Regulation of hepatic malic enzyme mRNAs during development.

Previous studies in vivo have demonstrated that in neonatal and adult rats hepatic malic enzyme expression is controlled at the pretranslational level by dietary and hormonal factors [ 1, 21. However, in order to demonstrate that particular effectors have a direct effect on gene expression in liver cells, techniques in vitro, such as the use of primary cultures of non-proliferating rat hepatocytes, have to be adopted. Unfortunately, the widely used practice of plating cells on type 1 collagen leads to decreased hepatocyte function as evidenced for example by diminished albumin production [3]. Recently, a new substratum, EHS matrix, has been employed which appears to result in greater maintenance of liver specific gene expression [4, 51. In the present study, the response of malic enzyme gene expression t o various putative effectors has been compared in adult hepatocytes plated on to type 1 collagen or Engelbreth-Holm-Swarm (EHS) matrix. Malic enzyme activity has been determined in cell lysate supernatant solutions and dot-blot and Northern blot analysis has been used to determine the amount and size of malic enzyme and albumin mRNA in hepatocytes maintained in culture for either 4 or 6 days in serum-free chemically defined medium containing 5 mM-glucose. Attachment of isolated hepatocytes was similar on collagen and EHS matrix, with approx. 80% of cells attached after 60 min. Hepatocytes plated on collagen flattened within 24 h to form confluent monolayers of polygonal, epitheliallike cells, whereas on EHS matrix cells remained spherical and aggregated within 24 h to form three-dimensional clusters of 10-20 cells as previously reported [ 5 ] . The levels of all parameters measured (malic enzyme mRNA and activity, albumin mRNA and lactate dehydrogenase activity) were very similar in cells assayed immediately after attachment to either substratum to those detected in the whole animal. Such post-plating values were standardized with those of the normal adult rat liver to facilitate a comparison of the magnitude of the responses of malic enzyme expression to the various effectors in vivo and in vitro. Hepatocytes plated o n collagen and maintained in culture for 4 days exhibited little response to tri-iodothyronine, insulin or dichloracetate with regard to malic enzyme activity or mRNA levels. In addition, the amounts of albumin mRNA were markedly reduced compared with the levels determined in cells immediately after isolation. However, after 6 days in culture the effectors did cause increases in enzyme activity and mRNA levels relative to control cells which were found to have markedly reduced lcvels of malic enzyme expression compared with freshly plated cells. The levels of malic enzyme induction observed were considerably less than those arising in vivo after treatment with the appropriate effector. Dramatic differences were observed when hepatocytes were incubated on EHS-coated culture dishes. The level of malic enzyme induction after 4 days of treatment with the appropriate effector corresponded to that recordcd after 6 days’ stimulation of cells on the collagenous substratum. After 6 days in culture using EHS matrix, increased induction of malic enzyme activity and mRNA was obtained with insulin, tri-iodothyronine and dichloroacetate such that ;I similar response was obtained to that recorded in t i w . In addition. EHS-maintained cells produced albumin mRNA at about twice the level found in cells immediately after isolation. thus indicating maintenance of hepatic function. In order to try to elucidate the nature o f potential mctabolic intermediates involved in the induction o f malic enzyme expression, hepatocytes platcd on EHS gel were maintained for 6 days in culture medium containing 5 mMglucose and either malate ( 5 mM), fructose ( 5 mM) o r high glucose concentrations (25 mM. final). Malate caused induction of malic enzyme in the absence of insulin. whereas fructose and high glucose concentrations only caused increased enzyme expression in the presence o f insulin. The direct or indirect role of malate in malic enzyme induction merits further investigation. Thus use of EHS matrix provides a cell-culture system for hepatocytes which resembles the situation in vivo for enzyme induction. However, a note o f caution is necessary, since Northern blot analysis demonstrates that whereas in vivo and in freshly isolated cells the ratio o f 2.0 kb t o 3.1 kb malic enzyme mRNA species is 3, it falls t o 0.5 after 3 days in culture. Paradoxically, hepatocytes cultured o n EHS matrix may provide an excellent system in which t o study the fiictors involved in the production of the various malic cnzymc mRNA species.