The genes that encode the five known enzymes of the mandelate pathway of Pseudomonas putida (ATCC 12633), mandelate racemase (mdlA), (S)-mandelate dehydrogenase (mdlB), benzoylformate decarboxylase (mdlC), NAD(+)-dependent benzaldehyde dehydrogenase (mdlD), and NADP(+)-dependent benzaldehyde dehydrogenase (mdlE), have been cloned. The genes for (S)-mandelate dehydrogenase and benzoylformate decarboxylase have been sequenced; these genes and that for mandelate racemase [Ransom, S. C., Gerlt, J. A., Powers, V. M., & Kenyon, G. L. (1988) Biochemistry 27, 540] are organized in an operon (mdlCBA). Mandelate racemase has regions of sequence similarity to muconate lactonizing enzymes I and II from P. putida. (S)-Mandelate dehydrogenase is predicted to be 393 amino acids in length and to have a molecular weight of 43,352; it has regions of sequence similarity to glycolate oxidase from spinach and ferricytochrome b2 lactate dehydrogenase from yeast. Benzoylformate decarboxylase is predicted to be 499 amino acids in length and to have a molecular weight of 53,621; it has regions of sequence similarity to enzymes that decarboxylate pyruvate with thiamin pyrophosphate as cofactor. These observations support the hypothesis that the mandelate pathway evolved by recruitment of enzymes from preexisting metabolic pathways. The gene for benzoylformate decarboxylase has been expressed in Escherichia coli with the trc promoter, and homogeneous enzyme has been isolated from induced cells.

The two preceding papers [Powers, V. M., Koo, C. W., Kenyon, G. L., Gerlt, J. A., & Kozarich, J. W. (1991) Biochemistry (first paper of three in this issue); Neidhart, D. J., Howell, P. L., Petsko, G. A., Powers, V. M., Li, R., Kenyon, G. L., & Gerlt, J. A. (1991) Biochemistry (second paper of three in this issue)] suggest that the active site of mandelate racemase (MR) contains two distinct general acid/base catalysts: Lys 166, which abstracts the alpha-proton from (S)-mandelate, and His 297, which abstracts the alpha-proton from (R)-mandelate. In this paper we report on the properties of the mutant of MR in which His 297 has been converted to asparagine by site-directed mutagenesis (H297N). The structure of H297N, solved by molecular replacement at 2.2-A resolution, reveals that no conformational alterations accompany the substitution. As expected, H297N has no detectable MR activity. However, H297N catalyzes the stereospecific elimination of bromide ion from racemic p-(bromomethyl)mandelate to give p-(methyl)-benzoylformate in 45% yield at a rate equal to that measured for wild-type enzyme; the unreacted p-(bromomethyl)mandelate is recovered as (R)-p-(hydroxymethyl)mandelate. At pD 7.5, H297N catalyzes the stereospecific exchange of the alpha-proton of (S)- but not (R)-mandelate with D2O solvent at a rate 3.3-fold less than that observed for incorporation of solvent deuterium into (S)-mandelate catalyzed by wild-type enzyme. The pD dependence of the rate of the exchange reaction catalyzed by H297N reveals a pKa of 6.4 in D2O, which is assigned to Lys 166.(ABSTRACT TRUNCATED AT 250 WORDS)

No abstract

16) The "C NMR chemical shifts of the 1-carbons of ethanol, 3methylbutanol, and 3,3-dimethylbutanol are 57.0, 60.2, and 58.9 ppm, respectively: Carbon-13 NMR Spectra Data; Ve-rlag Chemie: New York, 1979.(17) The uracil-DNA glycosylase is stored in a buffer containing dithiothreitol, and UV endo V is stored in a buffer containing 2-mercaptoethanol. (18) The I3C NMR chemical shift of carbon-1 in 1-(methylthio)cyclohexane is 44.9 ppm: Eliel, E. L.; Kandasamy, D.

The gene (pcaB) for 3-carboxymuconate lactonizing enzyme (CMLE; 3-carboxymuconate cycloisomerase; EC 5.5.1.2) from Pseudomonas putida has been cloned into pMG27NS, a temperature-sensitive expression vector, and expressed in Escherichia coli N4830. The specific activity and kinetic parameters of the recombinant CMLE were comparable to those previously reported. A comparison of the deduced amino acid sequence of CMLE with sequences available in the PIR and Genbank databases revealed that CMLE has highly significant sequence homology to the class II fumarase family, particularly to adenylosuccinate lyase from Bacillus subtilis. CMLE has no significant homology to muconate lactonizing enzyme (MLE) from P. putida, its sister enzyme in the beta-ketoadipate pathway. These findings fully corroborate a prediction made by us on the basis of mechanistic and stereochemical analyses of CMLE and MLE [Chari, R. V. J., Whitman, C. P., Kozarich, J. W., Ngai, K.-L., & Ornston, L. N. (1987) J. Am. Chem. Soc. 109, 5514-5519] and suggest that CMLE and MLE were recruited into this specialized pathway from two different enzyme families.

These findings suggest that apparent selective pituitary resistance and generalized resistance to thyroid hormone are not qualitatively different syndromes. Nevertheless, identification of selective pituitary resistance is a useful clinical distinction since such patients with clinical and biochemical features of hyperthyroidism appear to benefit from reduction in serum thyroid hormone concentrations. In contrast, patients with more conventional forms of thyroid hormone resistance require no treatment or may benefit from increased concentrations of thyroid hormone.

The plasmid pSCR1 containing the gene for mandelate racemase (EC 5.1.2.2) from Pseudomonas putida (ATCC 12633) allows Pseudomonas aeruginosa (ATCC 15692) to grow on (R)-mandelate as its sole carbon source [Ransom, S. C., Gerlt, J. A., Powers, V. M., & Kenyon, G. L. (1988) Biochemistry 27, 540]; the chromosome of the P. aeruginosa host apparently does not contain the gene for mandelate racemase but does contain genes for the remaining enzymes in the mandelate pathway and enables growth on (S)-mandelate as carbon source. However, in the presence of alpha-phenylglycidate, an active-site-directed irreversible inhibitor (affinity label) of mandelate racemase, P. aeruginosa transformed with pSCR1 can utilize (S)-mandelate but not (R)-mandelate as carbon source. This inhibition of growth on (R)-mandelate provides a metabolic selection for mutants that are resistant to alpha-phenylglycidate. When (R)-mandelate is used as carbon source and alpha-phenylglycidate is present, a few colonies of P. aeruginosa transformed with pSCR1 grow slowly and appear on plates after several days. The plasmid isolated from these cells confers resistance to alpha-phenylglycidate on newly transformed cells of P. aeruginosa. This resistance to the affinity label is not due to a mutation within the primary structure of the enzyme. A single base change (C----A) located 87 bp upstream of the initiation codon for the gene for mandelate racemase was detected in three independent isolates of alpha-phenylglycidate-resistant colonies and appears responsible for a 30-fold increase in the amount of mandelate racemase encoded by the gene contained in the plasmid.(ABSTRACT TRUNCATED AT 250 WORDS)

Mutations in the gene encoding the human beta 1 T3 receptor (hTR beta 1) have been associated with generalized resistance to thyroid hormone (GRTH). We measured the T3-binding affinity and transcriptional regulatory capacity of the mutant hTR beta 1 from four unrelated kindreds with GRTH. These mutations are contained in different functional regions of the ligand-binding domain. The T3 affinity of the mutant receptors correlated well with the degree of impairment of their trans-activating function in a transient cotransfection system in HeLa cells; two mutant receptors with undetectable ligand affinity showed no transcriptional activity, whereas the two other mutants characterized by a 2- and 5-fold reduction in T3 affinity required 5- and 15-fold higher T3 concentrations for half-maximal activity in the cotransfection assay, respectively. All of the mutant hTR beta 1s were able to inhibit the function of transfected normal hTR beta 1 and endogenous retinoic acid receptor in activating a palindromic positive T3 response element (TRE). In the partially functional mutants this dominant negative effect could be completely reversed by increased T3 concentrations. The dominant negative potency did not depend on the type of TRE used; mutant hTR beta 1s were able to inhibit normal receptor function to the same degree on a dimer-permissive palindromic TRE as on a nondimer-permissive inverted repeat of two identical half-sites separated by five spacer bases. However, the dominant negative potency was dependent on the absolute amount of receptor expression vector transfected. The expression of normal and mutant hTR beta 1 was assessed by immunocytochemistry. The hTR beta 1 protein levels in HeLa cells paralleled the amount of transfected expression vector. Moreover, all the mutant receptors were properly expressed in the nuclei of the transfected cells. These data suggest that different mutations in the ligand-binding domain of the human hTR beta 1 result in a variable degree of functional impairment, which may partially explain the phenotypic differences between kindreds with GRTH. Our findings suggest that competition for binding to the TRE and possibly the binding of limiting accessory factors may be more important in mediating the dominant negative effect than the formation of normal/mutant T3 receptor dimers.