Analytical Biochemistry 437 (2013) 17–19
Contents lists available at SciVerse ScienceDirect
Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio
Notes & Tips
Generation of multicellular tumor spheroids of breast cancer cells: How to go three-dimensional Anika Nagelkerke a,b,⇑, Johan Bussink a, Fred C.G.J. Sweep b, Paul N. Span a,b a b
Department of Radiation Oncology, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands Department of Laboratory Medicine, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands
a r t i c l e
i n f o
Article history: Received 27 October 2012 Received in revised form 25 January 2013 Accepted 8 February 2013 Available online 19 February 2013 Keywords: Breast cancer cells Spheroid culture Hanging drop Liquid overlay Matrigel
a b s t r a c t The multicellular tumor spheroid model is a widely used culture system to mimic the three-dimensionality of tumors. Several methods and an even larger number of protocols are available to prepare spheroids from regular monolayers. This paper describes the assessment of several techniques frequently used to culture spheroids of breast cancer cells. We found that some methods produced suboptimal results. Ultimately, an adapted liquid overlay technique generated tight, compact and robust spheroids of the breast cancer cells tested. Ó 2013 Elsevier Inc. All rights reserved.
An important characteristic of solid tumors is their rapid expansion. Due to the incapability of blood vessels to keep up with tumor growth, beyond a certain size poorly vascularized regions will occur. Consequently, areas with oxygen deficiency (hypoxia) and nutrient deprivation are present in most solid tumors. To study this characteristic tumor microenvironment, several methods are available, including the use of biopsy material from patients, animal models, and two-dimensional in vitro cell cultures. These models have their own advantages and disadvantages. Patient material has its own microenvironment, but manipulating experimentally is difficult. Material is prone to inter- and intra experiment variation, and does usually not allow for multiple measurements during experiments. Animal models have microenvironmental complexity and can be manipulated but are relatively expensive and time-consuming, and can require large numbers of animals. Two-dimensional cell cultures are easily used and manipulated but lack the complexity of a malignant tumor. Microenvironmental conditions, like hypoxia or nutrient deficiency can be artificially induced in cultures, but the characteristic regional pattern of tumors will always be absent. This leaves a tremendous difference between in vitro cultures and in vivo tumors, that can be partially filled by the use of three-dimensional cell cultures: spheroids. Multicellular tumor spheroids have been employed since the pioneering ⇑ Corresponding author. Address: Department of Radiation Oncology 874, Radboud University Nijmegen Medical Centre, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands. Fax: +31 24 3568350. E-mail address:
[email protected] (A. Nagelkerke). 0003-2697/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ab.2013.02.004
work of Sutherland and co-workers, who used spheroids to study the in vitro response to drugs and radiotherapy [1,2]. Ever since, spheroids have been applied to study several aspects of cell biology: metabolism, apoptosis, necrosis, proliferation, differentiation, angiogenesis and metastasis [3–7]. In literature, the term spheroid is not well defined. Several terms are used to describe threedimensional cell clusters: aggregates, spheres, spheroids, tumoroids and organoids [3]. Typically, spheroids are round or elliptic, globe-like structures. In our view, spheroids should be tight and compact structures, similar to in vivo tumors, and can be handled without causing mechanical dissociation of the cells. In proliferating spheroids, at a certain diameter cells in the centre will encounter limitations in their oxygen and nutrient supplies. Consequentially, a necrotic core will develop. This core is surrounded by a rim of quiescent, hypoxic cells, which itself is surrounded by a rim of proliferating cells. Numerous techniques are available to culture cancer cells as spheroids (reviewed in [3,8]). However, in literature there is little technical information available that compares different methods to generate spheroids. We have tried several techniques to make spheroids of two commonly used breast cancer cell lines, MCF-7 and MDA-MB-231. Notably, we found that some regularly used methods do not yield the compact, uniformly shaped spheroids we are looking for. Here, we report our findings on which methods we could use to make our spheroids tight and compact. To culture spheroids, cells must prefer to adhere to each other instead of to the substratum they are plated on. This can be achieved both mechanically, by continuous movement of the cells
18
Notes & Tips / Anal. Biochem. 437 (2013) 17–19
(spinner culture) or by defying gravity (hanging drop technique), and chemically, by using non-adhesive substrates to culture cells on (liquid overlay technique). Spinner cultures are widely used [3,5]. Permanent movement prevents cells from adhering to the flask. Consequently, cells will attach to each other and form spheroids. This method has the advantage that spheroids can be produced on a large scale. However, the mechanical forces exerted on the cells may cause damage. Spinner cultures require specialized equipment and for each cell line conditions need to be optimized. In addition, the spheroids generated tend to be of different size and shape, making standardization problematic. For example, the effect of a drug on spheroid size would be difficult to quantify if starting points already differ. Other methods are therefore worthwhile exploring. In the hanging drop technique gravitational force induces cell aggregation. We used this method on MDA-MB-231 and MCF-7 cells. 10,000 cells were plated in 25 ll drops on the lids of standard 10 cm cell culture plates. Lids are more hydrophobic, i.e. better able to hold a drop together, than the bottom of the dish. Often viscous components, like methylcellulose, are added to the medium to increase its thickness. Addition of 0.24% (w/v) methylcellulose to the culture medium was indeed necessary for cell aggregation (Supplementary Fig. 1A and B). To form clusters, we also tried adding 2.5% reconstituted basement membrane (matrigel) to the medium. This, however, did not result in the formation of cell clusters at all. In contrast, with addition of methylcellulose, within 24 h after plating cell clusters formed at the bottom of the drops. Still, tight and compact spheroids never formed. Instead, these aggregates were flat and disk-shaped (Fig. 1A and B) and cells adhered to each other very weakly. This made aggregates extremely delicate to handle and often they disintegrated by simply pipetting them. The disk-shape caused aggregates to lack the characteristic hypoxic core, evident by the absence of the exogenous hypoxia marker pimonidazole (Fig. 1C and D). Several studies on gliomas indicated that after aggregate formation in hanging drops, cell clusters should be placed on a non-adherent surface, such as agar, for several days to make them form true ball-shaped spheroids [9,10]. For our breast cancer cells this was not the solution. On agar, aggregates stuck to each other and did not compact to true spheroids. Instead, aggregates adopted a cup-shaped appearance, after which they disaggregated and cells died. We therefore continued our search for an optimal method to generate spheroids. Friedrich et al. suggested a liquid overlay technique in 96-wells plates using agarose as the non-adherent substrate [11]. The advantage of using 96 separate wells in comparison with one big plate is the generation of similar sized spheroids, which cannot adhere to each other. The authors tested the ability of several cell lines of different origin, including six breast cancer cell lines, to generate spheroids. Four days after seeding cells in coated plates, spontaneous spheroid formation was assessed. This study showed a vast difference in the ability of cell lines to compact into tight spheroids. Several cell lines did not compact at all and remained in a state of loose aggregation. MCF-7 cells did generate a compact spheroid, but of irregular shape [11]. We tested this method on our MDA-MB-231 and MCF-7 cells and confirmed the results found by Friedrich et al. Our MCF-7 cells formed firm aggregates, but uneven in appearance (Supplementary Fig. 2A and movie 1). Our MDA-MB-231 cells did not compact into the tight spheroids we seek and formed loose aggregates (Supplementary Fig. 2B). As this method could not provide us with the spheroid structure we are looking for, we attempted another protocol. A different liquid overlay technique in 96-wells plates was described by Ivascu and Kubbies [12]. The non-adherent substrate poly-HydroxyEthylMethAcrylate (poly-HEMA) was used to coat V-shaped 96-wells plates. V-shaped wells have the advantage that
Fig.1. (A) Schematic illustration of the effect of rotating a cell culture plate with hanging drops when microscopically imaging aggregates. (B) Microscopic image sequence of a rotating hanging drop with an aggregate of MCF-7 cells 24 h after initiation. Per drop 10,000 cells were plated. Time frame of the image sequence is 5 s. (C–D) Cross-sections of hanging drop aggregates with methylcellulose of MCF-7 (C) and MDA-MB-231 (D) cells. Per drop 10,000 cells were plated. 24 h after seeding aggregates were incubated with 200 lM pimonidazole (1-[(2-hydroxy-3-piperidinyl)propyl]-2-nitroimidazole hydrochloride, Natural Pharmacia International Inc, Burlington, MA, USA) for 1 h, after which aggregates were fixed and sectioned. Sections were stained as previously described [13] with mouse-anti-pimonidazole (Natural Pharmacia) diluted 1:800 and biotin-conjugated donkey-anti-mouse (Jackson ImmunoResearch Laboratories Inc, West Grove, PA, USA) 1:200. Notice the absence of hypoxia. Bars equal 100 lm.
cells can be collected in the cone of the V by a gentle centrifugation step, generating rounder spheroids [12]. This study also examined the cause of the difference between breast cancer cells that only aggregate and those compacting to true spheroids. Spheroid compaction of MDA-MB-231 cells was tested with addition of exogenous extracellular matrix compounds such as fibronectin, laminin, or matrigel. Matrigel addition gave compact, tight spheroids [12]. Our experiments confirmed that several breast cancer cell lines can use matrigel to form compact spheroids (Fig. 2A and B and movie 2). Below, we specify the precise protocol used. At least 3 days before initiating spheroid culture, V-shaped 96wells plates (Greiner Bio-one, Alphen a/d Rijn, the Netherlands) are coated with a 0.5% (wt/vol) solution of poly-HEMA (Sigma-Aldrich, St. Louis, MO, USA) in 95% ethanol. To dissolve poly-HEMA, we incubate the solution in a water bath of 65 °C and vortex it frequently. Each well is coated with 50 ll of this solution. Plates are incubated at 37 °C to evaporate the ethanol during 72 h. Next, 10,000 cells are added per well. Medium containing 2.5% matrigel (BD biosciences, San Jose, CA, USA) is used if cells do not compact into a spheroid on their own. The plate is centrifuged for 10 min at 1000g, after which it is placed in a standard cell incubator at 37 °C with 5% CO2 and left alone until the required spheroid size is reached. Using this technique we generated spheroids of MCF-7 and MDA-MB-231 cells, but also of other breast cancer cell lines:
Notes & Tips / Anal. Biochem. 437 (2013) 17–19
19
Fig.2. Effect of matrigel addition on spheroid formation with poly-HEMA coated plates in MDA-MB-231 cells. Without (A) and with (B) matrigel addition. Per well 10,000 cells were plated, images were acquired after 24 h. Bars equal 100 lm.
MDA-MB-175, MDA-MB-436, MDA-MB-468, OCUB-F, SKBR-3 and HCC-1937. All cell lines compacted into firm spheroids within 24 h (Fig. 2B and movie 2B), during which a striking tendency to attach to each other was observed (movie 2B and 3). Spheroids harvested 24 h after initiation already showed hypoxic cores (Supplementary Fig. 3A). An overview of all techniques tested and their outcomes is given in Supplementary Table 1. After spheroids were generated we used them originally for immunohistochemistry. We produced both cryo- and formalinfixed paraffin embedded (FFPE) sections. For cryosections, spheroids were collected in a test-tube and accumulated at the bottom using gravitational force. The supernatant was discarded before washing the spheroids with HBSS containing calcium and magnesium (Gibco, Carlsbad, CA, USA). Spheroids were embedded in TissueTek OCT (Sakura Finetek Europe, Zoeterwoude, the Netherlands) diluted with HBSS (1:1) and snap frozen. The generated spheroid block was then ready for sectioning. We made FFPE spheroids by again collecting and washing the spheroids. They were fixed overnight at room temperature in 4% paraformaldehyde (Sigma–Aldrich), after which they were washed once more with HBSS. Next, spheroids were embedded in 2% (wt/vol) agarose (Sigma–Aldrich) in HBSS. This agarose disk was embedded in paraffin, after which it could be sectioned and stained. Examples of stainings are shown in Supplementary Fig. 3A and B. Here, FFPE sections were used to stain for pimonidazole (hypoxia) and integrin b1 (cell-cell adhesion). Taken together we found that several methods to generate spheroids of breast cancer cells can provide the user with suboptimal results. Using the matrigel addition method with poly-HEMA coated V-shaped 96-wells plates, we were able to generate the firm spheroids with strong cell-cell contacts and a microenvironment that we were seeking instead of loosely coherent aggregates. This method allows for experimental manipulation and immunohistochemical analysis of the spheroids.
Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ab.2013.02.004. References [1] R.M. Sutherland, W.R. Inch, J.A. McCredie, J. Kruuv, A multi-component radiation survival curve using an in vitro tumour model, Int. J. Radiat. Biol. Relat. Stud. Phys. Chem. Med. 18 (1970) 491–495. [2] R.M. Sutherland, J.A. McCredie, W.R. Inch, Growth of multicell spheroids in tissue culture as a model of nodular carcinomas, J. Natl. Cancer Inst. 46 (1971) 113–120. [3] J. Friedrich, R. Ebner, L.A. Kunz-Schughart, Experimental anti-tumor therapy in 3-D: spheroids – old hat or new challenge?, Int J. Radiat. Biol. 83 (2007) 849– 871. [4] F. Hirschhaeuser, H. Menne, C. Dittfeld, J. West, W. Mueller-Klieser, L.A. KunzSchughart, Multicellular tumor spheroids: an underestimated tool is catching up again, J. Biotechnol. 148 (2010) 3–15. [5] J.B. Kim, R. Stein, M.J. O’Hare, Three-dimensional in vitro tissue culture models of breast cancer – a review, Breast Cancer Res. Treat. 85 (2004) 281–291. [6] L.A. Kunz-Schughart, J.P. Freyer, F. Hofstaedter, R. Ebner, The use of 3-D cultures for high-throughput screening: the multicellular spheroid model, J. Biomol. Screen. 9 (2004) 273–285. [7] W. Mueller-Klieser, Three-dimensional cell cultures: from molecular mechanisms to clinical applications, Am. J. Physiol. 273 (1997) C1109–C1123. [8] R.Z. Lin, H.Y. Chang, Recent advances in three-dimensional multicellular spheroid culture for biomedical research, Biotechnol. J. 3 (2008) 1172–1184. [9] A. Corcoran, L.I. De Ridder, D. Del Duca, O.J. Kalala, T. Lah, G.J. Pilkington, R.F. Del Maestro, Evolution of the brain tumour spheroid model: transcending current model limitations, Acta Neurochir. (Wien) 145 (2003) 819–824. [10] D. Del Duca, T. Werbowetski, R.F. Del Maestro, Spheroid preparation from hanging drops: characterization of a model of brain tumor invasion, J. Neurooncol. 67 (2004) 295–303. [11] J. Friedrich, C. Seidel, R. Ebner, L.A. Kunz-Schughart, Spheroid-based drug screen: considerations and practical approach, Nat. Protoc. 4 (2009) 309–324. [12] A. Ivascu, M. Kubbies, Rapid generation of single-tumor spheroids for highthroughput cell function and toxicity analysis, J. Biomol. Screen. 11 (2006) 922–932. [13] A. Nagelkerke, H. Mujcic, J. Bussink, B.G. Wouters, H.W. van Laarhoven, F.C. Sweep, P.N. Span, Hypoxic regulation and prognostic value of LAMP3 expression in breast cancer, Cancer 117 (2011) 3670–3681.