Area of Research: DNA Barcoding

There are so many examples of new animal species found through mitochondrial DNA analysis that I believe this should be a routine part of species descriptions. Using morphology alone, taxonomists have often overlooked species that are readily apparent on mitochondrial DNA analysis, including in what should be ideal circumstances using intact adult specimens of large, abundant, and/or economically important organisms. Morphologic characters have been found in some cases but only after DNA has led the way, indicating the discovery process would have been much faster if DNA were analyzed at the beginning. Speed is likely a good way to attract funding, as the public will want the fastest and therefore most economical approach. Mitochondrial DNA analysis can also help extinguish synonomies which have persisted in literature for decades (eg Siddall and Budinoff 2005. Conservation Genetics 6:467).

Under current practice, species recognition whether big or small can be slow (see also earlier post with timeline for discovery of New York Central Park centipede).

Do you see a new species anywhere? Baleen whale specimen collected in 1976, new species description 27 years later based in part on mitochondrial DNA characters (Wada et al. 2003. Nature 426:278)  

   

Does this look like 1904? Honeybee mite Varroa jacobsoni described in 1904. In 1970’s, worldwide epidemic infestation of honeybees presumed due to V. jacobsoni began in Asia. 30 years later, epidemic discovered to be due to a new species, V. destructor (Anderson and Trueman 2000. Exp Appl Acarology 24:165). Species description based on mitochondrial DNA divergence; no morphologic characters other than body size.

  

With funding from the Gordon and Betty Moore Foundation, researchers at Reveo, Inc. and the University of Washington are collaborating on developing a grapefruit-sized sequencer. It uses electronic and photonic effects rather than liquid chemistry and could potentially sequence an entire genome for pennies.

In 2002, Godfray recognized that “in 10 or 20 years time it will be simpler to take an individual organism and get enough sequence data to assign it to a “sequence cluster” (equivalent to species) than to key it down using traditional methods” (Godfray 2002 Nature 417:17). That future is getting closer.

Here is your sequencer, sir

Four whales stranded on the coast of California in the 1970s were identified as Hector’s beaked whale Mesoplodon hectori (Mead 1981 J Mammalogy 62:430). 20 years later mitochondrial DNA analysis revealed these 4 specimens, and a fifth stranded in 1997, to be a new species, designated Perrin’s beaked whale Mesoplodon perrini  (Dalebout et al 2002 Marine Mammal Sci 18:577). The presence of a new species was first recognized by researchers compiling a database of mitochondrial DNA to assist in species identification. The formal species description includes diagnostic molecular characters, helping integrate DNA sequence data with classical taxonomy. As emphasized by Rob DeSalle and others, there is a need to interweave phylogeny and classical taxonomy, which can be met by including DNA sequences routinely used for evolutionary analysis as diagnostic characters in species descriptions (DeSalle et al. 2005 Phil Trans Royal Soc B 360:1905). 

I find it remarkable that a species as large as a whale can languish unrecognized in a collection. How many new species lurk in museum drawers awaiting discovery?  A comprehensive DNA barcode survey of a tissue collection could be relatively inexpensive, as DNA isolation from tissues is generally simple and and sequencing costs are approaching a $1 a specimen. If whales can hide in museums for 20 years, there must be a multitude of new species already collected that will be uncovered by DNA analysis.

In 7 June 2006 Systematics and Biodiversity Andrew Brower examines application of DNA barcodes to identifying and defining new species. His Perspective piece “Problems with DNA barcodes for species delimitation: ‘ten species’ of Astraptes fulgerator reassessed (Lepidoptera: Hesperiidae)” reviews Hebert et al’s 2004 PNAS paper “Ten species in one: DNA barcoding reveals cryptic species in the neotropical skipper butterfly Astraptes fulgerator. My short summary is that we are still learning about how best to use DNA barcodes for identifying new species. In an April 2006 post responding to another worry piece I wrote: “DNA barcoding is a taxonomic tool for a) assigning specimens to known species and b) speeding discovery of new species. More work is needed to determine the best use of DNA barcodes in species discovery.” 

Brower’s article argues about how many Astraptes species are supported by the sequence data, worries about possible consequences of the rise of DNA barcoding, and misses the scientific opportunity to examine this very young species complex with a large set of morphologic, ecologic, and sequence data. He concedes “there are probably at least three species” but declines to put forth what criteria can be used to delimit species. I was surprised to read that the question of whether two sequences “belong to the ‘same species’ is a metaphysical one”. 

A brief summary may be helpful. The PNAS Astraptes paper reports on 25 years of ecological and morphological work on Astraptes fulgerator larva and adults in a conservation area in northwestern Costa Rica. Approximately 40% of the 2,592 A. fulgerator caterpillars collected over this interval were successfully raised to adult; DNA barcodes were derived from the adult specimens. Differences in larval morphology and food plants and in some cases subtle differences in adult morphology suggested the presence of 6-7 cryptic species. When ADDED to this very large natural history knowledge base, DNA barcoding supported these inferences and highlighted another possible 3 or 4 species.  The number of individuals sequenced is larger (4-103 per putative species; 7 of 10 with more than 40 specimens sequenced) than in any other single species study I am aware of. Despite the large sample sizes and detailed information on biological co-variation in ecological and morphological traits, Brower seems to read the paper as defining species based on DNA barcodes. According to his own analysis, NJ bootstrap recovered 7 species and cladistic haplotype analysis recovered 8, albeit with low statistical support. The argument over whether some of the putative species are established to be distinct biological entities is important for the Astraptes specialist, but misses the point of how well barcode lineages in the neighbor-joining tree match differences in biology in this VERY YOUNG species complex. To my reading, the findings suggest that if one were starting over, or starting with a unstudied group, using DNA barcoding as a first step would be the fastest way for taxonomists to sort specimens and find new species.  

In Barcoding Life, Illustrated we suggested that “standardization typically lowers costs and lifts reliability…and should help accelerate construction of comprehensive, consistent reference library of DNA sequences and development of economical technologies for species identification.”  However, some have been doubtful about the benefits of standardization. This reflects in part the balkanized nature of taxonomic science, as most systematists specialize on groups of related organisms and have limited scientific interactions with those studying other groups. Last year Jesse Ausubel and Paul Waggoner outlined some of the concerns raised by the growth of DNA barcoding in Barcoding Worries and Limits. The final item asks whether “it is too soon to standardize on a very few localities”. The salamander paper discussed in last week’s post seems to support this concern, as it implies that the relative rates of evolution of mitochondrial genes differ between animal groups. According to the paper, in salamanders COI evolves much more rapidly than other mitochondrial protein-coding genes.

However, as detailed below, my analysis indicates that salamanders are unexceptional in terms of COI differences.  I find evolutionary divergences are widely distributed in mitochondrial protein-coding genes and the patterning of these differences is similar in diverse invertebrate and vertebrate organisms, including salamanders. First, I used PipMaker (percent identity plot) to examine the distribution of sequence differences in complete mitochondrial genomes. A representative PipMaker plot, shown above, compares the mitochondrial genomes of 2 plethodontid salamanders, Desmognathus fuscus and D. wrighti, revealing a typical pattern of widely distributed sequence differences. As in most vertebrates, in salamanders COI is slightly LESS divergent (14%) than other protein-coding regions, including cytochrome b (CYTB) (17%).   

Second, I compiled differences between closely-related species pairs in CYTB and COI. For historical reasons, CYTB has been the single most common locus used to analyze differences in vertebrate species, and COI has been the single most common single locus for invertebrates. As shown below, these genes show highly correlated differences in deeply divergent lineages of invertebrates and vertebrates, including salamanders. 

Species examined include Demospongiae (sponges); Platyhelminthes (tapeworms); Echinodermata (starfish); Arthropoda (ticks; fruit flies; mosquitos; shrimp); Mollusca (octopi; mantis shrimp) Vertebrata (turtles; eels; coelecanths; elephants; wolves; apes; and 5 pairs of congeneric salamanders)  

These findings support standardizing on COI as the primary DNA barcode for multicellular animals. As discussed in earlier posts, there will of course be cases in which the primary barcode does not resolve closely-related species and a secondary barcode(s) may be needed.  

In “Estimating diversity of Indo-Pacific coral reef stomatopods through DNA barcoding of stomatopod larvae” (FirstCite early online publication in Proceedings Royal Society Biology) Paul Barber and Sarah Boyce, Boston University, look at a what is thought to be a well-understood group, stomatopods, commonly known as mantis shrimp. Stomatopods are marine crustaceans distinct from true shrimp and are thought to include about 400 species worldwide. Like many marine species, they have a bipartite life cycle where dispersal is achieved through a planktonic larval developmental stage. However, larval stages are notoriously difficult to identify morphologically. Barber and Boyce first established a database of COI DNA barcodes from adult forms of nearly all known species. They then applied DNA barcoding to planktonic larvae collected in light traps at locations in the Pacific Ocean and Red Sea. Comparison of sequences from 189 larval forms revealed 22 distinct larval operational taxonomic units (OTUs), but a minority of these matched known species, suggesting that stomatopod species diversity is underestimated by a minimum of 50% to more than 150%. Their results support general use of DNA barcoding as a rapid, relatively-inexpensive first step in cataloging marine species with planktonic larvae. A similar approach has been applied on land by Smith et al “DNA barcoding for effective biodiversity assessment of a hyperdiverse arthropod group: the ants of Madagascar“.

DNA barcode OTUs, such as found in these studies, are not equivalent to species descriptions and are not sufficient to establish systematic phylogeny. In my view, these studies indicate that DNA barcodes can be permanent indexers for filing and retrieving biologic information in the encyclopedia of life.  Routinely incorporating DNA barcoding into biological surveys will enhance the long-term value of expensive field work.