DNA Barcoding and Conservation Genetics

The number of species currently described is circa 1.7 million. Although this may seem a large number, it is relatively small when compared to the total number of eukaryote species estimated to populate the planet, in the region of 10 million [1]. While the large majority of species whose organisms have body sizes larges than 10 mm are thought to be classified, the same is not true for smaller living forms – the order of magnitude of the taxonomic deficit is likely to be several fold worse than for vertebrates and land plants [2].

This fact is more important when one considers the rate at which species, known or unknown to Mankind, are going extinct: last estimates put that number at around an astonishing 100 species per day [3]. Habitat destruction due to Human activity, environmental variability, natural catastrophes, demographic variability and genetic stochasticity are among the factors that contribute to the loss of important genetic data that might remain forever unrevealed and, consequently, unvalued.

Conservation Genetics is the science that aims to apply genetic methods to study the classification, maintenance, loss, and restoration of biodiversity. As a result of the continuous appearance of different molecular biology techniques since the late 1980’s, Conservation Genetics, a discipline that had its origins in Ecology and Evolutionary Biology, gained relevance and has helped to get into the spotlight systematics and population genetics.

When conservation scientists study a given species, they often use mathematical models derived from collected data (DNA included) in order to be able to make recommendations about how it should be managed. Quantifying the minimum viable population size, the breeding effective population size, the inbreeding depression, the levels of genetic variation and gene flow in a population gives the scientists a valuable insight on the processes affecting the species [4].

Conservation in the past has been addressed from a mathematical, evolutionary, or taxonomic point of view. Lately, due the rapidly decreasing Earth’s biodiversity and need to classify and study to fully understand and value it, along with the inherent limitations of morphology-based species identification, phylogenetic approaches to Conservation Biology have received more attention. Because molecular systematic techniques are simple to use, fast, relatively cheap, reliable and can often be used without harming organisms that are being studied, they are becoming more widely accepted [3-6].
Two different molecular systematic approaches scientists use to define the taxonomy of a species are chromosome analysis (used mostly when studying plants, because the number of chromosomes in two closely related species can be different), and the comparison of equivalent proteins (or protein-encoding DNA) between species.

Learning from errors made in the past, scientists have realized that the enormous variety of genes and techniques of analysis employed in DNA-based phylogenetic research diminishes the applicability of obtained results to other studies regarding different taxonomic groups. Thus, it became evident that a large-scale initiative with the objective of sequencing a small set of specific genes across all species was needed [1]. Theoretically, DNA-based identification of organisms up to the species level is possible, due to the fact that in any given protein-encoding portion of DNA, the mutations that occur in the third nucleotide of a codon are likely to suffer little selection restraints. Thereby, studying a 600 bp sequence, and even considering the third-position sites are strongly biased (e.g. A–T in arthropods, C–G in chordates), one can expect around 1 billion different sequences. Taking in consideration a modest rate of sequence change, there can be more than 10 nucleotide differences in a 600 bp comparison of closely related species with just a million year history of reproductive isolation [1, 6].

It was in this context that, early in 2003, Prof. Herbert and colleagues coined the term “DNA barcode” [6], drawing a comparison between the Universal Product Codes and genomic sequences – a species is identifiable analysing a single sequence of DNA. Intra-specific divergences (typically less than 2%) mean that a species can be characterized, or bar-coded, by a collection of similar sequences.

The ideal gene sequence was defined as one that is sufficiently conserved to be amplified with broad-range primers, yet is divergent enough to resolve closely related species [1].

For animals, the cytochrome c oxidase I gene (COI) was soon established as a good target: not only it is a mitochondrial gene (it does not have introns, the mode of inheritance is haploid and it has limited exposure to recombination), but also the universal primers for this gene are very robust, enabling recovery of its 5’ end from representatives of most, if not all, animal phyla, and COI appears to possess a greater range of phylogenetic signal than any other mitochondrial gene. In addiction to these characteristics, the third-position nucleotides of the COI gene have a high incidence of mutations, leading to a rate of molecular evolution that is about three times greater than that of 12S or 16S rDNA [6].

Unfortunately or not, DNA-barcoding using the COI gene is not suitable for other life forms. Mitochondrial DNA of plants is much more conserved than in animals, possibly due to hybridization and introgression. In this case, a nuclear gene with a high level of intra-specific variability, such as the internal transcribed spacer (ITS), is proving to be a better choice [1,7]. Fungi present many introns in their mitochondrial DNA, but DNA from this origin can still be used, provided scientists employ reverse transcription together with PCR. For archaea and bacteria, the small subunit ribosomal RNA gene (SSU rRNA) can be used for determination of major groups, but additional genes are needed for more detailed resolution [1].

All new techniques are subject to critics and met with mistrust by concerned scientists: DNA barcoding is no different.
Moritz and Cicero [8] emphasized the need to test DNA barcoding as an effective tool for the correct identification of characterized species, as well as for the discovery of new species, while pointing out that DNA barcoding might not be sufficient to successfully distinguish closely related species. Various studies have since proven the contrary; an example of that being the study conducted by Prof. Herbert [9], in which he and his colleagues discovered that a species of butterfly (Astraptes fulgerator), present from northern Argentina to southern United States and with an unusually wide range of feeding habits, was in fact composed by 10 different cryptic butterfly species that rely in morphologic similarities to escape predators – the theory postulates that all this species evolved to look the same as a particularly fast-flying butterfly that birds are unable to chase.

Moritz and Cicero [8] also criticized the DNA barcode calibration, stating that the values used to distinguish different species in another study conducted by Prof. Herbert [6] were fairly arbitrary and could not be used in other studies for other groups of animals. Once more, this proved not to be a consistent critic: comparison of COI sequences from 13,000 pairs of congeneric species showed a mean divergence of 11.3%, while the sequence variation for the same gene within species is less than 2% [10]; furthermore, another study of a group of an extinct avian species, unrelated from those focused in the study by Herbert, used the previously defined DNA barcode calibration successfully [11], proving once more the potentiality of this technique.

The same scientists point out that the generalizations made by Herbert and colleagues [6] are not valid, claiming that Johnson and Cicero [12] found that almost 74% of the DNA barcodes of closely-related species differences fell short of the defined 2.7% threshold. However, a close look into this study reveals that the presented results are misleading – the results were mainly obtained using different molecular biology methods and other DNA regions other than COI.

They also claim they found different species with the same DNA barcode, an affirmation that per se would make this technique useless. Again, this results are misleading, because they were obtained using an unpublished 723 bp sequence of ND6, which has not been suggested as a good DNA barcoding sequence, and for which there are no calibrations available.

After the publication of the landmark paper by Herbert et al [6], and building on the hype and excitement that soon surrounded this new technique, an initiative with the goal of coordinating a large-scale effort to determine the DNA barcodes of all of Earth’s species was set in motion. The primordial objective of this project was defined: promoting and initiating “a process for a rapid and inexpensive method for on-site identification by any user, irrespective of their background training, for the estimated 10 million species of eucaryote life on Earth” [10]. The foundations for this project, the Barcode of Life, were laid during a meeting sponsored by the Sloan Foundation held in New York in September 2003.

The scientists present at that meeting were able to precisely define the benefits, objectives and priorities of such a large-scale effort, and also set the guidelines on which future research should be based, while envisaging future application for this new technology.

The benefits of the Barcode of Life project include:

“facilitating species identification for all users, including flagging specimens that may represent species new to an area or new to science;

enabling rapid, inexpensive, on-site identifications where traditional methods are unfruitful or too complex;

driving technology development for DNA barcode analysis that can be applied in the field;

providing insight into the evolutionary history of life via quick and reliable species-level identifications. In specific cases, this sequence data will also contribute to multi-gene efforts to determine evolutionary relationships” [10].

The long-term objectives were set to seize the full potentiality of DNA barcoding, while relying on the capacity of networking of the scientists involved and on previous gathered knowledge to maximize the impact of this initiative. As such, it was established that an electronic database linking determined and compiled DNA barcodes with the corresponding “voucher specimens, images, and collateral information including taxonomy, geographic distributions, natural history, and relevant medical and economic concerns” should be created. Ultimately, efforts should be directed towards the creation of a low-cost, portable device for in situ determination of DNA barcodes. Scientists that set these objectives were also careful not to ostracize the centuries-old taxonomic scientific community, stating that they should play their part in this large effort, contributing with their knowledge to facilitate it.

Given that the characterization of the DNA barcodes of all eukaryotes is a large time and money-consuming effort, priorities were set to define which groups of living forms should be studied first. Being better known, more accessible to scientists and more likely to be funded, studies of large animals and plants (larger than 10 mm) were an obvious primary target, more so the ones belonging to endangered species or the ones belonging to invasive or pest species. A set of pilot projects scattered all over the world were set in motion, with the objective of proving the capabilities of DNA barcoding and drawing attention to potential investors, so that further studies of smaller, less well-known and (at least initially) less interesting organisms could be initiated in the near future.

Currently, the Barcode of Life is a well-funded initiative, and gathers the efforts of scientists all over the world in three main campains: the FishBol (the DNA barcoding of fish species, with a priority given to over-fished species), the ABBI (the DNA barcoding of bird species, specially those living in endangered habitats) and the INBIS (the network for barcoding invasive and pest species) [13].

To date, the results gathered and compiled demonstrate strong congruence between morphology-based taxonomy and COI barcode analysis, and the objectives set by the Barcode of Life project are in the process of being met. Was this new tool made available to scientists earlier, several wrong decisions concerning the management of species would have been avoided. Some case-studies are here presented [3, 4]:

The dusky seaside sparrow example: after a strong decline of the number organisms belonging to this birds species, leading to a total population of 6 males in the 1980’s, an unsuccessful captive breeding program was initiated. After the last male died, mitochondrial DNA was analyzed and it was found that the chosen females, which belonged to the Scott’s seaside sparrow species, were members of different clades of phylogeny from the males. An in-depth analysis of the males’ DNA prior to their death, using the DNA barcode technique, could have pinpointed another compatible mating species, and therefore saving the species from extinction;

The red wolf example: in the mid 1970’s, before being extinct in the wild, the red wolf was placed into a captive breeding program, when they were shown to hybridize with coyotes. Having studied mitochondrial DNA from wild, captive and museum skins of red wolves, scientists discovered that red wolves have either coyote or grey wolf mitochondrial DNA, suggesting that this population is in fact a hybrid, and not a species. Regardless of its species status, this study puts in evidence that there are other threatened species that may need a conservation effort much than do the red wolves. Having to choose to save one species in detriment of others, one should choose the species that is more likely to contain unique genetic data, which may prove valuable in the future.

Other interesting studies that suggest the power of DNA barcoding, either for its value in the implementation of species management strategies or for the discovery of new species include:

the tardigrades example [14]: the study, carried out using the SSU rRNA gene, suggested that many tardigarde species present in the United Kingdom remain cryptic and unclassified;

the beaked whales example [15]: the authors of this study claim this molecular approach made possible three important discoveries – a new whale species ( M. perrini), the correct identification of M. bahamondi and M. traversii as belonging to an unique species, and the identification of specimens of I. pacificus, formerly considered the rarest of all whale species;

The leech example [16]: the correct identification of a leech species present in various locations worldwide as belonging to a South American species, lead to the recognition of this invasive species, one that may have an impact on native annelid and mollusc faunas.

Although it as been demonstrated that DNA barcoding is a very powerful tool, its use does not render obsolete other more traditional taxonomic practices – using it only complements them. Scientists using DNA barcoding are able to gather more valuable data that, used together with data from morphology, ecology and behavioural studies, can sometimes prove to be of vital importance for the preservation of endangered species, or for the classification of previously unknown species.
In addition to its use in taxonomy and in the discovery of new species, it’s not hard to imagine that DNA barcoding has broad potential practical applications in the pharmaceutical industry (identification of pathogens and vectors), in the food industry (recognition of pest species, discovery of new species with genetic and nutritional value), cosmetics industry (discovery of species than can be used to obtain new natural products) and national security (monitoring for biowarfare

agents).