Lytic bacteriophages active in urine against multi-drug resistant clinically derived Klebsiella pneumoniae causing urinary tract infection

This study identifies and characterizes three genetically distinct lytic bacteriophages isolated from environmental samples that effectively target multidrug-resistant *Klebsiella pneumoniae* strains causing urinary tract infections, demonstrating enhanced lytic activity in human urine compared to standard laboratory media.

The Gist

Imagine your body is a bustling city, and sometimes, unwanted invaders (bacteria) set up camp in the wrong neighborhood. In this case, the invaders are Klebsiella pneumoniae, a tough, multi-drug-resistant bacteria that loves to cause infections in the urinary tract (the city's plumbing system).

For patients with specific health issues, like a "neurogenic bladder" (where the nerves controlling the bladder are damaged), this infection is a nightmare. The usual weapons we have—antibiotics—are like old, rusty keys that no longer fit the locks on these bacteria's doors. The bacteria have evolved to ignore them.

This paper tells the story of scientists trying to find a new kind of key: Bacteriophages (or just "phages").

What is a Phage?

Think of a phage as a tiny, microscopic robot or a specialized sniper. It's a virus that doesn't infect humans or animals; it only hunts bacteria. It lands on the bacteria, injects its own DNA, and turns the bacteria into a factory that builds more robots until the original bacteria explodes (lyses) and dies.

The Mission: Finding the Right Snipers

The scientists needed to find phages that could hunt down a very specific, super-tough strain of Klebsiella found in a patient with a neurogenic bladder. They didn't look in a lab; they went to the "wild," searching in sewage and river water. It's like looking for a specific needle in a haystack, but the haystack is a river.

They found three candidates: EDIRA083, EDIRA088, and EDIRA092.

The Character Profiles

The team gave these three "robots" a physical exam and a background check:

1. The Look: Under a powerful microscope, they all looked like little tadpoles with round heads and long tails (a shape called Siphoviridae).
2. The Background Check (Genome): The scientists read their "instruction manuals" (DNA). They wanted to make sure these robots weren't carrying any dangerous cargo.
   • Good news: None of them had instructions to make the bacteria stronger, nor did they carry any antibiotic-resistance genes. They were "clean" and safe to use.
3. The Speed Test:
   • EDIRA083: A slow starter. It takes 40 minutes to get to work, but when it does, it produces a huge army (170 new robots from one bacteria).
   • EDIRA088: A medium speed, medium output.
   • EDIRA092: The sprinter. It gets to work in just 8 minutes! However, it produces a smaller army (38 robots).

The Challenge: The "Target Range"

Here is the tricky part. Phages are like specialized locksmiths. A key that opens one door might not open another.

• EDIRA083 and EDIRA088 were very picky. They only worked on the specific bacteria they were trained on. They were like master locksmiths who only had one key.
• EDIRA092 was more versatile. It managed to open the doors of 29% of the other bacteria strains tested. While that sounds low, in the world of bacteria, that's actually a pretty broad range!

The Real-World Test: The "Urine Pool"

This is the most exciting part of the story. Usually, scientists test these robots in a petri dish full of rich, sugary food (called LB broth). It's like testing a car on a smooth, empty race track.

But the bacteria live in urine, which is a very different environment. It's salty, has a different pH, and lacks the rich nutrients of the lab.

• The Surprise: The scientists tested the phages in human urine.
• The Result: The phages actually worked better in the urine than in the sugary lab food!
• Why? It seems that when the bacteria try to build a shield to resist the phages, they become weak and clumsy in the urine environment. It's like the bacteria trying to wear heavy armor to fight the robots; the armor protects them from the robots, but it makes them too heavy to move or grow in the urine.

The Big Takeaway

The scientists found three new "robot hunters" that are safe, effective, and surprisingly good at their job in the actual environment where the infection happens (the bladder/urine).

• EDIRA088 turned out to be the star player in urine, performing better than the others.
• The study suggests that instead of just testing these cures in a lab, we need to test them in the "real world" (urine) to see if they truly work.

In short: When antibiotics fail, these tiny, nature-made robots might be the new heroes we need to clear the plumbing, especially for patients who are most vulnerable. The study gives us hope that we can engineer a "cocktail" of these robots to hunt down these super-bugs right where they live.

Technical Summary

1. Problem Statement

• Clinical Challenge: Multi-drug resistant (MDR) Klebsiella pneumoniae is a critical priority pathogen causing recurrent urinary tract infections (UTIs), particularly in high-risk populations such as patients with neurogenic bladders. Standard antibiotic therapies often fail due to resistance mechanisms (e.g., ESBL production), leading to increased morbidity, mortality, and hospitalization.
• Therapeutic Gap: While bacteriophage therapy is a promising alternative, there is a scarcity of pre-clinical models and characterized phages specifically targeting MDR K. pneumoniae strains derived from UTIs.
• Methodological Limitation: Most phage efficacy studies are conducted in nutrient-rich laboratory media (e.g., LB broth), which may not accurately reflect the physiological conditions of the urinary tract (e.g., pH, nutrient limitation, ionic composition). This discrepancy can lead to overestimation of efficacy or failure to predict the emergence of resistant mutants in vivo.

2. Methodology

The study employed a comprehensive workflow to isolate, characterize, and evaluate three novel lytic bacteriophages:

• Isolation & Strain Selection:
  • Host Strain: An ESBL-producing K. pneumoniae clinical isolate (RT535, ST1958, KL110 capsule) from a neurogenic bladder patient was used as the isolation host.
  • Sources: Phages were isolated from environmental samples: sewage water (EDIRA083), the Seine River (EDIRA088), and the Auvezère River (EDIRA092).
• Characterization:
  • Morphology: Transmission electron microscopy (TEM) was used to determine virion structure.
  • Genomics: Whole-genome sequencing (Illumina) followed by annotation (Pharokka, Phold) to assess safety (absence of lysogeny, virulence, or resistance genes) and taxonomy (VipTree, TaxMyPhage).
  • Stability: Assays tested phage stability across a pH range (3–11) and temperatures (37°C–70°C).
  • Kinetics: One-step growth curves determined latent periods and burst sizes.
• Host Range Analysis: Efficiency of plating (EOP) was measured against a panel of 79 clinical K. pneumoniae UTI isolates representing 57 sequence types (STs) and 43 capsule types.
• Activity Assessment in Physiological Conditions:
  • Media Comparison: Bacterial growth kinetics were monitored in both LB broth and pooled human urine.
  • Metric: A "Lytic Activity Score" (LAS) was calculated based on the Area Under the Curve (AUC) of bacterial growth over 46 hours to quantify the suppression of bacterial regrowth and resistance emergence.

3. Key Contributions

• Discovery of Novel Phages: Identification and characterization of three distinct lytic phages (EDIRA083, EDIRA088, EDIRA092) belonging to different genera within the Siphoviridae family.
• Safety Profile: Confirmation that the phage genomes lack integrases, transposases, antibiotic resistance genes, and bacterial virulence factors, making them suitable candidates for therapeutic development.
• Physiological Relevance: A critical methodological contribution is the direct comparison of phage activity in standard LB broth versus human urine. The study demonstrates that phage efficacy and resistance dynamics differ significantly between these environments.
• Cocktail Potential: Evaluation of single phages versus multi-phage cocktails to determine optimal strategies for broadening host range and delaying resistance.

4. Key Results

• Genomic & Morphological Features:
  • EDIRA083: Yonseivirus genus; 57.8 kb genome; 88 ORFs; Latent period: 40 min; Burst size: 170.
  • EDIRA088: Undescribed genus (Peekayseptimavirus-like); 48.1 kb genome; 63 ORFs; Latent period: 25 min; Burst size: 43. Notably, it has a longer tail (219 nm) than the others.
  • EDIRA092: Peekayseptimavirus genus; 50.6 kb genome; 91 ORFs; Latent period: 8 min (very short); Burst size: 38.
  • Safety: All three genomes were free of lysogeny-associated genes and antibiotic resistance markers.
• Stability: All phages were stable at physiological pH (5–9) and up to 55°C. EDIRA083 and EDIRA092 remained stable at 60°C, while EDIRA088 was inactivated.
• Host Range:
  • EDIRA083 & EDIRA088: Extremely narrow host range (EDIRA083 infected 2/79 strains; EDIRA088 infected only the isolation host).
  • EDIRA092: Broader host range, infecting 29% (23/79) of the clinical isolates.
  • Depolymerase Activity: EDIRA083 showed a slow capsule depolymerase activity against the KL110 capsule.
• Activity in Urine vs. LB:
  • Regrowth Dynamics: In LB, resistant bacteria emerged rapidly (within 8–10 hours), leading to lower Lytic Activity Scores (LAS) at 46 hours.
  • Urine Superiority: In human urine, overall lytic activity was higher and more sustained. The emergence of resistant mutants was less frequent or their fitness was reduced in urine, resulting in higher LAS values at 46 hours compared to LB.
  • Cocktail Performance: EDIRA088 performed best as a single agent in urine. Cocktails containing EDIRA088 showed superior activity compared to those without it.

5. Significance

• Therapeutic Potential: The study provides a validated pipeline for identifying phage candidates specifically for MDR K. pneumoniae UTIs. The isolation of phages active against a neurogenic bladder strain addresses a critical unmet clinical need.
• Paradigm Shift in Testing: The findings underscore that testing phages in nutrient-rich media (LB) may underestimate their therapeutic potential in the urinary tract. The "urine advantage" suggests that the physiological cost of resistance (fitness trade-off) is higher in urine, making phage therapy more durable in this specific infection site.
• Future Directions: The identification of EDIRA092 (broad host range) and the stability of the cocktail components support the development of personalized or semi-personalized phage cocktails. The data supports moving toward in vivo models that replicate urinary conditions to guide dosing and combination strategies with antibiotics.

In conclusion, this paper establishes a robust foundation for the clinical translation of phage therapy against MDR K. pneumoniae UTIs by characterizing safe, effective, and physiologically stable phages that perform optimally under infection-relevant conditions.