Antibiotic resistance is a major public health challenge worldwide, whose implications for global health might be devastating if novel antibacterial strategies are not quickly developed. As natural predators of bacteria, (bacterio)phages may play an essential role in escaping such a dreadful future. The rising problem of antibiotic resistance has revived the interest in phage therapy and important developments have been achieved over the last years. But where do we stand today and what can we expect from phage therapy in the future? This is the question we set to answer in this review. Here, we scour the outcomes of human phage therapy clinical trials and case reports and address the major barriers that stand in the way of using phages in clinical settings. We particularly address the potential of phage resistance to hinder phage therapy and discuss future avenues to explore the full capacity of phage therapy.

Citrobacter spp., although frequently ignored, is emerging as an important nosocomial bacterium able to cause various superficial and systemic life-threatening infections. Considered to be hard-to-treat bacterium due to its pattern of high antibiotic resistance, it is important to develop effective measures for early and efficient therapy. In this study, the first myovirus (vB_CfrM_CfP1) lytic for Citrobacter freundii was microbiologically and genomically characterized. Its morphology, activity spectrum, burst size, and biophysical stability spectrum were determined. CfP1 specifically infects C. freundii, has broad host range (>85 %; 21 strains tested), a burst size of 45 PFU/cell, and is very stable under different temperatures (−20 to 50°C) and pH (3 to 11) values. CfP1 demonstrated to be highly virulent against multidrug-resistant clinical isolates up to 12 antibiotics, including penicillins, cephalosporins, carbapenems, and fluroquinoles. Genomically, CfP1 has a dsDNA molecule with 180,219 bp with average GC content of 43.1 % and codes for 273 CDSs. The genome architecture is organized into function-specific gene clusters typical for tailed phages, sharing 46 to 94 % nucleotide identity to other Citrobacter phages. The lysin gene encoding a predicted D-Ala-D-Ala carboxypeptidase was also cloned and expressed in Escherichia coli and its activity evaluated in terms of pH, ionic strength, and temperature. The lysine optimum activity was reached at 20 mM HEPES, pH 7 at 37°C, and was able to significantly reduce all C. freundii (>2 logs) as well as Citrobacter koseri (>4 logs) strains tested. Interestingly, the antimicrobial activity of this enzyme was performed without the need of pretreatment with outer membrane-destabilizing agents. These results indicate that CfP1 lysin is a good candidate to control problematic Citrobacter infections, for which current antibiotics are no longer effective.

The human periodontium health is commonly compromised by chronic inflammatory conditions and has become a major public health concern. Dental plaque, the precursor of periodontal disease, is a complex biofilm consisting mainly of bacteria, but also archaea, protozoa, fungi and viruses. Viruses that specifically infect bacteria - bacteriophages - are most common in the oral cavity. Despite this, their role in the progression of periodontal disease remains poorly explored. This review aims to summarize how bacteriophages interact with the oral microbiota, their ability to increase bacterial virulence and mediate the transfer of resistance genes and suggests how bacteriophages can be used as an alternative to the current periodontal disease therapies.

Providencia stuartii is emerging as a significant drug-resistant nosocomial pathogen, which encourages the search for alternative therapies. Here, we have isolated Providencia stuartii phage Stuart, a novel podovirus infecting multidrug-resistant hospital isolates of this bacterium. Phage Stuart is a proposed member of a new Autographivirinae subfamily genus, with a 41,218-bp genome, direct 345-bp repeats at virion DNA ends, and limited sequence similarity of proteins to proteins in databases. Twelve out of the 52 predicted Stuart proteins are virion components. We found one to be a tailspike with depolymerase activity. The tailspike could form a highly thermostable oligomeric β-structure migrating close to the expected trimer in a nondenaturing gel. It appeared to be essential for the infection of three out of four P. stuartii hosts infected by phage Stuart. Moreover, it degraded the exopolysaccharide of relevant phage Stuart hosts, making the bacteria susceptible to serum killing. Prolonged exposure of a sensitive host to the tailspike did not cause the emergence of bacteria resistant to the phage or to serum killing, opposite to the prolonged exposure to the phage. This indicates that phage tail-associated depolymerases are attractive antivirulence agents that could complement the immune system in the fight with P. stuartii. IMPORTANCE The pace at which multidrug-resistant strains emerge has been alarming. P. stuartii is an infrequent but relevant drug-resistant nosocomial pathogen causing local to systemic life-threatening infections. We propose an alternative approach to fight this bacterium based on the properties of phage tailspikes with depolymerase activity that degrade the surface bacterial polymers, making the bacteria susceptible to the immune system. Unlike antibiotics, phage tailspikes have narrow and specific substrate spectra, and by acting as antivirulent but not bactericidal agents they do not cause the selection of resistant bacteria.

Staphylococcus epidermidis is a major causative agent of nosocomial infections, mainly associated with the use of indwelling devices, on which this bacterium forms structures known as biofilms. Due to biofilms’ high tolerance to antibiotics, virulent bacteriophages were previously tested as novel therapeutic agents. However, several staphylococcal bacteriophages were shown to be inefficient against biofilms. In this study, the previously characterized S. epidermidis-specific Sepunavirus phiIBB-SEP1 (SEP1), which has a broad spectrum and high activity against planktonic cells, was evaluated concerning its efficacy against S. epidermidis biofilms. The in vitro biofilm killing assays demonstrated a reduced activity of the phage. To understand the underlying factors impairing SEP1 inefficacy against biofilms, this phage was tested against distinct planktonic and biofilm-derived bacterial populations. Interestingly, SEP1 was able to lyse planktonic cells in different physiological states, suggesting that the inefficacy for biofilm control resulted from the biofilm 3D structure and the protective effect of the matrix. To assess the impact of the biofilm architecture on phage predation, SEP1 was tested in disrupted biofilms resulting in a 2 orders-of-magnitude reduction in the number of viable cells after 6 h of infection. The interaction between SEP1 and the biofilm matrix was further assessed by the addition of matrix to phage particles. Results showed that the matrix did not inactivate phages nor affected phage adsorption. Moreover, confocal laser scanning microscopy data demonstrated that phage infected cells were less predominant in the biofilm regions where the matrix was more abundant. Our results provide compelling evidence indicating that the biofilm matrix can work as a barrier, allowing the bacteria to be hindered from phage infection.