The use of next-generation sequencing technologies has accelerated in recent years, allowing for in-depth studies on the ecology of microbial communities (i.e., the microbiome) in both humans and animals. Recent studies have expanded past the traditional use of 16S rRNA gene amplicon data, from high-level characterization of host-associated communities to detecting specific bacterial pathogens. However, few studies have compared clinical versus sequencing methods for pathogen detection such as Salmonella, especially in wildlife. Furthermore, there is little research on the impact of the presence of Salmonella on the microbial communities of wildlife. Therefore, the objectives were to (1) examine the prevalence of Salmonella across bird species from 16S rRNA data, (2) compare microbial communities between birds positive and negative for Salmonella, and (3) evaluate the use of 16S rRNA data for pathogen surveillance by contrasting with the diagnostic results from clinical culture techniques. We collected 705 cloacal or fecal swabs from steppe buzzards (Buteo buteo), Eurasian cranes (Grus grus), barn swallows (Hirundo rustica), and barn owls (Tyto alba) throughout Israel and parts of Russia. Samples were cultured for Salmonella following a standard culturing approach, microbial community composition was characterized by 16S rRNA sequencing and analyzed in R using the phyloseq package. A subset of samples (279) with both culture and 16S data were compared for Salmonella detection.
Results/Conclusions
Sequencing data identified Salmonella in 48 out of 705 birds (6.8%): 5 buzzards (31.3%), 34 cranes (13.1%), 8 swallows (5.6%), and 1 owl (0.35%). For comparison of culture and 16S rRNA data, cultures revealed extremely low prevalence of Salmonella with only six positives all in swallows (7.0%) for a total of 6 out of 279 birds (2.2%). In contrast, sequencing data identified Salmonella in 13 out of 279 birds (4.7%): 5 buzzards (31.3%) and 8 swallows (9.4%) with no significant difference between the techniques in detection (p = 0.18). The results of this study highlight the potential value of 16S rRNA data for bacterial pathogen detection, its ability to identify multiple potential pathogens simultaneously, and as a more sensitive alternative to standard culturing methods.