A Novel Colorimetric Biosensor for the Detection of Catalase-Positive Staphylococcus aureus Based on an Onion-like Carbon Nanozyme

Abstract

Staphylococcus aureus is one of the leading causes of skin and soft tissue infections, and it is even life-threatening if it enters the bloodstream, lung or heart. In the present work, we proposed a novel colorimetric biosensor for the detection of S. aureus through hydrogen peroxide consumption. An onion-like carbon nanozyme with high peroxidase-like activity was prepared, which competed with the endogenous catalase of S. aureus in consuming hydrogen peroxide. This reaction was further characterized by the colorimetric reaction of 3,3′,5,5′-tetramethylbenzidine. The results showed that our approach allowed for the simple and rapid determination of S. aureus, with a linear range of 2 × 10⁴ to 2 × 10⁷ CFU/mL. Moreover, our method displayed good selectivity, with Bacillus subtilis and Escherichia coli showing negligible responses at the concentration of 2 × 10⁵ CFU/mL. The application of the as-prepared biosensor to analyze S. aureus in real water samples yielded recovery rates ranging from 95% to 112%, with relative standard deviations less than 7%. The method demonstrated good accuracy and specificity, which offers a novel approach for the simple and selective detection of S. aureus.

1. Introduction

Bacterial infections are among the leading causes of illness and death, which presents a growing risk to public health. Staphylococcus aureus is among the most prevalent pathogens, causing a spectrum of infections including severe inflammation, skin ulceration, and suppuration [1,2,3], as well as more serious conditions such as endocarditis [4], pneumonia [5], and meningitis [6]. Despite the fact that regulations have been enacted to restrict S. aureus contamination, persistent risks remain in environments such as medical facilities [7]. Moreover, the presence of food-borne pathogens in meat and meat products poses a significant risk of infection to consumers, with livestock-associated workers at heightened risk of colonization and potential transmission [8,9]. Therefore, there is an urgent need to develop rapid and efficient techniques capable of detecting S. aureus and promptly identifying potential outbreaks.

The plate-counting approach, which has been the standard method for S. aureus detection, is a conventional culture-dependent microbiological method that is used to observe and enumerate viable microorganisms. In this method, bacterial concentration is reported as colony-forming units (CFUs), which is presumed to originate from a certain microorganism, and the bacterial colonies can also reflect specific characteristic features, such as color, size and shape. However, it usually requires several days to achieve trustworthy results, which no longer meets the demands for rapid pathogenic detection [10]. Consequently, considerable efforts have been directed towards the development of rapid detection methods. Polymerase chain reaction (PCR) and fluorescence-linked immunosorbent assay (FLISA) have significantly reduced the detection time to a few hours [11,12]. Biosensors are capable of converting signals arising from interactions between biological components and analytes into detectable physicochemical signals, which have been extensively used in applications ranging from medical treatment to environmental monitoring. Among them, colorimetric biosensors have emerged as a promising platform for rapid bacterial recognition [13,14]. These biosensors enable the quantitative measurement of specific analytes through chromogenic or discoloration reactions, such as nanozyme-mediated colorimetric reactions or the localized surface plasmon resonance of noble metal nanoparticles. Their merits of cost-effectiveness, straightforward interpretation of results, and rapid response render them well-suited for the on-site detection of S. aureus [15,16,17].

S. aureus is one of the representative Gram-positive bacteria, with most strains being catalase (CAT)-positive and capable of catalyzing the decomposition of hydrogen peroxide (H₂O₂) into O₂ and water [18,19]. Catalase production represents a defense mechanism, allowing bacteria to better resist intracellular and extracellular killing by H₂O₂. The catalase-negative strain is observed to possess much lower pathogenic potential than the catalase-positive strain, which may contribute to the predominance of the catalase-positive strain in bacterial infections. Utilizing the endogenous catalase of S. aureus, Guarín et al. proposed an electrochemical sensor to monitor H₂O₂ consumption by S. aureus. Using a screen-printed gold electrode modified with cysteine and peroxidase (POD), this method demonstrated a sensitive response to H₂O₂, enabling the detection of S. aureus ranging from 3 × 10² to 3 × 10⁸ CFU/mL, with a detection limit of 10² CFU/mL [20]. Majumdar et al. applied an amperometric approach to detect H₂O₂ consumption for the quantification of S. aureus at a Pt microelectrode. This work realized the quantification of S. aureus in the range from 10 CFU/mL to 10⁶ CFU/mL within 10 min [21]. Abdelhamid et al. reported a fluorescent biosensor based on chitosan modified CdS quantum dots, in which S. aureus showed accelerated decomposition of H₂O₂, resulting in a significant quenching of fluorescence emission, thereby offering high selectivity for CAT-positive S. aureus over Bacillus subtilis and Escherichia coli [22]. In comparison to other assays for the detection of S. aureus, these three biosensors offer the benefits of short detection periods and broad detection ranges. Also, the sensor preparation process is simplified as no additional recognition units are required.

Nanozymes, defined as nanomaterials with enzyme-like catalytical activity, have attracted increasing attention in biosensing applications over the past few years [23,24,25]. These nanozyme-based enzymatic reactions can generate or amplify signals for precise target detection, among which POD-like and oxidase-like nanozymes are commonly employed in biosensor construction. POD is defined as a class of oxidoreductases that catalyze the oxidation of a substrate by hydrogen peroxide or an organic peroxide [26]. Nanozymes that exhibit mimic peroxidase activity are known as POD-like nanozymes [27]. In addition to catalytic properties, nanomaterials offer enhanced electrical conductivity and compatibility with biorecognition elements, thereby improving biosensor performance. Onion-like carbon (OLC) is a carbonaceous nanomaterial with stacked layers of sp² graphene sheets and exhibits high POD catalytic ability [28,29,30,31,32,33,34]. Its highly symmetric structure and large volume-to-surface ratio make it promising for applications in energy storage and pollutant degradation [35,36]. In the field of biosensing, OLC can be utilized as a modification material to increase the conductivity and biocompatibility of the electrodes [37,38,39,40]. OLC can serve as an effective linker between biomolecules and nanomaterials, facilitating electron transfer and signal amplification. Therefore, in combination with the intrinsic POD-like property of OLC, it is promising in the construction of biosensors with high sensitivity and specificity.

Herein, we proposed a novel approach for S. aureus detection utilizing its endogenous CAT activity. The accelerated consumption of H₂O₂ catalyzed by CAT was characterized using a colorimetric reaction involving 3,3′,5,5′-tetramethylbenzidine (TMB) in the presence of OLC. The analytical performance and anti-interference ability of the as-prepared colorimetric biosensor were examined. Additionally, the recovery experiments were conducted using real water samples to assess the feasibility of the present method in practical scenarios. The present work provided a simple and rapid approach for the detection of CAT-positive bacteria, which can be used in pathogen monitoring and environmental early warning.

2. Materials and Methods

2.1. Reagents and Materials

Staphylococcus aureus (ATCC 6538), Escherichia coli (ATCC 25922) and Bacillus subtilis (CGMCC 1.1086) were provided by the Testing Center for Antimicrobial Materials, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences. Additionally, 3,3′,5,5′-tetramethylbenzidine (TMB), acetic acid (>99.7%) and dimethyl sulfoxide (DMSO) were provided by Macklin Co., Ltd. (Shanghai, China). Sodium acetate and 2,2′-azinobis(3-ethylbenzothiazoline-6-sulfonic acid ammonium salt) (ABTS) were provided by InnoChem Co., Ltd. (Shanghai, China). Hydrogen peroxide (H₂O₂) was purchased from Aladdin Co., Ltd. (Shanghai, China). Lysostaphin was provided by Sangon Co., Ltd. (Shanghai, China). Nutrient broth and nutrient agar were purchased from Haibo Co., Ltd. (Qingdao, China). Nanodiamonds (NDs) were purchased from Carbodeon (Finland). Phosphate-buffered saline (PBS, 0.1 M, pH = 7) and Tris-HCl buffer (0.05 M, pH = 7.4, sterile) were provided by Regan Co., Ltd. (Beijing, China). All chemicals were of reagent grade and used as received. Aqueous solutions were prepared with Milli-Q water (Millipore Merck, Burlington, MA, USA, R > 18.2 MΩ·cm). Real water samples were collected from the tap-water pipe.

2.2. Preparation of Onion-like Carbon Nanozyme

OLC was obtained by annealing NDs in an N₂ atmosphere at 800 °C for 1 h. The morphology of as-prepared OLC was characterized by high-resolution transmission electron microscopy (TEM, JEOL JEM-2100F, Tokyo, Japan). The phase structure of OLC was obtained using an X-ray diffractometer (Bruker D8 focus, Karlsruhe, Germany). The composition of OLC was characterized using a Raman spectrometer (Renishaw InVia-Qontor, New Mills, UK) with an excitation light at 532 nm.

2.3. Detection of the POD-like Catalytic Activity of Onion-like Carbon Nanozyme

Chromogenic reactions based on TMB-H₂O₂ and ABTS-H₂O₂ double-substrate systems were used to detect the POD-like properties of the OLC nanozyme. Experiments on the TMB-H₂O₂ reaction were carried out with 250 μg/mL of OLC, 1 mM of H₂O₂ and 0.5 mM of TMB in 0.2 M NaAc-HAc buffer (pH = 4). The mixed solution was incubated at 37 °C for 10 min. Experiments on the ABTS-H₂O₂ reaction were performed in 0.2 M NaAc-HAc buffer (pH = 4) containing 250 μg/mL of OLC, 1 mM of H₂O₂ and 1 mM of ABTS. The mixed solution was placed at 37 °C for 10 min.

2.4. Catalytic Kinetic Determination of OLC’s POD-like Activity

The catalytic kinetic determination of OLC was carried out using TMB and H₂O₂ as substrates. The catalytic reaction was carried out in 0.2 M NaAc-HAc buffer (pH = 4). The concentration of OLC was 100 μg/mL. The concentration of TMB was 1 mM in the kinetic assay of H₂O₂ and the concentration of H₂O₂ was 1 mM in the kinetic assay of TMB. After reacting at room temperature for 10 min, the reaction mixture was filtered by a needle filter (0.22 μm × 13 mm), and the absorbance of the supernatant at 652 nm was measured on an ultraviolet-visible spectrophotometer (SECOMAM UVIKONXL). The Michaelis–Menten equation was used to calculate the Michaelis constant (K_M) and maximum reaction rate (V_max).

2.5. Bacterial Culture

The bacteria were cultured in LB broth that contained 10 g/L peptone, 3 g/L beef extract and 5 g/L NaCl. The broth was sterilized by autoclaving at 121 °C for 15 min. After cooling to room temperature, the bacteria were inoculated into the sterilized culture medium and cultured aerobically in a shaker at 37 °C. The culture time for S. aureus, E. coli and B. subtilis were 24 h, 16 h and 24 h, respectively. Subsequently, the microbial cells were harvested by centrifugation at 6000 rpm for 5 min and washed twice with PBS. The harvested bacteria were re-dispersed in Tris-HCl buffer, and the OD₆₀₀ value of the bacterial solution was adjusted to 1.5 using an ultraviolet-visible spectrophotometer (SECOMAM UVIKONXL).

The concentration of bacteria was further determined by the plate-counting method according to the following procedure: First, the agar liquid medium that contained 10 g/L of peptone, 5 g/L of sodium chloride, 3 g/L of beef extract and 15 g/L of agar was well mixed and sterilized at 121 °C. After cooling to 46 °C, the liquid agar medium was poured onto the culture dishes and allowed to solidify at room temperature. Then, 100 μL of bacteria solution was added to the agar plate and spread evenly using a cell spreader. The plate was then flipped over and incubated for 24 h at 37 °C. After incubating, a colony counter was used to read the total number of bacterial colonies.

2.6. Colorimetric Detection of S. aureus

The detection principle of the colorimetric biosensor is illustrated in Scheme 1. To enhance detection specificity, lysostaphin was employed, as it selectively hydrolyzed the cell wall of Staphylococcus species [41,42]. Following incubation with lysostaphin, CAT was released from S. aureus and catalyzed the decomposition of H₂O₂. With the increase in S. aureus concentration, CAT concentration in the bacterial lysate increased, leading to the accelerated consumption of H₂O₂. As a result, less H₂O₂ was involved in the subsequent TMB oxidation, resulting in a decrease in the final absorbance of TMB oxide and an increase in the inhibition rate.

The detailed detection procedure was as follows: lysostaphin (13 μg/mL) was mixed with S. aureus solution to lyse S. aureus and release endogenous CAT. At the same time, H₂O₂ (0.5 mM) was added and consumed under the catalysis of CAT. The reaction was carried out at 37 °C for 20 min. After incubation, TMB (1 mM), OLC (250 μg/mL) and NaAc-HAc buffer (0.2 M, pH = 4) were added to the above solution. The reaction mixture was placed at room temperature for 10 min. Then, the solution was filtered with a needle filter of 0.22 μm × 13 mm to remove the bacteria and OLC nanozyme. Then, the absorbance of the supernatant at 652 nm was tested and recorded as A_x. In the blank group, the S. aureus solution was replaced with Tris-HCl buffer solution while all other conditions remained the same. The absorbance value of the blank group acted as the internal reference and was recorded as A₀. The inhibition rate was used to characterize the bacterial concentration:

Inhibition rate (%) = (1 − A_x/A₀) × 100%

(1)

Compared with the absolute value of absorbance, the use of the inhibition rate eliminated the influence of subjective errors and fluctuations in nanozyme activity on detection accuracy, which improved the precision and reliability of the results.

3. Results and Discussion

3.1. Structural and Catalytic Activity Characterization of OLC Nanozyme

OLC was obtained by annealing NDs at high temperature under an inert atmosphere. As shown in Figure 1a, the TEM image showed the presence of (111) planes of sp³-hybridized cubic diamond and (002) planes of sp²-hybridized graphitized carbon, with lattice spacing of 0.206 and 0.34 nm, respectively. OLC exhibited a quasi-spherical shape with a diameter of 5–10 nm. The X-ray diffraction pattern showed broad peaks of graphitic carbon at 27° and diffraction peaks of diamond at 44° and 75°, suggesting the presence of both sp² and sp³ carbon in OLC (Figure 1b). We also observed peaks in the Raman spectra at 1320 cm⁻¹ and 1535 cm⁻¹, which belong to the D and G bands of graphene (Figure 1c). The above characterization confirmed the partial conversion from sp³ to sp² carbon in pristine NDs, suggesting the successful synthesis of OLC. The abovementioned structural characterization in the present work is also consistent with our previous works, which can verify the formation of OLC [35,36].

The POD-like property of OLC was validated by the chromogenic reactions with TMB/ABTS and H₂O₂. Under the catalysis of OLC, the •OH was generated from the decomposition of H₂O₂, which subsequently oxidized TMB/ABTS into charge-transfer complex TMB_OX (blue)/ABTS^•+ (green) (Equations (2) and (3)). As shown in Figure 2, in the presence of H₂O₂ and OLC, TMB and ABTS underwent oxidation to produce blue and green products, respectively. However, the experimental groups in the absence of OLC showed almost no color change, indicating that OLC acted as an effective catalyst for the oxidation of TMB and ABTS in the presence of H₂O₂. Considering that TMB exhibits greater sensitivity in biosensing when compared to other chromogenic substrates, including ABTS [43,44], TMB was chosen as the substrate for subsequent detection.

$$\mathrm{TMB}+{\mathrm{H}}_2{\mathrm{O}}_2 \stackrel{\mathrm{O}\mathrm{L}\mathrm{C}}{\to }{\mathrm{TMB}}_{\mathrm{OX}}+{\mathrm{H}}_2\mathrm{O}$$

(2)

$$\mathrm{ABTS}+{\mathrm{H}}_2{\mathrm{O}}_2 \stackrel{\mathrm{O}\mathrm{L}\mathrm{C}}{\to }{\mathrm{ABTS}}^{•+}+{\mathrm{H}}_2\mathrm{O}$$

(3)

The POD-mimicking activity of the OLC nanozyme was further characterized according to the Michaelis–Menten equation (Equation (4)) and the Lineweaver–Burk equation (Equation (5)), which reflect the relationship between reaction rate and substrate concentration:

$$v={\displaystyle \frac{V_{max}\left[S\right]}{K_M+\left[S\right]}}$$

(4)

$${\displaystyle \frac{1}{v}}={\displaystyle \frac{K_M}{V_{max}}}{\displaystyle \frac{1}{\left[S\right]}}+{\displaystyle \frac{1}{V_{max}}}$$

(5)

$$v={\displaystyle \frac{\Delta c}{t}}$$

(6)

$$A=\epsilon Lc$$

(7)

where [S] is the concentration of the substrate TMB or H₂O₂, $v$ is the reaction rate, and V_max refers to the maximum reaction rate when the enzyme is saturated by the substrate. The reaction rate v is determined by Equation (6), where $\Delta c$ is the concentration change of the chromogenic substance during the reaction, and t is the reaction time. The concentration of the chromogenic substance can be calculated from the absorbance value according to Lambert–Beer law (Equation (7)). A refers to the absorbance value measured by the UV-vis spectrophotometer. ε is the molar absorption coefficient. The ε value for TMB is 39,000 M⁻¹ cm⁻¹. c is the concentration of the chromogenic substance, and L is the optical path length. By calculating the $\Delta c$ values based on the changes of absorbance intensity, the reaction rates of TMB can be determined. K_M is the Michaelis constant, representing the substrate concentration at which the reaction rate reaches half of V_max. A smaller K_M indicates a higher affinity between the enzyme and the substrate. Figure 3 shows the Michaelis–Menten and Lineweaver–Burk curves of OLC. V_max and K_M were determined from the fitted straight line between 1/v and 1/[S]. The intercept of the line refers to 1/V_max. Therefore, V_max can be obtained from the intercept. The slope refers to K_M/V_max. The value of K_M can be calculated from the slope of the fitted line and V_max. The K_M and V_max of both substrates were calculated and are listed in

Table 1. Comparisons of kinetic parameters between OLC and other carbon nanozymes.

Nanozyme	KM (mM)		Vmax (10−8 M s−1)		Reference
	H2O2	TMB	H2O2	TMB
graphdiyne oxide	2.59	0.62	1.77	1.92	[45]
C60[C(COOH)2]2	24.58	0.23	0.40	0.35	[46]
GO-COOH	3.99	0.02	3.85	3.45	[47]
OLC	0.77	0.05	7.48	3.25	This work

. OLC exhibited a higher V_max when compared to other carbon nanomaterials with POD-like properties [45,46,47]. Furthermore, a smaller K_M value was observed for OLC compared with graphdiyne oxide and C₆₀[C(COOH)₂]₂, indicating that OLC had a high affinity to H₂O₂ and exhibited high POD-like catalytic activity.

3.2. Feasibility Verification of the As-Prepared Biosensor for S. aureus Detection

CAT is a biological enzyme capable of catalyzing the decomposition of H₂O₂ into water and oxygen [48,49], thereby playing a crucial role in protecting cells from oxidative damage induced by reactive oxygen species [50,51]. Firstly, the presence of CAT in S. aureus (ATCC 6538) cells was confirmed by incubating 10⁸ CFU/mL S. aureus with 5 mM H₂O₂ at 37 °C for 30 min. Clear bubbles were observed in the solution (Figure 4a), corresponding to the generation of O₂ under the catalysis of bacterial endogenous CAT.

Furthermore, to verify the feasibility of the as-prepared biosensor for S. aureus detection, the colorimetric responses of S. aureus solutions at different concentrations were recorded using TMB as the chromogenic agent. The bacteria were first incubated with 5 mM H₂O₂ for 30 min. Then, the bacteria were removed by filtration to avoid interfering with the subsequent absorbance detection. After filtration, OLC and TMB were added, and the colorimetric changes were observed after 5 min. In the presence of H₂O₂ and nanozymes, TMB easily loses one electron and produces a charge-transfer complex that displays a blue color. When TMB is exposed to an acidic condition or the amount of TMB is insufficient, it can then undergo further oxidation to produce a yellow diimine compound [52]. In present work, the TMB amount we used was sufficient, and the blue-colored reaction product TMB_ox was observed. As shown in Figure 4b, the intensity of the blue color decreased with increasing concentrations of S. aureus, owing to the higher consumption of H₂O₂ by S. aureus and less H₂O₂ being involved in subsequent TMB oxidation. In summary, the method proposed in the present work is feasible for S. aureus detection according to the colorimetric change of TMB.

To demonstrate the superior performance of the OLC nanozyme in our detection system, control experiments were performed using nanodiamonds and Fe₃O₄ nanoparticles as representative carbonaceous and non-carbonaceous nanomaterials, respectively. The control experiments were conducted at a S. aureus concentration of 10⁵ CFU/mL. The detection procedures and parameters were consistent with those described in Section 2.6. The results are presented in

Table 2. Results of different materials in S. aureus detection.

Materials	OLC	Nanodiamonds	Fe3O4 Nanoparticles
Absorbance of experimental group	1.18	0.13	0.45
Absorbance of blank group	1.45	0.14	0.53
Inhibition rates	18.6%	7.1%	15.1%

. Under the same concentration of nanozymes (250 μg/mL), the highest change in TMB absorbance intensity was observed for OLC, indicating that OLC possessed the strongest POD-like properties. Additionally, OLC also demonstrated the highest inhibition rate, reflecting its enhanced sensitivity for S. aureus detection.

3.3. Optimization of Detection Conditions

3.3.1. Optimization of the Incubation Time of Lysostaphin and S. aureus (t₁)

To enhance the detection efficacy of the as-prepared biosensor, experimental conditions were systematically optimized. Firstly, considering the potential interference posed by the bacterial cell wall on the interaction between CAT within the S. aureus cells and H₂O₂ in the solution, lysostaphin was employed to facilitate bacterial lysis and ensure the release and adequate reaction between endogenous CAT and H₂O₂. Evaluation revealed that at a concentration of 10⁶ CFU/mL S. aureus, lysostaphin treatment resulted in an inhibition rate of 49.11%, significantly higher than the 19.84% inhibition rate observed without lysostaphin, which proved the sensitivity enhancement facilitated by lysostaphin.

In the detection process, S. aureus was first incubated with lysostaphin to proceed bacterial lysis and release endogenous CAT, after which H₂O₂ was added. Considering a high concentration of H₂O₂ may hinder the lytic activity of lysostaphin, the incubation time of lysostaphin was first optimized by using 2 × 10⁵ CFU/mL S. aureus as the target analyte (Figure 5). The increase in t₁ from 0 to 15 min did not yield significant changes in the inhibition rate, suggesting a rapid and complete lysis of S. aureus in a short period of time. Given that the inhibition rate was not sensitive to t₁, lysostaphin and H₂O₂ were introduced to S. aureus simultaneously to simplify the detection process in the subsequent experiments.

3.3.2. Optimization of H₂O₂ Concentration and H₂O₂ Reaction Time (t₂)

Considering that the determination of S. aureus concentration relies on the enzymatic consumption of H₂O₂ by endogenous CAT, the initial concentration of H₂O₂ is essential in achieving optimal detection sensitivity. As depicted in Figure 6a, the inhibition rate exhibited an upward trend as the H₂O₂ concentration varied from 0.1 to 0.5 mM. However, the response of the biosensor to S. aureus exhibited a diminishing trend as the H₂O₂ concentration increased from 0.5 to 1 mM, with the maximum detection sensitivity observed at 0.5 mM. Consequently, 0.5 mM was selected as the optimal concentration.

Moreover, the influence of H₂O₂ reaction time was also investigated. As shown in Figure 6b, the inhibition rate increased, with the prolonged reaction time ranging from 5 to 20 min, and reached a plateau after 20 min, indicating that 20 min was sufficient for a complete reaction. Hence, 20 min was selected as the optimal reaction time for subsequent experiments.

3.4. Application of the As-Prepared Colorimetric Biosensor for S. aureus Detection

Under optimized conditions, the application of the as-prepared colorimetric biosensor for S. aureus detection was assessed. As depicted in Figure 7a, the final absorbance intensity decreased as the concentration of S. aureus was elevated. The relationship between the inhibition rate and the logarithm of bacterial concentration was fitted and is depicted in Figure 7b, revealing a strong linear correlation within the concentration range of 2 × 10⁴ to 2 × 10⁷ CFU/mL (R² = 0.983). Since the sensitivity of our method relies on the measurement accuracy of the UV-vis spectrophotometer, we determined that our detection limit was reached when the absorbance intensity difference between the experimental and blank groups was less than 0.02. The limit of detection (LOD) was calculated as 3 S_b/m, where m represents the slope at low bacterial concentrations and S_b is the standard deviation of the blank group. The LOD value of our method was 9.2 × 10³ CFU/mL.

3.5. Assessment of Anti-Interference Ability

Two other types of pathogenic bacteria, Gram-positive B. subtilis and Gram-negative E. coli, were selected to evaluate the anti-interference capability of the present method. The detection of other pathogenic bacteria followed the same procedure. The colorimetric results showed an inhibition rate of 28.76% for S. aureus at the concentration of 2 × 10⁵ CFU/mL, while the E. coli and B. subtilis under the same concentration exhibited nearly no response (Figure 8). The good selectivity of the present method is attributed to two key factors. Firstly, the strain of S. aureus is CAT-positive, so it exhibits specific CAT catalytic activity that is different to other bacteria strains [22]. Secondly, the S. aureus-specific lysostaphin can selectively lyse the cell wall of S. aureus without affecting other bacteria. Together with the specific enzymatic activity of target S. aureus and the selective cell processing method, the as-prepared biosensor exhibited good anti-interference ability against other pathogenic bacteria.

3.6. Detection of S. aureus in Real Water Samples

3.6.1. Single Bacterial Test

Recovery experiments in real water samples were conducted to confirm the feasibility of the present method for S. aureus detection. Three samples with varying S. aureus concentrations were added to the collected real water sample, and the colorimetric assay was performed. The inhibition rates of real water samples spiked with different concentrations of S. aureus were calculated, and the detected bacterial concentration was converted according to the fitting relationship in Figure 7b. Real sample results for S. aureus are shown in

Table 3. Results of S. aureus detection in real water samples.

Spiked Concentration of S. aureus (CFU/mL)	Inhibition Rate	Found Concentration (CFU/mL)	Recovery Rate	RSD
5 × 104	9.14%	5.16 × 104	103.20%	2.71%
2 × 105	27.16%	2.06 × 105	103.00%	6.45%
2 × 107	88.17%	2.24 × 107	112.00%	3.94%

. The recovery rates between the measured concentration and the spiked S. aureus concentration demonstrated the good accuracy of the present method in real water detection.

The practical anti-interference performance of our method was also evaluated by detecting the real water samples spiked with B. subtilis or E. coli at a concentration of 2 × 10⁵ CFU/mL, respectively. As listed in

Table 4. Results of 2 × 10⁵ CFU/mL E. coli or B. subtilis detection in real water samples.

	Inhibition Rate	RSD
E. coli	3.74%	0.97%
B. subtilis	1.25%	0.55%

, negligible responses were observed for samples containing B. subtilis and E. coli. The results proved the effectiveness of our approach in detecting CAT-positive bacteria such as S. aureus, thereby offering a promising solution for rapid pathogen detection.

3.6.2. Mixed Bacterial Test

Natural water bodies often contain a wide range of bacteria, which potentially complicates the detection of target S. aureus due to interferences among different species. To assess the robustness of our assay under such conditions, we added the mixtures of S. aureus, E. coli and B. subtilis to the real water samples with each at a concentration of 2 × 10⁵ CFU/mL, and the found concentrations of S. aureus were determined. The found S. aureus concentrations in the mixed bacterial samples, as well as recovery rates and RSDs, are listed in

Table 5. Results of the mixed bacterial test in real water samples.

Spiked Samples	Inhibition Rate	Found Concentration (CFU/mL)	Recovery Rate	RSD
S. aureus + E. coli	26.93%	2.03 × 105	101.50%	4.94%
S. aureus + B. subtilis	26.20%	1.92 × 105	96.00%	2.42%
S. aureus + E. coli + B. subtilis	26.14%	1.91 × 105	95.50%	4.00%

. The results indicated that the presence of E. coli and B. subtilis had a minor effect on the detection of S. aureus, with recovery rates ranging from 95.5% to 101.5%. The above experiments demonstrated the capability of the present method in effectively mitigating interferences from co-existing bacteria. However, given that our method determines the S. aureus concentration by measuring the H₂O₂ consumption, it is plausible that substances that can react with H₂O₂ may interfere with the detection results. This could include sulfur ions and iodine ions, which may undergo redox reactions with H₂O₂. Additionally, industrial wastewater may also contain inorganic metal ions, such as Fe³⁺, which can accelerate the decomposition of H₂O₂ and potentially interfere with detection accuracy.

4. Conclusions

In this study, we proposed a colorimetric method for the detection of CAT-positive S. aureus based on its endogenous CAT activity. The POD-like property of an onion-like carbon nanozyme and a classical H₂O₂-TMB reaction were employed to monitor H₂O₂ consumption, which also served as the indicator for S. aureus concentration. In contrast to conventional biosensors for specific bacterial detection, our approach did not require the use of antibodies and aptamers, which greatly reduced the detection cost and manufacturing complexity. The method demonstrated a linear response to S. aureus from 2 × 10⁴ to 2 × 10⁷ CFU/mL with an LOD of 9.2 × 10³ CFU/mL. Furthermore, it exhibited robust anti-interference capability, enabling us to effectively distinguish S. aureus from co-existing pathogenic bacteria in real water samples. The underlying mechanism of the present sensing method could be further expanded to the detection of other CAT-positive bacteria, providing a straightforward and specific approach for pathogen detection.