Use of length heterogeneity polymerase chain reaction (LH-PCR) as non-invasive approach for dietary analysis of Svalbard reindeer, Rangifer tarandus platyrhynchus

Abstract

To efficiently investigate the forage preference of Svalbard reindeer (Rangifer tarandus platyrhynchus), we applied length-heterogeneity polymerase chain reaction (LH-PCR) based on length differences of internal transcribed spacer (ITS) regions of ribosomal RNA (rRNA) to fecal samples from R. tarandus platyrhynchus. A length-heterogeneity (LH) database was constructed using both collected potential food sources of Svalbard reindeer and fecal samples, followed by PCR, cloning and sequencing. In total, eighteen fecal samples were collected between 2011 and 2012 from 2 geographic regions and 15 samples were successfully amplified by PCR. The LH-PCR analysis detected abundant peaks, 18.6 peaks on an average per sample, ranging from 100 to 500 bp in size and showing distinct patterns associated with both regions and years of sample collection. Principal component analysis (PCA) resulted in clustering of 15 fecal samples into 3 groups by the year of collection and region with a statistically significant difference at 99.9% level. The first 2 principal components (PCs) explained 71.1% of the total variation among the samples. Through comparison with LH database and identification by cloning and sequencing, lichens (Stereocaulon sp. and Ochrolechia sp.) and plant species (Salix polaris and Saxifraga oppositifolia) were detected as the food sources that contributed most to the Svalbard reindeer diet. Our results suggest that the use of LH-PCR analysis would be a non-invasive and efficient monitoring tool for characterizing the foraging strategy of Svalbard reindeer. Additionally, combining sequence information would increase its resolving power in identification of foraged diet components.

Figure 1. The study area, Ny-Ålesund (78°53'-78°55'N, 11°46'-12°11'E), northwest coast of Spitsbergen Island, Svalbard in Norway.

Eight different sites were selected as sampling sites around two different glaciers. First letter of abbreviation represented directions: A: east (Austre), M: middle (Midtre), V: west (Vestre). Last letter of abbreviation represented two different glaciers: B: Brøggerbreen, L: Lovénbreen.

Figure 2. LH profiles of 15 fecal samples collected from different regions in between 2011 and 2012.

(a) Lovénbreen, Ny-Ålesund in 2011; (b) Brøggerbreen, Ny-Ålesund in 2012; (c) Lovénbreen, N-Ålesund in 2012. Shaded bars on profiles represented major peaks, which were the most contributing variables on first two principal components in PCA analysis.

Figure 3. PCA results extracted from LH profiles of 15 fecal samples from collected from different regions in between 2011 and 2012.

Bars represented loading values of the first principal component (PC1) and the second principal component (PC2). Open circles: Lovénbreen, Ny-Ålesund in 2011; Closed circles: Brøggerbreen, Ny-Ålesund in 2012; Closed inverted triangles: Lovénbreen, Ny-Ålesund in 2012.