Degradation of Atrazine to Cyanuric Acid by an Encapsulated Enzyme Cascade

Abstract

Atrazine (2-chloro-4-ethylamino-6-isopropylamino-s-triazine CAS: 1912-24-9) is a widely used herbicide that has been connected to a variety of negative human health and environmental effects. Various bacterial strains utilize a six-step enzyme cascade to fully degrade atrazine. The third step in this pathway, N-isopropylammelide aminohydrolase (AtzC), produces the first non-toxic intermediate, cyanuric acid. As such, AtzC, paired with enzymes catalyzing the first two steps in this pathway, triazine hydrolase (TrzN) and hydroxyatrazine (2-(N-ethylamino)-4-hydroxy-6-(N-isopropylamino)-1,3,5-triazine) N-ethylaminohydrolase (AtzB), can effectively degrade atrazine. All three of these enzymes were successfully encapsulated in tetramethyl orthosilicate (TMOS) gels using the sol–gel method, producing active biomaterials. These materials showed increased protection against proteolytic digestion by the endopeptidase trypsin, as well as increased thermal and pH stability when compared to their non-encapsulated counterparts. AtzB:sol and AtzC:sol also showed increased stability over time compared to soluble enzyme. A combination of all three biomaterials, TrzN:sol, AtzB:sol, and AtzC:sol, was shown to be effective at fully degrading 50 µM atrazine to cyanuric acid in just over an hour and a half, thus establishing a potential bioremediation enzyme cascade for atrazine.

1. Introduction

Atrazine (2-chloro-4-ethylamino-6-isopropylamino-s-triazine CAS: 1912-24-9) is a chlorinated, aromatic herbicide that was originally patented in 1958 in Switzerland, followed by the Unites States in 1959 [1]. Today, more than 70 million pounds are applied in the United States alone, primarily to corn, sorghum, nuts, and sugarcane, making it the second most used herbicide nationwide [2]. Atrazine is selective against broadleaf weeds, such as pigweed, cocklebur, and velvetleaf, and works by entering through the roots of plants and inhibiting photosynthesis [3]. Because of this, it is applied to soil and must enter the water stream to be effective. While this allows the chemical to enter the target plants, it also increases off-target organism reactions and allows atrazine to spread through the watershed [4]. Since atrazine is highly mobile, it has been found in everything from drinking water across the country to remote glaciers [1]. This is problematic since it has been connected to hormone disruption in plants and animals [1,2,5]. It has also been linked to a wide range of adverse effects in humans, including birth defects, endocrine disruption, lung and kidney disease, muscle degeneration, low blood pressure, and increased risk of multiple cancers [2,5]. Even at levels below legal limits in the United States, it has been linked to disrupted menstrual cycles, including longer time between periods, irregular periods, and spotting between periods [6]. Another study found that even at these low concentrations, atrazine exposure can be linked to preterm birth and low birth weights [7]. Recognizing the toxicity and widespread contamination of atrazine, Italy and Germany banned its use in 1991 [8], followed by the rest of the European Union in October 2003. In the same month, the US EPA approved its continued use [9]. As of 2020, concerns regarding environmental impact and human safety have led to the ban of atrazine in 35 countries, the state of Hawaii, and 5 US territories [10]. Nonetheless, it remains one of the most widely used herbicides in the United States, Canada, and Australia and is among the most often reported contaminants in ground and drinking water, according to the EPA [11]. It is for these reasons that atrazine remediation is significant for both human and ecosystem health.

Bioremediation offers an effective, safe, and low-cost process for the degradation of atrazine. Multiple bacteria from a variety of genera have been shown to completely degrade atrazine to carbon dioxide and ammonia through the Atz pathway, which consists of six enzymes [12,13,14,15,16]. The final product of the first three enzymes within this pathway is cyanuric acid, a compound commonly used to stabilize chlorine in swimming pools, and while it is an eye irritant, it is considered non-toxic, making it the first non-toxic compound generated by the atrazine biodegradation pathway [17,18]. The first three enzymes are the atrazine chlorohydrolase (AtzA), the hydroxyatrazine N-ethylaminohydrolase (AtzB), and the N-iosopropylammelide amidohydrolase (AtzC) (Figure 1). In some bacteria, triazine hydrolase (TrzN) is utilized instead of AtzA as the first step of the enzyme cascade to produce 2-hydroxyatrazine (2-(N-ethylamino)-4-hydroxy-6-(N-isopropylamino)-1,3,5-triazine) [14,19]. TrzN is a monomer vs. the hexameric AtzA enzyme, for which its chemical properties are better understood. Furthermore, TrzN is a Zn-dependent protein, much like AtzB and AtzC, whereas AtzA is Fe-dependent [14,19]. For these reasons, TrzN was selected over AtzA to perform the first reaction [14].

In comparison to whole-cell reactions, isolated enzymes have fewer side reactions, do not risk altering the microbiota of the environment, and do not require nutrients or aeration [20,21,22]. However, isolated enzymes are susceptible to environmental degradation by high temperatures, proteases, and other factors [21]. Encapsulation was shown to protect enzymes from such environmental degradation [20,22]. TrzN has been previously encapsulated within silica glasses derived through the sol–gel process, which resulted in a biomaterial that was more robust compared to wild-type (WT) TrzN [23]. Reported herein, both AtzB and AtzC were encapsulated in a tetramethyl orthosilicate (TMOS)-based sol–gel, the best matrix for TrzN biomaterials [23]. The stability of both AtzB and AtzC, like TrzN, was enhanced when encapsulated within a sol–gel matrix under a variety of conditions, including long-term storage, non-physiological pH, increased temperature, and in the presence of proteases such as trypsin. A combination of all three biomaterials, TrzN:sol, AtzB:sol, and AtzC:sol, into an enzyme cascade where they are in sequence or combined, allowed for the efficient degradation of atrazine to cyanuric acid, the first non-toxic compound in the atrazine biodegradation pathway.

2. Results and Discussion

2.1. Expression and Purification of Atrazine Degrading Enzymes

TrzN from Arthrobacter aurescens strain TC1, with mutations Leu131Pro, Ala159Val, and Asp38Asn for better Escherichia coli expression, a polyhistidine (His₆) affinity tag engineered onto the N-terminus, and a tobacco etch virus (TEV) protease cleavage site, was transformed and overexpressed in E. coli BL21(DE3) as previously described [23]. AtzB and AtzC from Pseudomonas sp. strain ADP, also with His₆ affinity tags and TEV cleavage sites, were transformed and overexpressed in E. coli BL21(DE3) cells. All three enzymes were successfully purified via fast protein liquid chromatography (FPLC) using a nickel–nitrilotriacetic acid (Ni–NTA) Superflow Cartridge. Kinetic analysis of TrzN against atrazine, AtzB against hydroxyatrazine, and AtzC against N-isopropylammelide revealed k_cat values of 3.1 ± 0.6 s⁻¹, 1.1 ± 0.6 s⁻¹, and 12 ± 1 s⁻¹, respectively, consistent with previously reported values for each His₆-tagged enzyme [12,24,25].

2.2. Encapsulation of TrzN, AtzB, and AtzC

All three enzymes were encapsulated in TMOS-based sol–gels. Bradford analysis of the buffer used to make and wash the materials suggests that negligible enzyme loss occurred throughout this process. Substrate degradation was observed spectrophotometrically, and the percentage of degraded substrate was calculated. For all three proteins, their respective substrate concentration decreased over time, indicating activity upon encapsulation (Figure 2). TrzN:sol degraded 96 ± 3% of a 50 µM atrazine solution after ~90 min, consistent with previously reported data [23]. Similarly, AtzB:sol and AtzC:sol degraded 85 ± 3% of 50 µM hydroxyatrazine and 94 ± 1% of 50 µM N-isopropylammelide, respectively, to completion after ~90 min. These data indicate that all three enzymes exhibit activity upon sol–gel encapsulation.

2.3. Proteolytic Digestion of Free and Encapsulated TrzN, AtzB, and AtzC

To confirm that the encapsulated enzyme was the active catalyst and not surface-bound enzymes, TrzN:sol, AtzB:sol, and AtzC:sol were exposed to the protease trypsin from bovine pancreas. This endopeptidase cleaves polypeptides at the C-terminus of arginine and lysine residues unless a proline residue is present [26]. ExPASy predicts 39 such sites in one TrzN monomer, 38 in an AtzB monomer, and 47 in an AtzC monomer [27]. All three biomaterials retained nearly full activity after digestion compared to their unencapsulated free WT forms (Figure 3). Free TrzN only retained 20 ± 3% of its activity post-digestion compared to TrzN:sol, which reatained 74 ± 6% of its activity. A paired t-test with an alternative hypothesis that TrzN:sol retained more activity than free enzyme results in a p-value of 7.5 × 10⁻⁴, indicating a statistically significant difference between the two samples, showing that encapsulation successfully protected TrzN from digestion by trypsin. Similarly, free AtzB retained just 21 ± 1% of its activity, significantly less than the 92 ± 8% retained by the AtzB:sol, as indicated by p = 7.6 × 10⁻⁴. Finally, free AtzC retained 14 ± 2%, whereas AtzC:sol retained 95 ± 3%, resulting in a p value of 5.4 × 10⁻⁴ and thus again indicating that encapsulation significantly protects each of these enzymes from proteolytic degradation.

2.4. Thermostability of Free and Encapsulated TrzN, AtzB, and AtzC

The catalytic activity of each enzyme, both in their free soluble and sol–gel encapsulated forms, was examined at various temperatures (Figure 4). Each of the biomaterials was incubated in a water bath for 30 min at increasing temperatures in 50 mM Tris HCl, pH 7.5. Afterward, the biomaterials were allowed to cool and tested for activity at 25 °C. Encapsulation generally offered increased thermostability after exposure to temperatures greater than 50 to 60 °C. After exposure to 70 °C, soluble TrzN retained 4.5 ± 0.1% of its original activity compared to 22.1 ± 0.1% for the TrzN:sol, resulting in p = 1.7 × 10⁻³ and thus indicating significant thermal protection from encapsulation. Similar results were observed for AtzB, where the free enzyme retained 5.6 ± 0.3% of its activity compared to the 20.5 ± 0.3% retained by the AtzB:sol biomaterial after incubation at 70 °C. Significant protein precipitation and no activity were observed for AtzC after incubation at temperatures above 50 °C. However, AtzC:sol retained 22.3 ± 0.5% of its original activity after incubation at 70 °C, indicating that encapsulation significantly increased thermostability. In general, the thermostability of all three proteins increased at high temperatures when encapsulated in a sol–gel.

Enzyme loss was investigated for each of the biomaterials after being incubated at 60 °C or higher using a Bradford assay. Protein loss was observed, averaging 27 µg/mL at 60 °C up to 60 µg/mL at 80 °C. This was hypothesized to be due to surface-adhered proteins leeching off the material. To test this, a trypsin digestion was first performed to degrade surface-adhered proteins, followed by incubation at increasing temperatures. After digestion with trypsin, TrzN:sol exhibited similar activity compared to nondigested TrzN:sol after exposure to 80 °C (27 ± 2% compared to 31 ± 4%) with negligible protein loss. Similarly, trypsin-digested AtzB:sol and AtzC:sol samples exhibited negligible protein loss and nearly identical catalytic activities after exposure to high temperatures, suggesting that the protein lost in the initial thermostability tests was, in fact, a surface-bound enzyme.

2.5. pH Stability of Free and Encapsulated TrzN, AtzB, and AtzC

To test if encapsulation improved the stability of TrzN, AtzB, and AtzC at non-physiological pH values, the activity of the free and encapsulated enzymes was examined at pH 5 and 9. The activity of each sample was compared to a control at pH 7, and the percentage of retained activity was calculated. Overall, encapsulated TrzN, AtzB, and AtzC retained significantly higher activity levels at pH 5 and 9 compared to the free enzyme (Figure 5). Free TrzN only retained 12.8 ± 0.6% and 55 ± 1% of activity at pH of 5 and 9, respectively. This is in contrast to TrzN:sol, which retained 70 ± 1% and 102 ± 3% under the same conditions. A one-sided, paired t-test comparing activity retained by free enzyme and encapsulated enzyme shows a p-value of 4.4 × 10⁻³ and 4.9 × 10⁻⁴, respectively, indicating that the observed protection by the material is statistically significant. This is consistent with previous tests on TrzN encapsulation [23]. Free AtzB was virtually inactive at pH 5, retaining only 2 ± 1% of its activity, in contrast to AtzB:sol, which retained 53 ± 1% under the same conditions. Statistical analyses result in a p-value of 2.7 × 10⁻³, indicating that the encapsulation significantly protects AtzB under these conditions. In basic conditions (pH 9), free and encapsulated AtzC retained 19 ± 2% and 81 ± 3% in activity, respectively, resulting in p = 5.1 × 10⁻². At pH 5, soluble AtzC retained 57 ± 4% of activity when observed at pH 7, while AtzC:sol retained 83 ± 4%. At pH 9, the free and encapsulated AtzC enzyme retained roughly the same activity: 60 ± 3% and 65 ± 5%, respectively. Overall, these results show that encapsulation within a sol–gel matrix protects all three enzymes at both acidic and basic pH values.

2.6. Reusability of Free and Encapsulated Enzyme

Soluble and encapsulated enzymes were stored for a period of six weeks at 4 °C. Each week, activity was tested and compared to the initial activity on week 0 (Figure 6). The storage buffer was tested for enzyme loss using a Bradford assay. After five weeks, soluble AtzC had no detectable activity compared to the AtzC:sol, which retained 46 ± 1% of its original activity and 11 ± 1% after six weeks. Unencapsulated AtzB performed slightly better, retaining 12 ± 4% of its initial activity after six weeks compared to AtzB:sol, which retained 34 ± 3% of its original activity after six weeks. A paired t-test revealed p = 5.2 × 10⁻², which is just above α = 0.05; thus, encapsulation protects less stable proteins from degradation over time when stored at 4 °C. Soluble TrzN, however, was shown to remain active over the entire course of the study, retaining 95 ± 4% of its activity at the end of the six weeks. TrzN:sol retained 68 ± 2%, consistent with previous studies on TrzN [23]. This loss of activity can be attributed to protein loss observed in weeks three and four, rather than enzyme degradation.

2.7. Degradation of Atrazine by an Encapsulated Enzyme Cascade

The degradation of atrazine to cyanuric acid requires three enzymes, TrzN, AtzB, and AtzC, so three different versions of a biomaterial cascade were created (Figure 7). First, the TrzN:sol, AtzB:sol, and AtzC:sol biomaterials were tested sequentially by reacting atrazine with TrzN:sol at 25 °C until the reaction went to completion, followed by transferring the reaction mixture to AtzB:sol. Once the AtzB:sol reaction went to completion, the reaction mixture was transferred to the AtzC:sol biomaterial. The initial atrazine concentration was 50 µM, and the reactant and substrate concentrations were examined every 20 min spectrophotometrically. After ~40 min, the atrazine had fully reacted, and the reaction mixture was transferred to the AtzB:sol biomaterial. After ~80 min, the concentration of hydroxyatrazine was found to be zero, and the reaction mixture was then transferred to a flask containing the AtzC:sol biomaterial. AtzC:sol degraded N-isopropylammelide to cyanuric acid in just 20 min Next, all three biomaterials, TrzN:sol, AtzB:sol, and AtzC:sol, were mixed, and a 50 µM atrazine solution was added. Interestingly, a small build-up of hydroxyatrazine was observed with a consistent increase in cyanuric acid, and the reaction was complete at ~140 min (Figure 7, panel A). These data are consistent with the data from the sequential cascade experiments and the calculated k_cat values for each soluble enzyme. When an equimolar mixture of TrzN, AtzB, and AtzC was used to create a single sol–gel biomaterial, which is termed a “Combo gel”, neither intermediate hydroxyatrazine nor N-isopropylammelide was observed, and the reaction reached completion after ~100 min. While all three methods were effective at degrading atrazine to cyanuric acid, the combined sol–gel system provided the product faster without any physical steps needed to transfer solute or gels.

3. Materials and Methods

Chemicals: Atrazine, 2-hydroxyatrazine, cyanuric acid, tetramethyl orthosilicate (TMOS, ≥99%), and type I trypsin from bovine pancreas were purchased from Sigma-Aldrich, Burlington, MA, USA, in the highest purity available.

Enzyme expression: The genes encoding TrzN from A. aurescens TC1 (accession number: Q6SJY7) with D38N, L131P, and A159V mutations and AtzB (accession number: P95442) and AtzC (accession number: O52063) from Pseudomonas sp. strain ADP were codon optimized for E. coli expression and engineered into individual pET28a(+) plasmids with a polyhistidine (His₆) affinity tag. The resulting plasmids were transformed into E. coli BL21(DE3)-competent cells (Stratagene) and grown on agar plates containing 50 mM kanamycin for 16 h. at 37 °C. A single colony was then used to inoculate a 100 mL lysogeny broth–Miller starter culture with 50 μg/mL kanamycin and grown overnight at 37 °C, shaking at 180 rpm. In the case of TrzN, this culture was split between nine liters of culture supplemented with 5 µM isopropyl ß-d-1-thiogalactopyranoside (IPTG) and grown at 37 °C for 48 h [23,25]. For AtzB- and AtzC-expressing cells, the 100 mL starter culture was split between nine cultures of 1 L each and first grown to an optical density of ~0.8–1.0 at 600 nm before they were induced with 0.1 mM IPTG with the addition of 0.05 mM ZnCl₂ and expressed for 16 h at 20 °C. Cells were then pelleted by centrifugation at 7000 rpm at 4 °C for 25 min in a Beckman Coulter Avanti JLA-8.1 rotor (Beckman Coulter, Brea, CA, USA) and resuspended in Buffer A (TrzN: 50 mM NaH₂PO₄, 500 mM NaCl, 10 mM imidazole, glycerol, pH 7.5; AtzB: 50 mM 3-(N-morpholino)propanesulfonic acid (MOPS), 500 mM NaCl, 10 mM imidazole, pH 7.5; AtzC: 25 mM MOPS, 145 NaCl, 10 mM imidazole, 10% glycerol, pH 6.8).

Enzyme purification: Cells were lysed on ice using a Cole-Parmer 21 W Misonex sonicator 3000 (Vernon Hills, IL, USA) with 30 s on and 45 s off intervals for a total of 20 min. The lysed cells were centrifuged for 45 min at 17,000 rpm and 4 °C in a Coulter Avanti JA-17 rotor (Beckman Coulter, Brea, CA, USA) to remove cell debris. The supernatant was then loaded onto a 5 mL nickel–nitrilotriacetic acid (Ni–NTA) Superflow Cartridge (Qiagen, Hilden, Germany) attached to an ÄKTA FPLC Prime plus pre-equilibrated with the appropriate Buffer A. In total, 5–10 column volumes (CV) of Buffer A were run through the column to remove any unbound protein. To purify TrzN, a gradient of Buffer B (50 mM NaH₂PO₄, 500 mM NaCl, 500 mM imidazole, pH 7.5) from 0 to 100% over 6 CV was run across the column. For AtzB, a wash with 5% Buffer B (AtzB: 50 mM MOPS, 500 mM NaCl, 500 mM imidazole, pH 7.5) for 10 CV followed by a gradient from 5 to 100% Buffer B over 10 CV was utilized. AtzC was purified in a very similar way to AtzB: first with 5% Buffer B (25 mM MOPS, 145 NaCl, 500 mM imidazole, 10% glycerol), followed by a 5–80% gradient over 10 CV.

Synthesis of N-isopropylammelide: Since N-isopropylammelide was not commercially available, it was synthesized via a two-step reaction (Figure 8). The starting material (cyanuric chloride) was dissolved in tetrahydrofuran (THF). Isopropylamine was then added to form Compound 1. From there, a 2.5 M solution of NaOH was added, and the mixture was refluxed for 6 h to obtain the final product N-isopropylammelide. A detailed description of this synthesis and full characterization can be found in the Supplementary Materials. Steps of this synthesis were adapted from Zhao et al. and Talebian et al., respectively [28,29].

Kinetic Activity Assay: The catalytic activities of all three purified enzymes were tested in triplicate using a Shimadzu UV–Vis 2600i spectrophotometer (Shimadzu, Tokyo, Japan) equipped with a T1 temperature controller from Quantum Northwest in 1 mL quartz cuvettes (Shimadzu, Tokyo, Japan). TrzN activity was measured by monitoring a decrease in absorbance at 264 nm (ε₂₆₄ = 3.5 mM⁻¹ cm⁻¹) using a 150 µM atrazine solution in 0.1 M sodium phosphate buffer, pH 7.0, in the presence of 3% v/v methanol to assist with atrazine solubility [30]. AtzB activity was measured by monitoring a decrease in absorbance at 240 nm (ε₂₄₀ = 16.2 mM⁻¹ cm⁻¹ for hydroxyatrazine, ε₂₄₀ = 1.6 mM⁻¹ cm⁻¹ for N-isopropylammelide) using a 60 µM hydroxyatrazine solution in 50 mM sodium phosphate, pH 7.0, in the presence of 3% methanol [24]. Finally, AtzC activity was measured by monitoring a decrease in absorbance at 240 nm using a 1 mM N-isopropylammelide solution in 50 mM HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid) at pH 7.0 [12,31].

Encapsulation: TrzN, AtzB, and AtzC were all encapsulated in TMOS-based sol–gels at a final concentration of 2 mg/mL as previously reported [23,32]. A TMOS (Sigma-Aldrich) solution containing 813 µL of TMOS, 181.4 µL of nanopure water, and 5.6 µL 0.04M HCl was combined to make a 1 mL reaction mixture that was then sonicated on ice for 30 min using a Branson 2800 ultrasonic cleaner (Branson Ultrasonics, Brookfield CT, USA). Following sonication, the sol mixture was allowed to cool on ice before proteins were added. Proteins in 50 mM HEPES, pH 7, were added at a 2:1 ratio of protein–TMOS to a final concentration of 2 mg/mL [32]. The resulting mixture was left on ice until gelation occurred, resulting in TrzN:sol, AtzB:sol, and AtzC:sol monoliths that were subsequently washed 3 times with 100 µL of 50 mM Tris HCl, pH 7.5, and then stored in the same buffer at 4 °C overnight. Aged monoliths were crushed with a metal spatula and washed again with 300 µL of the same buffer. Wash and storage buffers were saved and tested for protein loss using a Coomassie (Bradford) Protein Assay Kit from Thermo Scientific (Waltham, MA, USA).

Kinetic characterization of sol–gel encapsulated TrzN, AtzB, and AtzC: The activity of each of the biomaterials was measured by observing a decrease in absorbance at the appropriate wavelength (TrzN:sol 264 nm, AtzB:sol 240 nm, and AtzC:sol 240 nm) on an Agilent 8453 UV-visible spectrophotometer. For TrzN:sol, a solution of 50 µM atrazine was added to 50 mM HEPES and 3% v/v methanol to improve atrazine solubility (TrzN:sol reaction mixture). In total, 1 mL of this reaction mixture was added to 100 µL of TrzN:sol and reacted at 25 °C, shaking at 180 rpm. Absorbance of the reaction mixture was measured every 15 min, and substrate concentrations were calculated using the molar absorptivity and compared to the initial concentration determined from the absorbance at time zero. The specific activity (U/mg) of the enzyme was calculated by accounting for the reaction rate, volume, and protein concentration. The standard deviation was calculated based on three trials. The same process was repeated for the AtzB:sol and AtzC:sol biomaterials with solutions of 50 µM hydroxyatrazine in 50 mM HEPES, 5% v/v methanol (AtzB:sol reaction mixture), and 50 µM N-isopropylammelide in 50 mM HEPES (AtzC:sol reaction mixture).

Proteolytic digestion of soluble and encapsulated enzymes: TrzN:sol, AtzB:sol, and AtzC:sol and their respective free enzymes were all exposed to the endopeptidase trypsin from bovine pancreas (Sigma-Aldrich) at a ratio of 4 mg/mL trypsin to 2 mg/mL enzyme for 18 h in a trypsin reaction buffer (50 mM Tris-HCl and 1 mM CaCl₂ pH 7.6) at 35 °C with shaking at 180 rpm [33]. Control samples were incubated in a trypsin reaction buffer without trypsin present under the same conditions. After 18 h, the trypsin-laced buffer was removed from the biomaterials through centrifugation at 15,000 rpm using Eppendorf centrifuge 5424 (Eppendorf, Oldenburg, Germany), and the biomaterials were washed with 300 µL of Tris-HCl, followed by placing them in 1 mL of the appropriate reaction mixture and reacting for 60 min at 25 °C while shaking at 180 rpm. Absorbance values were obtained for the corresponding product before and after reaction, allowing the concentration of the substrate and product to be calculated and permitting the specific activity to be determined. The activity of trypsin-digested biomaterials and soluble enzymes was then compared to the control to calculate the percentage of retained activity.

Thermostability of soluble and encapsulated enzymes: The thermostability of each free enzyme in its encapsulated form was examined by exposing each sample to 30 ± 3, 40 ± 3, 50 ± 3, 60 ± 3, 70 ± 3, and 80 ± 3 °C for 30 min in a water bath. After removal from the water bath, the free enzyme and the biomaterials were washed with 50 mM Tris-HCl, pH 7.5, which was saved for analysis via a Bradford assay. Absorbance values were obtained before and after a one-hour reaction with 1 mL of the appropriate reaction buffer (outlined above) at 25 °C while shaking at 180 rpm and used to calculate the change in substrate and product concentration. From there, specific activity was determined. This activity was compared to the activity of the biomaterials incubated at room temperature (20–23 °C) for 30 min.

pH stability of soluble and encapsulated enzymes: The activity of both soluble and encapsulated enzymes was tested at pH 5 and 9 to analyze their catalytic ability at various pH values. For pH 5, 50 mM citric acid was used. For pH 9, 50 mM borate was used. To each 100 µL of TrzN:sol, AtzB:sol, and AtzC:sol, 1 mL of each reaction buffer (outlined above) at the appropriate pH and substrate was added. Absorbance was measured before and after a one-hour reaction at 25 °C while shaking at 180 rpm, and the specific activity was calculated. Retained activity was calculated by comparing the specific activity to that observed at pH 7.

Reusability of free and encapsulated enzymes: The reusability of TrzN:sol, AtzB:sol, and AtzC:sol was tested over the course of six weeks. Sol–gels were stored in 300 µL Tris-HCl, pH 7.5, at 4 °C throughout the experiment. Once a week, they were removed from storage. The storage buffer was decanted and saved to test for protein loss using a Bradford assay. The catalytic activity of each biomaterial was determined using a one-hour reaction with 1 mL of the appropriate reaction buffer (outlined above) at 25 °C while shaking at 180 rpm. After testing, 300 µL Tris-HCl was added and stored again at 4 °C. Soluble enzyme stocks were stored in 50 mM HEPES, pH 7, at 4 °C over the course of the experiment, and aliquots were taken each week and tested for activity.

Degradation of atrazine to cyanuric acid by a three-enzyme cascade: Three different versions of a TrzN:sol, AtzB:sol, and AtzC:sol enzyme cascade were tested: sequential, mixed gels, and combo gels. With sequential gels, 3 mL of 50 µM atrazine in 50 mM HEPES at pH 7 was added to 100 µL 2 mg/mL of TrzN:sol. After the reaction was completed, the reaction mixture was decanted to 100 µL of 2 mg/mL of AtzB:sol, and again, the reaction was allowed to proceed to completion. Finally, the reaction mixture was decanted to 100 µL of 2 mg/mL of AtzC:sol, and the reaction was allowed to proceed to completion.

To make mixed enzyme biomaterials, TrzN:sol, AtzB:sol, and AtzC:sol were all prepared individually at a concentration of 2 mg/mL each. Individual gels, 100 µL of each, were mixed in one tube. From there, 3 mL of 50 µM atrazine in 50 mM HEPES pH 7 was added to the final 300 µL volume.

To make a combination of enzyme biomaterials, a solution of 50 mM HEPES at pH 7 with 2 mg/mL of TrzN, AtzB, and AtzC was prepared. This solution was used to make a sol–gel biomaterial containing all three enzymes, with a final volume of 300 µL. From there, 3 mL of 50 µM atrazine in 50 mM HEPES pH 7 was added, and the reaction was allowed to proceed to completion.

Throughout the experiments, sol–gels were reacted at 25 °C, with mixing at 180 rpm for 15 min, after which they were centrifuged for 5 min at 4000 rpm for a total time of 20 min between measurements. The absorbance at 224, 240, and 264 nm was then measured and recorded.

To calculate the concentration of each compound present, the molar absorptivity of each was calculated at all three wavelengths, and a system of equations was developed. This process is outlined in the Supplementary Materials.

4. Conclusions

In summary, the data presented herein represent the first enzymatic biomaterial cascade capable of degrading atrazine, a toxic herbicide, to cyanuric acid, which is non-toxic, in three steps. The three enzymes utilized are the first three within the biodegradation pathway for atrazine: TrzN, AtzB, and AtzC. Each of these enzymes was successfully encapsulated in TMOS-based sol–gels and was shown to retain its catalytic function. Encapsulation of each enzyme greatly enhanced their ability to withstand extreme environments, such as proteolytic digestion, thermal degradation, and acidic or basic conditions, and their performance was better than their corresponding free enzyme (

Table 1. Summary of retained activity of each enzyme, both soluble and encapsulated, under the conditions tested.

		Percent Activity
Enzyme	Material	After 18 h Digestion	pH 5	pH 9	70 °C	6 Weeks
TrzN	Soluble	20 ± 3%	13 ± 1%	55 ± 1%	5 ± 1%	95 ± 4%
	Sol–gel	74 ± 6%	70 ± 1%	102 ± 3%	22 ± 11%	68 ± 2%
AtzB	Soluble	21 ± 1%	2 ± 1%	19 ± 2%	6 ± 1%	13 ± 4%
	Sol–gel	92 ± 8%	53 ± 1%	81 ± 3%	21 ± 1%	34 ± 3%
AtzC	Soluble	14 ± 2%	57 ± 4%	60 ± 3%	0%	0%
	Sol–gel	95 ± 3%	83 ± 4%	68 ± 5%	22 ± 1%	11 ± 1%

). All three biomaterials were reusable over a six-week period and retained activity at greater levels than their free enzyme counterpart. Protection from these environmental factors, along with their reusability profiles, clearly demonstrates that sol–gel-encapsulated enzymes are better suited for bioremediation applications than their soluble counterparts. The beauty of sol–gel biomaterials is the fact that they can be cast into any desired shape or ground into a powder, which can be used in columns. Finally, the combination of TrzN:sol, AtzB:sol, and AtzC:sol, either in sequence or in a mixture, could degrade atrazine to its non-toxic derivative cyanuric acid in about two hours. Encapsulating all three enzymes into a single sol–gel material provided a “Combo gel” that was capable of fully degrading atrazine to cyanuric acid in 1.5 h. These data provide a new avenue for the design of bioremediation methodologies for the removal of atrazine from the environment.