Apoptosis by Gel-Entrapped Hepatocytes in a Bioartificial Liver D.J. Rivera, G.J. Gores, S.P. Misra, J.A. Hardin, and S.L. Nyberg B IOARTIFICIAL LIVER systems have been developed to provide short-term metabolic function to patients with acute hepatic failure. One such system involves the entrapment of hepatocytes in cylindrical collagen gels lo- cated in the fibers of a hollow fiber bioreactor. 1–3 An in vitro system utilizing gel entrapment rat hepatocytes (GERH) has been developed for static culture experimen- tation. 4,5 This in vitro model avoids the expenses of large scale testing. Also, multiple experiments can be performed simultaneously. Potential immune interactions present with in vivo models of BAL testing are avoided with an in vitro system. Although liver specific functions are present after several weeks of culture, a loss of hepatocyte viability begins at the time of hepatocyte isolation and continues at a slow rate after gel entrapment. 6 The decline in hepatocyte viability of over time limits the effectiveness and duration of BAL therapy. Two distinct morphological forms of hepatocyte death (apoptosis and necrosis) have been identified. 7–9 Necrosis is characterized by cell swelling, loss of cellular ATP, and membrane lysis. In contrast, apoptosis is characterized by cell shrinkage, condensation of nuclear chromatin, DNA cleavage, and ultimately, cell fragmentation. 10 Another distinguishing characteristic of these two forms of cell death is that pharmacologic use of the amino acid glycine atten- uates necrosis in hepatocytes 11,12 and other epithelial cells, 13 but has no effect on apoptosis. Currently, there are no studies regarding the mode of hepatocyte cell death in regards to the BAL and GERH. Thus, the specific aims of this study were to answer the following questions about GERH: (1) What is the primary mechanism of cell death (necrosis or apoptosis) by GERH in the static culture BAL system? (2) Does glycine increase viability and/or function- ality of GERH? MATERIALS AND METHODS Hepatocyte Isolation and Gel-Entrapment Rat hepatocytes were harvested from 4 to 6-week-old male Sprague-Dawley rats, weighing 200 to 250 grams by a two step in situ collagenase perfusion technique modified from the method described by Seglen. 1,14 Rat harvests yielded from 3.0 to 5.0  10 8 hepatocytes with an average viability of greater than 85% by trypan blue exclusion. Hepatocytes were entrapped in collagen gel at 2.5  10 6 cells/mL  0.1 mL as previously reported. 4 Control gels were transferred into 3 mL of serum-free culture media (William’s E medium supplemented with 5 g/L epidermal growth factor, 200 U/L insulin, 1 mg/L glucagon, 5 mg/L transferrin, 0.5 g/L albumin, 5 mg/L linoleic acid, 1 md/L dexamethasone, 6.25 g/L selenium, 40,000 U/L penicillin G, 400 mg/L streptomycin sulfate, and 15 mmol/L HEPES). Media for the experimental group was supple- mented with 3 mmol/L glycine. Gels were maintained at 37°C in a 5% CO 2 incubator. Samples of culture medium and GERH were collected at 2 hour and 24 hours after entrapment. Media samples were stored at -20°C prior to analysis. GERHs were analyzed at the time of sampling. Determination of Cell Viability Vital Staining. Hepatocyte viability was quantified using a digitized, video archival, epifluorescence microscopy system after vital staining with fluorescein diacetate:ethidium bro- mide (FDA:EB). 15 Stock solutions of FDA (5 mg/mL in acetone) and EB (10 g/mL in phosphate buffered saline) were stored at 4°C. The working FDA:EB (5 g/mL:10 g/mL) solution was prepared fresh before each use by adding 10 L of FDA stock solution to 10 mL of EB stock solution. GERH were incubated in FDA/EB stain for 5 minutes at 37°C and then washed twice in phosphate buffer solution. Viability of hepatocytes (100 to 200 cells in 3 fields at 400  magnification) was quantified with Image Pro Plus 3.0 software (Silver Spring, Md) and a epifluorescent mi- croscope (Axiovert; Carl Zeiss, Inc., Thornwood, NY) configured for FITC (450 to 490 nm excitation filter, 510 nm barrier filter). Cells with green staining cytoplasm are scored as viable. Cells with orange staining nuclei were scored as dead. Percent viability was expressed as: (viable cells/total cells)  100%. From the Department of Surgery, Mayo Clinic, Rochester, Minnesota (D.J.R., J.A.H., S.L.N.), Department of Internal Med- icine, Mayo Clinic, Rochester, Minnesota (G.J.G.), University of Nevada School of Medicine, Reno, Nevada (D.J.R.), USA; and Allahabad, India (S.P.M.). Financial Support of Research by the Mayo Foundation. Funded by NIH R25 GM 55252-01 (D.J.R.). International Travel Award from the American College of Gastroenterology (S.P.M.). Address reprint requests to Scott L. Nyberg, MD, PhD, Liver Transplantation Unit, Mayo Clinic, Eisenberg 3G, 200 First Street SW, Rochester MN, 55905, USA. © 1999 by Elsevier Science Inc. 0041-1345/99/$–see front matter 655 Avenue of the Americas, New York, NY 10010 PII S0041-1345(98)01743-6 Transplantation Proceedings, 31, 671–673 (1999) 671