|
- Worms in space
- Insights into birth defects from mice
- Signaling wars in worms
- Double danger for tadpoles
- The Secrets of Limb Developments in Flies
- New view of Phage Phylogeny
- Mouse genetics uncovers multiple roles for CtBP
- A New Look at Specificity
- Modified Proteins Found in Tumors
Pittsburgh Bacteriophage Institute
|
|
Bacteriophage Genomics - Bacterial Genomics - Yeast Genomics
Bacteriophage Genomics
Bacteriophages are the most abundant life form on the Earth, and in fact the majority (!) of genomes in the biosphere are those of
dsDNA phages. Profs. Hatfull, Hendrix, and
Lawrence and their research groups are working together to learn about the genetic structure of this largest
of all biological populations and to infer how it evolves. Aside from the intrinsic interest of this problem, it is clear that phages have a profound
influence on the evolution of their bacterial and archael hosts, and an understanding of how phages evolve is a prerequisite for understanding how prokaryotes do it.
Fig. 1. Most bacteriophages can become part of a bacterial genome by entering the lysogenic life cycle and becoming a prophage.
Prophages are very likely responsible for a large fraction of the movement of genes among phages, since they are available for recombination
with phages that infect the host where they are resident. As phages alternate between their infectious virion and prophage forms, they are undoubtedly
responsible for much of the movement of DNA sequences among bacteria as well (see Bacterial Genomics). Here we show
the genomic maps of two putative prophages discovered by Pittsburgh Bacteriophage Institute members in the genome of
Mycobacterium tuberculosis. Remarkably, these prophages contain genes that are clear homologs of genes found in phages of a phylogenetically
diverse group of bacterial hosts (Escherichia, Streptomyces, Mycobacterium). This is part of the evidence arguing that bacteriophages
are in the business of swapping genes with each other, and that they do so over a very wide phylogenetic range.
|
This group has used the Genome Center of the Pittsburgh Bacteriophage Institute to determine the genomic sequences of a
dozen dsDNA phages. They are comparing these sequences to each other and to another ~2 dozen from other laboratories around the world.
This "comparative genomics" approach is marvelously powerful, and much more so than the ordinary form of sequence gazing.
This work shows for the first time that all the dsDNA phages are part of one genetic population, partaking of the same gene pool through
promiscuous horizontal exchange, even though (paradoxically) two phages picked at random are unlikely to have any detectable similarity
in genome sequence; for more details, consult Hendrix et al., 1999.
Genome comparison is turning out to be a gold mine as well for unexpected novel insights into biochemical functions of phage genes
and the proteins they encode, and comparative phage genomics is becoming a platform for developing new techniques of genome analysis
that can be applied to organisms with bigger, clunkier genomes.
|
Fig. 2. As Shakespeare once said, "All the world's a phage". What Willie meant was that all of the dsDNA phages share common ancestry,
and that they are all partaking of a common gene pool. You would not be too far off to think of the global phage population as indulging in one big
orgy of genetic exchange that has been underway continuously for at least the last couple of billion years. The result of this process (and an important part
of the evidence for it) is that sequence comparisons of phage genomes show them to be genetic mosaics with respect to one other. This is illustrated in the
figure, where we do a "virtual DNA-DNA hybridization", probing in turn with the genomic DNA of each of the three phages shown.
The results are that even though HK97 (an E. coli phage) and L5 (infects Mycobacteria) have no detectable sequence similarity, they are
linked by the Rosetta Stone of the phage world, phi-C31 (infects Streptomyces),
which shares head gene sequences with HK97 and scattered early gene sequences with L5.
For a more detailed explanation, check out Hendrix et al., 1999. (We acknowledge our colleague and
Pittsburgh Bacteriophage Institute International Member Dr. Maggie Smith of the University of Nottingham, who provided the
phi-C31 genome sequence and collaborated in the analysis of these genomes.)
|
Bacterial Genomics
Research in the Lawrence laboratory includes the analyses of genomic nucleotide sequences to make
inferences about bacterial evolution. In the figure to the left, the genome of the enteric bacterium Escherichia coli is depicted where
genes that were introduced by lateral genetic transfer are shown as colored bars. By analyzing nucleotide composition (GC-content), patterns of codon usage bias,
dinucleotide frequencies, and other genomic "fingerprints," we can detect that almost 18% of the genome of this organism comprises foreign
DNA introduced in at least 234 separate events. Foreign genes are readily detected in bacterial genomes as those that fail to resemble native genes.
While this seems like a tautology, it is not, since all native genes have experienced the same set of direction mutation pressures for long periods
of time, giving them recognizable patterns of nucleotide composition, codon biases, codon contexts, and other features.
More importantly, we can assess how long each of these segments has been present in the E. coli genome by an even more clever analysis of the nucleotide
sequences (see Lawrence and Ochman 1998). Some of the patterns left by directional mutation
pressure vary in predictable ways across different genomes. When genes are transferred from one host to another, they move from experiencing one
set of direction mutation pressures to experienceing a different set. While horizontally-transferred genes are ameliorating to resemble native genes, their
"fingerprints" deviate from these universally predictable patterns. By examining these deviation, we can determine how long genes have resided in their new genome,
thereby determining when genes were introduced into the genome. From the amount of new DNA and its age in the genome, we can calculate the rate of
horizontal genetic transfer (about 16 kb/MYr for E. coli), and assess the patterns of species evolution (like
what the order of events was in Salmonella's acquisition of pathogenicity islands).
|
Yeast Genomics
A new type of novel genome-wide analysis of eukaryotic gene function has been initiated by a
consortium of labs in Europe and the US
These researchers are in the process of deleting all of the genes, one-at-a-time, from the yeast S. cerevisiae.
The function of the novel gene products will be determined by analyzing the phenotypes of the mutants.
The Saunders lab is participating in this effort by examining the gene deletion mutants for the ability to
undergo meiosis. We have screened ~ 1000 gene deletions and identified ~ 25 novel genes required for meiosis or sporulation. We are
currently testing the mutants further to define what part of the meiotic pathway requires the activity of these proteins. Two examples are give below.
The BMD1 gene product is required for the cells to initiate premeiotic DNA replication. In its absence the cells
exit the vegetative cell cycle, as shown by accumulation of 2C cells, but arrest in G1 without duplicating their genomes.
| |
Wild Type |
bmd1 |
| 0 Hours |
 |
 |
| 2 Hours |
| 4 Hours |
| 6 Hours |
| 12 Hours |
 The SWS1 gene is required for
complete sporulation of the haploid products. The sws1 mutants have only 1-2 spores instead of the 4 of wild type strains (arrows A&B).
The nuclei divide normally (C&D), but the membranous prospore wall only forms at 1-2 sites
in the sws1 mutant compared to the 4 sites of wild type cells (E&F). The spores are visualized with DIC microscopy (A&B), the
nuclei by DAPI (C&D), and the prospore wall by Spo14-GFP overexpression (E&F).
|
|
