Students in the Phage Hunters Integrating Research and Education program learn about science by doing research on phages
Graham F. Hatfull
Graham F. Hatfull is Eberly Family Professor of Biotechnology, HHMI Professor, Chair of the Department of Biological Sciences, and co-Director of the Pittsburgh Bacteriophage Institute, University if Pittsburgh, Pittsburgh, Pa.
Author Profile--Hatfull Believes in Students Doing, Not Merely Reading about, Research
● Studying bacteriophages proves an excellent way to introduce high school and undergraduate students to research.
● Bacteriophage populations provide a rich and diverse source of biological novelty.
● Mycobacteriophages are relatively simple to manipulate to determine gene functions and to explore biological applications.
When asked how they became involved in science, it is not unusual to hear microbiologists say that it was their early exposure to research that proved crucial. Doing research enhances science education and can promote a lifelong interest in science. However, academic institutions often select only the most gifted to do research. This bias is troubling because the skills needed to do research can be different from those used to measure academic success. Thus, while conformity and memorization are key at the high school and undergraduate levels, research success depends on challenging what we think we know.
The best way to determine aptitude for science is to do it, with curiosity and desire being the primary prerequisites. It also helps when that research is authentic, meaning it will lead to peer-reviewed, published reports.
The Phage Hunters Integrating Research and Education (PHIRE) program is particularly suited for providing students opportunities to do research. Students from wide-ranging backgrounds have participated in this program, where they have isolated and sequenced phage genomes, and then coauthored peer-reviewed reports. In turn, student involvement in such activities adds energy and momentum to the field of phage genomics. With many worthy research projects lined up, we count on PHIRE students to continue making research laboratories hotbeds of educational innovation.
Don't We Know Everything There Is To Know about Bacteriophages?
In terms of integrating our research and education missions, bacteriophages make great research subjects for our students to study. Phage research dates back to their discovery in the early 20th century, and a history of its impact includes the use of phages as therapeutic agents as well as their central role in the development of molecular biology and biotechnology.
Meanwhile, during the past decade, bacteriophage research brought two important surprises. First, phages are enormously abundant, with an estimated 1031 particles in the biosphere, making them an absolute majority of all biological entities. This population is not only large, but also highly dynamic, leading to an estimated 1023 infections per second across the planet. Second, they encompass enormous genetic diversity, and the approximately 600 sequenced phage genomes include huge numbers of novel genes.
The prototype phages, lambda, T4, and T7, taught us much about molecular biology. However, studying them provides only limited insight into the biology of the broader phage population. The viral genetic diversity outside these three phages reflects the large variety of other bacterial hosts as well as the diversity of phages infecting common hosts. Ongoing efforts to isolate and characterize new phages thus are expected to provide many new insights into viral diversity and evolution.
Baptism by PHIRE: Bacteriophage Isolation
Having students isolate bacteriophages provides a good starting point for introducing them to research (Fig. 1). First, a suitable bacterial host is chosen. For instance, we use Mycobacterium smegmatis mc2155 extensively, in part because of our interest in using mycobacteriophages to study the related human pathogen Mycobacterium tuberculosis. However, numerous other bacterial strains are there to choose and use.
Next, students collect samples from which the phages will be isolated. Because many mycobacteria are saprophytic, we ask the students to half-fill a15-ml plastic tube with soil or compost, to which a simple buffer is later added, mixed, and filtered. The filtrate is plated directly with bacterial cells onto agar plates. After incubation, a bacterial lawn forms, and the presence of phage particles is revealed as plaques-small (~1 mm diameter), round areas of clearing where cell death has occurred (Fig. 2A).
In our experience, about 10-20% of samples contain at least one plaque that can be further characterized, and although it is typical to see fewer than half-a-dozen plaques, occasionally there are hundreds. The proportion of positive samples and the numbers of plaques can be enriched by incubating samples with bacterial cells prior to plating for individual plaques. However, although we usually avoid this step for fear of biasing the types of phages that we isolate, it does prove useful at times. These initial steps require little prior knowledge of content or concept and can be performed by virtually anyone.
PHIRE in the Belly: Project Ownership through the Power of Discovery
Isolating phages from environmental samples proves encouraging and exciting for students, giving them a sense of accomplishment and a feeling of ownership of their new virus. These feelings help to motivate them to continue analyzing their new phage. Furthermore, students are encouraged to name their new viruses, a process for which we grant considerable latitude because it reinforces this project ownership. A nonsystematic naming system is also consistent with the mosaic nature of phage genomic architectures, which sets up a good "teachable moment" later in the project.
The next steps involve plaque purification and amplification to prepare high-titer stocks (Fig. 1). Afterwards, particles can be viewed by electron microscopy, a real "aha" moment (Fig. 2B). At this point, students have succeeded in isolating a virus from the environment, growing it, and seeing its morphology, but any sense of novelty is largely predicated upon what they have heard or read about phage diversity, in the absence of any genetic information. However, questions about genetic and evolutionary relationships depend on genomic sequence determination, and this requirement provides a clear direction for further investigations, including DNAisolation, genome sequencing, annotation, and comparative analysis. These efforts involve more sophisticated technical approaches and lead to a deeper understanding of microbiology and molecular biology (Fig. 1). Project ownership and the thrill of discovery provide a context and relevancy for moving forward to embrace these additional key concepts and skills. Learning ceases to be a classroom exercise, and becomes part of a personal mission.
Bacteriophage Genomic Analysis
The mycobacteriophages that we isolate are double-stranded DNA tailed phages, and their average genome length is about 70 kbp. While determining a phage genomic sequence was a substantial challenge 20 years ago when the first mycobacteriophage genome was sequenced, it is becoming increasingly simpler and cheaper. Until recently, we used a shotgun sequencing approach with automated fluorescent Sanger technologies. Now other sequencing technologies are supplanting that approach, and we are having success with 454 pyrosequencing.
With a genome sequence in hand, students can annotate it using readily available software to predict gene locations and possible gene functions. At this point the project moves from a type of science that is concrete to one that is abstract and representational (Fig. 1). Although annotation requires a further degree of understanding, the transition is manageable because the context is established and students remain motivated by their sense of project ownership. It is noteworthy that phage genomes are replete with genes of unknown function (Fig. 3).
Like a House on PHIRE: Parallel Projects and Peer Mentoring
Because the PHIRE program is built upon individual but similarly structured phage discovery projects, the more independent and open aspects of research are somewhat compromised. Nonetheless, it facilitates the involvement of larger numbers of students by having many projects running in parallel. Mentoring therefore becomes less hierarchical, with students training each other, rather than all students being dependent on a single instructor, and avoiding the need for individual instructors for each student.
Thus, a key advantage of the parallel project structure is that experienced students mentor novice students. This mentoring is important because it reinforces the integration of the research and educational missions, and emphasizes that with the opportunity to perform cutting- edge research comes the responsibility to educate and mentor other students. Further, the program sustains itself, with novice students progressing to become experienced student mentors. Mentor training such as with the Wisconsin Entering Mentoring system simplifies and optimizes this process.
Great Balls of PHIRE: Insights into Mycobacteriophage Genomics
By now, 70 mycobacteriophage genome sequences are deposited in GenBank. A substantial proportion of these were isolated and sequenced by PHIRE program students, and more than 25 students have been coauthors on peer-reviewed publications. As the number of mycobacteriophage genome sequences grows, we find that their genetic diversity is large.
For the purpose of analysis, we grouped those phages that are most closely related to each other into clusters, some of which are further divided into subclusters. In total, we sorted 65 of the 70 GenBank mycobacteriophages into nine major clusters (A-J), 5 of which can be further divided in a total of 12 subclusters; 5 of the 70 are singletons and have no close relatives. There are therefore at least 20 distinct types with little or no sequence similarity with each other (Fig. 4). Some of the clusters/subclusters are reasonably large, with the largest, subcluster A1, having nine members. We cannot predict when relatives of current singletons might be isolated or how many new singletons remain to be discovered for this strain of M. smegmatis.
This grouping of phages into clusters is pragmatic, and does not represent a phylogeny of the whole genomes. Indeed, phage genomes are characteristically mosaic, as revealed by comparative analyses (Fig. 3). Many of the genes-or groups of genes-are present in two or more genomes that are otherwise not closely related, and the relationships are evident through amino acid sequence similarities because the exchange events that gave rise to them occurred in distant evolutionary time. Thus many genes within a given genome have distinct phylogenetic relationships, and their precise evolutionary histories cannot be reconstructed because the phage genome sequence space is grossly underrepresented. Any mycobacteriophage genome can be considered the result of a particular assortment of individual exchangeable modules, justifying individualistic phage names rather than a systematic nomenclature.
The mycobacteriophage genes can be sorted into "phamilies" according to their sequence relationships using BlastP and ClustalW. Steve Cresawn of James Madison University wrote the program Phamerator to automate this process and to generate genome maps and gene phylogenies (Fig. 3). In an analysis of 60 sequenced genomes containing 6,858 genes, they can be sorted into 1,523 families of distinct sequences, of which about 45% contain only a single member, or "orpham." Database searching shows that only about 20% of these phamilies have significant matches to non-mycobacteriophage genes, about half of which are to genes annotated as being hypothetical. Biological functions can therefore be ascribed to only about 10% of gene phamilies.
A subset of genes can be considered homologues of bacterial genes of known function, but which have not been previously found in phage genomes. Assuming that these are functional and expressed, they are candidates for modifying the physiology of either lysogens or infected cells.
Adding Fuel to the PHIRE: Moving beyond Genomic Descriptions
Students discovering phages and analyzing their genomes make a multitude of intriguing observations, but these efforts are strongly descriptive. Yet, a key rationale for investigating mycobacteriophages is to advance our knowledge of their hosts, including by exploiting integration-proficient plasmid vectors, non-antibiotic selectable markers, mycobacterial-specific recombineering, and delivery systems for transposons and allelic exchange substrates. Moreover, phage amplification or reporter gene delivery systems could find use in clinical microbiology because of their potential for rapidly and sensitively diagnosing infections and determining susceptibility to antibiotics.
Genetically manipulating and exploring biological functions of mycobacteriophages is an important extension of genomic analysis. It has the notable advantage of introducing students to hypothesis-driven and experimentally open approaches, including more complex data analysis and experimental design (Fig. 1). Students find numerous questions to address or features to exploit.
Until recently, genetic manipulation of mycobacteriophages was awkward at best, but circumstances improved with development of bacteriophage recombineering of electroporated DNA (BRED), which enables simple construction of mycobacteriophage recombinants and mutants.
In brief, BRED involves constructing a 200-bp DNA substrate containing the desired mutation (typically by PCR) that can be coelectroporated with phage genomic DNA into recombineering-proficient M. smegmatis cells. Plaques are recovered and screened by PCR. If the mutant or recombinant is viable, then a mixed plaque containing both wild-type and mutant alleles can usually be identified by PCR of 12-18 primary plaques, and a homogenous mutant identified after plaque purification and a second round of PCR. Because the recombination efficiency in the recombineering cells is high, there is no need for a selectable marker and virtually any type of mutation can be introduced, including precise deletions, insertions of gene tags and whole genes, and point mutations.
The BRED technology therefore provides a powerful genetic strategy for discovering the biological secrets embedded in these amazing phage genomes. The PHIRE program can therefore grow beyond the descriptive aspects of viral genomics by inclusion of open-ended investigations into the biology of the phages that students have been inspired to discover, genomically characterize, and love.
Hanauer, D., G. F. Hatfull, and D. Jacobs-Sera. 2009. Active assessment: assessing scientific inquiry. Springer, New York.
Hanauer, D. I., D. Jacobs-Sera, M. L. Pedulla, S. G. Cresawn, R. W. Hendrix, and G. F. Hatfull. 2006. Inquiry learning. Teaching scientific inquiry. Science 314:1880-1881.
Handelsman, J., S. M. Lauffer, C. M. Pribbenow, and C. Pfund. 2005. Entering mentoring: a seminar to train a new generation of scientists. University of Wisconsin Press, Madison, Wis.
Hatfull, G. F., M. L. Pedulla, D. Jacobs-Sera, et al. 2006. Exploring the mycobacteriophage metaproteome: phage genomics as an educational platform. PLoS Genet. 2: e92.
Hatfull, G. F., D. Jacobs-Sera, J. G. Lawrence, et al. 2010. Comparative genomic analysis of sixty mycobacteriophage genomes: genome clustering, gene acquisition and gene size. J. Mol. Biol., in press.
Hendrix, R. W., M. C. Smith, R. N. Burns, M. E. Ford, and G. F. Hatfull. 1999. Evolutionary relationships among diverse bacteriophages and prophages: all the world's a phage. Proc. Natl. Acad. Sci. USA 96:2192-2197.
Jacobs, W. R., Jr. 2000. Mycobacterium tuberculosis: a once genetically intractable organism, p. 1-16. In G. F. Hatfull and W. R. Jacobs, Jr. (ed.), Molecular genetics of the Mycobacteria, ASM Press, Washington, D.C.
Marinelli, L. J., M. Piuri, Z. Swigonova, A. Balachandran, L. M. Oldfield, J. C. van Kessel, and G. F. Hatfull. 2008. BRED: a simple and powerful tool for constructing mutant and recombinant bacteriophage genomes. PLoS ONE 3, e3957.
Pedulla, M. L., M. E. Ford, J. M. Houtz, et al. 2003. Origins of highly mosaic mycobacteriophage genomes. Cell 113:171-182.
Piuri, M., W. R. Jacobs, Jr., and G. F. Hatfull. 2009. Fluoromycobacteriophages for rapid, specific, and sensitive antibiotic susceptibility testing of Mycobacterium tuberculosis. PLoS ONE 4, e4870.