Discussion
Challenges associated with plastid contamination represent a major concern in microbiome analyses (Beckers et al. 2016; Jackrelet al. 2017; Gaona et al. 2020). Our results indicate that the use of DNA-PNA clamps significantly improves the microbiome sequencing output of faecal samples obtained from species with a diet harbouring a large amount of chloroplast and mitochondrial DNA. This effect has also been shown by Fitzpatricket al. (2018) in plant surface microbiomes; however, our study is the first to test the usefulness of clamps in wildlife microbiome studies relying on faecal pellets. Microbiome studies have recently been growing at an accelerated pace. As we move away from model organisms, the diets of the animals under study become more and more diverse, As a rough estimate, 26% of bats and 33% of birds (Ko et al.2014) follow a plant-based diet. Therefore, techniques that allow us to bypass the remnant plant material in faecal samples are becoming more and more important for microbiome studies.
One important factor to keep in mind when using PNA-DNA clamps is the need to have some information about the diet of the study species. PNA-DNA clamp specificity varies between groups. In our case, we had previous knowledge that, in our study area, the diet of Tequila bats consists of almost 100% columnar cacti, particularly from one species, namely Carnigea gigantea (LV and MT, personal observation and unpublished data). In the bat case, visual inspection of the faecal pellets also revealed that a large percentage of the pellets was undigested pollen grain clusters. This facilitated the development of the cpPNA clamp thanks to the information available from other studies (Fitzpatricket al. 2018).
Our technique allows the more cost-effective use of sequencing capacity. By employing PNA-DNA clamps, we have been able to target the “true microbiome” more directly and waste fewer reads related to by-products from the diet of the animal. Having higher read numbers enables better statistical power in the analysis and decreases data losses in the subsequent steps in downstream processing. Even though the sequencing price tag is becoming cheaper every day (NHGRI 2020), without the PNA-DNA clamps, we would have had to double or triple or even increase by ten-fold our sequencing depth to make the latter reasonable enough to allow downstream analyses. The cost of the clamps varies between providers but, in general, the use of the clamps will always be more cost-effective than aiming at larger sequencing depth. With the expansion of microbiome studies to non-model organisms, we believe that additional tools like the one presented in this paper will streamline the future advancement of the field.