Vision

The PROBE program will make use of two novel methodologies in DNA-based biodiversity research: DNA barcoding and genome quantification through computerized image analysis. Both methods have been developed in large part by members of our research team over the past few years, and both provide the potential to carry out biodiversity surveys of unprecedented scope at the genetic,
genomic, and species levels. These methods are reviewed briefly below.

i) DNA barcoding.

The DNA barcoding approach is based on the premise that a short, standardized segment of the genome can enable species identification and discovery (10). There is now clear evidence that a 648 bp segment positioned near the 5' terminus of the mitochondrial cytochrome c oxidase I (COI) gene is extraordinarily effective in discriminating members of the animal kingdom,
allowing unambiguous identification of more than 98% of animal species in studies that have examined a wide range of taxonomic groups (11-13). Moreover, the few cases of incomplete resolution involve sister species pairs and these cases are more than offset by the discovery of species overlooked through past taxonomic study. These results have now provoked the activation of a vibrant international research program on DNA barcoding, co-ordinated by the Consortium for the Barcode of Life (CBOL) based at the Smithsonian Institution in Washington. Although CBOL was just launched in 2004, it now has more than 80 member organizations in 25 countries. It has already made impressive strides in advancing the institutional arrangements that are needed for large-scale DNA barcoding programs. For example, GenBank and the other genomics repositories (DDBJ, EMBL) have created special data
standards for barcode submissions. CBOL has also launched the first global barcode campaigns on animals and several research groups are now developing barcode protocols for fungi, protists, and plants. Interestingly, there is growing evidence that the 5' region of COI will also be effective in separating fungal and protist species. Most importantly for our project, there is good evidence that this gene region will be very effective for the identification of marine macroalgae (14).
DNA barcoding has now gained serious support in Canada. More than $10M has been invested to create state-of-the-art facilities, including both a high throughput sequencing centre and a sophisticated databasing system that is based in the Biodiversity Institute of Ontario’s new facilities at the University of Guelph. In addition, a group of nearly 50 researchers in university and government labs have joined forces to create the Canadian Barcode of Life Network (BOLNET.ca). This Network has now received $11M in support from NSERC and Genome Canada to gather barcode records for economically important species in southern Canada. PROBE will take full advantage of these prior investments in order to deliver a high volume of barcode results for Arctic life at very low cost. We emphasize that the presence of the Canadian Barcode of Life Network means that strong linkages are in place with most of Canada’s leading taxonomists, a critical asset in gaining identifications for the species gathered in our sampling programs.
The actual analytical protocols involved in DNA barcoding are now well standardized (15). The work flow involves the removal of a small tissue sample from each specimen selected for analysis (e.g., a single leg from any insect). DNA is then extracted from each tissue sample, the target gene region is amplified, and the resultant PCR product is sequenced to generate a barcode record. The specimen itself is photographed and information on its collection history is databased. All of this information migrates to a sophisticated data repository and analytic system for barcode records (www.barcodinglife.org) where it is archived for long-term access. Although the DNA barcode community plans to assemble a comprehensive barcode library for all animal species over the next 20 years, work has just begun. Global initiatives were launched in 2005 to gather barcode records for all bird and all fish species on the planet, but PROBE represents the first effort to build a comprehensive barcode library for specific sites. Thus, although our research program is clearly groundbreaking, we emphasize that the protocols and facilities are in place to ensure the delivery of results. As noted in sections 5B and 5C, we will gather barcode records for 20,000 specimens over PROBE’s three-year duration. Though remarkable by any other standard, their analysis will represent just a 10% increment in the scheduled production volume for the Guelph barcode facility.

ii) Genome quantification.

The amount of DNA contained within a single copy of the genome varies enormously among eukaryotic taxa, ranging more than 3,300-fold in animals alone (16). Similarly, whereas polyploidy has traditionally been considered important only in plants, it is evident that modulations in genome copy number (i.e., ploidy level) are also common in many animal groups. Both genome size and polyploidy exhibit intriguing associations with climate, with larger amounts of DNA generated through either mechanism becoming increasingly prevalent at higher latitudes. With genome size, this pattern is based on interspecific variation, whereas in the case of polyploidy this applies both among and within individual species (e.g., 3,16). The reasons for these geographical associations remain a subject of debate, but their existence means that DNA content analyses can serve as a highly complementary DNA-based method of tracking responses to climate change along with the DNA barcoding program. Furthermore, to the extent that polyploidy arises by hybridization in the Arctic (e.g., 3,17), and that hybridization can complicate barcode analyses, the ability to assess ploidy level is likely to be important for the execution of the polar barcoding program. Finally, the fundamental questions raised by such a pattern are also of considerable interest in attempts to understand the complex nature of large-scale genome evolution. Thus, an analysis of genome sizes and ploidy levels from a wide range of previously unstudied Arctic organisms will contribute importantly to both the practical and academic interest of the PROBE initiative.
The quantification of nuclear DNA has traditionally been carried out by one of two methods: Feulgen densitometry (since the 1950s) and flow cytometry (since the 1970s). The two methods differ fundamentally in their underlying physical bases, with Feulgen densitometry involving a calculation of the amount of light absorbed by stained DNA and flow cytometry representing a measurement of the amount of light emitted by fluorescently-stained nuclei. Both techniques utilize comparisons against samples of known DNA content to convert from arbitrary units of absorbance or intensity to absolute mass of DNA (in picograms, where 1pg = 10-12g = 978 Mbp). Each approach has particular advantages, such that both methods continue to be employed regularly in genome size and ploidy analyses. Feulgen densitometry, in particular, is advantageous because it involves the production of permanent microscope slide preparations that can be archived, it allows the direct analysis of targeted tissues, and it requires a relatively small number of nuclei to deliver an accurate DNA content estimate (which can be important when working with small specimens). On the other hand, the different sizes and heterogeneous nature of Feulgen-stained nuclear DNA from different species means that each nucleus must be analyzed as a collection of individual point densities, the sum of which (known as integrated optical density, or IOD) is compared across samples. In the past, this has imposed significant constraints on the speed with which the method could be employed, as each nucleus, and indeed each point density within a given nucleus, had to be analyzed sequentially. Very recently, a novel method of Feulgen image analysis densitometry has been developed by members of the present research team (TRG, PDNH) that makes use of computerized imaging techniques to simultaneously and instantaneously provide IODs for all nuclei within a microscope field, using each pixel as an individual point density. This greatly increases the efficiency of the Feulgen densitometry method while preserving all of its traditional advantages (18). Most of the genome size and ploidy analyses in the proposed genomic diversity surveys will be carried out via Feulgen image analysis densitometry using the Bioquant Nova Prime image analysis package, supplemented as appropriate by the use of flow cytometry equipment at the University of Guelph. Briefly, this will involve the preparation of monolayers of nuclei from pre-determined tissues, air-dried, fixed, and subsequently stained by an optimized Feulgen reaction procedure (hydrolysis with 5N hydrochloric acid to depurinate DNA and free aldehyde groups, followed by staining with Schiff leucofuchsin sulfurous acid reagent and a series of rinses [18]). In cases where DNA content analysis alone is insufficient to establish polyploidy (e.g., if no obvious diploid counterpart is collected), methods
such as allozyme electrophoresis will be used to establish the presence of duplicated gene loci (3,4). Three hundred slides can be stained in a single run using the protocol in the Gregory lab, which takes less than a day. The use of the Bioquant image analysis program is straightforward to learn, and can be used to analyze 25-50 slides per day, depending on the type of preparation. As such, genome sizes and/or ploidy estimates can be generated for thousands of specimens even by a single graduate student. It will also be possible for undergraduate field course students and summer students to process large numbers of samples even over a relatively short period of study.