Several putative Escherichia coli pseudouridine (Psi) synthases have been identified by iterative searching of genomic databases for ORFs homologous to known Psi synthases [Gustafsson et al. (1996) Nucleic Acids Res. 24, 3756-3762]. Of these, yceC and yfiI were proposed to encode Psi synthases which modify 23S rRNA. In the present work, yceC and yfiI were cloned and overexpressed in E. coli, and the encoded enzymes, YceC and YfiI, were purified to homogeneity. Both proteins converted Urd residues of rRNA to Psi, thus confirming their identities as Psi synthases. However, in in vitro experiments both enzymes extensively modified Urd residues of both 23S rRNA and 16S rRNA. Gene-disruption of yceCresulted in the absence of Psi modification at positions U955, 2504, and 2580 of 23S RNA, thus identifying these sites as in vivo targets for YceC. Likewise, yfiI disruption resulted in the absence of Psi modification at positions U1911, 1917, and possibly 1915 of 23S RNA. Disruption of yceC did not affect the growth under the conditions tested, whereas yfiI-disrupted cells showed a dramatic decrease in growth rate. Since YceC and YfiI hypermodify RNA in vitro, factors in addition to ribonucleotide sequence must contribute to the in vivo specificity of these enzymes.

The fmu gene product has been proposed to be an RNA methyltransferase [Koonin, E. V. (1994) Nucleic Acids Res. 22, 2476-2478]. Fmu has been cloned and expressed, and the encoded 47 kDa protein has been purified and characterized. The enzyme catalyzed specific methylation of C967 of unmodified 16S rRNA transcripts. A 16mer stem-loop structure containing C967 (nt 960-975) was also a good substrate for the enzyme in vitro. Methylation of C967 was confirmed by several methods including analysis of RNase T1 digests and nearest-neighbor analysis. Fmu did not catalyze methylation of transcripts of 23S rRNA. E. coli cells that contained kanr-disrupted fmu produced 16S rRNA that could be specifically methylated by Fmu in vitro at C967 but not C1407. Further, fmu disruption did not significantly alter the growth rate of E. coli in rich or minimal media. We propose renaming this ORF "rrmB" and the enzyme "RrmB" for rRNA methyltransferase.

In an effort to determine whether proteins with structures other than the immunoglobulin fold can be used to mimic the ligand binding properties of antibodies, we generated a library from the four-helix bundle protein cytochrome b562 in which the two loops were randomized. Panning of this library against the bovine serum albumin (BSA) conjugate of N-methyl-p-nitrobenzylamine derivative 1 by phage display methods yielded cytochromes in which residues Trp-20, Arg-21, and Ser-22 in loop A and Arg-83 and Trp-84 in loop B were conserved. The individual mutants, which fold into native-like structure, bind selectively to the BSA-1 conjugate with micromolar dissociation constants (Kd), in comparison to a monoclonal antibody that binds selectively to 1 with a Kd of 290 nM. These and other antibody-like receptors may prove useful as therapeutic agents or as reagents for both intra-and extracellular studies.The immune system can produce antibodies that bind virtually any molecule, from large proteins to small synthetic ligands, with high affinities and specificities (1). The basic fold of the antigen binding region is that of an eight-stranded ,B-sheet onto which hypervariable loops, or complementarity-determining regions (CDRs), are grafted, three each from the light-and heavy-chain variable domains (2, 3). Binding affinity and specificity are achieved by generating a large number of potential binding sites and selecting those with highest affinity for a given immunogen. Diversity results from the combinatorial association of the genes encoding the variable region segments, as well as variable recombination and somatic mutation events (4, 5). Recently, it has been shown that semisynthetic libraries can be generated in which the CDR loops of antibodies are randomized either independently or in combination (6, 7). Screening of Fab fragments containing these randomized sequences using filamentous phage display techniques results in antibodies that can bind either proteins or small molecules with high affinity and selectivity.The question that then arises is whether alternative protein frameworks can be used to display randomized regions that can be screened by similar methods to generate high-affinity, selective receptors. These frameworks may have enhanced properties for a number of applications, including increased stability, smaller size, improved expression, or altered pharmacokinetics. Moreover, the characterization of these antibody-like receptors might provide additional insight into the remarkable binding properties of the antibody molecule itself. As a first step in this direction, we have generated libraries in which the two loops of the four-helix bundle proteins cytochrome b562 and myohemerythrin are randomized. These libraries were screened for their ability to bind a ligand-protein conjugate, and the binding properties of individual mutants were characterized.

Our finding indicates the presence of an almost equimolar amount of 4'-phosphopantetheine covalently bound to the NRPS module PheAT(His6), and that the functional expression of NRPS modules in E. coli is possible, provided that they are coexpressed with an appropriate P-pant transferase.