Supramolecular Chiral Nanozymes with High and Switchable Enantioselectivity

Understanding and manipulation of catalytic enantioselectivity have emerged as a paramount challenge for decades. Inspired by nature, nanozymes with enantioselectivity have been designed. However, the role of the spatial arrangement interplayed in enantioselective catalysis is rarely paid attention, let alone the manipulation of enantioselectivity switch under this context. Herein, the supramolecular chiral nanozymes composed of P/M‐polyaniline (P/M‐PANI) nanotwists and Fe3O4 nanoparticles without any chiral molecules are constructed. Taking the catalytic oxidation of 3,4‐dihydroxy‐d / l‐phenylalanine (d / l‐DOPA) as a model reaction, P‐PANI–Fe3O4 nanozymes are demonstrated to show better catalytic efficiency toward D‐DOPA, whereas M‐PANI–Fe3O4 nanozymes show preference to L‐DOPA. Intriguingly, through the simple modulation of the Fe3O4 nanoparticle density on P/M‐PANI–Fe3O4, the unprecedented enantioselectivity switch of the as‐designed nanozymes is achieved. Moreover, the select factors of P/M‐PANI–Fe3O4 toward d / l‐DOPA are determined to be significantly larger than that of commonly used chiral ligands (such as d / l‐phenylalanine)‐modified Fe3O4 nanozymes, indicating the superiority of spatial arrangement‐dominated supramolecular chirality over molecular chirality in enantioselectivity. This finding discloses the role that the spatial arrangement‐directed chiral scaffolds interplay in the enzymatic catalytic process and open a new avenue for the facile design of smart nanozymes with controlled enantioselectivity.

the doped emeraldine salt state and the preferable helical polymer with desired optical activity. [27,28] Moreover, the chirality of the PANI obtained by the aforementioned method can be maintained based on macromolecular memory even after the complete removal of the CSA, which can provide supramolecular chiral scaffold in the absence of chiral molecule. [29,30] Based on these, here we report the successful synthesis of the supramolecular chiral nanozymes constructed from P/M-PANI nanotwists and Fe 3 O 4 nanoparticles without any chiral molecules and demonstrate their outstanding capability of enantioselective catalysis toward 3,4-dihydroxy-D/L-phenylalanine (D/L-DOPA) (Scheme 1). In comparison with the commonly used chiral ligands, such as D/L-phenylalanine (D/L-Phe)-modified Fe 3 O 4 nanozymes, the spatial arrangement-mediated supramolecular chirality is demonstrated to be superior in enantioselectivity, confirming its crucial role in asymmetric catalysis. Intriguingly, through the regulation of the amount of Fe 3 O 4 nanoparticles on P/M-PANI, the reversed enantioselectivity can be achieved. To the best of our knowledge, this is the first example of the nanozyme with its intrinsic active center density-guided switchable enantioselectivity.

Results and Discussion
To construct supramolecular chiral scaffolds, the polymerization of aniline was first conducted in the presence of D/L-CSA as chiral inducer and ammonium persulfate as oxidant. The twisted nanofibers with inversed directions of PANI-CSA were observed when the CSA with opposite chirality was employed ( Figure S1, Supporting Information), which agrees well with previous reports. [31,32] Subsequently, through dedoping in ammonia solution, the CSA was completely removed and the twisted nanofibers of PANI were maintained. However, the twisted nanofibers of PANI cannot effectively stabilize metal nanoparticles. Given that the carboxylate group of thioglycolic acid (TA) can be employed to interact with PANI and the thiol group of TA can be used to stabilize metal nanoparticles, [33,34] the TA was encapsulated into PANI. Intriguingly, the twisting feature was maintained for PANI-TA nanofibers and identical with that of original PANI-CSA. Specifically, as shown in Figure S1, Supporting Information, with the initial induction of D-CSA, right-handed helical nanostructures (P-PANI-TA) were obtained, whereas left-handed helical nanostructures (M-PANI-TA) can be observed with the addition of L-CSA. In addition, the PANI-TA prepared with HCl as acidic media without any chiral induction process was also obtained as control. In this case, as revealed from the transmission electron microscopy (TEM) images ( Figure S1, Supporting Information), straight nanofibers without any helical tendency were obtained. Although the resulting morphologies were different in chirality, the molecular structure of the three kinds of PANI-TA was demonstrated to be same, as evidenced by their identical ultravioletvisible (UV-Vis) spectra ( Figure S2, Supporting Information). From the circular dichroism (CD) spectra ( Figure S2, Supporting Information), the P-PANI-TA showed intensive negative signal centered at 450 nm, whereas the M-PANI-TA was demonstrated to exhibit a mirror imaged positive peak, which clearly indicates that the P-PANI-TA and M-PANI-TA were endowed with opposite supramolecular chirality. In addition, the CD spectrum of PANI-TA was measured as well and the absence of evident signals in the range of 200-800 nm suggests its achiral nature. Considering the absence of any chiral molecules and the identical molecular structure of the nanomaterials, the discrepancy of the various PANI-TA in chirality can be rationally ascribed to the versatile spatial arrangement encoded asymmetric information. Next, Fe 3 O 4 nanoparticles (known as the first discovered nanozyme with peroxidase activity [35][36][37] ) were prepared and protected with 3,4-dihydroxyhydrocinnamic acid (DHCA) according to the previous publication. [38] As shown in the TEM images ( Figure S3a, Supporting Information), homogeneous spherical particles with an average diameter of %20 nm were observed. From the UV-Vis spectrum ( Figure S3b [39,40] Furthermore, X-ray diffraction (XRD, Figure S5, Supporting Information) was carried out to exhibit the simultaneous presence of the characteristic peaks of PANI and Fe 3 O 4 for P-PANI-Fe 3 O 4 . Meanwhile, the X-ray photoelectron spectroscopy (XPS) was conducted for P-PANI-Fe 3 O 4 , and the Fe 2p signals were clearly detected ( Figure S6a, Supporting Information). From the S 2p spectrum ( Figure S6b, Supporting Information), in comparison with the S signals from the -SH group, the ones from the SO 3 2À (arising from CSA) are quite low. Furthermore, considering that the total atomic ratio of S element from P-PANI-Fe 3 O 4 is as low as 3.13% (Table S1, Supporting Information), it is safe to conclude that the CSA molecules have been completely removed in the currently designed catalytic systems.
As revealed from the UV-Vis spectra (Figure 1k), it is quite straightforward that the absorption intensity of all the three PANI-Fe 3 O 4 composites centered at 350 nm has been evidently increased, in comparison with that of PANI-TA. This result can be attributed to the successful conjugation of Fe 3 O 4 nanoparticles. With the careful speculation of Figure 1a-c, the P-PANI-Fe 3 O 4 was demonstrated to be featured with right-handed helical configuration and the M-PANI-Fe 3 O 4 was featured with lefthanded helical tendency, of which the twisting direction was identical with that of P-PANI-TA and M-PANI-TA. It is also noteworthy the distribution of Fe 3 O 4 nanoparticles even went along the specific direction of PANI (Figure 1b-d), providing the possibility of chirality transfer from the chiral scaffold of PANI to Fe 3 O 4 nanoparticles. [41,42] To verify this assumption, the CD spectra of P-PANI-  www.advancedsciencenews.com www.small-structures.com reasonable to assume that the newly emerging CD signals were originated from the chirality transfer from chiral PANI architectures to Fe 3 O 4 nanoparticles through the connecting of TA.
With the successful combination of chiral environment and catalytic center, it is rational to speculate that the PANI-Fe 3 O 4 nanomaterials with different chiral configurations may be endowed with enantioselective catalytic activity. Thus, the catalytic performance of the various PANI-Fe 3 O 4 nanomaterials was then tested. Given the fact that DOPA is a crucial biomolecule and its clinical function for treating Parkinson's disease is closely related to its chirality, [43,44] the selective oxidation of DOPA was taken as a model reaction for the exploration of the enantioselective catalysis capability of PANI-Fe 3 O 4 nanomaterials. As the oxidation product of DOPA (i.e., dopachrome) is featured with a characteristic UV-Vis absorption peak centered at 475 nm, the progress of this reaction can be evaluated through the monitoring of the UV-Vis absorption intensity at 475 nm. [19] First, the catalytic performance of P/M-PANI-Fe 3 O 4 with 5 wt% Fe 3 O 4 -loading amount (denoted as P/M-PANI-Fe 3 O 4 (5)) toward D/L-DOPA was tested in the presence of H 2 O 2 , respectively. After 6 min reaction time, the UV-Vis absorption of the mixture was subsequently measured (Figure 2c). Surprisingly, when P-PANI-Fe 3 O 4 (5) was mixed with D-DOPA, the resulting absorption intensity at 475 nm was significantly higher than that of L-DOPA, suggesting the better catalytic efficiency of P-PANI-Fe 3 O 4 (5) toward D-DOPA. In contrast, when M-PANI-Fe 3 O 4 (5) was employed as nanozyme and L-DOPA was taken as substrate, at the same condition, the UV-Vis absorption intensity was demonstrated to be %2 times higher than that of using D-DOPA as substrate, indicating that the M-PANI-Fe 3 O 4 (5) showed preference to L-DOPA. This exciting preliminary result prompted us to further investigate the kinetic performance of the catalytic process. Therefore, the time-dependent change of the intensity of the UV-Vis absorption at 475 nm was recorded in the time span of 6 min. Through plotting the absorption intensity as function of time (Figure 2a,b), the catalysis of P-PANI-Fe 3 O 4 (5) toward D-DOPA is significantly faster than that of L-DOPA, while the enantioselectivity was completely reversed when M-PANI-Fe 3 O 4 (5) was employed under the identical conditions. In addition, with PANI-Fe 3 O 4 (5), no catalytic preference was observed for D-DOPA and L-DOPA ( Figure S7, Supporting Information). Taken together, the aforementioned results well clarified that the supramolecular chiral scaffolds dominated by spatial arrangement can imitate the enantioselectivity of natural enzymes.
To better understand the enantioselectivity of P/M-PANI-Fe 3 O 4 , the kinetic parameters were further calculated according to the saturation curves (Figure 2d,e) and Michaelis-Menten equation. As provided in Table 1, the calculated K M for P/M-PANI-Fe 3 O 4 (5) toward D/L-DOPA was similar in contrast to L/D-DOPA. However, the K cat , which represents the catalytic activity of nanozyme to specific substrate, for the P/M-PANI-Fe 3 O 4 (5) showed a remarkable preference for D/L-DOPA. In addition, K cat /K M , which is a commonly used parameter for the evaluation of catalytic efficiency, was calculated as well and the obtained result further verified the better catalytic efficiency of P/M-PANI-Fe 3 O 4 (5) toward D/L-DOPA than L/D-DOPA. In further, to quantify the enantioselectivity of the   (1)). [45] As expected, the activation energy of P/M-PANI-Fe 3 O 4 (5) toward D/L-DOPA was demonstrated to be lower (Figure 2f and S8, Supporting Information), in comparison with L/D-DOPA. However, under the same condition, no obvious difference in activation energy was found for the catalysis of PANI-Fe 3 O 4 (5) toward D/L-DOPA ( Figure S8, Supporting Information). In some way, the discrepancy in activation energy is matched with the catalytic preference of the chiral nanozymes.
Considering that the enantioselective nanozymes constructed herein were composed of magnetic Fe 3 O 4 nanoparticles, their recycling performance was then investigated. First, the vibrating sample magnetometer (VSM) analysis was conducted with P-PANI-Fe 3 O 4 (5) as representative case. As revealed from Figure 3a, the hysteresis loop of P-PANI-Fe 3 O 4 showed no remanence or coercivity, suggesting its superparamagnetic nature. According to the VSM results, the saturation magnetization of P-PANI-Fe 3 O 4 (5) was determined to be 0.98 emu g À1 . Encouraged by the aforementioned result, the catalysis toward D/L-DOPA was repeated after re-collecting the P/M-PANI-Fe 3 O 4 using a magnet. As shown in Figure 3b, though the catalytic activity was slightly decreased after six cycles, no evident attenuation in enantioselectivity was observed as the discrepancy in absorption intensity at 475 nm was kept almost identical after recycling, which endowed P/M-PANI-Fe 3 O 4 (5) nanozymes with high potential in practical application.
According to our previous report, chiral PANI scaffolds with abundant functional groups including amino and secondary amino groups, can adsorb amino acid-based enantiomers in enantioselective manner through the typical doping process. [46,47] Therefore, to deeply understand the enantioselective catalysis in the molecular level, the enantioselective adsorption experiments were performed with P-PANI-Fe 3 O 4 (5) as a representative case. Specifically, the P-PANI-Fe 3 O 4 (5) and racemic DOPA solution were mixed for 4 h, followed with filtration to obtain the DOPA filtrate. The UV-Vis absorption and CD spectra of the resulting filtrate were then measured. As shown in Figure S9, Supporting Information, the characteristic CD spectrum of D-DOPA was acquired, which indicates the preferred adsorption of L-DOPA on the surface of P-PANI-Fe 3 O 4 (5). According to the calibration curves of UV-Vis and CD spectra ( Figure S10, Supporting Information), the enantiomeric excess (ee%) value was demonstrated to be 52.4%. Considering that the coverage ratio of Fe 3 O 4 nanoparticles was extremely low (5 wt%) according to the TEM image ( Figure S11a, Supporting Information), it is reasonable to speculate that the excessive absorption of L-DOPA by P-PANI-Fe 3 O 4 (5) was attributed to the enantioselective interaction between supramolecular chiral scaffold of PANI and   Figure S11b The enantioselectivity switchable supramolecular nanozymes presented herein could be indicative of a more universal phenomenon. In general, the combination of chiral environment and catalytic center is a prerequisite for the construction of nanozymes with enantioselectivity. If the enantioselective adsorption of chiral scaffolds toward enantiomeric substrates takes place, imbalanced distribution between substrates with opposite configuration will exist both in solution and on the surface of chiral scaffolds. Considering the uneven distribution, the discrepant accessibility of the enantiomers toward the catalytic center is the crucial factor that leads to the asymmetric catalysis as well. [48,49] Therefore, through the manipulation of the density of catalytic center, the favorable accessibility of the excessive enantiomer in solution or on the surface of chiral nanostructures can be modulated and the enantioselectivity can be switched as desired ( Figure 5). Undoubtedly, the discovery of the enantioselectivity switchable capability offers additional superiority for spatial arrangement-dominated chiral scaffolds.
For the enantioselective catalysis, especially the ones related to nanozymes, the asymmetric information was mostly originated from molecular chirality. Amino acids, such as Phe, were mostly employed for constructing the chiral environment and achieved high enantioselectivity. [17][18][19] As in the present study, the  supramolecular arrangement-dominated chiral scaffolds were demonstrated to offer outstanding enantioselectivity as well, the debate about the superiority between molecular chirality and spatial arrangement directed supramolecular chirality in the perspective of enantioselective catalysis was raised spontaneously. To clarify the argument, Fe 3 O 4 nanoparticles modified with commonly used chiral ligand of D/L-Phe were fabricated. As revealed from the TEM images (Figure 6a), the fabricated nanoparticles were homogeneously distributed with an average diameter of %20 nm. In addition, according to the UV-Vis spectra shown in Figure S13, Supporting Information, the characteristic peak of Phe was observed for Fe 3 (Table S3, Supporting Information), which is much lower than that obtained for P/M-PANI-Fe 3 O 4 (5). Considering that the catalytic center, namely, the Fe 3 O 4 core, and the related weight of Fe 3 O 4 nanoparticles was kept the same for both P/M-PANI-Fe 3 O 4 (5) and Fe 3 O 4 @D/L-Phe, the discrepancy in enantioselectivity can only be ascribed to the differentiation in the origin of chirality. Altogether, the case study presented herein indicated the superiority of spatial arrangement-dominated supramolecular chirality over molecular chirality in enantioselective catalysis.

Conclusion
In summary, we reported the successful design and construction of P/M-PANI-Fe 3 O 4 chiral nanozymes with the chiral supramolecular spatial arrangement without any chiral molecules. With the combination of the chiral environment provided by the PANI scaffolds and the catalytic center of Fe 3 O 4 nanoparticles, the enantioselective catalysis of P/M-PANI-Fe 3 O 4 (5) toward D/L-DOPA was actively achieved, clarifying the asymmetric information encoded in the supramolecular arrangement can be the origin for enantioselectivity. In addition, owing to the enantioselective adsorption of PANI scaffolds toward D/L-DOPA, the uneven distribution of D/L-DOPA existed both in solution and on the surface of chiral PANI scaffolds. Through the manipulation of the density of Fe 3 O 4 as catalytic centers loaded, the favorable accessibility between excessive enantiomers in solution or on the surface of chiral PANI was directly regulated, and the unprecedented enantioselectivity switch in enantioselective catalysis process has been achieved. The interesting discovery presented herein revealed the role that the spatial arrangement-directed chiral scaffolds interplayed in the enzymatic process and opened an alternative avenue for the design of smart nanozymes with controlled enantioselectivity.
Instruments and Characterization: Morphologies of the as-prepared nanomaterials were characterized with field-emission scanning electron microscope (FESEM, 300, ZEISS), TEM (HT7800), and HRTEM (Tecnai G2 F30 S-Twin TEM, FEI). The UV-Vis absorption spectra were recorded using UV-2550 UV/Visible spectrophotometer (JASCO International Co., Ltd.). The CD analysis was performed on a JascoJ-810 CD spectropolarimeter (JASCO International Co., Ltd.) with a resolution of 1 nm. The phase composition was measured using an Axis Ultra XPS (Kratos Analytical Ltd.) equipped with a standard monochromatic Al Kα source (hv = 1486.6 eV). The crystal phase was analyzed using XRD using a Bruker AXS D8 ADVANCE X-ray diffractometer.
Synthesis of P-PANI-D-CSA or M-PANI-L-CSA: The P-PANI-D-CSA or M-PANI-L-CSA was prepared according to the previous publication. [25] Specifically, 0.2 g (2.1 mmol) of aniline, 3.252 g of D-CSA or L-CSA (13.9 mmol), and 0.063 mmol of oligomer were dissolved in 1.5 mL of water. Subsequently, the aqueous solution of ammonium persulfate (2.1 mmol) was added incrementally in five separate portions to the mixture of aniline and CSA. After each addition, the mixed solution was shaken vigorously for 30 s. The reaction mixture was left standing for 20 h at room temperature. Then, the resulting crude product was washed with water and ethanol for further use.
Synthesis of PANI-HCl: The PANI-HCl was prepared according to a typical procedure. [50] Specifically, 0.6 mL of aniline and 0.36 g of ammonium persulfate were dissolved in 20 mL of 1.0 M HCl solution, respectively. Then, both solutions were poured rapidly into a 50 mL glass vial and shaken vigorously for %30 s. The resulting mixture was left still for 24 h. The products obtained were thoroughly purified with water for further use.
Synthesis of Hydrophobic Fe 3 O 4 Nanoparticles: The Fe 3 O 4 nanoparticles were prepared according to the previous publication. [51] First, to prepare iron oleate, 10.8 g of iron chloride and 36.5 g of oleate sodium were dissolved in a mixed solvent consisting of 60 mL of distilled water, 80 mL of ethanol, and 140 mL of hexane. Then, the mixed solution was heated at 60°C and kept refluxing for 4 h. After the reaction, the upper organic layer was washed with 90 mL of distilled water for three times. After the completely removal of solvent using a rotary evaporator, the iron oleate was obtained as reddish-brown viscous oil. Subsequently, 2.14 g of iron oleate and 0.379 g of oleic acid were dissolved in 10.6 g of 1-octadecene at room temperature under N 2 atmosphere. After 32 min, the solution was heated at 320°C (%10°C min À1 ) and kept at this temperature for 60 min. Then, the mixture was cooled to room temperature and washed with a mixture of hexane and acetone for three times, resulting in the hydrophobic Fe 3 O 4 nanoparticles. The as-obtained hydrophobic Fe 3 O 4 nanoparticles were then dispersed in THF (36 mL) for further use.
Ligand Exchange of Hydrophobic Fe 3 O 4 Nanoparticles with DHCA: The hydrophilic Fe 3 O 4 nanoparticles were prepared according to the previous literature. [38] Here, 50 mg of DHCA was dissolved in 6 mL of THF at 50°C. Then, the THF dispersion of hydrophobic Fe 3 O 4 nanoparticles (6 mL) was added dropwise. After stirring for 4 h, the reaction solution was cooled to room temperature. To precipitate the Fe 3 O 4 nanoparticles, 500 μL of NaOH (0.5 M) was added to the solution. The precipitate was collected by centrifugation (4000 rpm) and the product was dispersed in water for further use.
Preparation of Supramolecular Chiral P/M-PANI-Fe 3 O 4 and PANI-Fe 3 O 4 : The as-prepared P-PANI-D-CSA, M-PANI-L-CSA, and PANI-HCl were dispersed into 150 mL of 0.1 M NH 3 ·H 2 O, separately. After stirring at room temperature for 4 h, the P-undoped PANI, M-undoped PANI, and undoped PANI were obtained by washing with deionized water for four times. The asobtained products were then dispersed in deionized water for further use.
Here, 20 mg of P-undoped PANI, M-undoped PANI, and undoped PANI were separately dispersed in 10 mL deionized water followed with the addition of 10 mL of 1 M TA solution. After stirring for 4 h at room temperature, the P-PANI-TA, M-PANI-TA, and PANI-TA were collected by centrifugation after washing with water for three times. The as-obtained products were then dispersed in 20 mL deionized water for further use.
The as-prepared P-PANI-TA, M-PANI-TA, and PANI-TA, which separately dispersed in 20 mL deionized water, were mixed with the aqueous solution of 1 mg Fe 3 O 4 (10 mL). After stirring at room temperature for 4 h, the P/M-PANI-Fe 3 O 4 and PANI-Fe 3 O 4 were obtained by centrifugation after thoroughly washing with deionized water.