1. Functional Genomics of Bacillus Phages
Kelly Flounlacker, Samantha Foltz, Allison Johnson
Virginia Commonwealth University, Richmond VA
Over the past three years, VCU students have discovered and characterized the genomes of 12 bacteriophages infecting Bacillus thuringiensis subspecies Kurstaki. These phages can be grouped into several different clusters of genomes and yet these clusters contain a large set of shared proteins, suggesting these phages are both diverse but also contain a core set of functionally important proteins. Many of these proteins are functionally uncharacterized. We are working to apply a variety of functional genomics approaches to predict protein function of this core set of shared proteins, including homology studies, a protein overexpression assay, and computational modeling. This poster will describe some of that work as well as provide examples of Gene Ontology based functional annotations completed through the Spring “CACAO” (Intercollegiate Annotation Competition) competition.
Comparative Genomics of Bacillus Phages
We used a dot plot of whole genome sequences and Splitstree analysis of the Bacillus database pham file to organize the genomes of 83 Bacillus phage genomes into 7 clusters. From these 83 phage genomes Phamerator organized 14,922 phage proteins into 3,638 phamilies. These phages feature a large set of shared protein content that is our target for functional genomics studies, as many of these proteins are of unknown function.
CACAO: Intercollegiate Annotation Competition
This is a project run by Dr. Jim Hu from Texas A&M University, for more info, see gowiki.tamu.edu
We compiled evidence-based functional annotations for submission to public databases (i.e. Uniprot, QuickGO, etc.), and compete with other college students through alternating rounds of annotation and challenge.. The wiki features a scoreboard to keep track of team standings! Some example annotations are provided below. A B
Standard Annotation
Transfer Annotation
Standard annotations use data in the figures and tables from scientific papers to as experimental evidence for the function of proteins.
Transfer annotations use homology between sequences to conclude that a target protein has the same function as a protein in another organism. These are done by using genome homology programs such as HHPred, Blast, or iTasser and defined criteria described in the SEA PHAGES GO_REF:
Comparative genomics tools like dotplot (A) suggests low sequence similarity between Bacillus phage clusters, but analysis of shared protein content (B&C) using the Bacillus database pham table shows a significantly higher rate of similarity between clusters when considering proteins shared between a large number of genomes. Majority of the phage proteins are found in one to two genomes. However, the select few that are conserved among many genomes suggests they might be essential proteins to the phage life cycle. Of the top 55 conserved proteins we focused on, 22 have no known function, yet they are found across each of the 7 clusters. Since these unknown proteins are highly conserved, finding their function would provide a better understanding of these bacterial viruses. One such method of evaluating protein function is overexpression.
Annotation
Qualifier
GO ID
GO term name
Reference
Evidence code
with/from
Aspect
Notes
Contributes to; Is part of;
Is a
A unique number associated with each GO term
Predicted function, chosen to be as specific as possible
A single peer-reviewed scientific publication; or a GO_REF code
We have 8 codes to choose from, depending on data used as evidence
Used for a comparative annotation, required for transfer annotations
F = Molecular Function
P = Biological Process
C = Cellular Component
Free-text area for user notes
Standard Annotation of SPP1 helicase
Contributes to
GO:
DNA helicase activity
PMID:
IDA: Inferred from Direct Assay F
**evidence shown below
GO:
DNA duplex unwinding
IMP: Inferred from Mutant Phenotype P
Transfer Annotation from SPP1 helicase to Streptococcus pneumoniae helicase
GO_REF:
ISA: Inferred from Sequence Alignment
UniProtKB:Q38152 C
Developing a Phage Protein Overexpression Assay
Overexpression is the process of inducing the overproduction of a single protein. We are overexpressing highly conserved and functionally unknown Bacillus phage proteins to explore protein function. Expressing each gene within its host can yield abnormal bacterial phenotypes or growth. This assay will help identify proteins of unknown function that can potentially have a significant impact on phage/host interactions and be analyzed further using other methods.
Amplified phage genes
Standard annotation for the SPP1 phage helicase
Experiment used x-ray crystallography to determine the structure of the protein
Performed mutations on the gene and observed the changes using gel electrophoresis
Transfer Annotations for related helicases
Using BlastP to determine sequence homology
Provide the e-value, percent query coverage, and percent identity to prove the homology is significant enough to show the proteins have the same function
Transferred from SPP1 to: Streptococcus pneumoniae, Bacillus sp. FJAT , and Geobacillus phage GBK2
Step 1: PCR Amplification of phage ORFs yields multiple copies of a single phage gene. The phage genome acts as a DNA template where primers bind to the complementary strand at the beginning and end of the gene. DNA polymerase then begins synthesis of the gene using dNTPs, who provides the DNA base pairs for the reaction.
Step 4. Bacteria will be evaluated for change in growth rate, shape, etc., as a measure of phage proteins interacting with bacteria. In this published example, researchers evaluated colony growth as well as bacterial cell shape to identify proteins that impacted bacterial growth.
Step 3: After ligation, the construct can be transformed into E. coli or B. thuringiensis for expression. IPTG will be used as a catalyst for transcription. This causes an overproduction of the single phage gene within the bacteria. This experiment uses plasmid pDG148- Stu. Specific to this plasmid is the HindIII and SphI, which makes up the Stu 1 site. It is a multiple copy plasmid.
Step 2: Restriction enzyme cleaves DNA of PCR construct and target plasmid, creating ‘sticky ends’ or ‘blunt ends’ for ligation.
Fig 1. shows hexameric structure of protein, and separation of N-terminal and ATPase domains.
Sticky
Fig 4 shows several helicase assays including a) N-terminal mutants with loss of activity as measured in DNA unwinding assay; and
Blunt
Synthesis of the plasmid
Blastp alignment between SPP1 helicase protein and Streptococcus pneumoniae helicase satisfies requirements of GO_REF:
Query coverage: 99.32%
Amino acid identity: 51.4%
E value: 1.4e-144
Wagemans et al., Functional elucidation of antibacterial phage ORFans targeting Pseudomonas aeruginosa. Cell Microbiol. 2014 Dec;16(12):
Table 1 shows results of an ATPase assay with the WT and several mutant forms of the protein.
Acknowledgements: