<b><i>Introduction:</i></b> Liver failure is characterized by compromised hepatic detoxification, protein synthesis, and metabolic derangements leading to an accumulation of a broad spectrum of water-soluble and lipophilic toxins as well as immune system mediators. Exploring complex detoxification mechanisms to therapeutically target those components, this article will focus on similarities, differences, and potential synergies in the mechanism of albumin dialysis and hemoperfusion. <b><i>Methods:</i></b> An in vitro two-compartment model for the comparison of liver support techniques was used to compare MARS albumin dialysis modified with novel charcoal adsorbents to CytoSorb hemoperfusion with added hemodialysis for effects on marker molecule removal. <b><i>Results:</i></b> MARS and CytoSorb performed similar in the removal of water-soluble toxins. Ammonia removal was increased using CytoSorb. CytoSorb lead to a statistically significant reduction of albumin-bound toxins, total bilirubin and subfractions. Bile acid removal was comparable. MARS demonstrated no removal of cytokines interleukin (IL)-6 and tumor necrosis factor-alpha (TNF-α), whereas CytoSorb allowed for near complete removal. Notably, CytoSorb displayed 50% of lipophilic substance and cytokine removal during the first hour of treatment. <b><i>Conclusion:</i></b> Compared to MARS, CytoSorb hemoperfusion leads to an initially fast removal of cytokines, TNF-α and IL-6, as well as reduction of albumin-bound toxins such as indirect bilirubin and bile acids in our model. The initial removal is also associated with removal of albumin.

The removal of small water soluble toxins and albumin-bound toxins in acute liver failure patients (ALF) or acute-on-chronic liver failure (AocLF) patients has been established using extracorporeal liver support devices (e.g. Molecular Adsorbents Recirculating System; MARS). However, reduction of elevated cytokines in ALF/AocLF using MARS is still not efficient enough to lower patients' serum cytokine levels. New membranes with larger pores or higher cut-offs should be considered in extracorporeal liver support devices based on albumin dialysis in order to address these problems, as the introduction of super-large pore membranes could counterbalance high production rates of cytokines and further improve detoxification in vivo. Using an established in vitro two compartment albumin dialysis model, three novel membranes of different pore sizes were compared with the MARS Flux membrane for cytokine removal and detoxification qualities in vitro. Comparing the membranes, no improvement in the removal of water soluble toxins was found. Albumin-bound toxins were removed more efficiently using novel large (Emic2) to super-large pore sized membranes (S20; HCO Gambro). Clearance of cytokines IL-6 and tumor necrosis factor-α was drastically improved using super-large pore membranes. The Emic2 membrane predominantly removed IL-6. In vitro data suggest that the usage of larger pore sized membranes in albumin dialysis can efficiently reduce elevated cytokine levels and liver failure toxins. Using large to super-large pore membranes might exert effects on patients' serum cytokine levels. Combined with increased detoxification this could lead to higher survival in ALF/AocLF. Promising membranes for clinical evaluation have been identified.

Albumin dialysis in extracorporeal organ support is often performed in the treatment of liver failure as it facilitates the removal of toxic components from the blood. Here, we describe a possible effect of albumin dialysis on proinflammatory cytokine levels in vitro. Initially, albumin samples were incubated with different amounts of cytokines and analyzed by enzyme-linked immunosorbent assay (ELISA). Analysis of interleukin 6 (IL-6) and tumor necrosis factor alpha (TNFα) levels indicated that increased concentrations of albumin reduce the measureable amount of the respective cytokines. This led to the hypothesis that the used proinflammatory cytokines may interact with albumin. Size exclusion chromatography of albumin spiked with cytokines was carried out using high-performance liquid chromatography analysis. The corresponding fractions were evaluated by immunoblotting. We detected albumin and cytokines in the same fractions indicating an interaction of the small-sized cytokines IL-6 and TNFα with the larger-sized albumin. Finally, a two-compartment albumin dialysis in vitro model was used to analyze the effect of albumin on proinflammatory cytokines in the recirculation circuit during 6-h treatment. These in vitro albumin dialysis experiments indicated a significant decrease of IL-6, but not of TNFα, when albumin was added to the dialysate solution. Taken together, we were able to show a putative in vitro interaction of human albumin with the proinflammatory cytokine IL-6, but with less evidence for TNFα, and demonstrated an additional application for albumin dialysis in liver support therapy where IL-6 removal might be indicated.

Albumin is the most abundant plasma protein and functions as a transport molecule that continuously interacts with various cell types. Because of these properties, albumin has been exploited by the pharmaceutical industry to improve drug delivery into target cells. The immediate effects of albumin on cells, however, require further understanding. The cell interacting properties and pharmaceutical applications of albumin incentivises continual research into the immediate effects of albumin on cells. The HepG2/C3A hepatocellular carcinoma cell line is used as a model for studying cancer pathology as well as liver biosynthesis and cellular responses to drugs. Here we investigated the direct effect of purified albumin on HepG2/C3A cell proliferation in the absence of serum, growth factors and other serum originating albumin bound molecules. We observed that the reduced cell counts in serum starved HepG2/C3A cultures were increased by the inclusion of albumin. Cell cycle analysis demonstrated that the percentage of cells in G1 phase during serum starvation was reduced from 86.4 ± 2.3% to 78.3 ± 3.2% by the inclusion of albumin whereas the percentage of cells in S phase was increased from 6.5 ± 1.5% to 14.3 ± 3.6%. A significant reduction in the cell cycle inhibitor protein, P21, accompanied the changes in the proportions of cell cycle phases upon treatment with albumin. We have also observed that the levels of dead cells determined by DNA fragmentation and membrane permeabilization caused by serum starvation (TUNEL: 16.6 ± 7.2%, ethidium bromide: 13.8 ± 4.8%) were not significantly altered by the inclusion of albumin (11.6 ± 10.2%, ethidium bromide: 16.9 ± 8.9%). Therefore, the increase in cell number was mainly caused by albumin promoting proliferation rather than protection against cell death. These primary findings demonstrate that albumin has immediate effects on HepG2/C3A hepatocellular carcinoma cells. These effects should be taken into consideration when studying the effects of albumin bound drugs or pathological ligands bound to albumin on HepG2/C3A cells.

No abstract