A test for the “physiological phagemia” hypothesis—natural intestinal coliphages do not penetrate to the blood in horses

Abstract

Soon after the discovery of bacteriophages in 1917, Felix d'Herelle demonstrated that they are present in humanand animal-associated microbiota, and speculated that these phages may play certain roles in the immunity against bacterial infections (d'Herelle 1926). That remains an intriguing but not yet proven possibility up to this day (reviewed in Gorski and Weber-Dabrowska 2005). It has been often reported that phages administrated orally or rectally in the course of phage therapy can penetrate through mucosa of gastrointestinal tract (GIT) to the blood (see Letarov et al. 2010; Gorski et al. 2006 for review) and it was suggested that natural intestinal bacteriophages might also translocate continuously from GIT to the blood. This state has been termed a “physiological phagemia” (Gorski et al. 2006). Coliphages, for example, were detected in a large fraction of commercial lots of animal sera used for eukaryotic cell cultures (reviewed in detail in Gorski et al. 2006). Since the bacteriophage communities present in different subjects are highly individual (Reyes et al. 2010; Golomidova et al. 2007; see also Letarov and Kulikov 2009), it is crucial to perform phage detection in the blood and in the intestinal contents simultaneously, using bacterial test strains able to detect the intestinal phages present in a particular subject to determine the extent to that intestinal phages in fact can routinely reach the blood. The aim of the work presented here is to quantify the presence of natural coliphages in the blood of the animals using a highly sensitive culturing approach. The choice of animal model was based on our data indicating that in contrast to humans and some other animal species (Reyes et al. 2010; see also refs in Letarov and Kulikov 2009), the intestinal viral community in horses is dominated by virulent phages that may occur in high titers in some healthy individuals (Golomidova et al. 2007; Kulikov et al. 2007; and our unpublished observations). Here, we present the analysis of bacteriophages in feces and in the blood of three horses. To select the animals for this study, we screened available population of horses for the titers of the fecal coliphages detectable on E.coli NM522 (Stratagene, USA) strain and selected three animals that yielded more than 10 plaque forming units (PFU)/g of coliphages. The indigenous coliform strains (ICSs) frequently detect different types of the coliphages in the same sample (Golomidova et al. 2007). In order to select a subset of these bacteria for subsequent phage detection, a series of 30 ICS isolates were obtained from each animal and subjected to whole-cell PCR fingerprinting using IS1tr (5′ATCAGTAAGTTGGA(G/A)(T/G)CATTACC-3′) and TR8834 (5′-ATCGGCGATGCGTTGACGAAT-3′) primers as described elsewhere (Isaeva et al. 2010; see also Golomidova et al. 2007). Briefly, the conventional PCR reactions (30 cycles consisting of 94°C for 15 s, 56°C for 30 s, 72°C for 45 s), using boiled cells from the plate as a template, were set up to generate strain-specific banding patterns. This analysis confirmed that all of the isolates were distinct, indicating M. Letarova and D. Strelkova contributed equally to this work. M. Letarova : S. Nevolina :A. Letarov (*) Winogradsky Institute of Microbiology RAS, Prospect 60-letiya Oktyabrya 7/2, 117312 Moscow, Russia e-mail: letarov@gmail.com