Introduction
Owing to ever-changing climate and land use patterns, biodiversity is increasingly under threat (Pimm et al., 1995; Jetz et al., 2007). In the last 50 years, especially, agricultural expansion has brought about changes in land use that have severely compromised biodiversity (Meyer & Turner, 1992; Matson et al., 1997; Pain & Pienkowski, 1997; Newton, 1998; Chamberlain et al., 2000; Benton et al., 2003). Among other consequences, expanding croplands are diminishing plant species abundance and richness (Wilson et at., 1999; Storkey et al., 2012; Fonderflick et al., 2020), and consequently seed banks (Wilson et at., 1999; Andreasen et al., 2018) and the seasonality of seed availability (Newton, 2004). Further, as the seeds used in intensive agriculture are mostly treated, this also has devastating repercussions on the birds that ingest these seeds (Prosser & Hart, 2005; Lopez-Antia et al., 2015; Millot et al., 2017; Eng et al., 2019). The intensification of agriculture has been also identified as the main driver of declining insect populations (Benton et al., 2002), which are also a dietary component of many bird species. This decline in insect populations is already having a notable impact on insectivorous bird communities (Bowler et al., 2019). Land use changes and reduced food availability are a stress factor for birds, affecting their health and immune response (Kitaysky et al., 1999; Hoi-Leitner et al., 2001; Kitaysky et al., 2007; Pigeon et al., 2013; Almasi et al., 2015) also lowering their defences against parasites (Korschgen et al., 1978; Wakelin, 1996; Nordling et al., 1998). In effect, synergistic effects of food availability, parasitism and stress have been observed on population densities (Chapman et al., 2006). Disease/parasitism and nutrition often interact to determine the abundance of wildlife populations. Helminth and protozoan parasites can impact host survival and reproduction directly through pathological effects and indirectly by compromising the host’s health state (Coop & Holmes, 1996; Murray et al., 1998). Hence, parasitism is also an important factor to consider as a potential risk for both wild and farm-land bird species (Lafferty, 1997; Dunn et al., 2014; Stockdale et al., 2015; Cabodevilla et al., 2020).
To date, studies designed to address the diet of farmland birds have paid most attention to available plants and arthropods, without confirming whether these were consumed or not (Salamolard & Moreau, 1999; Holland et al., 2006; Faria et al, 2012; Holland et al., 2012). Other studies focusing more on the prey ingested have examined this issue through visual identification of the remains of prey exoskeletons present in the faeces of these birds (Jiguet, 2002; Browne et al., 2006; Holland et al., 2006; Bravo et al., 2017). This identification method, widely used for diet analysis, does not generally go beyond the ordinal taxonomic level and few individuals are identified at deeper levels (Jiguet, 2002; Browne et al., 2006; Bravo et al., 2017). In addition, some prey remains could go undetected or unidentified because of difficulties in identifying diet components after they have been digested (Moreby, 1988; Pompanon et al., 2012). The situation for parasitological studies conducted in farmland birds is similar as most have involved the visual identification of parasites (Browne et al., 2006; Okulewicz & Sitko, 2012; Rengifo-Herrera et al., 2014; Presswell & Lagrue, 2016) often via the dissection of dead animals (Villanúa et al., 2008; Santoro et al., 2010; Okulewicz & Sitko, 2012). However, in studies based on faeces samples (non-invasive), the most abundant parasites can be visually identified but low intensity parasitism is hardly discernible (Rengifo-Herrera et al., 2014; Presswell & Lagrue, 2016). Moreover, taxonomic identification based only on the morphological appearance of the eggs of many species is also limited or even impossible (Browne et al., 2006; Presswell & Lagrue, 2016).
The introduction twenty years ago of environmental DNA analysis opened a new avenue for ecology science (Taberlet et al., 2012b). This method has quickly gained importance in parallel with new‐generation sequencing technology (Shokralla et al., 2012; Taberlet et al., 2012a, b) and is becoming increasingly popular for the study of diet, microbiology and parasitology (Pompanon et al., 2012; Shokralla et al., 2012; Taberlet et al., 2012a; Bass et al., 2015; Kerley et al., 2018). Through eDNA analysis, useful information can be obtained from non-invasive samples such as faeces (Srivathsan et al., 2016), thus avoiding the need to examine dead animals. Besides parasites, faeces samples can also provide useful information on host genetics, gut microbiota and diet (Srivathsan et al., 2016). However, faeces usually contain various substances that act as PCR inhibitors (Lantz, 1997; Wilson, 1997; Rådström et al., 2004) and faecal DNA is also fairly degraded (Deagle et al., 2006). This determines a need to design proper primers for eDNA analysis.
To date, many different barcodes, both specific and broad-spectrum primers, have been designed for DNA metabarcoding (Prosser et al., 2013; Hadziavdic et al., 2014; Van Steenkiste et al., 2015; Cheng et al., 2016; Krehenwinkel et al., 2018). However, the large size of the DNA fragments targeted sometimes makes these primers impractical for environmental metabarcoding studies. In effect, DNA from water, soil, air or faeces tends to be quite degraded (Deagle et al., 2006; Yu et al., 2012; Taberlet et al., 2012a) so there is a need to be careful with the length of the barcode used (Hajibabaei et al., 2006; Deagle et al., 2006; Deagle et al., 2007; Taberlet et al., 2012b). Studies that have focused on the metabarcoding of eukaryotic eDNA have already described specific mini-barcodes for different taxa (Epp et al., 2012; Pompanon et al., 2012). However, using these specific mini-barcodes, analysis focuses only on the target taxa, and no information is provided on the importance of these taxa in the sample. In contrast, broad-spectrum primers (usually with lower resolution) do provide information on both target taxa and on their overall contribution to a higher category taxon. For example, a eukaryotic broad-spectrum primer pair used on faeces samples could simultaneously provide information on host genetics, diet and intestinal parasites. However, as far as we know, the eukaryotic mini-barcodes available have not been tested with this purpose in mind. We propose that broad-spectrum primers could be useful tools for comparative studies. While, so far, metabarcoding cannot be considered a quantitative tool, several studies have shown some quantitative capacity of this method (Evans et al., 2016; Lamb et al., 2019; Piñol et al., 2019). Accordingly, it should be possible to compare the proportions of each taxon among similar samples, e.g., water samples from different ponds of faeces collected in different seasons (Pompanon et al., 2012). The use of candidate primers to identify taxa, nevertheless, depends on the existence of reference sequence data (Clarke et al., 2014) so broad-spectrum barcodes need a robust reference dataset. Moreover, for an accurate estimate of biodiversity, a sufficiently variable DNA region needs to be amplified. The most used DNA genes for barcode design are COI, cytb, 12S, 16S, 18S, ITS1, ITS2 and rbcL (Hajibabaei et al., 2007; Pompanon et al., 2012; Andújar et al., 2018; Djurhuus et al., 2020). Among these, the 18S rRNA gene of the small ribosomal subunit (SSU) spans an especially variable region for which there exists a robust and constantly expanding reference dataset (Hadziavdic et al., 2014).
The present study was designed to identify a mini-barcode for use in metabarcoding that can provide information from faeces samples on both the diet and intestinal parasites of birds. We assessed existing primers as well as self-designed primers targeting the 18S rRNA gene in terms of their suitability for use on bird faeces samples. Once we had identified a suitable mini-barcode, it was also tested empirically using a high throughput sequencing approach on faeces samples of different steppe bird species.