Preparation of bowl-shaped polydopamine surface imprinted polymer composite adsorbent for specific separation of 2′-deoxyadenosine

Pan Wang,Mengdei Zhou,Zhuangxin Wei,Lu Liu,Tao Cheng,Xiaohua Tian,Jianming Pan

School of Chemistry and Chemical Engineering,Jiangsu University,Zhenjiang 212013,China

Keywords:Molecularly imprinted polymers (MIPs)Bowl-shaped Base complementary pairing interaction Selective separation 2′-deoxyadenosine(dA)

ABSTRACT Molecularly imprinted polymers (MIPs) have great potential as adsorbents for selective adsorption and separation of nucleoside compounds,but effectively enhancing the affinity of recognition sites by adjusting the forces between template molecules and functional monomers remains an important challenge.In this work,a surface imprinting strategy was used to construct bowl-shaped molecularly imprinted composite sorbents(BHPN@MIPs)based on polydopamine(PDA)particles and have achieved selective separation and purification of 2′-deoxyadenosine(dA).Where by the base complementary pairing interaction of the combined template molecule dA and the pyrimidine functional monomer can enhance the preassembly force,and the hydrophilic bowl-shaped PDA can provide a larger storage space contact efficiency of dA in the test solution,causing the site utilization much higher and improving the kinetic adsorption performance.The equilibrium adsorption time and maximum adsorption capacity of 60 min and 328.45 μmol.g-1 were observed by static adsorption experiments,and the selectivity experimental results showed an imprinting factor IF of 1.30.After four adsorption-desorption cycles,the initial adsorption equilibrium adsorption capacity of BHPN@MIPs still retained 91.14%.By evaluating the selective adsorption of dA in spiked human serum solutions,BHPN@MIPs can be used to selectively enrich and analyze target dA in complex biological samples.

1.Introduction

2′-Deoxyadenosine(dA)is a purine nucleoside with good physiological activity,and it is an important raw material or intermediate for many nucleoside analogues such as antitumor and AIDS drugs [1-3].In addition,dA derivatives are an important component of nucleic acids,which are closely related to signal transduction,metabolism of cells and many enzymatic reactions,and have been recognized as potential markers for many diseases [4-6].So,there is a great need to separate and enrich the target dA before qualitative analysis and quantification.However,impurities in the raw materials severely limit the pharmacological studies and the further use of high-quality dA and its derivatives[7].The detection method for 2′-deoxyadenosine includes high-performance liquid chromatography(HPLC)with UV,mass spectrometry(MS),and capillary electrophoresis(CE)detection[8].The methods of dA separation commonly employed at home and abroad include column chromatography [9],crystallization [10] and solvent extraction[11],but there are problems such as high separation conditions and slow generation rate,where adsorption may be more suitable to separate dA because of the simple operation,low cost and good selectivity.However,the development of new adsorbents with excellent selectivity remains a challenge,given the diversity of biological matrices [12].Therefore,it is crucial to develop adsorbents with high affinity,high selectivity and fast kinetics.

Molecularly imprinted polymers (MIPs)have the advantages of simple preparation,high recognition and stability in a certain pH and temperature range and high selectivity for target molecules,making them become one of the most important media in the field of adsorption and separation [13-15].Surface molecular imprinting refers to graft polymer on the surface of micro-and nanocarriers,which is a limited method to improve mass transfer efficiency and adsorption capacity of imprinted adsorbents[16-18].Conventionally,materials such as polydopamine(PDA)particles[19],carbon nanotubes [20],mesoporous carbon or silica nanoparticles[21-23],metal-organic frameworks [24,25],and superparamagnetic Fe3O4particles[26]have been developed as substrates to fabricate MIPs.Recently,PDA and its derived materials have existence of abundant active functional groups such as catechol,amine and aromatic groups,it has a strong scavenging capacity toward numerous moleculesviaa versatile set of interaction mechanisms including coordination or chelation,electrostatic interaction,hydrogen bonding,or π-π interactions [27-29].However,most PDA materials still have several disadvantages,such as low surface area with insufficient porosity,severe aggregation,and poor stability under acidic conditions.We develop hollow bowl-like PDA,which have a large surface area with porous and strong chemical stability in a wide pH range.And the bowl-shaped PDA has extra and large internal space on the nanometer scale,which can improve the efficiency of mass transfer as a sorbent [30].

Nucleoside molecules have abundant functional groups such as purine,amino,cis-diols,and phosphate moieties[31-33].Recently,Bernadetteetal.[34]immobilized template molecularly imprinted polymers by metal chelation of the phosphate portion of nucleotides with functional monomers to prepare MIPs.To improve the interaction between nucleoside molecules and functional monomers,the Liu group [35] used the boron-affinity covalent and hydrogen-bonded non-covalent bonding properties of 3-acryloxyphenylboronic acid (AAPBA) and acrylic anhydride with adenosine aptamers for copolymerization to produce MIPs with better selectivity.Panetal.[33]used the enhanced affinity interaction through the base-complementary pairing of the template molecule adenosine monophosphate with the functional monomer 1-(vinylbenzyl) thymine (VBT) to improve the intermolecular interaction between the imprinting recognition site and the template molecule.Therefore,it is very important to choose the appropriate affinity force for the formation of imprinting recognition sites.

In this work,we prepared surface-imprinted adsorbents on the surface of bowl-shaped PDA nanoparticles (BHPN@MIPs) by a surface molecular imprinting strategy that used base complementary pairing between adenine of dA and uracil of the functional monomer 5-(2-carbomethoxyvinyl)-2′-deoxyuridine (AcrU) and the affinity of dopamine.First,PDA-coated silica particles (PDA@SiO2)were prepared by a simple one-step St?ber method and hydrolysis of tetraethyl silicate,then bowl-shaped PDA (BHPN) was obtained by hydrothermal treatment and HF etching,and the initiator Br atoms were preserved on modified surface of the particles with amino groups.Then,the pre-synthesized AcrU with complementary thymine bases was used as the functional monomer.After pre-assembly by forming hydrogen bonds with the adenine bases of dA,the imprinted polymer of dA was grafted onto the BHPN surface by ARGET ATRP construction to obtain the composite adsorbent BHPN@MIPs.In the latter process,the adsorption equilibrium,kinetics,selectivity and regeneration of this adsorbent were evaluated through static adsorption experiments.

2.Experimental

2.1.Materials and equipments

Dopamine(DA,98%),tetraethylorthosilicate(TEOS,99%),N,N,N’,N’’,N’’’-pentamethyl diethylenetriamine (PMDETA,99%),ascorbic acid (VC,99%),2′-deoxycytidine (dC,99%),2′-deoxyadenosine(dA,99%),adenosine 5′-monophosphate disodium salt (AMP,97%) and dimethyl sulfoxide (DMSO,99.5%) were purchased from Aladdin (Shanghai,China).α-Bromoisobutyryl bromide (BIBB,98%),triethylamine (TEA,98%),acetonitrile (98%),copper bromide monohydrate (CuBr2.H2O,99 %),tetrahydrofuran,ammonia (NH3-.H2O,28% (mass)),disodium hydrogen phosphate (NaH2PO4,99 %) and sodium dihydrogen phosphate(K2HPO4,99%) were purchased from Sinopharm Chemical Reagent Co.,Ltd.(Shanghai,China).Ethyl glycol dimethacrylate(EGDMA,97%)and hydrofluoric acid(HF,40%(mass))and adenosine triphosphate(ATP,98%)were obtained from Macklin Chemical Reagent Co.,Ltd.(Shanghai,China).All reagents were not purified before use and deionized water was obtained through a Milli-Q water-purification system.The details regarding the characterization instruments used in this study were provided in Supplementary Material (SM).”

2.2.Synthesis of bowl-shaped polydopamine particles (BHPN)

Ethanol(12 ml),deionized water(40 ml)and ammonia(0.5 ml)were mixed in a beaker and stirred for 1.0 h.Then TEOS (0.5 ml)was added to the mixed solution and stirred for 30 min.Subsequently,dA(0.2 g)was added into the mixture solution and stirred for 24 h,and the mixture was transferred into the autoclave.The autoclave was finally put into an oven at 140 °C for 24 h.The PDA@SiO2nanoparticles were collected by centrifugation and washed with water and ethanol alternately to make the supernatant clear.Finally,the PDA@SiO2nanoparticles were mixed in HF (40%) for 2.0 h at room temperature,washed with plenty of water until the supernatant became neutral,and dried under vacuum at 40 °C to give the product BHPN.

2.3.Study of binding interactions between functional monomers AcrU and template molecule AMP

Interactions between AcrU and the template molecule dA were determined by titrating an increasing amount of the function monomer AcrU into a constant amount dA,the mixture titration solutions were conducted by ultraviolet-visible spectrometry(UV-Vis) adsorption measurements,all UV-Vis titrations were performed carried out in PBS solution [33].The procedure was described as follows: the dA/AcrU mixtures with the molar ratios 50:10,50:20,50:30,50:40,50:40,50:50,50:60,50:70,50:80,50:90,50:100 were prepared in centrifuge tubes and assembled in the dark at room temperature for 1.5 h,the respective adsorption peaks were measured with a UV-Vis spectrophotometer.This was used to find the optimal binding ratio of dA and AcrU.

2.4.Synthesis of bowl-mounted surface-imprinted adsorbents(BHPN@MIPs)

Firstly,bromine modified bowl-shaped polydopamine particles(BHPN@Br) were prepared as follows: 0.3 g of BHPN and 1.0 ml of TEA were dispersed in 20 ml of THF,and nitrogen was added to the ice water bath for 30 min to remove oxygen.Subsequently,0.6 ml of BIBB was added into the mixture solution under magnetic stirring,and the solution was stirred under nitrogen protection at 0 °C for 2.0 h.Afterwards,the obtained product was washed with THF and deionized water two times respectively and dried by freeze-drying.

Secondly,BHPN@MIPs were preparedviaARGET ATRP.The detailed steps were listed as follows: The functional monomer AcrU was synthesized by a previously reported method [2],which structure was confirmed by 1H NMR (Supplementary Material,Fig.S1).After then,0.054 g (0.2 mmol) of dA and 0.087 g(0.28 mmol) of AcrU were dissolved in a mixture solution that DMSO/acetonitrile(2.0 ml/18 ml,volume ratio),after purging with N2for 30 min and in dark at 35°C for 12 h to assembly.Then,0.1 g of BHPN@Br and 280 μl of EGDMA were added into the above mixture under stirring for 30 min.Subsequently,10 mg of CuBr2.H2O and 0.012 ml of PMDETA and 20 mg of ascorbic acid were added into the mixture solution,after purging with N2for 30 min,then preheated at 80°C for 12 h.Then the product collected by centrifugation was washed with methanol to remove the unreacted reducing agent and the template molecules were eluted with deionized water/acetic acid (9:1,volume ratio) until no AMP was tested by UV-Vis in the eluent.Finally,the purified BHPN@MIPs were dried under vacuum at 45 °C for 24 h.As compared,non-imprinted BHPN-based polymers(BHPN@NIPs)were also synthesized in parallel without the addition of dA.

2.5.Batch mode adsorption experiments

The adsorption behavior of dA on the BHPN@MIPs,including adsorption kinetics,concentration and competition compounds at room temperature (25 ± 0.2) °C,was determined by batch adsorption experiments [36,37].In general,the optimal pH environment for dA adsorption is neutral,while samples such as human serum and urine are generally at room temperature and physiological pH.Therefore,a phosphate buffer solution (PBS) with pH=7.4 and an environment temperature of 25 °C was chosen for the adsorption tests of BHPN@MIPs and BHPN@NIPs.

The equilibrium binding of dA experiment: 2.0 mg of BHPN@MIPs and BHPN@NIPs as sorbents were added to the centrifuge tube,and then added to 2.0 ml of PBS with constant dA solution of the initial concentration ranging from 10 to 1000 μmol.L-1for 2.0 h in a thermostatic water bath oscillation at 25 °C.After incubating for the desired time,the BHPN@MIPs or BHPN@NIPs were both collected by centrifugation,and the supernatant was immediately filtered through a microporous nitrocellulose membrane(pore size 0.27 μm)to remove suspended powders.Then the amount of dA in the filtrate was determined using UV-Vis at 259 nm.This adsorption process has to be carried out in triplicate,the adsorption capacityQe(μmol.g-1) was calculated according to the Eq.(1) [38].

where theC0andCe(μmol.g-1) are the initial and equilibrium concentration of dA in PBS solution,respectively.V(ml) stands for the volume of dA solution,andM(mg) is the mass of adsorbents.

The binding kinetic study of the dA experiment: 2.0 mg of BHPN@MIPs and BHPN@NIPs as adsorbents were added to the centrifuge tube,then added to 2.0 ml of PBS with constant AMP solution of the initial concentration is 300 μmol.L-1,the samples were taken at different time intervals (i.e.5 to 120 min) in a thermostatic water bath oscillation at 25 °C.After incubating for the desired time,both the BHPN@MIPs or BHPN@NIPs were collected by centrifugation,and the supernatant was immediately filtered through a microporous nitrocellulose membrane (pore size 0.27 μm) to remove suspended powders.Then the amount of dA in the filtrate was determined using UV-vis at 259 nm.By using the concentrationCtof the standing timetinstead ofCe,the adsorption capacityQt(μmol.g-1) at timetwas also calculated according to the Eq.(1).

The binding selectivity of the BHPN@MIPs:To test the selectivity recognition of dA,2.0 mg of BHPN@MIPs and BHPN@NIPs were added to 10 ml centrifuge tube,each containing 2.0 ml of PBS solution (pH=7.4) containing 300 μmol.L-1concentration of dA,dC,and AMP,respectively.After shaking at 25°C for 2.0 h,BHPN@MIPs and BHPN@NIPs were collected by centrifugation,and then the supernatant was filtered through a microporous nitrocellulose membrane (pore size of 0.27 μm).The amount of each compound in the filtrate was then determined by UV-Vis spectroscopy,with the adsorption wavelengths of dA,dC,AMP and ATP being 259,253,270 and 260 nm,respectively.The specificity of BHPN@MIPs towards dA was estimated by imprinting factor (IF) depending on the following Eq.(2).

whereQBHPN@MIPsandQBHPN@NIPs(μmol.g-1) stand for the adsorption amounts of the template and the competitive compounds onto BHPN@MIPs and BHPN@NIPs,respectively.

2.6.Regeneration experiment

In this work,four cycles of adsorption/desorption experiments on BHPN@MIPs were performed.Briefly,2.0 mg of BHPN@MIPs were added into 10 ml centrifuge tube,which containing 2.0 ml of PBS solution (pH=7.4) with 300 μmol.L-1concentration of dA.After 2.0 h of shaking,the concentration of the dam in the solution was measured and the adsorption capacityQi(μmol.g-1) calculated.After each adsorption,the adsorbed BHPN@MIPs were eluted with deionized water/acetic acid (9:1,volume ratio) at room temperature for 4.0 h.Then,the regenerated adsorbent was washed neutral with deionized water and reused for the next cycle.

2.7.Actual sample analysis

Human serum samples were provided by three volunteers.The samples were centrifuged at 5000 r.min-1for 10 min and filtered through a microporous nitrocellulose membrane (pore size 0.27 mm) to remove other impurities.The actual serum samples were prepared by adding 0.2 ml of serum solution to 1.8 ml of PBS solution (pH=7.4).Then,0.2 ml of serum solution and 0.2 ml of dA solution (700 μmol.L-1) were added to 1.6 ml of PBS to prepare spiked samples.Next,2.0 mg BHPN@MIPs and BHPN@NIPs were added to the 2.0 ml actual serum samples and spiked samples.After incubating for 2.0 h at 25 °C,BHPN@MIPs and BHPN@NIPs were collected by centrifugation,and the supernatant was filtered through a microporous nitrocellulose membrane.The filtrate was determined by HPLC analysis,the mobile phase consisted of PBS (pH=7.4) and methanol in a volume ratio of 85:15,the mobile phase flow rate was 0.8 ml.min-1,and the oven temperature was adjusted to 45 °C [39,40].

3.Results and Discussion

3.1.Analysis of the optimal binding ratio of functional monomers to template molecules

For MIPs with ‘‘tailored” binding sites,the optimal ratio between the template molecule and the functional monomer is one of the most influential factors to ensure a precise and flexible assembly process [33].In this work,a continuous titration UV-Vis strategy was used to determine the optimal ratio of the functional monomer AcrU to the template molecule dA.First,the functional monomer AcrU is assembled with the template molecule dA,and AcrU with pyrimidine bases is base-complementary paired with dA with purine basesviahydrogen bonds.In addition,the solvent used for the imprinting polymerization process was a dimethyl sulfoxide/acetonitrile mixture.Therefore,the binding performance of dA to the functional monomer AcrU was tested in this ration mix,and the results are shown in Fig.1.The n-π*transition caused by the hydrogen bonding between the pyrimidine base of AcrU and dA is proved,which is in agreement with the slightly blue shift of Δλ=2 nm [41],confirming a preferred stoichiometric ratio of 50:70 for dA and functional monomer.Therefore,grafting of imprinted polymers(BHPN@MIPs)onto the BHPN surface by reactive controlled polymerization at this ratio.

Fig.1.UV-Vis titration spectra of dA and functional monomer (AcrU) at different binding ratios (a),and under the molar ratio of 50:60 and 50:70 (b).

3.2.Preparation and recognition mechanism of BHPN@MIPs

The preparation of BHPN@MIPs is schematically illustrated in Fig.2.Firstly,the PDA@SiO2were synthesized by a simple onestep St?ber process,which produced PDA@SiO2nanoparticles through a combination of rapid condensation of TEOS and slow polymerization of dopamine under alkaline conditions,using ethanol as the solvent to limit the rate of self-polymerization and control dopamine,to produce a uniform PDA layer [30].Then,a hightemperature hydrothermal treatment was performed in a reactor for 24 h to stabilize the structure.The target product BHPN were obtained after selectively removing of SiO2cores with HF.Secondly,the nucleophilic substitution reaction between the acyl bromide group and the hydroxyl group in BIBB was used to immobilize the initiator Br atom on the surface of BHPN [42].While conventional atom transfer radical polymerization requires extremely stringent anaerobic conditions and relatively large amounts of copper catalyst,the ARGET ATRP system selected in this work improves oxygen tolerance by using a reducing agent called ascorbic acid,which reactivates the catalyst from Cu(II) to Cu(I)and reduces the required copper dose to the ppm level.Thus,low concentrations of metal-based catalysts are sufficient to ensure the continuation of the ARGET ATRP reaction [43].

Fig.2.Synthesis route of BHPN@MIPs by ARGET ATRP.

3.3.Characterization of BHPN@MIPs based on BHPN

The BHPN before etching by HF was shown in Fig.S2(a),it can be seen that the diameters are (160 ± 20) nm.The morphological features of BHPN (a,b) and BHPN@MIPs (c,d) were characterized by SEM and TEM,and the results were illustrated in Fig.3,respectively.As can be seen in Fig.3(a),and Fig.S2(b),it can be clearly seen that the BHPN surface is rough and has a large open pore on the surface,and the TEM image (Fig.3(b) and Fig.S2(c)) shows that the BHPN has a typical bowl-shaped hollow nanostructure with an average PDA shell layer thickness of about 30 nm and a particle size of about 175-200 nm.This bowl-shaped structure is formed by the HF etching between the inside and outside of the BHPN osmotic pressure difference,the spherical nanoparticles undergo bending deformation through the partial collapse of the PDA layer [44].The morphology of the BHPN@MIPs composites was investigated by TEM,as shown in Fig.3(c) and (d),where the imprinted polymer layer on the inner and outer surfaces of the BHPN is more uniform and the thickness of the polymer is about 20 nm.And it can be found that the structure of the BHPN remains stable in the polar organic solvent environment during the ARGET ATRP process.This is attributed to the fact that the hydrothermal treatment effectively increases the chemical stability of the PDA shell,while the PDA structure prepared by the many processes is highly susceptible to damage and collapse in the presence of organic solvents [45].Even after 2.0 h of etching in a 20%(mass) HF solution,the PDA shell retains its integrity (Fig.3(a),and Fig.S2(b)).To evaluate the specific surface area and porosity of BHPN,the N2adsorption-desorption isotherms of the BHPN are shown in Fig.3(e).Similar IV-type were observed,suggesting the mesoporous property of the BHPN.Moreover,the BET results also confirm that BHPN material showed an increase in surface area and average pore sizes (48.59 m2.g-1,12.85 nm) compared to pure PDA in the literature(13.77 m2.g-1)[46].The infrared spectra of BHPN,BHPN@Br and BHPN@MIPs were measured and shown in Fig.4(a).The characteristic peak at 3370 cm-1in BHPN is assigned to the phenolic hydroxyl group and the stretching vibration of -N-H [47],while the characteristic peak at 1620 cm-1is assigned to the N-H bending vibration.After modification with BIBB,BHPN@Br showed a new peak at 550 cm-1[43],and the decrease in peak intensity of the N-H characteristic peak indicated the successful introduction of the C-Br bond to the surface of BHPN through the nucleophilic substitution reaction.Also,the shift of the-OH peak in BHPN@MIPs and the appearance of new peaks such as the amide bond(1620 cm-1)and the C=C stretching vibration peak confirm the ARGET ATRP imprinting polymerization of AcrU and dA in the presence of cross-linking agents [48].In the EDX energy spectrum analysis map of BHPN@Br (Fig.4(b)),the C,N,O,Br,and Si elemental signals are clearly visible and the element content of Br is about 5.64%.The extremely low Si content (0.2%)also indicates that the silicon core was almost completely etched.Thus,the above data confirm that the BIBB molecule was successfully modified on the BHPN surface [47].

Fig.3.SEM (a) and TEM (b) images of BHPN,and TEM images of BHPN@MIPs (c,d),and N2 adsorption and desorption isotherms (pore size distribution) of BHPN (e).

In an effort to demonstrate the success reaction of each step,XPS spectra were obtained for the BHPN and BHPN@MIPs,and the results were depicted in Fig.5.As shown in Fig.5(a),the full range XPS survey spectrum points out the existence of C 1s,O 1s and N 1s in BHPN,a new peak signal of Br 3d appears in BHPN@Br,which proves the successful modification of BIBB.Fig.5(b1) shows the high-resolution spectrum of the C 1s of BHPN,which five small peaks were fitted,located at binding energies of 283.9,284.5,285.4,287.4 and 287.8 eV corresponding to C-H,C-C,C-N/C=C,C=N and C-O,respectively[43],which mainly originate from the amino group of PDA,the benzene ring structure and catechol groups [49].While the same five corresponding characteristic peaks of C=N(398.57 eV),C-N(399.38 eV),N-H(400.2 eV),C-O(532.8 eV) and -OH (521.5 eV) were fitted in the N 1s and O 1s spectra of Fig.5(c1) and Fig.5(d1),all confirm that the BHPN was successfully synthesized.In contrast,the appearance of new peaks for C-OH (285.97 eV),NH-C=O (288.78 eV) and C=O (532.0 eV)in the high-resolution O 1s spectra of BHPN@MIPs and clearly observed can be attributed to the successful grafting of MIPs with AcrU as the functional monomer (Fig.5(b2),(c2) and (d2)) [50].In addition,the characteristic peak shifts of C-C,C=N,C-O,and N-H of BHPN@MIPs were all changed at lower intensities due to the introduction of special functional groups such as amide,ether,and carbon-carbon double bonds in AcrU molecules.Notably,from the atomic percentages given by XPS,we can roughly estimate the C,N,O and Br amounts (see Table 1).The Br content for BHPN@Br was 1.62%,after the imprinting modification,the Br content for BHPN@MIPs decreased to 0.3%,which could be attributed to the grafting of MIPs by the functional monomers AcrU and crosslinker,confirming the chemical changes during the MIPs grafting process [51].The TGA curve are used to investigate the relative composition of BHPN and BHPN@MIPs,by the difference in thermal stability between BHPN and BHPN@MIPs,and the related results are displayed in Fig.6.The slight mass loss around 100 °C of BHPN and BHPN@MIPs can be ascribed to the evaporation of water,around 10%-12%,implying the hydrophilic nature.The second mass loss of BHPN occurred at 150 °C,with a mass loss of about 26%,due to the dehydroxylation of the catechol group in the molecular structure of the polydopamine [52].At 150-800 °C,BHPN@MIPs underwent a rapid mass loss (about 32%)due to the instability of the polymer network at high temperatures,and the 6%more mass loss compared to BHPN was partially attributed to the high-temperature decomposition of the surface imprinting polymer layer.All these results of characterization imply that BHPN@MIPs sorbents are successfully prepared.

Table 1 Atomic percentages of BHPN,BHPN@Br and BHPN@MIPs from the XPS measurements

Fig.5.XPS survey spectra of BHPN,BHPN@Br and BHPN@MIPs(a),and high resolution XPS spectrum of C 1s,N 1s and O 1s from BHPN(b1,c1,d1)and BHPN@MIPs(b2,c2,d2).

Fig.6.TGA curves of BHPN and BHPN@MIPs.

3.4.Adsorption kinetics of BHPN@MIPs towards dA

The adsorption kinetics were closely related to the mass transfer of dA from the test solution phase to the binding site,in which the data were fitted by the corresponding models to explore the adsorption mechanism and adsorption equilibrium time.Therefore,the adsorption kinetics experiments of BHPN,BHPN@MIPs and BHPN@NIPs for dA were carried out using static adsorption experiments to determine the adsorption efficiency of the above three adsorbents at different adsorption times by UV-Vis.The kinetic data and model fitting results are shown in Fig.7.As shown in Fig.7,due to the strong hydrophilicity of PDA,the adsorption solution of dA molecules with PDA contact efficiency was greatly improved.The bowl-shaped structure of the adsorbent provides a large storage space for the test solution,making the site utilization in the adsorption process much higher and is conducive to improving the adsorption kinetic performance.As can be seen in Fig.7(a),the adsorption capacity of these three sorbents followed the order of BHPN@MIPs ＞BHPN@NIPs ＞BHPN,which was due to the presence of imprinting polymers made more specific recognition sites that could capture dA on the inner and outer surfaces of the sorbent BHPN@MIPs.BHPN@MIPs had better site capture capacity than BHPN@NIPs,which is mainly attributed to the enhanced recognition of dA by the imprinted cavities [53].In addition,the surface adsorption of dA by BHPN@MIPs,BHPN@NIPs and BHPN increases rapidly from 0 to 10 min and can capture up to about 93% of the total adsorption capacity of dA,and then reaches an equilibrium at 60 min due to the slow decrease available active binding sites,indicating fasting the rapid adsorption kinetics of the adsorbent.Several kinetic models in Table 2 [33,54],such as non-linear and linear forms of pseudo-first-order and pseudosecond-order kinetic model considered to fit the data,and the correlation coefficients(R2)and corresponding kinetic parameters are summarized in Table 2.It can be seen that theR2values of the quasi-secondary kinetic equations are relatively higher and the associated theoretical calculations ofQe(μmol.g-1) are closer to the experimental values,which implies that the kinetic data are better fitted to the pseudo-second-order kinetic model.Therefore,it can be concluded that the main mechanism of binding of BHPN@MIPs to dA is chemisorption,which dominates the adsorption process,that is the base complementary pairing between the pyrimidine bases of AcrU and the purine rings of dA molecules[54].For BHPN@MIPs,the rate constantk2of the linear pseudosecond-order kinetic model (18.8 × 10-3g.μmol-1.min-1) was slightly higher than that of BHPN@NIPs (10.8 × 10-3g.μmol-1-.min-1) (Table 3),indicating that the mass transfer rate of the imprinted adsorbent was faster,which also indicated that the presence of specific recognition sites could enhance the binding kinetic performance of the composite adsorbent.

Table 2 Several kinds of kinetic models

Table 3 Linear and nonlinear kinetic model parameters obtained in adsorption of dA onto BHPN@MIPs and BHPN@NIPs

Fig.7.Non-linear kinetic model fitting curves of BHPN@MIPs,BHPN@NIPs and BHPN for dA (a),linear kinetic model evaluation for dA (b).

3.5.Adsorption isotherms of BHPN@MIPs for dA

Adsorption isotherm model is significantly important to explain the equilibrium between dA captured by BHPN@MIPs to the dA molecules present in the solution.Adsorption amounts of BHPN@MIPs and BHPN@NIPs at different initial concentration are shown in Fig.8.As shown in Fig.8(a),whenCeis below 600 μmol.L-1,the adsorption capacity of the sorbent for dA increases rapidly,followed by increases slowly to stabilize.To analyze the effect of the initial dA concentration on the adsorption behavior,the adsorption equilibrium data were fitted using the Langmuir,Freundlich and Langmuir-Freundlich equations,which represent three different binding behaviors for monolayer adsorption and multilayer adsorption based on non-homogeneous sites,respectively,assuming a finite number and identical sites.The nonlinear forms of the three equations can be expressed by Eqs.(3)-(5),respectively [55,56].

Fig.8.Non-linear Langmuir,Freundlich and Langmuir-Freundlich model fitting curves of BHPN@MIPs,BHPN@NIPs and BNPN(a),and selectivity adsorption capacity of BHPN@MIPs and BHPN@NIPs for dA,AMP,dC and ATP (b).

whereQm(μmol.g-1)represents the maximum adsorption amount,andkL(L.μg-1) is the Langmuir adsorption constant.Furthermore,kF(μmol.g-1).(L.μmol-1)1/nis the adsorption constant in the Freundlich isotherm model,kLF(L.mg-1) is the adsorption constant of heterogeneous in the Langmuir-Freundlich isotherm model,and 1/nis the heterogeneity parameter.

The adsorption isotherm constants for BHPN@MIPs and BHPN@NIPs are listed in Table 4.From the corresponding parameters in Table 4,for BHPN@MIPs the correlation coefficientR2by Langmuir-Freundlich models provides a better fit (0.9998) than the Langmuir (0.9963) and Freundlich (0.9992),indicating that the Langmuir-Freundlich model is more suitable for describing the adsorption equilibrium performances.Also,implying that at low dA concentration the model would become Freundlich isotherm model with multilayered adsorption,while at high dA concentration it transforms into Langmuir isotherm with monolayer adsorption [57,58].According to the Langmuir-Freundlich equation,the adsorption capacity of BHPN@MIPs,BHPN@NIPs and BHPN were 328.45,252.74 and 175.16 μmol.g-1at 25 °C,respectively.In addition,the 1/nof the heterogeneity parameters obtained from the calculations,the value of BHPN@MIPs (0.776)is the lowest among three sorbents,indicating that the distribution of affinity on the BHPN@MIPs adsorbent surface is heterogeneous[57].

Table 4 Parameters of Langmuir,Freundlich and Langmuir-Freundlich isotherm model for dA onto BHPN@MIPs,BHPN@NIPs and BHPN

3.6.Adsorption selectivity and regeneration capability

To further evaluate the selectivity of BHPN@MIPs,some structural analogues such as dA,AMP,dC and ATP were used as competitive nucleoside analogues for binding experiments under the same conditions (300 μmol.L-1,pH=7.4).As shown in Fig.8(b),BHPN@MIPs showed the highest adsorption of dA(74.58 μmol.g-1),which was much higher than that of AMP (8.319 μmol.g-1),dC(22.54 μmol.g-1) and ATP (7.416 μmol.g-1).The reason is mainly that the imprinting sites prepared by the molecular imprinting strategy have specific recognition effects on template molecules in terms of shape,size and functional groups [32].ATP and AMP have the same base complementary pairing that adenine functional groups as dA,but they carry phosphate groups and have different sizes,shapes and recognition sites,so the adsorption effect is not high,when comparing with dA.dC is also a nucleotide compound,but compared to dA,they have different shapes and functional groups,it has different shapes and functional groups.Therefore,the memory of imprinting sites in terms of shape,size and functional group is significantly important for the specific recognition of the template molecule dA.In addition,the imprinting factor (IF) of BHPN@MIPs for dA was calculated to be 1.30,which also indicates that although the introduction of imprinting sites improves the selection performance,the non-specific adsorption of PDA is also always present.The regenerability of the sorbents is an important indicator of the economic value and stability.The adsorption-desorption of dA on BHPN@MIPs from four sequential cycles are illustrated in Fig.9(a),the result shows that the good uptake number of BHPN@MIPs for dA is maintained after four cycles,and the value of the fourth regeneration is 91.14%of the initial value that the adsorption capacity decreased from 76.08 to 70.29 μmol.g-1.This result indicates that BHPN@MIPs have ideal adsorption capacity and reusability.The adsorption capacity in previous study on the nucleotide imprinted sorbents are summarized in Table 5 [2,59-61],indicating the BHPN@MIPs have excellent selectivity towards dA.

Table 5 Comparison of the adsorption capacities for dA with other reported sorbents

Fig.9.Regeneration analysis of BHPN@MIPs via four sequential adsorption/desorption cycles(a),and HPLC chromatogram analysis of dA spiked human serum samples(b).

3.7.Spiked human serum samples recognition experiment

To further investigate availability,human serum samples spiked with dA at 700 μmol.L-1are extracted by BHPN@MIPs and BHPN@NIPs and then analyzed by HPLC.As show in Fig.9(b),a small amount of dA was detected in the actual serum samples.In the spiked samples,a significantly higher dA peak appeared about 3.8 min.The peak intensities of dA molecules were significantly lower after sorbent extraction.The magnitude of dA peak areas in the three samples followed the order of spiked human serum samples ＞extraction after by BHPN@NIPs ＞extraction after by BHPN@MIPs,which showed that BHPN@MIP was still effective in selective identification and separation of complex dA in complex real samples [62].

4.Conclusions

In summary,the bowl-shaped PDA particles BHPN were prepared by the St?ber strategy,and the BHPN@MIPs were obtained by grafting onto the imprinted polymerized particles with specific recognition sites through the ARGET ATRP strategy for the selective adsorption of dA.The results showed that the BHPN@MIPs have excellent adsorption capacity.The equilibrium time and maximum adsorption capacity of BHPN@MIPs were 60 min and 328.45 μmol.g-1.Meanwhile,the adsorption results revealed that the mechanism of selective separation of dA was mainly attributed to the base complementary pairing between the pyrimidine group of the functional monomer AcrU and the purine of the dA molecule.Therefore,this work provided a new method for separation and purification of nucleosides.

Data Availability

Data will be made available on request.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (22078132 and 22108103),Open Funding Project of the National Key Laboratory of Biochemical Engineering (2021KF-02),China Postdoctoral Science Foundation(2021M691301),Jiangsu Agricultural Independent Innovation Fund Project(CX(21)3079)and Graduate Research Innovation Program of Jiangsu Province (KYCX20-3040),China Postdoctoral Science Foundation (2021M691301).

Supplementary Material

Supplementary data to this article can be found online at https://doi.org/10.1016/j.cjche.2023.01.009.