No abstract

We have sequenced the structural gene and flanking regions for lanosterol 14 alpha-demethylase (14DM) from Saccharomyces cerevisiae. An open reading frame of 530 codons encodes a 60.7-kDa protein. When this gene is disrupted by integrative transformation, the resulting strain requires ergosterol and, as expected, grows only in the absence of oxygen. The deduced amino acid sequence of 14DM includes a hydrophobic segment near the amino terminus which may be a transmembrane domain. The deduced sequence has been compared with those of eight other eukaryotic P450s, each from a different family within the P450 superfamily. These comparisons indicate that this yeast gene is the first member of a new P450 family, P450LI. The P450, designated P450LIA1, is more closely related to mammalian P450s than to the bacterial P450cam. In fact, both the yeast P450 and several mammalian P450s have equivalent alignment scores when each is compared with the bovine P450scc. Matrix comparisons of the amino acid sequence of this P450 with those of mammalian P450s reveal three conserved regions. The DNA region 5' to the structural 14DM gene includes poly(dA:dT) sequences and a repeating hexamer sequence.

Proteins from eight eukaryotic families in the cytochrome P-450 superfamily share one region of sequence similarity. This region begins 275-310 amino acids from the amino terminus of each P-450, continues for =170 residues, and ends 35-50 amino acids before the carboxyl terminus. The region can be divided into four domains of sequence similarity, each possessing its own pattern of invariant, conserved, and variable amino acids. The four domains are 56,20, 59, and 28 residues long and are connected by three shorter segments of limited sequence similarity. The number of residues in these short segments varies with the P-450 protein but ranges from 0 to 20 residues. Consensus sequences based on these similarities can be used to determine whether the sequence of an unidentified peptide resembles that expected for a P-450. Sequence similarities between proteins sometimes reflect constraints imposed by the requirements of a common function. The fourth domain of the P-450s, for example, contains an invariant cysteine that provides the axial thiolate ligand to the heme iron. Other relationships between the four domains and P-450 function can be examined by in vitro mutagenic procedures that alter the conserved amino acids or modify the distance between domains.The P-450 cytochromes are components in monooxygenase systems that catalyze a wide variety of oxidative reactions in prokaryotes and eukaryotes. These reactions include steps in the synthesis and degradation of such compounds as cholesterol, steroid hormones, and prostaglandins. P-450 proteins are also involved in the metabolism of drugs and in the activation and inactivation of carcinogens. A recent analysis of >60 P-450s from eukaryotes and one prokaryote led to the organization of the known P-450s into 10 different families with a total of 15 subfamilies. Amino acid sequences within a P-450 family are >36% similar, while sequences within a subfamily are :70% similar (1). Although sequence relatedness between P-450 families is low, these families are still grouped into one P-450 superfamily according to standard criteria (2).The eukaryotic branch of the cytochrome P-450 superfamily arose >1000 million years ago (1, 3) and sequence similarities among these ancient families might identify conserved domains of structure or function. Since complete amino acid sequences had been published for proteins from eight eukaryotic P-450 families, we used a representative sequence from each family in a series of sequence comparisons. A multiple alignment of the eight sequences revealed one region of similarity shared by all. Sequence similarities in this region reside in four domains connected by short segments of variable length. The region will provide a focus for experiments that probe the relationship of amino acid sequence to structure and function in the P-450 superfamily. METHODS Amino Acid Sequences. The source for each eukaryotic P-450 sequence used in these comparisons was as follows: LAw, rat liver (4); c, rat liver (5); 17a, bovine adrenal cortex (6); scc,...

Approximately 39% of the clones from a BALB/c mouse genomic library hybridized with polyadenylated cytoplasmic RNA extracted from anemic mouse spleen. The DNA sequence of a portion of one such clone revealed the presence of three repetitive sequence elements within a 700 bp span. All three elements contain putative RNA polymerase Ill control regions oriented in the same direction and oligo(dA) tracts at their 3' ends. The first element is a member of the murine BI family. A comparison of this element with other BI family members indicates that the BI family can be divided into two subclasses besed on commonly held base changes and deletions. The second element within this 700 bp region may be a member of a new murine Alu family. Its structure is analogous to other murine Alu-equivalent sequences with respect to overall length, the presence of a 3' oligo(dA) tract and putative RNA polymerase Ill control regions. The third element is a murine type 2 Alu-equivalent sequence.

No abstract