Cells growing in a three-dimensional (3-D) environment represent a marked improvement over cell cultivation in 2-D environments (e.g., flasks or dishes). Here we describe the development of a multicellular 3-D organotypic model of the human intestinal mucosa cultured under microgravity provided by rotating-wall-vessel (RWV) bioreactors.
Because cells growing in a three-dimensional (3-D) environment have the potential to bridge many gaps of cell cultivation in 2-D environments (e.g., flasks or dishes). In fact, it is widely recognized that cells grown in flasks or dishes tend to de-differentiate and lose specialized features of the tissues from which they were derived. Currently, there are mainly two types of 3-D culture systems where the cells are seeded into scaffolds mimicking the native extracellular matrix (ECM): (a) static models and (b) models using bioreactors. The first breakthrough was the static 3-D models. 3-D models using bioreactors such as the rotating-wall-vessel (RWV) bioreactors are a more recent development. The original concept of the RWV bioreactors was developed at NASA's Johnson Space Center in the early 1990s and is believed to overcome the limitations of static models such as the development of hypoxic, necrotic cores. The RWV bioreactors might circumvent this problem by providing fluid dynamics that allow the efficient diffusion of nutrients and oxygen. These bioreactors consist of a rotator base that serves to support and rotate two different formats of culture vessels that differ by their aeration source type: (1) Slow Turning Lateral Vessels (STLVs) with a co-axial oxygenator in the center, or (2) High Aspect Ratio Vessels (HARVs) with oxygenation via a flat, silicone rubber gas transfer membrane. These vessels allow efficient gas transfer while avoiding bubble formation and consequent turbulence. These conditions result in laminar flow and minimal shear force that models reduced gravity (microgravity) inside the culture vessel. Here we describe the development of a multicellular 3-D organotypic model of the human intestinal mucosa composed of an intestinal epithelial cell line and primary human lymphocytes, endothelial cells and fibroblasts cultured under microgravity provided by the RWV bioreactor.
The first breakthrough in building a 3-D model was reported in the early of 1980s when scientists started to investigate different types of the scaffold (e.g., laminin, collagen type I, collagen IV, and fibronectin) and cocktails of growth factors to improve cell-to-cell and ECM interactions of "static" 3-D models1-7. Since then, the main problem with these models has been limitations in the transfer of nutrients and oxygen within the medium and tissue constructs8. In contrast to cells in the in vivo environment that receives a steady flow of nutrients and oxygen from surrounding networks of blood vessels, the static nature of these models hinders the effective distribution of them to the cells. For example, cell aggregates generated in in vitro static models that exceed a few millimeters in size will invariably develop hypoxic, necrotic cores9. The RWV bioreactors might circumvent this problem by providing fluid dynamics that allow the efficient diffusion of nutrients and oxygen 10-12. However, to date, work using RWV bioreactors have been limited to the inclusion of one or two cell types 13-17. Moreover, instead of a spatial orientation similar to native tissues, those cells formed cell aggregates. The main reason for these limitations has been the lack of a scaffold able to incorporate cells in an integrated fashion. The scaffolds used in the RWV bioreactors to date consist, with few exceptions 16-18, mainly of synthetic microbeads, tubular cylinders or small sheets 13-15,19-23. These are stiff materials whose composition and flexibility cannot be manipulated, and to which cells are attached to their surface. Thus, it is unlikely that these models will provide a system in which to evaluate, in an integrated fashion, the various cell components such as stromal cells (e.g., fibroblasts, immune and endothelial cells) that should be dispersed within the scaffold to closely mimic human tissue.
Here we describe the development of a multicellular 3-D organotypic model of the human intestinal mucosa composed of an intestinal epithelial cell line and primary human lymphocytes, endothelial cells, and fibroblasts24. These cells were cultured under microgravity provide by the RWV bioreactor 13,25-30. In our 3-D model, the ECM possesses many distinct properties, such as an osmolality similar to the culture medium (e.g., negligible diffusional restraints during culture) and the capability to incorporate cells and other relevant extracellular matrix proteins, as well as the appropriate stiffness to be used in bioreactors24. Biological systems are very complex, and over the past few years, there has been a shift in the focus of mucosal research toward the examination of cell interactions with their surroundings rather than studying them in isolation. In particular, the importance of cell-cell interactions in influencing intestinal cell survival and differentiation is well documented 31-34. Specifically, the communication between epithelial cells and their niche has a profound influence on the epithelial cell expansion and differentiation 35. Indeed, it is widely accepted that not only cell-to-cell but also cell-to-ECM interactions are critical to the maintenance and differentiation of epithelial cells in 3-D culture models. Previous studies have demonstrated that gut ECM proteins such as collagen I 24,36,37, laminin 38 and fibronectin 39 are instrumental in influencing intestinal epithelial cells to acquire spatial orientation similar to the native mucosa. Thus, the development of new technologies, like our 3-D model24, that can mimic the phenotypic diversity of the gut is required if researchers intend to recreate the complex cellular and structural architecture and function of the gut microenvironment. These models represent an important tool in the development and evaluation of new oral drugs and vaccine candidates.
Ethics statement: All blood specimens were collected from volunteers that participated in protocol number HP-00040025-1. The University of Maryland Institutional Review Board approved this protocol and authorized the collection of blood specimens from healthy volunteers for the studies included in this manuscript. The purpose of this study was explained to volunteers, and all volunteers gave informed, signed consent before the blood draw.
Note: See Table 1 for medium supplement preparation. See Table 2 for the preparation of the 3-D culture media.
1. Preparation of Culture Vessels
2. Preparation of the Cells
3. Preparation of Collagen-embedded Cells
4. Harvesting 3-D Cultures for Histology
Previously we have engineered a multicellular 3-D organotypic model of the human intestinal mucosa comprised of an intestinal epithelial cell line and primary human lymphocytes, endothelial cells and fibroblasts cultured under microgravity conditions24 (Figure 1). Fibroblasts and endothelial cells were embedded in a collagen I matrix enriched with additional gut basement membrane proteins45 (i.e., laminin, collagen IV, fibronectin and heparin sulfate proteoglycan) and added to RWV bioreactors. After 10 – 15 days, histological staining analysis demonstrated the presence of villus-like structures in the constructs. Approximately 60 – 80% of these epithelial cells were organized as a monolayer of polarized cells with their nuclei located in a basal position near the ECM, which is a major feature of well-differentiated cells (Figure 2).
Figure 1: Diagram of the Construction of the 3-D Model. Four steps are necessary to build the 3-D system. First, to obtain the required number of cells to generate the 3-D model, a human intestinal enterocyte epithelial cell line (HCT-8) and primary human endothelial cells and fibroblasts are grown as 2-D confluent monolayers. Second, the ECM composed of collagen-I matrix enriched with additional gut basement membrane proteins (i.e., laminin, collagen IV, fibronectin and heparin sulfate proteoglycan) is prepared. Third, fibroblasts and endothelial cells are embedded in a collagen-I mixture and added to RWV bioreactors followed by the addition HCT-8 epithelial cells. Fourth, primary lymphocytes are added to the culture at days 4 and 9 (±1 day). Please click here to view a larger version of this figure.
Figure 2: Comparison Between the Normal Human Intestine and the Organotypic Models. Hematoxylin and eosin staining of cells cultured in the 3-D microgravity model: tissues were stained purple and scaffold stained pink. Cells from the 3-D model were cultured for 14 ("a" and "b") and 20 days (c). Twenty days after seeding, total cell numbers are maintained, and cells were well differentiated showing villus-like features. Images are displayed at 100X ("a" and "b") and 40X ("c") magnification. Please click here to view a larger version of this figure.
Product Name | Pkg. Size | Reconstitution | Working Concentration | Storage |
Fibroblast Growth Factor-Basic (bFGF) | 25 mg | To prepare 25 μg/ml stock solution; add 25 μg of FGF into 1 ml of sterile medium (RPMI plus 1% FCS), swirl to dissolve, add 100 μl/aliquot | 5 ng/ml | powder -20 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
Stem Cell Factor (SCF) | 10 mg | To prepare 10 μg/ml stock solution; add 10 μg of SCF into 1 ml of sterile medium (RPMI plus 1% FCS), swirl to dissolve, add 250 μl/aliquot | 5 ng/ml | powder -20 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
Hepatocyte Growth Factor (HGF) | 5 mg | To prepare 5 μg/ml stock solution; add 5 μg of HGF into 1 ml of sterile medium (RPMI plus 1% FCS), swirl to dissolve, add 200 μl/ aliquot | 2 ng/ml | powder -20 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
Endothelin 3 | 50 mg | To prepare 50 μg/ml stock solution; add 50 μg of Endotelin into 1 ml of sterile medium (RPMI plus 1% FCS), swirl to dissolve, add 100 μl/aliquot | 10 ng/ml | powder -20 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
Laminin | 1 mg | Thaw slowly at 2 – 8 oC, swirl and add 100 μl/aliquot | 10 mg/ml | liquid -20 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
Vascular Endothelial Growth Factor (VEGF) | 10 mg | To prepare 5 μg/ml stock solution; add 10 μg of VEGF into 2 ml of sterile water, swirl to dissolve, add 100 μl / aliquot | 1 ng/ml | powder -20 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
Leukemia Inhibitory Factor (LIF) | 50 mg | To prepare 20 μg/ml stock solution; add 50 μg of LIF into 2.5 ml of sterile water, swirl to dissolve, add 100 μl/aliquot | 4 ng/ml | powder -20 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
Adenine | 25 g | To prepare 0.18 M stock solution; add 121.5 mg og adedine into 50 ml of 0.05 M HCl (250 μl of 37.1% HCl into 50 ml of water)), swirl to dissolve, filter sterilize and add 1 ml/aliquot | 1.8 x 10-3 M | powder 4 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
Insulin | 50 mg | To prepare 5 mg/ml stock solution; add 50 mg of insulin into 10 ml of 0.005 M HCl (5 μl of 37.1% HCl into 10 ml of water)), swirl to dissolve, filter sterilize and add 0.5 ml/aliquot | 5 mg/ml | powder -20 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
T3 | 100 mg | To prepare 2 x 10-8 M stock solution; add 3.4 mg of T3 into 25 ml of 1 M NaOH, dilute 0.1 ml of this solution into 9.9 ml of PBS, dilute again 0.1 ml of this solution into 9.9 ml of PBS, swirl to dissolve, filter sterilize and add 0.5 ml/aliquot | 2 x 10-11 M | powder -20 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
Cholera Toxin | 5 mg | To prepare 10-7 M stock solution; add 5 mg of Cholera toxin into 5 ml sterile ddH2O (store at -20 oC); dilute 50 μl of this solution into 5 ml ddH2O, swirl to homogenise, filter sterilize and add 0.5 ml/aliquot | 10-10 M | powder 4 – 8 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
Fibronectin | 1 mg | To prepare 1 mg/ml stock solution; add 1 mg of fibronectin into 1 ml of sterile distiled water. Allow 30 min. for material to go into solution. DO NOT AGITATE OR SWIRL. Add 100 μl/aliquot | 10 mg/ml | powder 4 – 8 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
apo-Transferrin | 100 mg | To prepare 5 mg/ml stock solution (1,000x); add 100 mg of transferrin into 20 ml of PBS swirl to dissolve, filter sterilize and add 0.5 ml/aliquot | 5 mg/ml | powder -20 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
Heparin | 50 KU | To prepare 50 mg/ml stock solution (1,000x); add 50 mg of heparin into 1 ml of water swirl to dissolve, filter sterilize and add 1 ml/aliquot | 0.1 mg/ml | powder -20 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
Heparan sulfate proteoglycan | 1 mg | To prepare 0.1 mg/ml stock solution; add 1 mg of heparan sulfate into 10 ml of sterile water swirl to dissolve, and add 0.2 ml/aliquot | 2 mg/ml | powder -20 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
Collagen IV | 5 mg | To prepare 2 mg/ml stock solutin: add 2 mg of collagen V into 2.5 ml of sterile 0.25% acetic acid. Allow 1 – 2 hr for material to go into solution. Swirl for better dillution. Add 400 μl/aliquot | 80 mg/ml | powder -20 oC, working aliquots -20 oC, avoid repeat freeze/thaw |
Table 1: Defined Medium Supplement Preparation.
F-12 Medium Supplemented with | |||
Reagent | Stock Solution | Final Solution | Amount |
Fetal Calf serum | 10% | 50 ml | |
Sodium pyruvate | 100 mM | 1 mM | 5 ml |
L-Glutamine | 200 mM | 2 mM | 5 ml |
Hepes | 1 M | 10 mM | 5 ml |
Gentamicin | 50 mg/ml | 50 μg/ml | 500 μl |
Penicillin/streptomycin | 10,000 U/ml/10 mg/ml | 100 U/ml/100 μg/ml | 5 ml |
Insulin | 5 mg/ml | 5 μg/ml | 500 μl |
T3 | 2 x 10– 8 M | 2 x 10-11 M | 500 μl |
Adenine | 1.8 x 10– 1 M | 1.8 x 10– 3 M | 5 ml |
Transferrin | 5 mg/ml | 5 μg/ml | 500 μl |
heparin | 50 mg/ml | 0.1 mg/ml | 1 ml |
ECGS | 3 mg/ml | 3 μg/ml | 5 ml |
bFGF | 25 μg/ml | 5 ng/ml | 100 μl |
SCF | 10 μg/ml | 5 ng/ml | 250 μl |
HGF | 5 μg/ml | 2 ng/ml | 200 μl |
Endothelin 3 | 50 μg/ml | 10 ng/ml | 200 μl |
LIF | 20 μg/ml | 4 ng/ml | 100 μl |
VEGF | 5 μg/ml | 1 ng/ml | 100 μl |
Choleran toxin | 10-7 M | 10-10 M | 500 μl |
Note: The amount cited above is for preparation of 500 ml of 3-D culture media. Media must be stored at 4 °C for no more than 2 weeks |
Table 2: Preparation of 3-D Culture Media.
In this manuscript, we describe the development of a bioengineered model of the human intestinal mucosa comprised of multiples cell types including primary human lymphocytes, fibroblasts, and endothelial cells, as well as intestinal epithelial cell lines24. In this 3-D model, cells are cultured within a collagen-rich extracellular matrix under microgravity conditions24.
As described previously, the major features of this model are: (i) the ability to mimic the epithelial tissue monolayer organization, (ii) the induction of appropriate polarity of epithelial cells, tight junctions, desmosomes and microvilli, (iii) a long-term culture (up to 20 days) with high viability of the primary cells (i.e., fibroblasts and endothelial cells), (iv) the expression the tissue-like differentiation markers including villin, cytokeratin, E-cadherin and mucin, (v) the capability to produce considerable amounts of cytokines (e.g., IL-8) and alkaline phosphatase upon antigenic stimulation, (vi) the transport of nutrients such as glucose (i.e., expression of disaccharidases, and presence of sugar transporters), and (vii) the multi-lineage differentiation of intestinal epithelial cells (i.e., absorptive enterocyte, globet and M cells)24.
It is important to highlight that to achieve reproducible results using our 3-D model the investigator must adhere to good cell culture guidelines46-48. It is crucial to systematically control cells for viability, mycoplasma contamination and changes in cell growth behavior. If a problem is identified, first ensure that no unauthorized changes have been introduced to the protocol. If the problem persists, switch to a new batch of the various medium components (including serum) and/or cells. A limitation of our 3-D model is the use of tumorigenic HCT-8 epithelial cell line. However, it is important to consider that the presence of HCT-8 line offers the advantage of being available commercially. Moreover, these epithelial cells do not express either classical or non-classical human leukocyte antigen (HLA)-class I molecules49,50. Thus, they allow the culture of epithelial cells with PBMC of different HLA-class I haplotypes in the absence of epithelial cell-PBMC alloreactivity. Furthermore, when comparing this system with systems such as the gut stem cell organoids51-53, this model offers many benefits. Although gut stem cell organoids provide invaluable information about cell biology and intestinal differentiation 51-53, this model permits direct apical exposure to nutrients, drugs and pathogens. This model also provides easy access to luminal contents such anti-microbial peptides and cytokines. In contrast, organoids are compact units with a luminal surface facing inwards. Consequently, there is a restricted amount of product that can be introduced into the organoid lumen 54.
We believe that our multicellular 3-D organotypic model of the human intestinal mucosa has wide-ranging potential as a tool for discovery in both health and disease, including interaction with pathogens, antigen trafficking, and inflammatory and metabolic processes24. Finally, due to the multicellular nature of our 3-D model, our model could enable gain- and loss-studies using immune cell types that are likely to influence epithelial cell behavior in vivo.
The authors have nothing to disclose.
This work was supported, in part, by NIAID, NIH, DHHS federal research grants R01 AI036525 and U19 AI082655 (CCHI) to MBS and by NIH grant DK048373 to AF. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Allergy And Infectious Diseases or the National Institutes of Health.
Quad Rotator/Independent Rotating Wall Vessel (RWV) bioreactor | Synthecon | RCCs-4DQ | For up to 4 vessels. Models with more or less vessels are also available. |
Disposable 50 ml-vessel | Synthecon | D-405 | Box with 4 vessels |
HCT-8 epithelial cells | ATCC | CCL-244 | |
CCD-18Co Fibroblasts | ATCC | CRL-1459 | |
Human Umbilical Vein Endothelial Cells | ATCC | CRL-1730 | HUVEC |
Fibroblast Growth Factor-Basic | Sigma | F0291 | bFGF |
Stem Cell Factor | Sigma | S7901 | SCF |
Hepatocyte Growth Factor | Sigma | H1404 | HGF |
Endothelin 3 | Sigma | E9137 | |
Laminin | Sigma | L2020 | Isolated from mouse Engelbreth-Holm-Swarm tumor |
Vascular Endothelial Growth Factor | Sigma | V7259 | VEGF |
Leukemia Inhibitory Factor | Santa Cruz | sc-4377 | (LIF |
Adenine | Sigma | A2786 | |
Insulin | Sigma | I-6634 | |
3,3',5-triiodo-L-thyronine | Sigma | T-6397 | T3 |
Cholera Toxin | Sigma | C-8052 | |
Fibronectin | BD | 354008 | Isolated from human plasma |
apo-Transferrin | Sigma | T-1147 | |
Heparin | Sigma | H3149 | |
Heparan sulfate proteoglycan | Sigma | H4777 | Isolated from basement membrane of mouse Engelbreth-Holm-Swarm tumor |
Collagen IV | Sigma | C5533 | Isolated from human placenta |
Heat-inactivated fetal bovine serum | Invitrogen | 10437-028 | |
D-MEM, powder | Invitrogen | 12800-017 | |
10% formalin–PBS | Fisher Scientific | SF100-4 | |
Bovine type I collagen | Invitrogen | A1064401 | |
Trypsin-EDTA | Fisher Scientific | MT25-052-CI | |
Sodium pyruvate | Invitrogen | 11360-070 | |
Gentamicin | Invitrogen | 15750-060 | |
Penicillin/streptomincin | Invitrogen | 15140-122 | |
L-Glutamine | Invitrogen | 25030-081 | |
Hepes | Invitrogen | 15630-080 | |
Ham's F-12 | Invitrogen | 11765-054 | |
Basal Medium Eagle | Invitrogen | 21010-046 | BME |
RPMI-1640 | Invitrogen | 11875-093 | |
Endothelial Basal Medium | Lonza | CC-3156 | EBM-2 |
Endothelial cell growth supplement | Millipore | 02-102 | ECGS |