ARTICLE IN PRESS
Biomaterials 26 (2005) 555–562
A multifunctional bioreactor for three-dimensional cell (co)-culture Artur Lichtenberga,*, Goekhan Dumlub, Thorsten Wallesc, Michael Maringkaa, Stefanie Ringes-Lichtenbergd, Arjang Ruhparwara, Heike Mertschingb, Axel Havericha a
Division of Thoracic and Cardiovascular Surgery, Hannover Medical School, Carl-Neuberg-Str. 1, Hannover 30625, Germany b Leibniz Research Laboratory for Biotechnology and Artificial Organs (LEBAO), Hannover, Germany c Department of Thoracic and Vascular Surgery, Heidehaus, Hannover, Germany d Department of Cardiology and Angiology, Hannover Medical School, Hannover, Germany Received 24 October 2003; accepted 24 February 2004
Abstract Investigation of cell abilities to growth, proliferation and (de)-differentiation in a three-dimensional distribution is an important issue in biotechnological research. Here, we report the development of a new bioreactor for three-dimensional cell culture, which allows for co-cultivation of various cell types with different culture conditions in spatial separation. Preliminary results of neonatal rat cardiomyocyte cultivation are shown. Isolated neonatal rat cardiomyocytes were cultured in spatial separated bioreactor compartments in recirculating medium on a biodegradable ﬁbrin matrix for 2 weeks. Glucose, lactate, and lactate dehydrogenase (LDH), pO2, pCO2, and pH levels were monitored in the recirculated medium, daily. Morphological characterization of matrix and cells was assessed by hematoxylin and eosin staining, and MF-20 co-immunostaining with 40 ,6-diamidino-2-phenylindole (DAPI). Cell viability was determined by LIVE/ DEAD staining before cultivation and on day 3, 7, and 14. The optimized seeding density in the matrix was 2.0 107 cells retaining cellular proportions over the cell culture period. The bioreactor allows the maintenance of physiologic culture conditions with aerobic cell metabolism (low release of lactate, LDH), a high oxygen tension (pO2–183.7718.4 mmHg) and physiological pH values (7.470.02) and a constant level of pCO2 (43.172.9) throughout the experimental course. The cell viability was sufﬁcient after 2 weeks with 8276.7% living cells. No signiﬁcant differences were found between spatial separated bioreactor compartments. Our novel multifunctional bioreactor allows for a three-dimensional culture of cells with spatial separation of the co-cultured cell groups. In preliminary experiments, it provided favorable conditions for the three-dimensional cultivation of cardiomyocytes. r 2004 Elsevier Ltd. All rights reserved. Keywords: Bioreactor; Co-culture; Cardiomyocyte; Cell viability
1. Introduction Cell culture has generated considerable interest in recent years due to the revolution in the molecular cell biology and tissue engineering. The invention of a suitable system for in vitro investigation of threedimensional cell growth, regenerative abilities and counteraction between various cell types is an essential aspect of biotechnological research. Current in vitro and in vivo studies have shown limited regeneration properties of the heart in adult mammals [1–3]. In contrast, *Corresponding author. Tel.: +49-511-906-3461; fax: +49-511-9063460. E-mail address: [email protected]
(A. Lichtenberg). 0142-9612/$ - see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2004.02.063
many groups presented excellent cardiac proliferative abilities in zebraﬁsh or amphibian [4,5]. In this light, understanding the differences of molecular mechanisms regulating the eukaryotic cell cycle between these species should provide important issue that may open the potential to stimulate cell cycle re-entry in terminally differentiated cells. However, mechanisms for these differences remain unclear [1,5,6]. For investigators whose ﬁeld of research addresses the interaction between mammalian and amphibian cells in a co-culture system, the different physiological temperature conditions for cell co-cultivation (37 C vs. 25 C) represents a major difﬁculty . Although the ideal conditions for in vitro cultivation of cells such as hypoxia-sensitive cardiomyocytes are yet unknown, current publications demonstrated the
ARTICLE IN PRESS 556
A. Lichtenberg et al. / Biomaterials 26 (2005) 555–562
positive effects of a three-dimensional cell distribution in a biological matrix [8–14]. Compared to static cell cultivation, recent studies have also evidenced the positive impact of dynamic cell culture conditions in a bioreactor on growth and survival of cells due to superior transfer of nutrients and gasses [9,12,13] as well as shear stress  by continuous medium perfusion [12,13]. Hence, the use of bioreactor technology for studies of mammalian cell proliferation may have beneﬁcial effects on the regenerative cell properties. In light of these concerns, we present in this work a development of our novel bioreactor for three-dimensional cell culture that allows for a co-cultivation of cells with diametrical different physiological conditions such as mammalian and amphibian cardiomyocytes.
In this ﬁrst study, functionality of the novel bioreactor was investigated by cultivation of neonatal rat cardiomyocytes [13,16].
2. Methods 2.1. Bioreactor The bioreactor consists of four transparent glass chambers generating two separate compartments (Figs. 1 and 2). The dimensions of chambers were: (1) The inner size of the upper chamber: length 20 mm, width 20 mm, height 4 mm; (2) The inner size of the lower chamber: length 20 mm, width 20 mm, height 7 mm; (3) The thickness of the chamber wall: 1 mm. The compartments and chambers were connected with silicone tubes of 2 mm inner diameter. Chambers were divided by a permeable polycarbonate membrane (NuclporeTM Membranes, Whatman Inc.) allowing a separation of co-cultures. Membranes with various pore sizes (0.1–12.0 mm) can be applied. In the ﬁrst experiments, we used a polycarbonate membrane with the biggest available pore size (12.0 mm). Such a pore size allows the ﬁltering only of cultured cells and not of molecules that are important for an interaction between co-cultures. The surface of the membrane in each bioreactor compartment measured 4.0 cm2. All components of the bioreactor were sterilized by a conventional autoclave.
Fig. 1. Schematic drawing of the novel multifunctional membrane bioreactor for a cell co-culture. Silicon tubes connected two spatial separated bioreactor compartments in one common system. The pump performs the medium perfusion between compartments with possibility for variable ﬂow and pulsation rate. The multiple inputs and outputs permit optimized medium ﬂow distribution. The passive aeration of the medium is performed inside the medium reservoir by continuous circulation of the gas (95% air, 5% CO2).
Fig. 2. The bioreactor compartment consists of two glass chambers. The chamber dimensions: Inner size of the upper chamber: length 20 mm, width 20 mm, height 4 mm; inner size of the lower chamber: length 20 mm, width 20 mm, height 7 mm; chamber wall thickness: 1 mm. A permeable polycarbonate membrane divides the chambers.
ARTICLE IN PRESS A. Lichtenberg et al. / Biomaterials 26 (2005) 555–562
2.2. Medium and gas perfusion Both compartments of the bioreactor were connected in one common circulatory system (Fig. 1). The medium perfusion in each compartment of bioreactor was performed by multiple inﬂow and outﬂow connections (6 connectors per chamber; Figs. 1 and 2). Initially, the medium was perfused via inﬂow connectors between cell matrix and membrane into the lower chamber of the ﬁrst compartment, and then through the membrane inside the upper chamber (Fig. 1). Then, from the upper chamber the ﬂuid was transported via outﬂow lines to the lower chamber of the second bioreactor compartment by a commercially available pump allowing pulsatile ﬂow (Ismatec SA Inc.). The perfusion design in the second compartment was repeated and after the aeration process in the medium reservoir the medium was returned again to the lower chamber of the ﬁrst bioreactor compartment. Pulsatile perfusion ﬂow rate was 2 ml/min with 60 pulsations/min. Gas (95% air, 5% CO2) exchange occurred by medium surface aeration inside the medium reservoir (Fig. 1). Active transport of fresh gas into the reservoir was available through a continuously active gas circulation by a roller pump (Ismatec SA Inc.) 2.3. Animal experiments and cell isolation All animal experiments and surgical procedures followed guidelines outlined in ‘‘Principles of Animal Care’’ (NIH publication No. 86-23, revised 1985) and protocols set forth by the Medical School Hannover Animal Care and Use Committee. The isolation of the neonatal rat ventricular cardiomyocytes and non-cardiomyocytes was performed as published by Wollert . Brieﬂy, hearts from 1- to 3day-old Wistar neonatal rats were removed and the ventricles were dispersed by digestion with collagenase II (Worthington Inc.) and pancreatin (Sigma Inc.). The cell suspension was puriﬁed by centrifugation through a discontinuous Percoll gradient to achieve a myocardial cell culture with more than 95% cardiomyocytes.
5 mm. Each lower chamber of compartments was ﬁlled with the cell matrix to 71% (2.0 cm3) of the total chamber volume (2.8 cm3). The mixing of matrix components and cells was performed under static conditions at room temperature (20–21 C). The matrix in the single compartment was then immediately removed and embedded in Tissue-Teks O.C.T. Compound (Sakura Finetek Europe, Netherlands), which was used for further histological and immunohistochemical evaluation as baseline. 2.4.1. Cell seeding In preliminary experiments we tested already different cell concentrations in our matrix: 0.5 107, 1.0 107, 1.5 107, and 2.0 107 cells per 2.0 ml total matrix volume. The latter seemed to be more suitable for our work with the superior spatial cell density in the matrix compared to others (Fig. 6b, c). 2.4.2. Cell culture conditions The compartments of bioreactors were ﬁlled and deaired by plating high glucose (4.5 g/l) medium consisted of Dulbecco’s modiﬁed Eagle’s medium (DMEM), and medium 199 in 4:1 proportion supplemented with 10% horse serum, 5% fetal bovine serum, 1% glutamine and antibiotics (Penicillin/Streptomycin) (Invitrogen Inc.). After ﬁlling of the bioreactors by 100 ml medium the pulsatile recirculation was immediately started. The bioreactors were also connected with the roller pump to achieve gas circulation. The plating medium was then changed on day 1 for further incubation by a high glucose (4.5 g/l) medium (DMEM-199) with glutamine and antibiotics without other components. Throughout the cultivation, the bioreactor was maintained in the dark in the conventional incubator at 37 (MCO-20AIC, Sanyo Scientiﬁc, Bensnville, IL). 2.4.3. Observation of the cell scaffold The cell scaffolds were observed through transparent glass chambers for stability of the matrix and detection of a potential bacterial or fungal contamination twice a day.
2.4. Experimental procedure For culture experiments, three bioreactors containing two compartments per each bioreactor were established. In addition, 1 single compartment was used for baseline evaluation. Per each compartment one milliliter of cell suspension containing 2.0 107 cells were added in the lower chamber and mixed to the biodegradable ﬁbrin consisting of ﬁbrinogen (1000 IU) and thrombin (10 IU) (Baxter Inc.) with 2.0 ml total volume of all components. Within 2 min, the ﬁbrin matrices were stabilized and became a gel structure. Dimensions of the ﬁbrin matrix were as follows, length 20 mm, width 20 mm, thickness
2.4.4. Lactate, glucose, pH, pO2, and pCO2 measurements Lactate, glucose, pH, pO2, and pCO2 levels were measured at the same time daily by three samplings of 1 ml of medium from the medium reservoir in a blood– gas syringe and determined on a blood–gas analyzer (Radiometer, ABL 300, Copenhagen). During the measurements the medium samplings were hermetically sealed and had no exposure to atmosphere. The measurements during the ﬁrst hour after perfusion beginning were performed for evaluation of the baseline data.
ARTICLE IN PRESS 558
A. Lichtenberg et al. / Biomaterials 26 (2005) 555–562
2.4.5. Lactate dehydrogenase measurements To analyze the metabolic properties of the cell culture the medium lactate dehydrogenase (LDH) levels were determined by three samplings also of 1 ml circulated medium from medium reservoir at the same time daily by routine clinical analyzers (Konelab 60, Thermo Electron Corporation, Finland). The LDH measurements during the ﬁrst hour of perfusion were performed for evaluation of the baseline value. 2.4.6. Histology and Immunohistochemistry On days 3, 7, and 14 the medium circulation was stopped in bioreactors 1, 2 and 3, respectively. After dismantling of bioreactor compartments each cell matrix (2 matrices per bioreactor) was removed and immediately frozen and embedded in Tissue-Teks O.C.T. Compound (Sakura Finetek Europe, Netherlands). The matrices were then serially sectioned in 10mm slices in the 6 areas: 2 central zones—each 0–0.3 mm away from the midline of the matrix, 2 intermediate zones—each 0.3–0.6 mm away from the midline and 2 outer zones—each 0.6–10.0 mm away from the midline. One half of each zone was sectioned in a longitudinal plane and the other half in a transversal plane. Cell culture sections were analyzed by routine bright ﬁeld and immunoﬂuorescent microscopy (Olympus Optical Co., BX-40). Hematoxylin and eosin (H–E) staining: Longitudinal and transversal sections in 6 matrix zones were analyzed by standard H–E staining for assessment of cell density, integrity, distribution, and porosity of the ﬁbrin matrix. Additionally, we performed a H–E staining of the native cross-section of the neonatal rat heart as reference for the comparison of cell density in the matrix. 2.4.7. Characterization of cardiomyocytes and quantification of total nuclei number The proportion of myocytes to non-myocytes was assessed by immunostaining with a primary mouse monoclonal sarcomeric myosin-speciﬁc MF20 antibody (LEBAO, Medical School Hannover) and goat antimouse IgG (H+L) secondary red-ﬂuorescent antibody (Alexa Fluors 594, Molecular Probes Inc.). Cardiomyocytes were distinguished by the presence of cells with red ﬂuorescence. The total nuclei number of myocytes and non-myocytes in the culture was quantiﬁed by co-immunostaining with 40 ,6-diamidino-2-phenylindole (DAPI, Sigma-Aldrich Inc.). DAPI binds to double-strained DNA by forming a stable ﬂuorescent complex with a blue stain (Fig. 3). The mean percentage value of cardiomyocytes was determined by counting at least 800 cells under 400 magniﬁcation, in a minimum of 10–15 ﬁelds in each ﬁbrin matrix section. The mean percentage of cardiomyocytes in each bioreactor was compared to the baseline data.
Fig. 3. Metabolic parameters during 14 days cultivation: (a) medium pO2 and pCO2; (b) medium pH.
2.4.8. Viability assessment using LIVE/DEADs assay Cell survival was assessed by LIVE/DEADs Viability/Cytotoxicity assay (Molecular Probes Inc.) based on the simultaneous determination of viable and dead cells with 2 probes (calcein AM and ethidium homodimer (EthD-1)) that measure intracellular esterase activity and plasma membrane integrity. Viable cells were distinguished by the presence of intracellular esterase activity, determined by the enzymatic conversion of the non-ﬂuorescent cell-permeant calcein AM to the intensely green ﬂuorescent calcein. Dead cells with damaged membranes were stained red-orange by entering EthD-1 into the cells and binding to nucleic acids. Cell viability was calculated as a percentage of living cells by counting at least 800 cells under 400 magniﬁcation, in a minimum of 10–15 ﬁelds in each matrix section. The mean percentage of viable cells in each bioreactor was compared to the baseline data. Additionally, we compared the cell viability between the compartments of each bioreactor system.
2.5. Statistics Results data were expressed as mean 7 standard deviation of the mean. The SPSS statistical software package 11.0 for Windows (SPSS Inc., Chicago, USA) was used for analysis. An unpaired t-test (Student’s t-test) was performed, considering a po0.05 as statistically signiﬁcant.
ARTICLE IN PRESS A. Lichtenberg et al. / Biomaterials 26 (2005) 555–562
3. Results During our experiments, the mechanical function of the bioreactors were excellent with constant pulsatile medium ﬂow. No medium leakage from the bioreactor was observed. Repeated observation of the scaffold through glass chamber wall by directly vision showed a good stability of the matrix and no macroscopic evidence of a bacterial or fungal contamination during two weeks of culture. The evaluation of pO2, pCO2, and pH in the medium of bioreactors during 2 weeks culture period showed physiological rates with a high oxygen tension (pO2 – 183.7 7 18.4 mmHg) and physiological pH values (7.4 7 0.02) and a constant level of pCO2 (43.1 7 2.9) (Fig. 3a, b). Low release of lactate without statistical signiﬁcant differences compared to baseline data was maintained during the total time of experiments in all three bioreactors, which implies aerobic cell metabolism in the cultures (Fig. 4). Glucose concentration decreased slightly throughout cultivation without achieving statistical signiﬁcance compared to baseline (Fig. 4). The content of cardiomyocytes in the cell culture before (baseline) and after 3, 7, and 14 days (Fig. 5) of cultivation was 98 7 3.8%, 95 7 6.8% (p = n.s. to baseline), 88 7 7.7% (p o 0.05 to baseline), and 80 7 8.2% (p o 0.01 to baseline), respectively. In preliminary experiments we tested the different cell seeding density (0.5, 1.0, 1.5, and 2.0 107 cells) by histological assessment. Fibrin constructs seeded with 2.0 107 cells per chamber had a more similarity regarding spatial cell density compared to native histological sections of the neonatal rat heart (Fig. 6a, c). The histological cell observation of the ﬁbrin construct showed uniform cell distribution with a good integrity and regular density of cells embedded in the matrix ﬁbers (Fig. 6c). The porosity of the matrix was similar in all sections with the pore size range 40–120 mm (Fig. 6b, c). The viability tests using LIVE/DEAD assay demonstrated initial cell viability of 98 7 2.5% (baseline).
Fig. 5. MF20 and DAPI co-immunoﬂuorescent image of a longitudinal ﬁbrin-matrix section after 2 weeks of cultivation. MF20 cardiomyocytes (arrow) were observed by the presence of cells with the red ﬂuorescence (Alexa Fluors 594 dye). Immunostaining with DAPI (blue dye) showed the total nuclei number of cardiomyocytes as well as non-cardiomyocytes (chevron). Magniﬁcation 400. Scale bar is 50 mm.
Fig. 6. Demonstration of the morphology with H–E stain of transversal sections of (a) native neonatal left ventricular cardiac rat tissue, (b) ﬁbrin matrix with seeding of 1.0 107 cells per 2 ml total matrix volume, and (c) ﬁbrin matrix with 2.0 107 cells also per 2 ml total matrix volume. Magniﬁcation 100. Scale bars are 200 mm.
Fig. 4. Metabolic parameters during 14 days cultivation: release of lactate in medium and concentration of the medium glucose.
Following 3 days cell culture a slight decrease towards 95 7 3.6% (p = n.s.) was detectable. After 1 and 2 weeks cell viability was 87 7 5.6% (p o 0.05 to
ARTICLE IN PRESS 560
A. Lichtenberg et al. / Biomaterials 26 (2005) 555–562
Fig. 7. The mean percentage value of the live cells before (day 0) and after 3, 7, and 14 days of culture.
Fig. 8. Three ﬂuorescent images of cells immunostained by LIVE/ DEADs viability/cytotoxicity assay (Molecular Probes Inc.) on 14 day of cultivation regarding three matrix zones: (a) central zone (0–0.3 mm away from the matrix midline), (b) intermediate zone (0.3–0.6 mm away from the midline), (c) outer zone (0.6–10.0 mm away from the midline). Transversal sections. Viable cells (arrow) appear green. Dead cells (chevron) are red-orange. Magniﬁcation 200. Scale bars are 100 mm.
baseline) and 82 7 6.7% (p o 0.05 to baseline), respectively (Fig. 7). The analysis of cell survival in respect to matrix zones showed no signiﬁcant differences (Fig. 8). We found also no signiﬁcant differences regarding viability between cells seeded in the spatial separated compartments of each bioreactor. Cumulative release of LDH into the medium of 9.2 7 4.5 U/l also implies a low cell damage during culture period.
The general consensus is that adult mammalian cardiomyocytes retain very low activity to re-enter the intrinsic cell cycle. Efforts of researchers to improve these properties, especially in humans, were yet not successful. Unfortunately, several methods such as application of various growth factors and cytokines or genetic manipulation could not signiﬁcantly overcome this important task. Current investigations have already shown the importance of co-cultures with different cell types to induce of various desired effects on cell differentiation , cell activities , cell death , and proliferative properties  due to reciprocal inﬂuences. However, the intensive use in a co-culture system of species (newts, zebraﬁsh) with excellent capacity for repair of injured myocardium [4–6] may open new possibilities to the understanding of regulatory cell cycle mechanisms in mammals with a potential for development of novel therapies for various diseases resulting in loss of cardiac muscle cells. For the ﬁrst time we demonstrated in this work a novel bioreactor design, which might allow a cocultivation of cells from warm-blooded species with cells of cold-blooded animals, which have diametrically different physiological temperature conditions (37 C vs. 25 C). As the compartments of the presented bioreactors could be placed in different distances and in different incubators the temperature for each cell population could be regulated variably. Such a concept could permit the retaining of physiological conditions for all cultures. Our bioreactor could provide an individual physiological environment for co-cultivated cells with constant communication and continues substance exchange between co-cultures. Our system could also be used for different studies in a monoculture with the advantages of cultivation in the three-dimensional distribution with dynamic perfusion conditions [9,12,13] The advantage of cell cultivation in spatial separation also of single cell types offers the possibility of selected application of different physical, chemical, or biological stimuli on each separate culture with potential impact on the cell growth and proliferation [3,21,22]. By use of such culture concept the interaction effects between isolated cell groups can be studied. In the presented study we ﬁrst of all showed the practicability of the described bioreactor in a monoculture although the future aim of our work is the development of a system for co-culturing with the possibility for creation of optimized individual environment for mammalian and amphibian cardiomyocytes with different temperature culture conditions. However, such concept seems to exclude the possibility of a direct contact between co-cultured cells and also static culture conditions. In our opinion, a common cultivation of
ARTICLE IN PRESS A. Lichtenberg et al. / Biomaterials 26 (2005) 555–562
cells with different temperature conditions is possible only in spatial separation of cultures. Furthermore, as guaranty for mass transfer between co-cultures in such concept should serve a sufﬁcient continuous perfusion of medium. Hence, our bioreactor system combines these both modalities: spatial separation of cells as well as constant medium perfusion between cell groups. In addition, current investigations have also shown that active continuous transport of nutrients and gases to the culture by medium perfusion promoted a more sufﬁcient aerobic metabolism than in static cultures where only diffusional transfer resulted in anaerobic cell metabolism [9,13]. Previous reports from our institution regarding viability of three-dimensional cardiomyocytes in a bioreactor conﬁrmed this statement . Therefore, the important issue with respect to our further proliferation studies was to provide a system with maximal optimized distribution of medium perfusion over the matrix surface in the bioreactor and hence process optimization of mass transfers [23,24]. This has been possible by multiple inﬂow and outﬂow lines, which generated a homogenous distribution of oxygenated culture medium inside the bioreactor chambers. This design proved valuable in our study, as there was a sufﬁcient metabolism and high survival rate of neonatal rat cardiac cells over 2 weeks in culture. The functionality of our system was tested by using of neonatal rat cardiac monoculture, which was already established in our laboratory by previous works regarding isolation, cultivation of cardiomyocytes as well as methods for viability investigation . Although for the establishment of our bioreactor system only single cell type was used, we emphasize that two spatial separated cell populations seeded in isolated bioreactor compartments were presented. However, both cell groups in tested compartments of each bioreactor showed similar viability that conﬁrmed a sufﬁcient functionality of the presented system. The potential interaction between the spatial divided groups of cells will be investigated in our future works. The co-culture concept of different cell types is scheduled for our further investigation. In light of it, we chose the low ﬂow rate for bioreactor perfusion already in our preliminary experiments with respect to intended co-culture of amphibian and mammalian cells. The low medium ﬂow (2 ml/min) allows for the desired temperature levels in each bioreactor compartment. There, the thermal change can occur during perfusion by passive medium heating or cooling between spatial separated compartments in the silicon tubes due to different temperatures inside two incubators. A potential use of special thermal devices would permit the use of any other perfusion models also with higher ﬂow rate. Current investigation showed already the advantages of three-dimensional cultures [9–14]. The choice of ﬁbrin glue in our experiments as scaffold was based on the
several advantages of ﬁbrin: Fibrin is a full biodegradable natural polymer, which can be formed through enzymatic interaction between ﬁbrinogen and thrombin . Various ﬁbrin structures with different ﬁber thickness and pore size can be achieved by selective adjustment of the thrombin–ﬁbrinogen mix. The low ratio of ﬁbrinogen–thrombin leads to formation of a ﬁbrin structure with thin ﬁbers and large pore size within few minutes. The latter proved to be ideal for homogenous incorporation of cells and bioactive molecules as well as for diffusion and washout of the components from and into circulated medium. Our histological tests demonstrated results similar to other groups [13,25] with excellent integrity, and density of cells in the matrix as well as regular and uniform pore size. Moreover, we believe that large cell matrix surfaces permit a superior interaction between recirculating medium and cells. The sufﬁcient survival rate of cells in our experiments acknowledges these advantages of ﬁbrin as biological scaffold. However, the behavior of cardiomyocytes embedded in ﬁbrin in respect to proliferation properties is unclear and should be investigated in the further studies. In the bioreactor module we used a commercially available polycarbonate membrane for separating the co-cultured cells. Polycarbonate membranes have an open cell structure and a broad pore size distribution with low protein binding. This membrane was established especially for cell (co)-culture conditions and has sharply deﬁned pore sizes, high ﬂow rates and is biologically inert with excellent chemical and thermal resistance . In this work we used the biggest industrially available pore size of 12.0 mm. Such pore size should allow for the passage of only cultured cardiac cells and not molecules (proteins, lipids, etc.). In this work the presented bioreactor has also several technical and practical advantages, which may facilitate the operational process with this system in a laboratory. Our bioreactor is compact in size with low volume, easy to handle, which allows the cultivation in a conventional incubator. The bioreactor was constructed from materials that withstand the high thermal stress of autoclaving without the use of toxic chemical agents for sterilization. The compartments can be disconnected separately that permitted an isolated access to the cell cultures. Additionally, through spatial separation of cell groups and transparent texture of the bioreactor a separate observation of cultures is made possible. Sterilization and medium change was easy to perform. Hence, our system is applicable in most laboratories. It is particularly important that costs of bioreactor production are low. In conclusion, our presented model serves as a useful and simple bioreactor for different culture studies that allows for a monoculture of cells in three-dimensional distribution as well as for a co-culture of cell types with
ARTICLE IN PRESS 562
A. Lichtenberg et al. / Biomaterials 26 (2005) 555–562
diametrically different cell culture conditions. We emphasize that the sufﬁcient viability of cells in our experiments with especially hypoxia-sensitive cardiomyocytes proved an excellent functionality of our system. The investigation of the ability of mammalian cardiomyocytes for growth, (de)-differentiation, and regeneration in the co-culture with amphibian cardiac cells and their mutual inﬂuences will be one of the main challenges in our future work by use of our presented bioreactor. In our opinion, also any other cell types could be investigated in such system that may be essential for various co-culture concepts of cells with diametrically different requirements for environment.
References  Dowell JD, Field LJ, Pasumarthi KB. Cell cycle regulation to repair the infracted myocardium. Heart Fail Rev 2003;8(3): 293–303.  Oberpriller JO, Oberpriller JC, Mauro A. The development and regenerative potential of cardiac muscle. New York: Harwood Academic; 1991 p. 463.  Pasumarthi KB, Field LJ. Cardiomyocytes cell cycle regulation. Circ Res 2002;90(10):1044–54.  Poss KD, Wilson LG, Keating MT. Heart regeneration in zebraﬁsh. Science 2002;298(5601):2141–2.  Oberpriller JO, Oberpriller JC, Matz DG, Soonpaa MH. Stimulation of proliferative events in the adult amphibian cardiac myocyte. Ann NY Acad Sci 1995;752:30–46.  Brockes JP, Kumar A. Plasticity and reprogramming of differentiated cells in amphibian regeneration. Nat Rev Mol Cell Biol 2002;3(8):566–74.  Velloso CP, Simon A, Brockes JP. Mammalian postmitotic nuclei reenter the cell cycle after serum stimulation in newt/mouse hybrid myotubes. Curr Biol 2001;11(11):855–8.  Pei M, Solchaga LA, Seidel J, Zeng L, Vunjak-Novakovic G, Caplan AI, Freed LE. Bioreactors mediate the effectiveness of tissue engineering scaffolds. FASEB J 2002;16(12):1691–4.  Carrier RL, Papadaki M, Rupnick M, Schoen FJ, Bursac N, Langer R, Freed LE, Vunjak-Novakovic G. Cardiac tissue engineering: cell seeding, cultivation parameters, and tissue construct characterization. Biotechnol Bioeng 1999;64(5):580–9.  Eschenhagen T, Fink C, Remmers U, Scholz H, Wattchow J, Weil J, Zimmermann W, Dohmen HH, Schafer H, Bishopric N, Wakatsuki T, Elson EL. Three-dimensional reconstitution of embryonic cardiomyocytes in a collagen matrix: a new heart muscle model system. FASEB J 1997;11(8):683–94.  Eschenhagen T, Didie M, Munzel F, Schubert P, Schneiderbanger K, Zimmermann WH. 3D engineered heart tissue for replacement therapy. Basic Res Cardiol 2002;97(Suppl 1):146–52.  Carrier RL, Rupnick M, Langer R, Schoen FJ, Freed LE, Vunjak-Novakovic G. Perfusion improves tissue architecture of engineered cardiac muscle. Tissue Eng 2002;8(2):175–88.
 Koﬁdis T, Lenz A, Boublik J, Akhyari P, Wachsmann B, MuellerStahl K, Hofmann M, Haverich A. Pulsatile perfusion and cardiomyocytes viability in a solid three-dimensional matrix. Biomaterials 2003;24(27):5009–14.  van Luyn MJ, Tio RA, Gallego y van Seijen XJ, Plantinga JA, de Leij LF, DeJongste MJ, van Wachem PB. Cardiac tissue engineering: characteristics of in unison contracting two- and three-dimensional neonatal rat ventricle cell (co)-cultures. Biomaterials 2002;23(24):4793–801.  Papadaki M, McIntire LV, Eskin SG. Effects of shear stress on the growth kinetics of human aortic smooth muscle cells in vitro. Biotechnol Bioeng 1996;50:555.  Ross PD, McCarl RL. Oxidation of carbohydrates and palmitate by intact cultured neonatal rat heart cells. Am J Physiol 1984;246:H389–97.  Wollert KC, Taga T, Saito M, Narazaki M, Kishimoto T, Glembotski CC, Vernallis AB, Heath JK, Pennica D, Wood WI, Chien KR. Cardiotrophin-1 activates a distinct form of cardiac muscle cell hypertrophy. Assembly of sarcomeric units in series VIA gp130/leukemia inhibitory factor receptor-dependent pathways. J Biol Chem 1996;271(16):9535–45.  Mummery C, Ward D, van den Brink CE, Bird SD, Doevandans PA, Opthof T, Brutel dela Riviere A, Tertoolen L, van der Heyden M, Pera M. Differentiation of human embryonic stemm stells to cardiomyocytes: role of coculture with visceral endodermlike cells. Circulation 2003;107(21):2733–40.  Guo W, Kamiya K, Yaui K, Kodama I, Toyama J. Paracrine hypertrophic factors from cardiac non-myocyte cells downregulate the transient outward current density and Kv 4.2 K+ channel expression in cultured rat cardiomyocytes. Cardiovasc Res 1999;41(1):157–65.  Imanishi T, Murry CE, Reinecke H, Hano T, Nishio I, Liles WC, Ho L, Kim K, O’Brian KD, Schwart SM, Han DK. Cellular FLIP is expressed in cardiomyocytes and down-regulated in TUNELpositive grafted cardiac tissues. Cardiovasc Res 2000;48(1): 101–10.  Dekens MP, Santoriello C, Vallone D, Grassi G, Whitmore D, Foulkes NS. Light regulates the cell cycle in zebraﬁsh. Curr Biol 2003;13(23):2051–7.  Pakhomov AG, Akyel Y, Pakhomova ON, Stuck BE, Murphy MR. Current state and implications of research on biological effects of millimeter waves: a review of the literature. Bioelectromagnetics 1998;19(7):393–413.  Galban CJ, Locke BR. Analysis of cell growth kinetics and substrate diffusion in apolymer scaffold. Biotechnol Bioeng 1999;65(2):121–32.  Petrov M. Tzonkov St. Modeling of mass transfer and optimization of stirred bioreactors. Bioprocess Eng 1999;21: 61–3.  Jockenhoevel S, Zund G, Hoerstrup SP, Chalabi K, Sachweh JS, Demircan L, Messmer BJ, Turina M. Fibrin gel—advantages of a new scaffold in cardiovascular tissue engineering. Eur J Cardiothorac Surg 2001;19(4):424–30.  De Bartolo L, Morelli S, Bader A, Drioli E. Evaluation of cell behavior related to physico-chemical properties of polymeric membranes to be used in bioartiﬁcial organs. Biomaterials 2002;23(12):2485–97.