The biosphere is dominated by microorganisms (32), yet most microbes in nature have not been studied. Traditional methods for culturing microorganisms limit analysis to those that grow under laboratory conditions (14,25). The recent surge of research in molecular microbial ecology provides compelling evidence for the existence of many novel types of microorganisms in the environment in numbers and varieties that dwarf those of the comparatively few microorganisms amenable to laboratory cultivation (7,13,31). Corroboration comes from estimates of DNA complexity and the discovery of many unique 16S rRNA gene sequences from numerous environmental sources (8,10,28). Collectively, the genomes of the total microbiota found in nature, which we termed the metagenome (11), contain vastly more genetic information than is contained in the culturable subset. Given the profound utility and importance of microorganisms to all biological systems, methods are needed to access the wealth of information within the metagenome.Cloning large fragments of DNA isolated directly from microbes in natural environments provides a method to access soil metagenomic DNA. Previously, we investigated the use of the bacterial artificial chromosome (BAC) vector to express Bacillus cereus genomic DNA (20). The advantage of BAC vectors is that they maintain very large DNA inserts (greater than 100 kb) stably in Escherichia coli (23), facilitating the cloning of large fragments of DNA. Our results demonstrated that expression of heterologous DNA from B. cereus in an E. coli BAC system was detectable at a reasonable frequency (20), validating the idea that the low-copy BAC vector (one to two per cell) (23) could be used to express foreign DNA from foreign promoters in E. coli.Here we describe the construction and initial screening of two BAC libraries made with DNA isolated directly from soil. We found detectable levels of several biochemical activities from BAC library clones. Sequence analysis of selected BAC plasmids encoding such activities and of 16S rRNA genes in one of the libraries confirms the novelty of the genomic information cloned in our libraries. The results show that DNA extracted directly from soil is a valuable source of new genetic information and is accessible by using BAC libraries. Our results demonstrate that both traditional and functional genomics of uncultured microorganisms can be carried out by this approach and that screening of metagenome libraries for activities or gene sequences can provide a basis for conducting genomic analyses of uncultured microorganisms. MATERIALS AND METHODSBacterial strains and plasmids. E. coli strain DH10B and the BAC vector pBeloBAC11 were provided by H. Shizuya (15). Bacillus subtilis strain BR151(pPL608) is strain 1E32 (lys-3 metB10 trpC2) from the Bacillus Genetic Stock Center, Ohio State University. -TnphoA was used as described before (20).

FKBP ligand homodimers can be used to activate signaling events inside cells and animals that have been engineered to express fusions between appropriate signaling domains and FKBP. However, use of these dimerizers in vivo is potentially limited by ligand binding to endogenous FKBP. We have designed ligands that bind specifically to a mutated FKBP over the wild-type protein by remodeling an FKBP-ligand interface to introduce a specificity binding pocket. A compound bearing an ethyl substituent in place of a carbonyl group exhibited sub-nanomolar affinity and 1,000-fold selectivity for a mutant FKBP with a compensating truncation of a phenylalanine residue. Structural and functional analysis of the new pocket showed that recognition is surprisingly relaxed, with the modified ligand only partially filling the engineered cavity. We incorporated the specificity pocket into a fusion protein containing FKBP and the intracellular domain of the Fas receptor. Cells expressing this modified chimeric protein potently underwent apoptosis in response to AP1903, a homodimer of the modified ligand, both in culture and when implanted into mice. Remodeled dimerizers such as AP1903 are ideal reagents for controlling the activities of cells that have been modified by gene therapy procedures, without interference from endogenous FKBP.

Growth factors and cytokines act through cell surface receptors with different biochemical properties. Yet each type of receptor can elicit similar as well as distinct biological responses in target cells, suggesting that distinct classes of receptors activate common gene sets. Epidermal growth factor, interferon-gamma, and interleukin-6 all activated, through direct tyrosine phosphorylation, latent cytoplasmic transcription factors that recognized similar DNA elements. However, different ligands activated different patterns of factors with distinct DNA-binding specificities in the same and different cells. Thus, unrelated receptors may activate a common nuclear signal transduction pathway that, through differential use of latent cytoplasmic proteins, permits these receptors to regulate both common and unique sets of genes.

We tested sequences flanking the mouse c-fos gene for the ability to form specific DNA-protein complexes with factors present in crude nuclear extracts prepared from mammalian cells. Three such complexes were detected. One complex formed in a region necessary for the induction of c-fos expression by serum growth factors. Two additional complexes formed at sequences that contribute to basal c-fos promoter activity in vivo. These complexes represent three novel sequence-specific DNA-binding activities which appear to participate in the regulation of c-fos transcription.

Gene therapy was originally conceived as a medical intervention to replace or correct defective genes in patients with inherited disorders. However, it may have much broader potential as an alternative delivery platform for protein therapeutics, such as cytokines, hormones, antibodies and novel engineered proteins. One key technical barrier to the widespread implementation of this form of therapy is the need for precise control over the level of protein production. A suitable system for pharmacologic control of therapeutic gene expression would permit precise titration of gene product dosage, intermittent or pulsatile treatment, and ready termination of therapy by withdrawal of the activating drug. We set out to design such a system with the following properties: (1) low baseline expression and high induction ratio; (2) positive control by an orally bioavailable small-molecule drug; (3) reduced potential for immune recognition through the exclusive use of human proteins; and (4) modularity to allow the independent optimization of each component using the tools of protein engineering. We report here the properties of this system and demonstrate its use to control circulating levels of human growth hormone in mice implanted with engineered human cells.

In vertebrates, many cytokines and growth factors have been identified as activators of the JAK/STAT signaling pathway. In Drosophila, JAK and STAT molecules have been isolated, but no ligands or receptors capable of activating the pathway have been described. We have characterized the unpaired (upd) gene, which displays the same distinctive embryonic mutant defects as mutations in the Drosophila JAK (hopscotch) and STAT (stat92E) genes. Upd is a secreted protein, associated with the extracellular matrix, that activates the JAK pathway. We propose that Upd is a ligand that relies on JAK signaling to stimulate transcription of pair-rule genes in a segmentally restricted manner in the early Drosophila embryo.

In mammals, many cytokines and growth factors stimulate members of the Janus kinase (JAK) family to transduce signals for the proliferation and differentiation of various cell types, particularly in hematopoietic lineages. Mutations in the Drosophila hopscotch (hop) gene, which encodes a JAK, also cause proliferative defects. Loss‐of‐function alleles result in lethality and underproliferation of diploid tissues of the larva. A dominant gain‐of‐function allele, Tumorous‐lethal (hopTum‐l), leads to formation of melanotic tumors and hypertrophy of the larval lymph glands, the hematopoietic organs. We show that a single amino acid change in Hop is associated with the hopTum‐l mutation. Overexpression of either wild‐type hop or hopTum‐l in the larval lymph glands causes melanotic tumors and lymph gland hypertrophy indistinguishable from the original hopTum‐l mutation. In addition, overexpression of Hop in other tissues of the larva leads to pattern defects in the adult or to lethality. Finally, overexpression of either hop or hopTum‐l in Drosophila cell culture results in tyrosine phosphorylation of Hop protein. However, overexpression of hopTum‐l results in greater phosphorylation than overexpression of the wild‐type. We conclude that hopTum‐l encodes a hyperactive Hop kinase and that overactivity of Hop in lymph glands causes malignant neoplasia of Drosophila blood cells.

Binding of interferons IFN-alpha and IFN-gamma to their cell surface receptors promptly induces tyrosine phosphorylation of latent cytoplasmic transcriptional activators (or Stat proteins, for signal transducers and activators of transcription). Interferon-alpha activates both Stat91 (M(r) 91,000; ref. 1) and Stat113 (M(r) 113,000; ref. 2) whereas IFN-gamma activates only Stat91 (refs 3, 4). The activated proteins then move into the nucleus and directly activate genes induced by IFN-alpha and IFN-gamma. Somatic cell genetics experiments have demonstrated a requirement for tyrosine kinase-2 (Tyk2) in the IFN-alpha response pathway and for Jak2 (ref. 6), a kinase with similar sequence, in the IFN-gamma response pathway. Here we investigate the tyrosine phosphorylation events on Stat and Jak proteins after treatment of cells with IFNs alpha and gamma and with epidermal growth factor (EGF). Stat91 is phosphorylated on Tyr701 after cells are treated with IFN-alpha and EGF, as it was after treatment with IFN-gamma (ref. 8). We find that Jak1 also becomes phosphorylated on tyrosine after cells are treated with these same three ligands, although each ligand is shown to activate at least one other different kinase. Jak1 may therefore be the enzyme that phosphorylates Tyr 701 in Stat91.