Gerlach J

Hepatocyte culture between three dimensionally arranged biomatrix-coated independent artificial capillary systems and sinusoidal endothelial cell co-culture compartments.

Freshly isolated parenchymal liver cells, the hepatocytes, retain many of the metabolic activities of the tissue in vivo. As isolated cells, they are generally in a catabolic stage and therefore care is essential in fully maintaining their activity. Bioreactors are in development to address this problem. Since they enable an upscale of hepatocyte in vitro culture, they could be used as the main device in hybrid liver support systems. A basic problem in the development of special bioreactors for such systems is that with current techniques of hepatocyte culture, external hepatocyte function in vitro is limited to only a few weeks. In the majority of recently developed culture models for bioreactors, substrate exchange to the hepatocytes is performed by diffusion. This brings a remarkable change in the local situation for the cells in comparison to the in vivo situation. Even the gradients of oxygen and carbon dioxide, as well as the gradients of metabolites are not sufficiently considered in current culture models. An organisation of hepatocytes along such gradients is therefore ruled out, which impairs the typical metabolic function of hepatocytes dependent on their localisation in the sinusoid. A bioreactor concepts, which will allow the cells to operate under physiological flow conditions, may result in a macro environment more similar to the physiological situation. We have recently developed a culture model together with a newly developed bioreactor construction, which address this problem. This model has the following characteristics: perfusion of the cells between independent capillary membrane systems; identical units (Fig. 1) for few hepatocytes in parallel, analog to the liver lobuli; decentral metabolite inand outflow with low gradients; decentral oxygen supply and carbon dioxide removal with low gradients; cell adhesion on hepatocyte membranes coated with biomatrix and aggregation between the capillaries; co-culture compartment for sinusoidal endothelial cells; possibility of upscaling the construction to a large cell mass bioreactor for therapeutic liver support; possibility of further capillary functions, such as dialysis and heat-exchange for DMSO-freezing of the reactor with the cells. In several studies (1 ), this culture model has been investigated over a period of three weeks. With the present report, the initial results of a long-term study over seven weeks are described.

Hybrid liver support system in a short term application on hepatectomized pigs.

A short term application of a hybrid liver support system in circuits with continuous plasma-separation was investigated in a model of hepatectomized pigs under general anesthesia. Primary pig hepatocytes were immobilized in a bioreactor with three independent capillary systems. An immune barrier is achieved by avoiding the direct contact of blood cells with the hepatocytes by a plasmaseparation step and by an outflow filtration within the reactor. In three groups (hepatectomized pigs and system with- or without hepatocytes as well as untreated pigs with system without hepatocytes), the short term metabolism of the reactors was positively demonstrated by investigating ammonia detoxification, phenylalanine- and lactate metabolism. Limitations of the presented model are discussed.

Membranes as substrates for hepatocyte adhesion in liver support bioreactors.

Fourteen membranes out of cellulose (CuprophanR), polyamide and polypropylene were compared in a cytocompatibility test using the cytokinetics and cytomorphology of primary hepatatocytes as parameters. Additionally, the impact of coating the membranes with collagen or fibronectin was investigated. Hepatocytes were not able to attach in acceptable amounts on investigated cellulose membranes. On polyamide and polypropylene membranes a sufficient cell seeding was possible. Coating with collagen or fibronectin improves the attachment and spreading on all membranes. Differences between collagen and fibronectin were detected, observing the morphology of the cells: on collagen, most of the cells spread, whilst on fibronectin, most of the cells spread and flattened polygonally. If the adhesion of hepatocytes prolongs their metabolic function, a large adhesion surface in bioreactors is necessary. To reach a high surface area for cell adhesion in bioreactors one possibility is the use of polyamide and polypropylene membranes.

Hepatocyte aggregate culture technique for bioreactors in hybrid liver support systems.

Utilizing a modified culture technique for hepatocytes, a high performance suspension culture is possible in which hepatocytes spontaneously form cell aggregates. The aggregates of 20-100 cells have been histologically confirmed to hold a three-dimensional structure, they show a long-term external metabolism and a survival time comparable with standard adhesion cultures. This technique has several advantages in the construction of large scale bioreactors for hybrid liver support systems.

Side effects of hybrid liver support therapy: TNF-alpha liberation in pigs, associated with extracorporeal bioreactors.

During acute liver failure, hybrid liver support therapy could serve as a bridge to liver transplantation. In this desired temporary use, immune competent cell responses, such as the production of cytokines, might be of limiting relevance. We have investigated the Tumor Necrosis Factor-α (TNF) liberation in two models using pigs, connected with an extracorporeal bioreactor with homologous hepatocytes: TNF liberation was measured in arterial plasma during a 4 day perfusion time in untreated animals, model (i), and during short term perfusion of hepatectomized pigs in model (ii). Animals four days after catheter implantation in model (i) had TNF values of < 5 pg/ml. After connecting the system without hepatocytes, TNF rose to 9.7 ± 2 within 120 min and rose further to 32.6 ± 6 pg/ml within 4 hours after filling the system with the homologous hepatocytes. After 24 hours of continuous perfusion and during four days of perfusion, the TNF levels were lowered to baseline levels. In model (ii), TNF rose to 220 ± 130 pg/ml within 180 min and decreased to 110 ± 10 pg/ml within six hours, whereas controls without hepatocytes showed mean levels with a maximum of 120 ± 20 pg/ml. In both models, there was no correlation between TNF levels and clinical abnormalities such as fever or shock symptoms. There is evidence for an activation of blood cells during experimental extracorporeal hybrid support. No typical side effects were, however, observed. Thus, TNF mediated extracorporeal cell activation does not appear to limit the application of homologous hybrid liver support therapy.

Nonenzymatic versus Enzymatic Hepatocyte Isolation from Pig Livers for Larger Scale Investigations of Liver Cell Perfusion Systems

A comparison of nonenzymatic and enzymatic hepatocyte isolation was performed on pig livers. The collagenase perfusion showed superior results: mean viability 72 ± 10% versus a maximum viability of 21% using EDTA-perfusion. A five-step collagenase perfusion technique, developed for pig livers enables larger scale investigations, in order to develop methods for hepatocyte cultures in therapeutical liver cell perfusion systems.

Amino acid metabolism by hepatocytes in a hybrid liver support bioreactor.

The amino acid patterns of medium perfusate in a liver cell bioreactor developed for a hybrid liver support system have been measured. There were considerable changes in the concentrations of glutamic acid, glutamine, alanine, arginine, ornithine and branched chain amino acids during the first 10 days which is indicative of dynamic cellular metabolism. From day 15, steady state conditions of nitrogen metabolism are reflected by stable amino acid turnover. Monitoring of urea, K+, and P-450 activity suggests that hepatocytes have switched to a stable protein synthesis with a general amino acid uptake and keto acid release following cell volume increase

Alpha-keto acid metabolism by hepatocytes cultured in a hybrid liver support bioreactor.

Isolated pig liver cells cultured using a perfusion technique were analyzed over 39 days to test their ability to change the perfusate α-keto acid profile. While the pyruvate concentration in the culture medium decreased as of the first day, the α-ketoglutarate (KG), α-ketoisocaproate (KIC), α-ketoisovalerate (KIV) and α-ketoβ-methyl-n-valerate (KMV) were synthesized immediately and released by the liver cells. The metabolic capacity of the cell culture system increased up to day 10, decreased during the following 5 days and reached a steady state beyond day 15, which was maintained for at least 30 days. The branched chain α-keto acid release, in particular α-ketoisocaproate, reflects an effective transamination capacity of the newly developed culture system and shows an intact protein biosynthesis for at least 30 days in vitro.

Hepatocyte isolation from pig livers after warm ischaemic injury.

Abstract Hepatocyte cultures have been used extensively for a wide variety of physiological, pharmacological and experimental studies. The warm ischaemic period before isolation is kept to a minimum to achieve a high yield of cells isolated and a good viability for culture. We have recently introduced a new concept of liver resuscitation after warm ischaemia that is based on a 3‐h reperfusion period with an improved perfusate and simultaneous dialysis. In this study, we applied the new technique for hepatocyte isolation from livers subjected to 80 min of complete ischaemia at 37 °C. Cell yield was improved by a resuscitating perfusion from 58% to 73% and viability from 39% to 76%.

Polyurethanes and Their Cytocompatibility for Cell Seeding

The present studies were done to investigate the cell behaviour on polymers as parameter for testing the cytocompatibility of biomaterials for cell coating. Endothelial cell kinetics and -morphology during initial adhesion, initial spreading and proliferation of the cells were examined as sensitive reaction of cells to the polymers. All investigated smooth polyurethanes had a similar cytokinetic and proliferation behaviour.Sprayed Tecoflex showed a faster initial adhesion and spreading of cells, but not a better cell proliferation than smooth Tecoflex. The results have shown that surface modifications have an influence on cell behaviour. Using bovine aortic EC, polyurethane as a basis for cell seeding can be used without any coating.