In mammalian cytochrome c oxidase (COX) three of the ten nuclear coded subunits (VIa, VIIa, VIII) occur in tissue-specific isoforms. The isoform distribution, however, varies in liver and heart of different species. Subunit VIII is different in liver and heart of bovine, dog, rat and chicken, but identical in human (liver-type) on one hand, and sheep, rabbit and rainbow trout (heart-type) on the other hand, as determined by N-terminal sequencing. Two moles of trinitrophenyl-ATP bind to monomeric COX from bovine heart and one to COX from bovine liver with dissociation equilibrium constant (Kd) values of about 3 microM. One binding site at the heart enzyme is blocked by a monoclonal antibody to subunit VIa-H. ATP (and/or ADP) interact with COX at two or three high-affinity binding sites, as shown by titration of the spectral changes of COX. Isolated COX from bovine heart was reconstituted with variable intraliposomal ATP/ADP ratios. By measuring the RCR (respiratory control ratio) and RCRVal (related to the valinomycin-respiration), which is a direct measure of the H+/e(-)-stoichiometry (Wilson and Prochaska, Arch. Biochem. Biophys. 282 (1990) 413-420), almost complete inhibition of the proton pump activity of COX by high intraliposomal ATP concentrations was found. The vectorial of protons for the formation of water, however, appears to be unaffected by nucleotides. This regulatory mechanism is assumed to have physiological significance for thermogenesis in muscle at rest. COX of fibroblasts from patients suffering from Leigh's syndrome, which is associated with a decreased COX activity, are suggested to have an incompletely assembled enzyme complex. This suggestion is further corroborated by the higher temperature-sensitivity of the enzyme when compared with COX from normal control fibroblasts. Defective regulation of COX via nuclear coded subunits is also proposed to cause mitochondrial diseases.

No abstract

Cytochrome c oxidase was isolated from heart and liver of rainbow trout (Salmo gairdnerii). SDS/PAGE analysis showed the presence of 11 different polypeptide subunits in the fish enzyme. The nuclear‐coded subunits IV, Va, Vb, VIc, VIIa, VIIc and VIII could be identified by their N‐terminal amino acid sequences. The mammalian subunits VIa and VIlb appear to be absent (or blocked at the N‐terminal) in cytochrome c oxidase from trout. For subunit Vb, two polypeptides of different electrophoretic mobilities were found which differed in their N‐terminal sequences, and represent a new pair of cytochrome‐c‐oxidase subunit isoforms, not found in mammalia. Both isoforms of subunit Vb were found in cytochrome c oxidase from heart and liver, but at different ratios. Subunit VIIa also seemed to occur in different isoforms, whereas subunit VIII had the same N‐terminal amino acid sequence in cytochrome c oxidase of liver and heart, similar to the human‐type subunit but different from rat, bovine and chicken.

Epidermal hyperproliferation (psoriasis, wound repair) is the result of quiescent (G0) keratinocytes being recruited into the cell cycle. A detailed characterization of the cell cycle kinetic parameters of the mouse keratinocyte line (Balb/MK) has been carried out with the aid of bivariate iododeoxyuridine (IdUrd) and DNA analysis using flow cytometry, in order to establish whether it might provide a useful model for the study of the biochemical events controlling recruitment into the cell cycle. Balb/MK keratinocytes were cultured using low Ca2+ Dulbecco's modified Eagle's medium/F 12 in the presence of 10% dialysed fetal bovine serum and 4 ng/ml epidermal growth factor (EGF). IdUrd labelling followed by flow cytometric analysis of trypsinized cells showed that about 95% of the population were actively cycling, with a cell cycle time of around 14h and no significant contact inhibition. Omission of serum and EGF followed by IdUrd pulse-labelling indicated that the cells progressively withdrew from the cycle and, after 16h, less than 10% incorporated IdUrd. Subsequent restimulation with serum resulted in a synchronized cohort of cells being recruited. Entry into the S phase of the cell cycle (IdUrd incorporation) started at 8 h and was maximal between 12 h and 16h after stimulation. Restimulation with EGF alone reached a growth fraction of 87% after 24 h continuous labelling compared with 97% using serum together with EGF. Epidermal growth factor already showed a significant stimulation at 10 pg/ml compared with the controls. There is a broad plateau centred on 5 ng/ml, followed by a slight decline above this level. We conclude that the use of a cell line with defined cell cycle kinetic parameters which can be switched between the quiescent and cycling states in a fully defined medium will provide an ideal model for biochemical studies of the relevant signal transduction pathways in keratinocytes.

No abstract