Pseudomonas aeruginosa PA5oct jumbo phage reduces planktonic and biofilm population and impacts its host virulence through a pseudolysogeny event

In this work we assess critical parameters to assess the in vitro capacity of the novel “jumbo” phage PA5oct for phage therapy by studying its impact on the planktonic and biofilm population of P. aeruginosa. PA5oct is the third largest (in terms of the genome size) Pseudomonas phage with a specific spectrum of lytic activity (infects 24% of strains from the international P. aeruginosa reference panel and 68% of clinical strains from the Belgian Military Hospital Neder-Over-Heembeek). In vitro studies using the airway surface liquid (ASL) model demonstrate that PA5oct effectively inhibits the growth of planktonic population of P. aeruginosa. Interferometric measurements of a 72-hour biofilm also prove the contribution of PA5oct in biofilm matrix degradation. Biofilm-derived PA5oct phage-resistant mutants often show cross-resistance to both LPS-dependent and only closely related fimbriae-dependent phages. This phage resistance is mostly associated with PA5oct phage pseudolysogeny event which strongly reduces bacterial in vivo virulence. These properties can be considered as key assets in using this bacterial virus in phage therapy settings. Importance An important aspect of phage therapy is the efficacy of bacterial eradication and emergence of phage-resistant mutants. Here we present the consequences of PA5oct jumbo phage treatment of planktonic and sessile P. aeruginosa cells. Apart of efficient bacteria eradication and biofilm degradation, PA5oct introduces stable pseudolysogeny in most of phage-resistant clones. Moreover, phage episome is related to reduced bacterial virulence and clearance by innate immune system of infected host. This is in contrast to the current knowledge about pseudolysogeny as an temporary stage in dormant cells, mostly having no influence on bacteria. In treatment, the pseudolysogeny is considered as an adverse effect giving phage-resistance, but our results prove that this event may lead to diminished bacterial virulence and clearance from infected host. The above findings provide new insights into general knowledge of the host-parasite interactions between lytic phages and bacteria and can impact the application of phage therapy in general.


61
Finding a solution to the antibiotic resistance problem is one of the greatest challenges of 62 modern science and medicine, and the search for alternative methods of antibacterial therapy 63 has led to the reappraisal of bacteriophages (1). Phage therapy efficacy studies (2-4) and recent 64 advances in the regulatory frame, in which phage therapy can be adopted as part of 'Magistral 65 preparations' (5) have shifted the focus from proving efficiency to its implementation and the 66 expansion of available phage for phage therapy cocktail design. 67 Among phages being evaluated for phage therapy applications, are the jumbo phages, which 68 contain dsDNA genomes in excess of 200kb (6). Generally, Jumbo phages can be found in 69 various historical commercial phage therapy products, as they have a broad host range. This is 70 explained by the large coding potential of jumbo phage, allowing them to be (partly) 71 independent from the bacterial enzymes (7). However, some are marked with a high frequency 72 of transduction, impacting the evolution of bacteria and raising questions on the safety of use 73 of this phages in therapy (8). Although the first "jumbo phage" has been discovered over 40 74 years ago (Bacillus bacteriophage G), the frequency of giant phages isolation remains rather 75 low (less than 90 complete genomes in GenBank database), but is surging in recent years. This 76 group is incredibly diverse, since newly isolated jumbo phages generally show low similarity 77 to those present in public databases. Furthermore, annotated genomes contain a vast majority 78 of genes with undefined function (9). 79 The first fully sequenced "jumbo phage" specific for Pseudomonas aeruginosa was the phiKZ     After infection of both epithelial cell lines for 3 h, the colony count shows that all P. aeruginosa 148 strains efficiently propagate in both ASLs (10 7 -10 9 cfu/ml). Phage treatment significantly (p < 149 6 0.05) reduces the CFU counts for normal NuLi-1, where a 4.5 log, 6.5 log and 3 log decrease 150 is observed for PAO1, nonCF0038 and CF708, respectively (Fig. 1A). The phage application 151 for CuFi-1 epithelia infection is also very effective (p < 0.05) giving 5 log, 2.5 log and 5 log 152 reductions in CFU of PAO1, nonCF0038 and CF708, respectively (Fig. 1B).   whereas 20 clones become resistant. In the next step, isolates are examined for cross resistance 207 to other phages lytic to PAO1 wild type strain (Table S1), recognizing different receptors: Type 208 IV pili-dependent (phiKZ, KTN4, LUZ19), LPS-dependent (KT28, KTN6, LUZ7) and phage 209 with an unverified receptor (LBL3). Control isolates and PA5oct-sensitive clones taken from 210 the biofilm after phage treatment retain the same phage typing pattern as PAO1 wild type. In 211 general, we can distinguish five different phage typing patterns (Table 2).    (Table 3).  The principal aim of this study was to assess the suitability of Pseudomonas virus PA5oct for 273 phage therapy applications, using our suite of preclinical evaluation methods.

274
The ASL experiments confirm PA5oct can efficiently access and infect planktonic cells in consequence, the bacterial population that survived the phage therapy becomes sensitive to the 313 immune system that effectively removes the pathogenic agent (4).

14
A somewhat forgotten phenomenon of pseudolysogeny for many years was regarded as a 315 temporary stage of phage particle dormancy. However, it appears that the presence of episome 316 inside the host cell influences its phenotype and contributes the cross-resistance to other 317 bacteriophages as well (31, 32). The giant phages are known to easily undergo episome 318 formation as was previously reported for phiKZ phage (33). In this study we indicated the 319 presence of a pseudolysogeny event of PA5oct giant phage which strongly affect the virulence 320 of bearing PAO1 population. These pseudolysogens show an increased pro-inflammatory 321 stimulation in monocytic cells enhancing the clearance mechanisms of innate immune system.

322
As the consequence significant survival improvement was seen in infected moth larvae. that PA5oct has an interesting potential as part of a therapeutic cocktail, based on its versatility, 331 its specific activity towards mature biofilms and its targeting of virulence factors as receptor.

332
However, its specificity towards groups of isolates also focuses its applicability and efficiency The bacterial susceptibility to phage-mediated lysis was determined by spot test technique 349 applying 10 5 pfu/ml phage titer (38 City, IA). The ASL model was prepared according to methods described by Zabner (19). The 361 experiment was conducted as previously described (15). In short, cell lines were infected with 362 the P. aeruginosa PAO1 reference strain (6.2 × 10 7 cfu/ml), nonCF0038 isolate from burn 363 wound (6.5 × 10 7 cfu/ml) and CF708 small colony variant (1.0 × 10 6 cfu/ml) and incubated for  Table   403   S1). All tests were performed in triplicate.

405
Pyocyanin and pyoverdine production 406 The level of pyoverdine and pyocyanine production was analyzed on a 72h-biofilm formed on