N-acetylcysteine prevents catheter occlusion and inflammation in catheter associated-urinary tract infections by suppressing urease activity

Abstract

Introduction: Proteus mirabilis is a key pathobiont in catheter-associated urinary tract infections (CA-UTIs), which is well known to form crystalline biofilms that occlude catheters. Urease activity alkylates urine through the release of ammonia, consequentially resulting in higher levels of Mg²⁺ and Ca²⁺ and formation of crystals. In this study, we showed that N-acetyl cysteine (NAC), a thiol antioxidant, is a potent urease inhibitor that prevents crystalline biofilm formation.

Methods: To quantify urease activity, Berthelot's method was done on bacterial extracts treated with NAC. We also used an in vitro catheterised glass bladder model to study the effect of NAC treatment on catheter occlusion and biofilm encrustation in P. mirabilis infections. Inductively-coupled plasma mass spectrometry (ICP-MS) was performed on catheter samples to decipher elemental profiles.

Results: NAC inhibits urease activity of clinical P. mirabilis isolates at concentrations as low as 1 mM, independent of bacterial killing. The study also showed that NAC is bacteriostatic on P. mirabilis, and inhibited biofilm formation and catheter occlusion in an in vitro. A significant 4-8_log10 reduction in viable bacteria was observed in catheters infected in this model. Additionally, biofilms in NAC treated catheters displayed a depletion of calcium, magnesium, or phosphates (>10 fold reduction), thus confirming the absence of any urease activity in the presence of NAC. Interestingly, we also showed that not only is NAC anti-inflammatory in bladder epithelial cells (BECs), but that it mutes its inflammatory response to urease and P. mirabilis infection by reducing the production of IL-6, IL-8 and IL-1b.

Discussion: Using biochemical, microbiological and immunological techniques, this study displays the functionality of NAC in preventing catheter occlusion by inhibiting urease activity. The study also highlights NAC as a strong anti-inflammatory antibiofilm agent that can target both bacterial and host factors in the treatment of CA-UTIs.

Keywords: N-acetyl cysteine (NAC); Proteus mirabilis; UTI; biofilms; catheter-associated urinary tract infections (CA-UTI); urease.

Figure 1

Laboratory set up of in vitro glass bladder model. (A) Schematic of the glass bladder model arrangement. Arrows depict the flow of artificial urine media through the bladder. The blue lines represent the flow of water through silicon tubes into the glass jacket around the bladder compartment, while the yellow represents flow of AUM through silicon tubes into the bladder. (B) Photo showing the actual arrangement in use.

Figure 2

NAC inhibits the activity of urease from clinical P. mirabilis isolates and JBU. (A) The minimum urease inhibitory concentration was identified using P. mirabilis 67 as a representative strain. (B) The urease inhibitory concentrations tested on Jack bean urease (JBU) as a control urease. Tukey’s multiple comparisons test was used for statistical analysis, **p<0.01 compared to the respective untreated control, *p<0.05 compared to the respective untreated control. ****p<0.0001, ***p<0.001. (C) The quantity of ammonia produced was used as a measure of urease activity after 2 h exposure to NAC of P. mirabilis clinical isolates. Tukey’s multiple comparisons test used for statistical analysis, **p<0.01 compared to respective untreated controls, *p<0.05 compared to 0.5 mM #p<0.05 compared to 5 mM. (D) Bacterial counts enumerated as CFU/mL corresponding to data in (C) Tukey’s multiple comparisons test used for statistical analysis, *p<0.05, **p<0.01. Data represents the mean of n=5 replicates.

Figure 3

The influence of NAC on the rate of urea degradation by Jack Bean urease. (A, B) Hyperbolic curve fitting (Least squares fit) was used to determine Michaelis-Menten parameters and change in V_max of JBU and its substrate, urea, in the presence of NAC. (B) A summary of these parameters. (C, D) Sigmoidal curve fitting to determine the Hills coefficient of urea binding to JBU in the presence of NAC. (E) The urea breakdown rate by JBU over 10 minutes, recorded at 15mM urea in the presence of varying concentrations of NAC. (F) The urea breakdown rate by JBU over a 10-minute period, recorded at 25mM urea in the presence of varying concentrations of NAC. Data represents the mean of n=3 replicates. The error bars show standard deviation in absorbance values. **p<0.01.

Figure 4

NAC affected the structure and ellipticity of urease. The appearance of the far ultraviolet region of the CD spectrum of urease when treated with NAC (in black). The far-UV region of the CD spectra of NAC and urease alone are also displayed (in red, green, and blue).

Figure 5

NAC prevented P. mirabilis adhesion to substratum and disrupted preformed biofilms in vitro. NAC significantly reduced initial adhesion of P. mirabilis isolates to substratum (A-C) compared to its untreated control. (A, B) show a representative image of the visual difference in adhesion as displayed by P. mirabilis 67. Scale bar: 10 μM (D) Significant disruption of 48 h biofilms and reduction in viable bacteria was recorded when treated with different concentrations of NAC. **p<0.01 relative to untreated controls. (E-H) Confocal images of biofilms stained with live/dead stain showing differences in biofilm architecture and level of disruption when treated with NAC and NAC + ciprofloxacin. (E) untreated control of P. mirabilis 67, (F) 30 mM NAC, (G) 6 μg/mL ciprofloxacin, (H) 30 mM NAC + 6 μg/mL ciprofloxacin (I) Quantification of surface area covered by biofilms under each treatment condition. (J) Quantification of live/dead biomass under each treatment condition. Sidak’s multiple comparisons testing was used for data in (C) and Tukey’s multiple comparison testing was used for the remaining statistical analyses reported. Unless otherwise stated, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. Data represents the mean ± SD of n = 3 biological replicates.

Figure 6

NAC prevented bacterial migration through catheter tubing. (A) Viable bacterial count obtained from catheter “bridges” that were scraped, sonicated, and plated on TSA plates for bacterial enumeration. (B) Growth as determined by OD, from a 1x1 cm² square of agar on the uninoculated side, to measure bacterial presence and migration. (C) Summary of P. mirabilis migration through catheter bridges in presence of NAC. (D, E) Representative images of P. mirabilis 67 migration and growth in 0-, and 30 mM NAC, respectively. Tukey’s multiple comparisons testing used for statistical analysis. *p<0.05, **p<0.01, ****p<0.0001, compared to untreated controls. Data represent the mean of ± SD of n = 4 biological replicates, with three technical replicates in each experiment.

Figure 7

NAC prevented catheter encrustation by P. mirabilis in vitro. (A) Comparison of time taken to cause catheter occlusion by P. mirabilis 67 in the presence of 10 mM AHA and 30 mM NAC. (B) Visual comparison of Untreated, NAC and AHA treated catheters respectively at time of occlusion or experimental endpoint. (C) Enumeration of viable bacteria present at time of catheter harvest (corresponding to time of occlusion or experimental endpoint), in catheter biofilms. (D, E) Measurement of artificial urine pH and bacterial counts in artificial urine, respectively, in the bladder, taken at 24 h intervals until time of catheter occlusion or experimental endpoint. Dunnett’s T3 multiple comparison’s testing was performed for statistical analysis. *p<0.05, ****p<0.0001. Data represent the mean ± SD of n = 3 biological replicates. ns, not significant.

Figure 8

Elemental profiles of biofilms formed in catheters in vitro under different treatment conditions. Elemental analysis of the biofilms formed in the eye of catheters harvested at occlusion or at experimental endpoint was profiled using ICP-MS. (A) Concentration of magnesium, calcium and phosphorus present in crystalline biofilms formed in catheter eye. (B) Concentration of Nickel, Iron, Zinc and Aluminium present in crystalline biofilms formed in catheter eye. Dunnett’s T3 multiple comparisons test performed for statistical analysis. *p<0.05, **p<0.01, ***p<0.001. Data represent the mean ± SD of n = 3 biological replicates.

Figure 9

Inflammatory profiles of 5637 BECs to P. mirabilis infection or JBU in presence of NAC. (A-D) IL-6, IL-8, IL-1b and TNF-α production in response to P. mirabilis infection in the presence of varying concentrations of NAC (in blue). Uninfected conditions are in pink. BECs were infected with P. mirabilis and incubated for 2 h prior to supernatant collection for cytokine analysis. (E-H) IL-6, IL-8, IL-1b and TNF-α production in response to JBU in presence of AHA and MBIC concentrations of NAC. The difference in cytokine production between AHA and NAC treatments were also observed. Dunnett’s T3 multiple comparisons test was performed for statistical analysis. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. Data represent the mean ± SD of n = 3 biological replicates. ns, not significant.

Figure 10

Treatment with NAC significantly reduced bacterial invasion of BECs. (A) BECs infected with either P. mirabilis 44 or P. mirabilis 67 for 2 h displayed a significant reduction in internalised bacteria in the presence of NAC, in a concentration dependent manner. (B) Percentage of infecting bacteria internalised by cells. (C) Visualisation of invasion of BECs by fuGFPb-tagged P. mirabilis 44 in the absence of NAC, with the inset image depicting a 3-D visualisation of invaded BECs. (D, E) Invasion of BECs in the presence of 20mM and 30mM NAC, respectively Scale bar: 50 μM. (F) Mean fluorescence intensity of different channels through an infected BEC. Tukey’s multiple comparisons test was used for statistical analysis. **p<0.01, ***p<0.001. For (A), data represent the mean ± SD of n = 3 biological replicates.