Uricase (urate oxidase; EC 1.7.3.3) is an enzyme involved in purine metabolism, which catalyzes the conversion of uric acid (UA) into allantoin with concomitant reduction of molecular oxygen to hydrogen peroxide (Kahn and Tipton, 1998). It is expressed in a wide range of organisms, including bacteria, fungi, plants, and animals (Kahn and Tipton, 1998). In humans, uricase is not functional due to multiple evolutionary events, resulting in 3- to 10-fold higher serum UA levels compared to other mammals (Kratzer et al., 2014) and predisposing humans to hyperuricemia. Hyperuricemia is a metabolic disturbance affecting ~20% of the United States population (Zhu et al., 2011) and causes gout, chronic kidney disorders, and cardiovascular diseases (Wu et al., 1994; Culleton et al., 1999). Hyperuricemia is exacerbated by western diet (Siener and Hesse, 2003) or other diets rich in purines (Brulé et al., 1992), thus therapeutic strategies often include nutritional and lifestyle interventions though evidence for their efficacy is scarce (Moi et al., 2013). On the other hand, UA-lowering drugs (e.g., xanthine oxidase inhibitors such as allopurinol and febuxostat) are restricted to patients with severe or recurrent gout (Engel et al., 2017), as they exhibit potential adverse effects (Pacher et al., 2006) and limited long-term efficacy (Pérez-Ruiz et al., 2019). Therefore, novel approaches are desired.

The gastrointestinal tract plays an important role in UA metabolism and excretion (Méndez-Salazar and Martínez-Nava, 2021). In healthy people, UA is primarily cleared via the kidneys, though ~30% is excreted via the intestine (Sorensen and Levinson, 1975) where it can be further degraded by the gut microbiota (Chu et al., 2021). In line, recent studies have shown that the gut microbial community composition is significantly altered in rat models of hyperuricemia (Liu et al., 2020; Pan et al., 2020) or in patients with gout (Chu et al., 2021; Lin et al., 2021). The gastrointestinal tract may therefore constitute a promising therapeutic target for hyperuricemia, for instance using oral probiotic strains, a concept supported by recent in vivo studies with microbial uricase (Szczurek et al., 2017) or uricolytic bacterial strains (García-Arroyo et al., 2018; Wu et al., 2021) that showed reduced serum UA levels in hyperuricemic rodent models.

To date, uricase activity was biochemically confirmed in various bacterial taxa, either extracellularly in Pseudomonas strains (Abdel-Fattah et al., 2005; Jagadeesan et al., 2019), or intracellularly in Arthrobacter (Suzuki et al., 2004), Microbacterium (Zhou et al., 2005; Kai et al., 2008), Streptomyces (Watanabe et al., 1969), Saccharopolyspora (Khucharoenphaisan and Sinma, 2011), Micrococcus (Olivieri et al., 1983), Metabacillus (Zhang et al., 2010), and Bacillus (Mahler, 1970; Bongaerts and Vogels, 1976; Bongaerts et al., 1978; Huang and Wu, 2004; Lee et al., 2005; Lotfy, 2008) strains. In few lactobacilli strains, uricase activity was suggested to be present both intracellularly and extracellularly (Handayani et al., 2018), thus prompting further investigation of uricase in taxa associated with probiotic strains. The determination of bacterial uricase activity can be assessed either qualitatively, via the observation of UA consumption on UA-containing solid media (Shaaban et al., 2015; Pustake et al., 2019), or quantitatively, via the measurement of H₂O₂ formation using fluorescent dyes (Fraisse et al., 2002) or the determination of UA degradation using spectrophotometry (Koyama et al., 1996; Fraisse et al., 2002; Huang and Wu, 2004) or liquid chromatography (Safranow et al., 2000). Because of their simplicity and compatibility with microplate systems allowing parallel analysis of multiple samples, spectrophotometric and fluorescent techniques appear promising to develop a high-throughput screening assay for bacterial uricase activity. However, their non-specific nature may be source of analytical artefacts, especially in presence of bacterial cells or complex media such as those used for the growth of probiotic strains.

Here, we evaluated both spectrophotometric and fluorescent uricase activity assays and highlighted abiotic and biotic factors potentially impacting the measurements’ accuracy and bias. We reported on the stepwise optimization of a simple, fast, and robust high-throughput UV-spectrophotometric method to screen bacterial uricase activity using whole bacterial suspension, thereby assessing both cell-associated and extracellular uricase. Finally, a panel of 319 QPS (Qualified Presumption of Safety) strains of lactobacilli (nine genera), Bacillus, and Bifidobacterium, respectively representing 18, 3, and 4 species, were screened for uricase activity using the optimized UV-spectrophotometric method.

In this work, we evaluated two quantitative methods to develop a high-throughput screening approach of bacterial uricase activity in microtiter plates, with particular focus on conditions relevant to lactobacilli, Bacillus, and Bifidobacterium strains. Using a stepwise approach, we aimed at developing a sensitive, yet robust method using whole bacterial suspensions to enable the simultaneous detection of both extracellular and cell-associated uricase activities, thereby accelerating the screening of large microbial collections and the identification of uricolytic strains independently from the enzyme’s partition (Watanabe et al., 1969; Olivieri et al., 1983; Suzuki et al., 2004; Kai et al., 2008; Lotfy, 2008; Zhang et al., 2010; Khucharoenphaisan and Sinma, 2011; Handayani et al., 2018; Jagadeesan et al., 2019).

Our data revealed that the fluorescent assay, based on the quantification of H₂O₂ released during UA degradation, was the least suitable method because of its susceptibility to biases derived from both abiotic and biotic factors. In absence of bacteria, many media exhibited strong background fluorescence intensities (=false positive), as exemplified by the fluorescence rates of MRS#2-Aut and RCM-SF which corresponded to ~20% of the fluorescence rate of the uricase positive control Bacillus sp. DSM 1306 (Supplementary Table S2, Figure 3). This abiotically-generated fluorescence might have arisen from the spontaneous generation of H₂O₂ as byproduct from the reactions between sugars, phosphate salts and proteinaceous components (e.g., tryptone, peptone, and yeast extract), especially during heat treatment (Finkelstein and Lankford, 1957; Carlsson et al., 1978; Nakashima et al., 2010). More importantly, the fluorescent assay was prone to false positive signals in presence of bacterial cells (Figure 3). As previous studies showed that H₂O₂ can be produced by various Bifidobacterium (Kawasaki et al., 2009) and lactobacilli (Pridmore et al., 2008; Martín and Suárez, 2010; Hertzberger et al., 2014) strains, the H₂O₂-based fluorescent assay appears unsuitable to screen uricase activity using whole bacterial suspensions.

In contrast, the UV-spectrophotometric method evaluated and optimized here can be used for the high-throughput screening of bacterial uricase activity. By monitoring UA degradation at 300 nm, we maximized the absorbance of UA and minimized the background absorbance of most tested complex media (Figure 1), while enabling the use of standard polystyrene 96-well plates (Supplementary Figure S1, Figure 4). To ensure the sensitivity of the spectrophotometric measurements, medium selection and preparation should be carefully considered though. Thermal sterilization by autoclaving consistently increased the background absorbance of most media (Figure 1), probably due to the reaction between reducing sugars (e.g., glucose) and proteinaceous components resulting in Maillard reaction products (Hemmler et al., 2017) that absorbed in the UV–Vis wavelengths range (Wang et al., 2016; Liu et al., 2018). The absence or low concentration of glucose in WC, ½WC, AM, and LB (Supplementary File 1) could explain the minor variations of absorbance observed between these media when autoclaved or filter sterilized (Figure 1). It is worth noting that autoclaved MRS, a media heavily used in laboratories to grow lactobacilli strains, were highly absorbing at UV wavelengths, thus hindering its compatibility with UV-spectrophotometric assays in general (Figure 1); nonetheless, those effects can be attenuated by using a diluted and sterile filtered version of this medium (Supplementary File 1).

The screening of 319 QPS strains of lactobacilli (nine genera), Bacillus, and Bifidobacterium using the optimized UV-spectrophotometric assay did not result in the identification of uricase positive strains, except for the positive control Bacillus sp. DSM 1306. Although we cannot fully discard the possibility of few false negative results from the high-throughput screening, previous reports on the regulation of bacterial uricase activity support the conditions tested here. Nitrogen- or carbon-limitation was shown to stimulate uricase activity in resting cells of Streptomyces spp. (Watanabe et al., 1976), thereby validating the use of protein-depleted media to grow and assess bacterial uricase activity (i.e., ½MRS#1-SF, AM-Aut, and ½WC-SF; Supplementary File 1). In various Bacillus species, UA was shown to serve as a nitrogen source (Bongaerts and Vogels, 1976), and while no UA was supplemented in the culture media of the 319 tested strains, cells were grown in protein-depleted media, harvested in the stationary growth phase, and ultimately exposed to 0.5 mM UA for 60 min during uricase activity measurement (Figure 4). Besides, uricase activity was not promoted by the direct supplementation of UA in the growth media of four randomly tested strains, and the positive control Bacillus sp. DSM 1306 (Figure 3).

The data obtained from the 319 screened strains, representing a total of 11 genera and 25 species with several strains of the same species (Supplementary Table S1), strongly suggest that uricase activity is a rare feature in lactobacilli, Bacillus, and Bifidobacterium species, especially in strains isolated from fermented food products or from the gastrointestinal tract. Beside previous reports on uricase-positive Bacillus strains from soil samples (Mahler, 1970; Bongaerts and Vogels, 1976; Huang and Wu, 2004; Lotfy, 2008), no study have yet reported the presence of uricase in Bifidobacterium, and only two studies reported uricase activity in lactobacilli food isolates, i.e., Limosilactobacillus fermentum (Wu et al., 2021) and Lactiplantibacillus plantarum (Handayani et al., 2018). Considering that 40 L. fermentum and 29 L. plantarum strains (Supplementary Table S1) were tested uricase-negative in our screening, and that no uricase-encoding gene has yet been identified in lactobacilli genomes (in contrast to the uricase-encoding genes in Bacillus; Schultz et al., 2001), we question whether previous reports on lactobacilli uricase activity (Handayani et al., 2018; Wu et al., 2021) are the results of potential methodological biases as highlighted here for a broad panel of tested strains from 11 genera, or a less likely highly strain-specific feature.

Altogether, the whole bacterial suspension screening method developed here provides various advantages compared to the qualitative agar plate-based assay, as it is quantitative, specific, and scalable. Our approach can serve as a guide for the assessment of uricase activity in large and taxonomically diverse microbial collection, and therefore potentially supporting the identification of novel uricolytic strains with therapeutic potential for the treatment of hyperuricemia and associated comorbidities.

The original contributions presented in the study are included in the article/Supplementary Material; further inquiries can be directed to the corresponding author.

BP, SP, DM, and CL designed the experiments. RN provided critical feedback on the study design. SP and DM performed the experiments. BP, SP, and DM performed data analysis. BP and CL wrote the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by a grant from the Swiss Innovation Agency (40722.1 IP-LS), in collaboration with Beo Therapeutics AG. The research was conducted in the absence of any commercial relationships that could have influenced or biased the work presented here.

RN works for the company Beo Therapeutics AG.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.