Search

We previously reported the construction of an overexpression system of Candida tropicalis peroxisomal catalase (CTC) gene in the Saccharomyces cerevisiae cytoplasm using the GAL7 promoter (Kinoshita et al. , Appl. Microbiol. Biotechnol., 40: 682–686, 1994). To study the mechanism of the subcellular localization of peroxisomal catalase which is composed of four identical subunits and four heme molecules, the system was improved by changing the promoter used in order to express the catalase subunit accompanying the proliferation of peroxisomes in S. cerevisiae cells ( S. cerevisiae ΔAΔT strain), in which catalase A and catalase T genes were disrupted. On native-PAGE, besides the functional tetrameric form (220 kDa), other larger multimeric forms of recombinant CTC were observed in the cell-free extract. It was clearly demonstrated that only the tetrameric form, among the various forms, contained heme and exhibited catalase activity. Subcellular fractionation and protease protection experiments indicated that only the functionally active tetrameric form was present inside the peroxisomes even under conditions of overexpression of the catalase subunit, in which different multimeric forms including the active tetrameric form were produced outside the peroxisomes. The results indicated that C. tropicalis catalase subunit and heme stoichiometrically localize even in S. cerevisiae peroxisomes, suggesting a common regulatory role of peroxisomes themselves for the transport of the oligomeric form of catalase beyond yeast species.

In this paper, we derive an exact test for linear restrictions on regression coefficients in seemingly unrelated regressions with the same regressors in each equation. The test statistic is distributed as the T 2 distribution, and thus critical values can be obtained from the F distribution. Since the test for symmetry in demand systems is a special case of the test for linear restrictions, our test gives an exact test for symmetry.

This paper considers small sample properties of estimators for an autocorrelation coefficient ( p ) and of test statistics based on Prais-Winston (henceforth, P-W) estimators in a linear model with AR(1) errors. We propose two types of new modified P-W estimators (i.e., the bias-corrected P-W estimators) making use of the second-order approximation suggested by Ullah et al. (1983). It is shown by Monte Carlo experiments that one of the modified estimators performs remarkably better in hypothesis testing that the original P-W estimator especially in highly autocorrelated cases.