Optimization of Liquid Overlay Technique to Formulate Heterogenic 3D Co-Cultures Models Elisabete C. Costa, 1 Vı´tor M. Gaspar, 1 Paula Coutinho, 1,2 Ilı´dio J. Correia 1 1 CICS-UBI—Health Sciences Research Centre, Universidade da Beira Interior, 6200-506, Covilh~ a, Portugal; telephone: þ351 275 329 0055; fax: 351 275 329 099; e-mail: icorreia@ubi.pt 2 UDI-IPG—Unidade de Investiga S c~ ao para o Desenvolvimento do Interior, Instituto Polite ´cnico da Guarda, Guarda, Portugal ABSTRACT: Three-dimensional (3D) cell culture models of solid tumors are currently having a tremendous impact in the in vitro screening of candidate anti-tumoral therapies. These 3D models provide more reliable results than those provided by standard 2D in vitro cell cultures. However, 3D manufacturing techniques need to be further optimized in order to increase the robustness of these models and provide data that can be properly correlated with the in vivo situation. Therefore, in the present study the parameters used for producing multicellular tumor spheroids (MCTS) by liquid overlay technique (LOT) were optimized in order to produce heterogeneous cellular agglomerates comprised of cancer cells and stromal cells, during long periods. Spheroids were produced under highly controlled conditions, namely: (i) agarose coatings; (ii) horizontal stirring, and (iii) a known initial cell number. The simultaneous optimization of these parameters promoted the assembly of 3D characteristic cellular organization similar to that found in the in vivo solid tumors. Such improvements in the LOT technique promoted the assembly of highly reproducible, individual 3D spheroids, with a low cost of production and that can be used for future in vitro drug screening assays. Biotechnol. Bioeng. 2014;9999: 1–14. ß 2014 Wiley Periodicals, Inc. KEYWORDS: cancer; in vitro models; LOT; 3D MCTS; tumor microenvironment Introduction In tumors, cancer cells are typically in contact with stromal cells (e.g., fibroblasts, endothelial cells, pericytes, adipocytes, and immune cells) and enclosed in a complex extracellular matrix (ECM) (Hanahan and Coussens, 2012; Koontongkaew, 2013). Such environment influences the establishment of cancer hallmarks, such as limitless cell proliferation, resistance to death inducing signals and metastization, among others (Hanahan and Coussens, 2012; Koontongkaew, 2013). Nowadays, the development of new cancer therapies demands the establishment of new in vitro methodologies that can markedly reduce the use of animal models during pre-clinical research (Ferdowsian and Beck, 2011). These methodologies must however assure that the tumor heterogeneity is correctly mimicked, in order to assure that the data gathered in vitro is suitable for in vivo applications. Thus, since the 1970s, organotypic and 3D cell culture models have been emerging as viable options to replace traditional 2D cell cultures and preliminary assays in animal models (Rimann and Graf-Hausner, 2012). A leading 3D platform for testing pharmaceutical formulations are spheroid models (Hirschhaeuser et al., 2010; Mehta et al., 2012). These unique structures can be comprised by multicellular aggregates of different cell types, thus opening the possibility to closely mimic the tumor microenvironment heterogeneity and also the complex cell–stromal interactions in vitro (Almendro et al., 2013). Moreover, the architecture and cellular organization of these 3D models are very similar to that found in solid tumors (Mehta et al., 2012). Characteristically, spheroids form a necrotic and hypoxic core, surrounded by an outer layer of actively proliferating cells (Box et al., 2010; Mehta et al., 2012; Yeom et al., 2012). Moreover, they also reproduce the biological properties presented by solid tumors, including cell morphology, growth kinetics, gene expression and response to drugs (Burdett et al., 2010; Mehta et al., 2012). However, until now none of the methodologies [Micro- arrays, Gyratory Rotation, Hanging Drop, LOT (Friedrich et al., 2007; Lin and Chang, 2008; Page et al., 2013)] currently The authors have no conflict of interest to declare. Correspondence to: I.J. Correia Contract grant sponsor: Portuguese Foundation for Science and Technology (FCT) Grant numbers: PTDC/EBB-BIO/114320/2009; PEst-OE/EGE/UI4056/2011; PEst-C/SAU/ UI0709/2011 Received 15 November 2013; Revision received 29 January 2014; Accepted 31 January 2014 Accepted manuscript online xx Month 2014; Article first published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/bit.25210 ARTICLE ß 2014 Wiley Periodicals, Inc. Biotechnology and Bioengineering, Vol. 9999, No. xxx, 2014 1