Chemostat culturing reduces fecal eukaryotic virus load and delays diarrhea after virome transplantation

Fecal virome transfer (FVT) shows promise in reducing necrotizing enterocolitis (NEC), likely due to donor bacteriophages preventing the gut dysbiosis preceding disease. However, concurrent transfer of eukaryotic viruses may carry a risk of infection for the recipient. To increase safety, we investigated chemostat propagation as a method to eliminate eukaryotic viruses from donor feces while maintaining a diverse and reproducible bacteriophage community. Donor feces was collected from healthy suckling piglets and inoculated into a fermenter containing growth media supplemented with lactose and milk oligosaccharides (MOs). During continuous medium exchange (20% volume/h), dilution significantly reduced eukaryotic viruses. Viral richness was concurrently reduced although still preserving a stable community of 200-250 bacteriophages. Inclusion of MOs in the medium ensured higher bacterial richness and a bacterial community closer resembling donor feces. Fecal Lactobacillaceae bacteria were lost during cultivation but partially replaced by members of the Bacteroidota phylum in MO-supplemented cultures, accompanied by phages predicted to have Parabacteroides as host. After cultivation, virus-like particles (VLPs) were isolated, and their ability to reduce NEC incidence tested in vivo. Preterm piglets were delivered by cesarean section and received either the lactose- or MO-propagated viromes by oral route (n = 14-15/group). These were compared with groups receiving the same dose of donor fecal virome (1010 VLPs/kg) or vehicle control. The piglets were subsequently fed infant formula for 96 hours followed by euthanasia and tissue sampling. Both chemostat-propagated viromes effectively mitigated diarrhea compared to the donor virome. The donor virome partially engrafted in recipients and led to higher levels of Lactobacillaceae bacteria and Lactobacillaceae targeting phages. However, these signatures were lost in recipients of chemostat-propagated viromes, and only minor microbiome effects and no NEC prevention were observed. To conclude, we provide in vivo proof-of-concept for chemostat propagation of fecal viruses as a means to deplete eukaryotic viruses and in turn reduce side effects in newborn virome recipients. However, chemostat culture conditions need further optimization to preserve the donor phageome.