid model.Pharmacology and Therapeutics. 2016; 163: 94–108. DOI: 10.1016/j.pharmthera.2016.03.013.

6. Reynolds, D. S., Tevis, K. M., Blessing, W. A., et al. Breast cancer spheroids reveal a differential cancer stem cell response tochemotherapeutic treatment. Scientific Reports. 2017; 7(1): 10382. DOI: 10.1038/s41598-017-10863-4.

7. Sangster-Guity, N., Conrad, B. H., Papadopoulos, N., et al. ATR mediates cisplatin resistance in a p53 genotype-specificmanner. Oncogene. 2011; 30(22): 2526–2533. DOI: 10.1038/onc.2010.624.

8. Wang, S., Xie, J., Li, J., et al. Cisplatin suppresses the growth and proliferation of breast and cervical cancer cell lines byinhibiting integrin β5-mediated glycolysis. American Journal of Cancer Research. 2016; 6(5): 1108–1117. PMID: 27294003.

Contact Us US phone: (+1) 833-235-5465

Email: sales@cellink.com

© 2020 CELLINK AB. All rights reserved. Duplication and/or reproduction of all or any portion of this document without the express written consent of CELLINK is strictly forbidden. Nothing contained herein shall constitute any warranty, express or implied, as to the performance of any products described herein. Any and all warranties applicable to any products are set forth in the applicable terms and conditions of sale accompanying the purchase of such product. CELLINK provides no warranty and hereby disclaims any and all warranties as to the use of any third-party products or protocols described herein. The use of products described herein is subject to certain restrictions as set forth in the applicable terms and conditions of sale accompanying the purchase of such product. CELLINK may refer to the products or services offered by other companies by their brand name or company name solely for clarity and does not claim any rights to those third-party marks or names. CELLINK products may be covered by one or more patents. The use of products described herein is subject to CELLINK’s terms and conditions of sale and such other terms that have been agreed to in writing between CELLINK and user. All products and services described herein are intended FOR RESEARCH USE ONLY and NOT FOR USE IN DIAGNOSTIC PROCEDURES.The use of CELLINK products in practicing the methods set forth herein has not been validated by CELLINK, and such nonvalidated use is NOT COVERED BY CELLINK’S STANDARD WARRANTY, AND CELLINK HEREBY DISCLAIMS ANY AND ALL WARRANTIES FOR SUCH USE. Nothing in this document should be construed as altering, waiving or amending in any manner CELLINK’s terms and conditions of sale for the instruments, consumables or software mentioned, including without limitation such terms and conditions relating to certain use restrictions, limited license, warranty and limitation of liability, and nothing in this document shall be deemed to be Documentation, as that term is set forth in such terms and conditions of sale. Nothing in this document shall be construed as any representation by CELLINK that it currently or will at any time in the future offer or in any way support any application set forth herein.

Website: www.cellink.com

European phone: +46 31-128 700