In situ and perfused rat livers showed a spon- .) The existence of such light emission, which should be termed "low-level chemiluminescence" to differentiate it from the more effective photoemission of the luciferin/luciferase systems (5, 6), was soon related to oxygendependent chain reactions involving biological lipids (3-5). This early work lay fallow for years, notwithstanding the reports by Stauff and Ostrowski on the chemiluminescence of mitochondria (7) and Howes and Steele on the chemiluminescence of microsomes (8, 9), both isolated from rat liver. The more recent reports by Nakano et al. (10) and Sugioka and Nakano (11) of light emission during lipid peroxidation and other oxidative reactions (12) in microsomes revived interest in the phenomenon and suggested chemiluminescence as a tool for the investigation of the radical reactions of lipid peroxidation under physiological conditions. We have recently reported that maximal light emission in isolated mitochondria and microsomes (13) and in submitochondrial particles (14) requires an electron transfer system, hydroperoxide, and oxygen, and that hydroperoxide-supplemented cytochrome c provides a chemiluminescent model system suitable for the elucidation of some of the molecular mechanisms responsible for light emission (15). On the other hand, isolated cells such as amoebae (16) and phagocytizing leukocytes (17) also have been found to be effective chemiluminescent sources.The most important aspect of the organ chemiluminescence is that it gives readily detectable, continuously monitorable, noninvasive signals of oxidative metabolism. This article explores the possibility of continuously monitoring the metabolism of exposed or fiberoptic probed organs in vivo by the chemiluminescent technique. In this paper we report the spontaneous and hydroperoxide-induced chemiluminescence of the in situ and perfused rat liver, as well as a partial spectral analysis of the chemiluminescence of the perfused liver. Light emission seems to indicate the generation of shkrt-lived free radicals and excited states derived from the side reactions of the free radical process of lipid peroxidation. A preliminary report on light emission has been published elsewhere (13).
MATERIALS AND METHODSPhoton Counting. A single-photon-counting apparatus was used (Fig. 1). Both an EMI 9658 photomultiplier, responsive in the range 300-900 nm, with an applied potential of -1.2 kV (dark current: 20-30 counts per second), and an RCA 8850 photomultiplier, responsive in the range 300-650 nm, with an applied potential of -1.8 kV (dark current: 300-400 counts per second) were used. Phototube output was connected to an amplifier-discriminator (model 1121; Princeton Applied Res., Princeton, NJ) adjusted for single photon counting and connected to both a frequency counter (Heathkit IB 1100, Heath, Benton Harbor, MI) and a recorder. The EMI phototube, cooled .down to -400C by a thermoelectric cooler (EMI-Gencom, Plainview, NY) and the RCA phototube were placed in an Ortec housing, sealed and su...