Portal de Periódicos da CAPES

Light-Controlled Chemical Reaction Pathways in Disulfide-Based Nonequilibrium Systems

2023; Chinese Chemical Society; Volume: 6; Issue: 6 Linguagem: Inglês

10.31635/ccschem.023.202303331

2096-5745

Yingshuai Zhao, Peng Zhao, Jingyi Zhang, Yuanfeng Zhao, Bohan Li, Guoxiu Hao, Yuanyuan An, Wei Zhou, Yan Lü, Lien‐Yang Chou, Yijun Zheng,

Spectroscopy and Quantum Chemical Studies

Open AccessCCS ChemistryRESEARCH ARTICLES26 Oct 2023Light-Controlled Chemical Reaction Pathways in Disulfide-Based Nonequilibrium Systems Yingshuai Zhao†, Peng Zhao†, Jingyi Zhang†, Yuanfeng Zhao, Bohan Li, Guoxiu Hao, Yuanyuan An, Wei Zhou, Yan Lu, Lien-Yang Chou and Yijun Zheng Yingshuai Zhao† , Peng Zhao† , Jingyi Zhang† , Yuanfeng Zhao , Bohan Li , Guoxiu Hao , Yuanyuan An , Wei Zhou , Yan Lu , Lien-Yang Chou and Yijun Zheng *Corresponding author: E-mail Address: [email protected] https://doi.org/10.31635/ccschem.023.202303331 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Pathway selection in a complex chemical reaction network (CRN) enables organisms to adapt, evolve, and even learn in response to changing environments. Inspired by this, herein we report an artificial system, where light signal was used to manipulate the reaction pathways in a disulfide-based nonequilibrium CRN. By changing the photon energy and irradiation window, the anion or new radical-mediated pathways were selectively triggered, resulting in a user-defined evolution pathway. Additional photodissipative cycles were achieved by UV (365 nm) irradiation, increasing the total number of reactions from 3 to 7. The emerging pathway selection of the CRN is accurately predictable and controllable even in complex organo-hydrogel materials. We demonstrate up to five-state autonomous sol–gel transitions and the formation of fuel-driven dissipative organo-hydrogel through both chemical and light input. This work represents a new approach to allowing CRNs to communicate with the environment that can be used in the development of materials with lifelike behaviors. Download figure Download PowerPoint Introduction Chemical reaction networks (CRNs) within living organisms act as the "machinery" of life.1–5 These CRNs are normally preprogrammed precisely in these systems for inducing various predictable emerging behaviors.6–11 When the environment varies, living organisms create new reaction pathways within CRNs that allow them to adapt to the variation in the environment and better compete for resources as a result of mutation or other genetic changes.12,13 Examples of this include the emergence of photosynthesis in early bacteria and the evolution of methanotrophic bacteria, which give these organisms a competitive advantage to thrive in a specific environment.14,15 This capability of generating new reaction pathways plays a significant role in the evolution of life.16 Recently, different artificial CRNs have been designed as a means of endowing materials with lifelike properties, such as out-of-equilibrium behaviors,17–21 oscillations,22–24 and dissipative self-assembly.6,17,25–30 To enable their ability to communicate with the environment, several recent systems have been created to be responsive to external signals such as catalysis,16,31,32 pH signals,33–38 and light.39–44 However, these CRNs can only vary resident reaction rates rather than reaction pathways, which limits their ability to sense, respond, and even evolve in response to changing conditions.1,45,46 The development of CRNs with the ability to select their reaction pathways has the potential to enable these systems to exhibit even more complex and dynamic behaviors, but this remains a challenge. Herein, we present a strategy for reaction pathway selection in a CRN enabled by light signals (Figure 1). A small-molecule model beginning with thiol and thiuram disulfides (TDS) was initially used, where 365 nm UV light (10 mW/cm2) as an environmental signal facilitated the new radical-mediated reaction pathways. Substance identification, product distribution, and the kinetics profile in a model reaction were comprehensively compared with the same characteristics in dark conditions. We found that the photogenerated intermediates disturbed the equilibration of disulfide metathesis (R2), and the photolysis and radical recombination contributed to an additional nonequilibrium pathway (R2hv). Light can also introduce a fast photolytic reaction (R3hv) that replaces the slow thionate decomposition-coupling step (R3) and allows the active intermediate (diethylammonium diethyldithiocarbamate, DDDC) to persist and react with the photolytic product [2-hydroxyethyl disulfide (HEDS), R4]. The user-defined pathway selection of the CRN was accurately achieved, even in complex organo-hydrogel, by varying the photon intenisty and irradiation window. Notably, up to five-state autonomous sol–gel transitions and the fuel-driven dissipative organo-hydrogel were achieved through both chemical and light input. Figure 1 | Schematic representation of thiol-based CRN with pathway selectivity. The starting compounds of the system are MCE and TDS. The size of each pie chart slice represents the relative proportion of the intermediates along the reaction pathways (the blue bulb indicates the presence of 365 nm UV illumination). Download figure Download PowerPoint Experimental Methods Experimental materials N,N,N′,N′-tetraethylthiuram disulfide (TDS, 97%; Shanghai Macklin Biochemical Co., Ltd., Shanghai, China), 4-arm-PEG-SH (95%, Mn = 20,000; Hunan Huateng Pharmaceutical Co., Ltd., Hunan, China), 2-hydroxyethyl disulfide (HEDS, 90%; Shanghai Macklin Biochemical Co., Ltd.), diethylammonium diethyldithiocarbamate (DDDC, 97%; Bide Pharmatech Ltd., Shanghai, China), 2-mercaptoethanol (MCE, 98%; Alfa Aesear China Chemical Co. Ltd., Shanghai, China), carbon disulfide (CS2, 99%; Shanghai Macklin Biochemical Co., Ltd.) were used without further purification unless specified. 2-Hydroxyethyl diethylcarbamo(dithioperoxo)thioate (HEDT) was synthesized according to literature procedures.47 After mixing MCE (18.0 mg, 0.23 mM) and TDS (0.24 mM) in acetonitrile, the system was stirred at room temperature for 1 h. Then the mixture was concentrated and purified by column chromatography (Hexane/EtOAc, 3/2) to provide light yellow powders (yield: 60%). High-performance liquid chromatography (HPLC)-grade CH3CN was purchased from BOJN Scientific (GD) Co. Ltd. (Guangdong, China) and used without further purification. All solvents used for proton nuclear magnetic resonance (1H NMR) were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China) and used without purification. Small-molecule model reactions All small-molecule model reactions were performed in Milli-Q water and CH3CN (500 μL, 50/50, v/v). The reaction mechanism and reaction kinetic constants of the small-molecule model reaction (R1–R3,48 R4, R1hv–R4hv) were characterized and calculated by 1H NMR. R1hv and R4 complemented the HPLC characterization results. The reactant concentration of all model reactions was 0.02 mol/L, R1–R4 was performed without UV light mediation, and R1hv–R4hv was performed under ultraviolet light irradiation with different illumination time (R1hv–R4hv; 10, 30, 70, and 180 min). The reactions involving UV illumination were performed using UV lamps (15 W, 100 V, 0.37 A, PHILIPS) irradiated with a light intensity of 10 mW/cm2. For more information, please see Supporting Information Figures S1–S10. Preparation of PEG organo-hydrogel All organo-hydrogels were prepared in Milli-Q water and CH3CN (50/50, v/v). 4-arm-PEG-SH (10 wt %) and TDS (0.02 mol/L) were respectively dissolved in water and CH3CN, and the two solutions were mixed well in the volume ratio of 1:1 to obtain the organo-hydrogels. The demonstration of the organo-hydrogel mediated by light was performed in a quartz cuvette with a light range of 1 mm. The reaction pathway for the required light was performed under UV irradiation with an intensity of 10 mW/cm2. HPLC characterization The concentration of reactive substances was calculated according to the standard concentration curve of each reactant determined by HPLC. The rate constant k was obtained according to the second-order opposite reaction kinetic equation. The HPLC data were recorded at 254 nm using the mobile phase of CH3CN-water (65:35, v/v) at a flow rate of 1 mL/min, and the concentration of reactive substance was 0.02 mol/L. If not otherwise specified, all experiments were performed at ambient temperature for a run time of 15 min. NMR spectroscopy NMR were recorded at 25 °C on a Bruker Avance III HD500 spectrometer (Bruker). All NMR characterization was performed in CD3CN and D2O (600 μL, 50/50, v/v), and the concentration of reactants was 0.02 mol/L. All the chemical shifts were reported in ppm relative to the signals corresponding to the residual nondeuterated solvent. δ: 2.04 for CD3CN and δ: 4.79 for D2O in 1H NMR. Multiplets were reported as s (singlet), d (doublet), dd (double doublet), t (triplet), and m (multiplet). Simulation modeling to predict the substance concentration fluctuations The programmable gel was prepared according to the established kinetics of the cascade reactions, and all the components were simulated quantitatively using numerical integration methods.49 The kinetic equations of the system were constructed using ordinary differential equations (ODEs) based on the proposed elementary steps and rate constants with or without UV light, respectively. The variation of component concentrations with time was calculated by numerically solving the ODEs through a backward differential formulation method implemented in the python2/scipy library. ODEs for the substances were constructed based on the elementary steps and rate constants as given in Supporting Information Figure S3. Rheological characterization of organo-hydrogels Oscillatory experiments were carried out using a rheometer (Discovery Hybrid Rheometer (DHR); TA Instruments; Waters Technologies (Shanghai) Ltd., Shanghai, China) equipped with the OmniCure S2000 Spot UV Curing System in a frequency-controlled, amplitude-controlled, or time-controlled mode. UV light exposure was applied through an OmniCure® S2000 UV curing lamp system with a 365 nm filter (5–100 mW/cm2). Measurements were performed at room temperature with the Peltier solvent trap and evaporation blocker to prevent solvent evaporation. The temperature of the Peltier-heated plate was always controlled at ∼25 °C. A strain scan with an oscillation frequency of 1 rad/s in the range of 0.01–100% was performed to determine the linear region. After determining the linear viscoelastic region, a frequency scan of 0.01–10 Hz was performed on the test sample. A time scan was performed at a strain amplitude of γ0 = 1% and a frequency of 1 Hz. Results and Discussion Characterization of small-molecule model reactions As revealed in our recent work,48 when MCE (0.02 mol/L) and TDS (0.02 mol/L) were mixed, MCE was oxidized by TDS under mild conditions to form HEDS and DDDC (Figure 2, R1). This reaction can be understood as a fast interchange between the thiolate anion and the alkylthiuram disulfide. In the presence of the residual TDS, the obtained HEDS was subsequently converted into HEDT via disulfide–disulfide metathesis reaction R2 (k2 = 0.014 L•mol−1•s−1). HEDT was unstable and slowly underwent decomposition R3 (k3 = 0.0014 L•mol−1•s−1) to produce HEDS and volatile by-products (CS2, SO2, and CO2).48 The presence of additional TDS caused the HEDS to undergo metathesis (k2) and decomposition-coupling (k3) reactions again, creating a fuel-driven reaction cycle. These reactions (R1, R2, and R3) involved an intermediate thiolate anion and were accelerated under basic conditions. The MCE/TDS systems provided us with an ideal CRN model with lifelike autoevolved and dissipative behaviors. Considering that disulfide bonds can be broken and reformed through radical-mediated pathways under UV irradiation, the distinct bond dissociation energies of –S–S– (BDE: 204–272 kJ/mol) led us to expect that UV light can selectively regulate its photodissociation and alter the kinetics and pathways of the CRN ( Supporting Information Table S1).50,51 To test this idea, we used 365 nm UV light (327 kJ/mol) as the environmental input, and a Peltier device was used to maintain a constant temperature of 25 °C ( Supporting Information Figure S1). The generated reactive thiyl radicals in these disulfides were expected to participate in thiyl-centered processes to create new pathways. Figure 2 | Kinetics of subreactions in CRN. Comparison of the reaction equations with (R1hv–R4hv) or without UV illumination (R1–R4). Download figure Download PowerPoint Compared to the dark conditions (k1 = 10.08 L•mol−1•s−1), the oxidation reaction in the small-molecule model was slightly accelerated under UV exposure, with a rate constant of k ‾ 1hv = 10.26 L•M−1•s−1 (Figure 3a, see also Supporting Information Figure S2), suggesting that the oxidation reaction was still predominantly governed by the thiolate-centered process. Due to the BDE difference ( Supporting Information Table S1), 365 nm UV irradiation caused the preferential homolytic –S–S– photocleavage of TDS,52 and the negligible acceleration was likely due to the slow dissociation kinetics ( Supporting Information Figure S3, kdecomTDS = 0.02 L•M−1•s−1), in which the combination of thiol proton and the thiyl radicals provided the same products with R1 ( Supporting Information Figure S4). In contrast, UV light altered the pathway of the disulfide metathesis reaction (R2). Without light exposure, R2 was reversible, with a typical nonequilibrium steady-state of [TDS]:[HEDS]:[HEDT] = 0.25:0.23:0.5 after 16 h (Figure 3b, lower panel). However, under UV light, the disulfide metathesis by photogenerated thiyl recombination in both HEDS and HEDT was significantly accelerated. A rate constant k2hv of 0.074 L•M−1•s−1 was determined by 1H NMR measurements ( Supporting Information Figure S5), a fivefold increase compared to that for R2 in the absence of light (k2 = 0.014 L•M−1•s−1). In addition, the TDS is consumed faster than HEDS due to light-induced self-dissociation, with a [TDS]:[HEDS]:[HEDT] = 0.10:0.21:0.40 after just 10 min of UV irradiation (Figure 3b, upper panel). Figure 3 | (a) Comparison of the rate constants of oxidation (k1, k1hv), exchange (k2, k2hv), and decomposition reactions (k3, k3hv) with or without UV illumination. (b) Comparison of the molar fractions of HEDS, TDS, HEDT, and concentration of DDDC as a function of the reaction time with or without UV illumination determined using HPLC. (c) HPLC of the mixed solution ([DDDC]0 = 0.04 mM; [HEDS]0 = 0.02 mM, the mobile phase ratio was CH3CN-water (50:50, v/v)) over time. The retention time of DDDC, HEDS, and HEDT were 1.9, 4.3, and 8.9 min, respectively. The inset shows the reaction rates (k4) collected from the changes. (d) Stack of 1H NMR spectra obtained at different times during HEDS (at 2.91 ppm) decomposition under UV illumination. The new peak at 2.69 ppm was assigned to MCE, and k4hv was obtained by the changes. UV light intensity is 10 mW/cm2 unless otherwise stated. Download figure Download PowerPoint Prolonged UV irradiation ( Supporting Information Figure S6a) can cause the formation of HEDS through a process, involving the S–S cleavage in HEDT (BDE: 179 kJ/mol) and the radicals recombination (R3hv). In addition, long-term UV irradiation (1–2 h) can also dissociate the C–S (=S) bond (BDE: 193 kJ/mol), producing diethoxy trisulfidetrisulfide (HETS), sulfur, and N,N-diethylmethanethioamide (DDC) with the release of oxygen ( Supporting Information Figure S6). The rate constant for R3hv, which forms HEDS/HETS is much larger than the thionate disproportionation reaction (R3) that reforms HEDS without light (k3hv = 0.0438 L•M−1•s−1 vs k3 = 0.0014 L•M−1•s−1). When UV irradiation was applied for 40 min, the content of HEDS/HETS increased to 50% while the HEDT decreases to 5%. The HEDS/HETS produced through the R3hv process was active and stayed out of equilibrium due to the existence of DDDC (HEDS/HETS activation). In the absence of UV light, the sulfide can react with DDDC to reform HEDT through a thiolate anion intermediate in a slow oxidation reaction (R4) that consumes atmospheric oxygen (HEDS/HETS deactivation). The rate constant for this reaction, called R4, has been determined to be 5 × 10−3 M−2•s−2 using 1H NMR and HPLC ( Supporting Information Figure S7a and Figure 3c). To confirm the oxidation mechanism, the same reaction was conducted under a nitrogen atmosphere, and no change was observed. HEDS/HETS was reformed through the R3hv process with the application of UV light. And the input energy was dissipated at a lower rate through the deactivation reaction R4 (k4 = 5 × 10−3 M−2•s−2 > k3hv = 0.0438 L•M−1•s−1). The R3hv (activation) and R4 reactions (deactivation) formed a light-dissipating cycle, with DDC and sulfur as waste. When UV light was applied for 40 min, the concentration of DDDC remains constant during the R2hv and R3hv processes (Figure 3b). However, when the R3 reaction was conducted without light, the final products reached an equilibrium state with only HEDS remaining. SO2 and CO2 were produced slowly through the R3 process, and the formed acid decomposed the basic DDDC.48 The presence of SO2/CO2 with a molar ratio of 2:1 was demonstrated by precipitation assay ( Supporting Information Figure S8). As a result, the concentration of DDDC decreased slowly to 55% after 27 days. We also studied how continuous light exposure in addition to the irradiation-free condition (R4) affects the reaction pathway of the mixture obtained from R3hv. We found that prolonged UV irradiation slowly decomposed HEDS/HETS through a process called R4hv, producing photoproducts including MCE and sulfur as a result of S–S/S–C bond cleavage (k4hv = 1.57 × 10−4 s−1, Figure 3d and Supporting Information Figure S9). DDDC also slowly decomposed under UV light exposure (kdecomDDDC = 2.24 × 10−4 s−1, Supporting Information Figure S10). It is worth noting that similar photocleavage reactions also occur in mixtures obtained from the R3 process, although the compositions may be different. Overall, the complexity of the disulfide CRN is increased through the induction of new reaction pathways and light-driven reaction cycles with nonlinear kinetics. Simulation of the concentration of CRN components over time Using the identified reaction kinetics ( Supporting Information Table S2), we developed simple models in Pathon to quantitatively describe the behavior of the CRN based on the reaction equations (see Supporting Information Figures S3 and S11). The concentration of each component as a function of time was calculated. Figure 4a shows the concentration fluctuations of the components without light. The concentration of HEDS initially increased to 3.9 mM at 16 min, then decreased to 2.7 mM at 4 h, before increasing again to 5 mM at 278 h. The concentration of HEDS eventually reached a steady state after a single oscillation. With continuous light input, it is clear that a higher degree of control can be achieved over the evolution of the system. Full-time light irradiation significantly accelerated the process, completing it within 66 h, and changed the profile (Figure 4b). After two cycles of oscillation, the concentration of HEDS/HETS decreased. Light can also be selectively applied at user-defined intervals to regulate individual rates of the network without disrupting others. For example, when light irradiation was applied to the system for 4 h (after the disulfide metathesis reaction), the HEDS/HETS quickly increased to 4.5 mM in 8 h and then slowly decreased over 83 h (Figure 4c). More complex behaviors can also be created by applying light at specific intervals. Notably, when irradiation is applied in the range of 5–7.5 h, 28–36 h, and 139–145 h (R3hv) to drive the reformation of HEDS/HETS to high levels (4.5, 4.2, and 4.0 mM), the system underwent dissipative oscillation (Figure 4d). This significant difference of k3hv and k4 with a ratio of deactivation/activation time of 0.12 (2.5 h/20.5 h) is key to the dynamically controlled dissipating cycle. The system can be recycled at least three times before it decays because of photodegradation. Along with the fluctuation of HEDS/HETS, HEDT underwent the opposite fluctuation while the concentration of DDDC increased and decreased at varied rates. These profiles indicate that the emerging behaviors of our CRN in response to light are predictable and reprogrammable. Figure 4 | Simulations of the time-dependent concentration profiles of key intermediates (HEDS/HETS, blue; HEDT, orange; DDDC, green) in a mixture of TDS and MCE (10 mM/10 mM). (a) The intermediates concentration fluctuation in the dark. (b–d) User-defined reaction pathway selection enabled by light. The blue background represents UV light irradiation windows. HEDS/HETS produced through the R3hv and HEDT from R4 respectively represent the dissipating and equilibrium state. Download figure Download PowerPoint User-defined pathway selection in organo-hydrogels enabled by light regulation To visualize the user-defined pathway selection in CRN and demonstrate the generality, the organo-hydrogels starting with 4-arm-PEG-SH precursor (20 kDa, [–SH]: 0.02 mol/L) and TDS (0.02 mol/L) were prepared. Mixing these materials should result in the consecutive formation, dissociation, and rebonding of S–S/S–S–S cross-links along the reaction pathways, leading to multistate macroscopic autoevolution as predicted by our modeling. As shown in the inseted images in Figure 5, we indeed observed the complex autonomous evolution. Figure 5 | Programmable organo-hydrogel evolution following distinct pathways under light regulation. The starting materials are a thiol end group functionalized 4-arm-PEG polymer (4-arm-PEG-SH) precursor and TDS. Insets show images of a typical sample in an inverted quartz cuvette. Download figure Download PowerPoint Without UV exposure, the system underwent a four-state independent transition from sol (I) to a soft gel (I), back to sol (II), and then to a stiff gel over 90 h through the R1–R2–R3 pathway (Figures 5 and 6a). When UV irradiation was applied afterward, the system converted from a stiff gel to sol (V) in 70 h. In contrast, under continuous light irradiation, the system underwent autonomous five-state evolution: sol (I) → soft gel (I) → sol (II) → soft gel (II) → sol (V) over 75 h through the R1hv–R2hv–R3hv–R4hv pathways (k1hv > k2hv > k3hv > k4hv). The stiff gel formed through the R3 process reached an equilibrium state, while the sol (II) and soft gel (II) created through the R2 and R3hv processes remained out of equilibrium with lower moduli (Figure 6b). The autoevolution pathways were also influenced by the light intensity (Figure 6c). At 20 mW/cm2, five-state transitions were still observed, but with a shorter lifetime for the sol (II) state. Increasing the light intensity to 100 mW/cm2 results in transitions from soft gel (I) to soft gel (II) without an intermediate sol (II) state because the rate of reformation of S–S/S–S–S cross-links was greatly enhanced and ultimately outpaced the dissociation rate ( Supporting Information Figure S12). Figure 6 | (a and b) Time-dependent storage modulus (G′) and loss modulus (G″) of the system under varied irradiation conditions. (c) Transition of states in the system as a function of light intensity. (d) Time-dependent G′ of the system with different irradiation intensity. (e) Light intensity-dependent G′max of soft gel (II), second gelation rate and gel lifetime from (d). (f) Time-dependent G′ and G″ of the system undergoing light-driven dissipation cycles. The blue area indicates the presence of UV light. UV light intensity is 10 mW/cm2 unless otherwise stated. Download figure Download PowerPoint By selectively applying UV light, it was possible to achieve a five-state transition of sol (I) → soft gel (I) → sol (II) → soft gel (II) → sol (V) through the R1–R2–R3hv–R4hv pathways (Figures 5 and 6d). Light input offered the possibility of independently tuning the stiffness (G′max), gelation rate, and the lifetime of the soft gel (II). A higher dose of UV exposure (from 5 to 20 mW/cm2) resulted in a higher G′max, faster gelation rates, and shorter lifetime for the soft gel (II) (Figure 6e). The G′max of the soft gel (II) can be tuned in the range of 300–1500 Pa. The gelation rate was 0.011 Pa/s at 5 mW/cm2 but increased to 0.12 Pa/s at 20 mW/cm2. The lifetime decreases from 70 to 42 h with increasing light intensity. These results indicate that the evolution rate and mechanical properties of the gel can be selectively adjusted by the user-defined timing and duration of light exposure. To demonstrate the light-powered dissipative cycles, we conducted a final rheology experiment using light-induced cross-linking. By applying UV irradiation to the intermediate sol (II) state, a transient soft gel (II) was formed, as shown in Figure 6f. When the light was removed, the system returned to equilibrium through the R4 process (Figure 5). The cross-links were gradually dissociated as monitored with rheology. The reversible formation and dissociation of the S–S/S–S–S cross-linking sites are evident through at least three sol (IV)-soft gel (II) transitions during consecutive cycles of 2–3 h of UV light followed by 20 h in the dark. The decrease in the moduli (G′max) of the soft gel (II) in the final cycle was due to irreversible photodecomposition. However, transient materials can still be obtained that only function with an energy supply and break down spontaneously without energy input. Based on these results, we conclude that UV light exposure controls both autoevolution pathways and light-driven cycles, providing external perspectives to understand the adaption and evolution of organism systems in response to changing environments, which also inspire the design of dynamic materials with emergent behaviors that can be preprogrammed by the user-defined presence of light such as recyclable polymers and controlled drug delivery. Conclusion In summary, we have shown how light can control the reaction pathways in complex CRNs. Different from the anion-mediated CRN based on the disulfide bond, light can selectively trigger disulfide and C–N bond dissociation to varying degrees. New reactions with accelerated reaction rates were generated through radical-mediated mechanisms, resulting in specific reaction pathways and product compositions. The process is predictable, allowing for the creation of autonomous and adaptive materials with user-defined evolution pathways in CRN. Our work provides a simple method for regulating the reaction pathways in out-of-equilibrium CRN, which advances our understanding of how organism systems adapt and evolve in response to changing environments and the thiol-thiuram chemistry also provided a useful toolbox to design complex "living"/interactive materials for some interesting applications such as recyclable polymers and controlled drug delivery. Supporting Information Supporting Information is available and includes model reactions, kinetic studies, simulation, and rheology measurements (Figures S1–S12 and Tables S1 and S2). Conflict of Interest There is no conflict of interest. Acknowledgments This work was supported by the National Natural Science Foundation of China (grant no. 52073175), National Natural Science Foundation of Shanghai (grant no. 23ZR1442700), and ShanghaiTech University. The authors gratefully acknowledge the support from the Analytical Instrumentation Center (contract no. SPST-AIC10112914), SPST, ShanghaiTech University. References 1. van Roekel H. W. H.; Rosier B. J. H. M.; Meijer L. H. H.; Hilbers P. A. J.; Markvoort A. J.; Huck W. T. S.; de Greef T. F. A.Programmable Chemical Reaction Networks: Emulating Regulatory Functions in Living Cells Using a Bottom-Up Approach.Chem. Soc. Rev.2015, 44, 7465–7483. Google Scholar 2. Ashkenasy G.; Hermans T. M.; Otto S.; Taylor A. F.Systems Chemistry.Chem. Soc. Rev.2017, 46, 2543–2554. Google Scholar 3. Wong A. S. Y.; Huck W. T. S.Grip on Complexity in Chemical Reaction Networks.Beilstein J. Org. Chem.2017, 13, 1486–1497. Google Scholar 4. Shinar G.; Feinberg M.Structural Sources of Robustness in Biochemical Reaction Networks.Science2010, 327, 1389–1391. Google Scholar 5. Semenov S. N.; Wong A. S. Y.; van der Made R. M.; Postma S. G. J.