Sparing spiders: faeces as a non-invasive source of DNA

Introduction Spiders are important arthropod predators in many terrestrial ecosystems, and molecular tools have boosted our ability to investigate this taxon, which can be difficult to study with conventional methods. Nonetheless, it has typically been necessary to kill spiders to obtain their DNA for molecular applications, especially when studying their diet. Results We successfully tested the novel approach of employing spider faeces as a non-invasive source of DNA for species identification and diet analysis. Although the overall concentration of DNA in the samples was very low, consumer DNA, suitable for species identification, was amplified from 84% of the faecal pellets collected from lycosid spiders. Moreover, the most important prey types detected in the gut content of the lycosids were also amplified from the faecal samples. Conclusion The ability to amplify DNA from spider faeces with specific and general primers suggests that this sample type can be used for diagnostic PCR and sequence-based species and prey identification such as DNA barcoding and next generation sequencing, respectively. These findings demonstrate that faeces provide a non-invasive alternative to full-body DNA extracts for molecular studies on spiders when killing or injuring the animal is not an option. Electronic supplementary material The online version of this article (doi:10.1186/s12983-015-0096-y) contains supplementary material, which is available to authorized users.


Multiplex PCR development
Primers targeting DNA of P. nigra and P. saturatior, respectively, were designed based on available sequences of the cytochrome c oxidase subunit one gene (COI) of high alpine species [2,3].
Out of a larger set of primers developed for species identification of linyphiid and theridiid spiders in glacier forelands [3], primer pairs targeting five common European species of Linyphiidae (Diplocephalus helleri, Entelecara media, Erigone tirolensis, Janetschekia monodon, Agyneta nigripes) were selected and combined with a new reverse primer for E. media to a multiplex system (LIN). This new multiplex PCR system amplifies only DNA fragments shorter than 300 bp length and is thus suitable to detect also semi-digested DNA [4]. Primer concentrations for each multiplex PCR system were adjusted on the basis of standardized DNA templates [2], so that all targets were amplified at approximately the same efficiency. This yielded detection signals of comparable strength if the same amount of template DNA is present during PCR. The sensitivity of both optimized multiplex systems was determined with and without the addition of approx. 300 ng of non-target DNA (Mitopus glacialis) using standardized numbers of DNA templates. For the Pardosa duplex it was also verified that with 20-times excess of DNA of one species the other species is still detectable. The specificity of the multiplex PCR systems was tested with 53 different target and non-target taxa found/collected in high alpine environments. A complete list of all tested taxa and their corresponding amplification success is available in Table A1.  The optimized reactions were performed using the Multiplex PCR Kit (Qiagen) in a volume of 10 µl and contained 1.5 µl DNA extract, each primer at its corresponding concentration (Table A2) species present at a 20 times higher concentration.

Multiplex PCR system for linyphiid prey (LIN)
For the multiplex PCR system targeting five species of linyphiid spiders, the primer sequences, corresponding fragment length, and primer concentrations in the PCR are given in Table A2. Each 10 µl reaction contained 1.5 µl DNA extract, each primer at its corresponding concentration (Table   A2) Detection limits were not negatively influenced by the presence of a high amount of non-target DNA.
If a higher amount of target DNA was present, D. helleri and J. monodon might result in an additional weaker band of ~390 and ~400 bp lengths, respectively. As those fragments are considerably longer than the longest diagnostic fragment included in the multiplex system (298 bp), they can be well distinguished from target bands.   Samples where no DNA was detected with any of the multiplex PCR systems were re-tested with general primers C1-J-1859 [6]and HCO2198 [7] to check for the presence of amplifiable DNA. If no DNA was detected with these primers, both samples from the respective individual (full-body DNA extract and faecal sample) were excluded from the analysis.
From a total of 189 spiders, faecal samples and full-body DNA extracts were tested with different multiplex PCR systems to investigate the suitability of faecal samples for molecular analysis.
In four faecal samples no amplifiable DNA was present, and these were excluded from the analysis together with the respective full-body DNA extracts, leaving 185 individuals for the comparison between the two sample types.