Preparation and Application of a Molecularly Imprinted Monolith for Specific Recognition of Domoic Acid

Analytical and Bioanalytical Chemistry https://doi.org/10.1007/s00216-017-0843-3

RESEARCH PAPER

Preparation and application of a molecularly imprinted monolith for specific recognition of domoic acid

Fan Yang1 & Ruirui Wang1 & Guangshui Na 1 & Qilun Yan1 & Zhongsheng Lin1 & Zhifeng Zhang1

Received: 14 September 2017 /Revised: 19 November 2017 /Accepted: 18 December 2017 # Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract In this work, a molecularly imprinted monolithic column was synthesized by a facile procedure and was applied for specific recognition of domoic acid, an amnesic shellfish poison. The poly(4-vinylpyridine-co-ethylene glycol dimethacrylate) molecularly imprinted monolith was synthesized in a stainless steel column by in situ polymerization. Pentane-1,3,5-tricarboxylic acid was used as a dummy imprinting template instead of the highly toxic and expensive target molecule. It is the first time that a molecularly imprinted monolith is introduced for separation and detection of domoic acid. After optimizing the preparation conditions, the prepared imprinted monolith was systematically characterized and exhibited excellent stability and permeability as a HPLC stationary phase. The results of chromatographic analysis demonstrated that the molecularly imprinted monolith exhibited specific retention and selective recognition toward domoic acid, with an imprinted factor up to 3.77. Furthermore, the molecularly imprinted monolith was successfully applied for selective enrichment of domoic acid from biological samples.

Keywords Molecular imprinting . Monolith . Domoic acid . High performance liquid chromatography

Introduction low in seawater and biological samples, a reliable and sensitive analytical method for DA detection is urgently needed. To date, Domoic acid (DA) is a neurotoxinaminoacidwhichispro- several analytical methods have been reported for monitoring of duced by several species of marine single-celled diatoms of the DA in shellfish, phytoplankton, and seawater, such as enzyme- genus Pseudo-nitzschia and accumulates in crustaceans, fish, linked immunosorbent assay (ELISA) [6, 7], capillary electro- and shellfish [1]. As a major amnesic shellfish poisoning toxin, phoresis (CE) [8, 9], capillary electrochromatography (CEC) DA can bioaccumulate in filter-feeding marine organisms, and [10], high performance liquid chromatography (HPLC) the ingestion of these DA-contaminated seafood may lead to coupled with ultraviolet detection or mass spectrometry poisoning symptoms such as vomiting, diarrhea, abdominal [11–14], and so on. Compared with other DA detection cramps, headache, and dizziness. In severe cases, the victim methods, HPLC had the advantages of being accurate, sensi- may experience trouble breathing, seizures, permanent loss of tive, stable, and reliable, and was considered to be the most short-term memory, even coma or death [2–4]. DA is consid- widely used analytical method for DA detection. ered as a significant risk to not only public health but also the Molecular imprinting is known as a technique for synthesis stability of aquaculture and shellfish harvesting industries [5]. of tailor-made recognition materials by polymerization of suit- Many countries have carried out DA monitoring programs to able functional monomers and cross-linkers in the presence of ensure consumer protection and sustainable development of template [15]. After removal of the template from the polymer, marine economy. Since the concentration of DA is extremely molecularly imprinted polymers (MIPs) are obtained with recognition cavities complementary to the template in shape, size, and functional groups, which can selectively recognize the template. Due to the remarkable selectivity and specific recognition * Fan Yang toward the target molecule, MIPs have gained great interest and [email protected] been widely applied in fields of separation processes, micro reactors, catalysis, biosensors, and so on [16–18]. In recent 1 Key Laboratory for Ecological Environment in Coastal Areas (SOA), years, molecularly imprinted materials were specifically de- National Marine Environmental Monitoring Center, No.42, Linghe Street, Dalian, Liaoning 116023, China signed as affinity absorbents to remove low-level DA from Yang F. et al. complex samples [19–24]. Lotierzo et al. [19], using DA as Experimental template, synthesized a molecularly imprinted polymer film by direct photo-grafting on a gold chip. Considering the high Materials and chemicals toxicity and high cost of DA, some dummy imprinting tem- plates were chosen as DA substitutes. Kubo et al. [20]usedo- 4-Vinylpyridine (4-VP) and DA were obtained from Sigma phthalic acid as an alternative template to prepare molecularly Aldrich (St. Louis, USA). Ethylene glycol dimethacrylate imprintedpolymerstowardtoDA,whichprovidedtwosimilar (EDMA) was obtained from Acros Organics (NJ, USA). neighboring carboxylic acid groups to simulate DA during the 2,2′-Azobisisobutyronitrile (AIBN) and benzoic acid were ob- imprinting process. Subsequently, they investigated a Bthree- tained from J&K Chemical (Beijing, China). PTA was from point^ recognition of DA by using pentane-1,3,5-tricarboxylic Tokyo Chemical Research Institute (Tokyo, Japan). Toluene acid (PTA) as imprinting template. By comparing with other and dodecanol were obtained from Kemiou Chemical structural analogues of DA, PTA was proved to be the opti- Reagent Co., Ltd. (Tianjin, China). Deionized water was pu- mized dummy template. The PTA-MIPs exhibited the highest rified using a Milli-Q water purification system (Millipore, selective recognition ability for DA and was successfully used Milford, MA, USA). Other reagents used were of analytical as solid-phase extraction (SPE) medium to selective enrichment grade or better. Stainless steel columns (100 mm × 4.6 mm of DA [21]. Zhou et al. [22] prepared MIP toward to DA by i.d.) were supplied by Hongtai Chromatogram Equipment bulk polymerization method, and used it as SPE sorbent for Co., Ltd. (Dalian, China). selective extraction of DA from seafood samples. These resulted MIP materials have shown good selectivity toward to DA. Instruments However, when using MIPs as SPE sorbent before further analysis, a tedious preparation procedure of packed columns is HPLC was used to investigate the adsorption and recognition needed, such as grinding, sieving, and column packing, which properties of the MIP monoliths. All the chromatographic ex- leads to time-consuming, poor reproducibility, and material periments were performed by using a Thermo Finnigan loss. Surveyor HPLC system (Thermo Fisher Scientific, USA), Monolith is a separation media that consist of a single, which consisted of a Surveyor LC pump, an Auto Sampler, continuous, integrated interconnecting porous skeleton with- and a PDA detector. Data acquisition and processing were out interparticular voids. As a new generation of stationary controlled by Xcalibur software system. The scanning elec- phase, monolithic column has shown significant advantages tron micrographic (SEM) images of the monoliths were ob- comparing with conventional particle-packed column, such as tained by a Nova NanoSEM 450 microscope (FEI, USA). easy preparation, versatile surface modification, fast mass- Pore size distribution measurements of the monoliths were transfer kinetics, and higher permeability [25]. With a bimodal conducted on an AutoPoreIV 9500 mercury intrusion porous structure comprising macropores and mesopores, porosimetry (Micrometritics, Norcross, GA, USA). Fourier- monoliths show both high permeability and large surface area transform infrared (FT-IR) spectra (4000 to 400 cm−1)in for interaction with analytes. Molecularly imprinted monolith- KBr were recorded using a NEXUS FT-IR spectrophotometer ic column (MIP monolith), which integrates high selectivity of (Thermo Nicolet, USA). Elemental analysis of the monolith molecularly imprinted technology and high efficiency of was performed by a CHNOS elemental analyzer (Vario monolithic column, has attracted significant interest as station- MICRO, Elementar, Germany). ary phase of HPLC and CEC for the fast and online isolation of desired analytes [26, 27]. It is important to explore MIP Preparation of organic polymer-based MIP monolith monolith in sample pretreatment and chromatographic separation of amnesic shellfish poisoning toxin. The MIP monolith was prepared using PTA as dummy tem- In this work, a MIP monolith was synthesized and applied plate in a stainless steel column by in situ polymerization for specific recognition of DA. To the best of our knowledge, under the optimal conditions. Briefly, the template molecule it is the first time to introduce MIP monolith for the separation PTA (10 mg) and functional monomer 4-VP (0.1 mL) were and detection of DA. Due to the high toxic and high cost of dissolved in a binary porogenic solvent, which consisted of DA, PTA was chosen as an alternative template during the 175 μL toluene and 1.575 mL dodecanol. Then cross-linker imprinting process. 4-Vinylpyridine and ethylene glycol EDMA (0.70 mL) and initiator of the polymerization reaction dimethacrylate were used as functional monomer and cross- AIBN (1 wt%) were added in the above solution stepwise. The linker respectively during the in situ polymerization process. polymerization mixture was sonicated for 3 min in order to The synthesis procedure was simple, facile, and cost efficient. obtain homogeneous solution, and purged with nitrogen for The resultant monolith was systematically characterized and 1 min to eliminate soluble oxygen. Subsequently, the reaction evaluated, and then was applied as stationary phase of HPLC mixture was injected into a stainless steel column, with both for the online analysis of DA in biological samples. ends of the column sealed. After polymerization at 60 °C for Preparation and application of a molecularly imprinted monolith for specific recognition of domoic acid

24 h in an oven, the monolith was connected to a HPLC pump chromatography method [27]. The elution was monitored at and washed thoroughly with methanol-acetic acid (9:1, v/v), 242 nm and the flow rate was 0.5 mL·min−1. The template water, and acetonitrile sequentially, to remove the template PTA was chosen as the testing analyte and the acetone was molecules and other residual reagents. chosen as the dead time marker. After equilibrating monolith In comparison, a non-imprinted monolith (NIP monolith) with mobile phase before measurement, a loading buffer con- for control experiments was prepared using the same process taining 1 mg·mL−1 PTA was pumped through the monoliths described above, without imprinted template. and the dynamic binding capacity Q (μmol·g−1) was calculated according to the following formula:

Chromatographic conditions Q ¼ ðÞÂVB−V0 C=m

All chromatographic experiments were performed on the MIP where VB (mL) is the 10% breakthrough volume, V0 (mL) is monolith prepared via in situ polymerization in a stainless steel the void volume of HPLC system, C is the concentration of − column (100 mm × 4.6 mm i.d.). The UV detection wavelength testing molecule (μmol·mL 1), and m is weight of dry mono- was set at 242 nm, and the sample injection was auto using an lithic rod (g). injection valve of 10 μL.Aflowrateof0.5mL·min−1 was used during the HPLC analysis. The acetonitrile/0.05% aqueous acetic acid (7:3, v/v) was employed as the mobile phase and Results and discussion the monolith was adequately equilibrated with the mobile phase until a stable baseline was achieved. All mobile phases and Preparation of poly(4-VP-co-EDMA) MIP monolith sample solutions were filtered by a 0.45-μm membrane (Nihon Millipore) before using. In this work, the organic polymer-based MIP monolith was k The retention factor ( ) was calculated from the formula: fabricated by in situ polymerization, using 4-VP as functional k − =(tR t0)/t0,wheretR is the retention time of the target mol- monomer, EDMA as cross-linker, PTA as dummy template to ecule and t0 is the elution time of the void marker. The retention replace the toxic template DA. The monolith was prepared in a time of the void marker, t0, was measured by injecting acetone. stainless steel column, which could be connected to HPLC for α α The separation factor ( ) was calculated from the formula: = online analysis. The synthesis method is proposed to be con- k k k k tem/ com,where tem and com are the retention factors of tem- venient, safe, and effective. The general scheme for the MIP plate and other competitive molecule in MIP monolith or NIP monolith preparation was illustrated in Fig. 1. monolith, respectively. The specific recognition of the MIP To date, the dummy molecular imprinting technique, using monolith was evaluated by imprinted factor (IF), which is de- a structural analog of the targeted compound as template, has fined as the ratio of the retention factor of imprinted molecule in been considered as a feasible and effective method to prepare k k the MIP monolith and NIP monolith (IF = MIP/ NIP). MIPs for imprinting of those high toxic, high cost, or rare targets [15]. In previous work, Bthree-point^ recognition of Dynamic binding capacity COOH groups on DA has been achieved by using PTA as alternative template during the imprinting process [21, 22]. The specific binding capacities of the resultant MIP monolith As a structural analog of DA, PTA has a similar molecular andNIPmonolithwereinvestigatedbyfrontal structure and spatial distance of the COOH groups with DA.

Fig. 1 Schematic illustration of the preparation of poly(4-VP-co-EDMA) MIP monolith Yang F. et al.

Table 1 Preparation conditions of the poly(4-VP-co-EDMA) Monolith 4-VP EDMA Monomer-cross- Toluene content in Porogenic Template- MIP monolith (mL) (mL) linker ratio (mol) porogenic mixture mixture monomer ratio (v%) (mL) (mol)

M1a 0.1 0.53 1:3 10% 1.75 1:20 M2abcd 0.1 0.70 1:4 10% 1.75 1:20 M3a 0.1 0.88 1:5 10% 1.75 1:20 M4a 0.1 1.05 1:6 10% 1.75 1:20 M5b 0.1 0.70 1:4 5% 1.75 1:20 M6b 0.1 0.70 1:4 15% 1.75 1:20 M7c 0.1 0.70 1:4 10% 1.55 1:20 M8c 0.1 0.70 1:4 10% 2.00 1:20 M9d 0.1 0.70 1:4 10% 1.75 1:10 M10d 0.1 0.70 1:4 10% 1.75 1:30

a Ratio of 4-VP to EDMA b Content of toluene in porogenic mixture c Amount of porogenic mixture d Ratio of PTA to 4-VP

Coupled with the flexible conformational changes of DA, the A proper ratio of template to monomer was necessary for the recognition sites constructed by PTA on MIPs were proved to specificity of MIP monolith. Refer to the conditions used in be able to selective recognition of DA. PTAwas considered as literature [21, 22], the influence of template-monomer molar an optimized dummy template for DA and was chosen as the ratio (1:10 to 1:30) on the separation performance was evaluat- imprinting template in this work. ed. The ratio of 1:20 (M2) was optimal for the preparation of In order to acquire a monolith with high selectivity, stable MIP monolith with separation factor α reached to 4.90, while mechanical property, and high throughput (low backpressure), higher or lower ratios provided relatively lower selectivity. some preparation conditions must be taken into account. The selection of the porogenic solvents is a crucial factor Several factors were systematically optimized, including ratio during preparation of MIP monolith. In order to obtain high of monomer to cross-linker, ratio of template to monomer, and selectivity of MIPs, the polarity of porogenic solvents should composition and amount of porogenic mixture. The composi- be relatively low to reduce the interferences during complex tions of polymerization mixture were indicated in Table 1. formation between template and monomer. A classic First of all, an appropriate monomer is the key factor for porogenic mixture of toluene and dodecanol was chosen in imprinting effect. For the acidic template molecule such as this work, in which good solvent toluene leads to small size PTA, the basic monomer 4-VP, which can interact with tem- pores and bad solvent dodecanol leads to large size pores. It plate by hydrogen bonding and π-donor/acceptor interactions, was chosen as functional monomer. The amount of cross- linker EDMA should also be well considered to maintain the stability of the recognition sites and recognition cavities in the polymerization. Several molar ratios of monomer to cross- linker ranging from 1:3 to 1:6 were investigated. When ratio of 4-VP to EDMA was 1:3 (M1), the obtained monolith was slack and easy to collapse at a high flow rate of mobile phase, which was not a suitable stationary phase material for analysis on HPLC. With increasing the proportion of EDMA, the ri- gidity of the resultant monolith was improved and the backpressure was increased as well. When increasing the ratio to 1:6 (M4), the MIP monolith exhibited a tough appearance with bad permeability. 1:4 was chosen as appropriate ratio in this work to synthesize imprinted monoliths with good permeability and stable mechanical property for a long-time HPLC analysis. Fig. 2 FT-IR spectra of poly(4-VP-co-EDMA) MIP monolith Preparation and application of a molecularly imprinted monolith for specific recognition of domoic acid

Fig. 4 Pore size distribution of poly(4-VP-co-EDMA) MIP monolith

The amount of porogenic mixture addition in the reaction mixture was also investigated. A series of monoliths were synthesized with the amount of porogenic mixture ranging from 1.55 to 2 mL. The results showed that the MIP monolith obtained at amount of 1.75 mL had good permeability and stable structure. Therefore, 1.75 mL porogenic mixture was adopted as the optimized condition. Based on these detailed investigations, a polymerization mixture consisting 0.1 mL 4-VP, 0.70 mL EDMA, 10 mg PTA, 175 μL toluene, and 1.575 mL dodecanol was adopted in the preparation process. By adding 1 wt% AIBN and poly- merized at 60 °C for 24 h, a homogeneous and durable MIP monolith with good mechanical strength and permeability could be obtained. Moreover, comparing with the conventional MIP particle-packed columns, the in situ polymerization procedure of poly(4-VP-co-EDMA) MIP monolith was easy and timesaving. The tedious procedure of grinding, sieving, and column packing was avoided and the loss of materials was reduced, and the reproducibility of preparation process of MIP Fig. 3 Scanning electron microscopy photographs of poly(4-VP-co- chromatographic column could be improved. EDMA) MIP monolith under the optimized preparation conditions at different magnifications. a ×5000; b ×10,000 The characterization of MIP monolith was found that, by adjusting the ratio of toluene and In this work, FT-IR spectroscopy, elemental analysis, SEM, dodecanol, the pore structure of monolith was changed and and mercury intrusion porosimetry were used to characterize the appearance and mechanical stability of monoliths were the prepared monolithic column. Firstly, the resulted MIP correspondingly affected. When less toluene added (volume monolith was characterized by FT-IR spectroscopy. As shown content of 5%), the resulted MIPs monolith had a slack and in Fig. 2, the stretching vibration peak of C-H was found at soft texture with good permeability but bad mechanical 2980 cm−1, and the stretching vibration of C-O-C was found strength. With increasing proportion of toluene, the mechani- at 1160 cm−1. There was a characteristic stretching band of cal strength of the monolith was increased. When the toluene C=O at 1734 cm−1 from the ester groups of EDMA. The reached 15% in the porogenic mixture, the monolith had a characteristic peak of 4-VP was around at 1600 and denser structure that the mobile phase was hard to pass 1425 cm−1, which corresponded to the C=N and C=C through. Considering the requirements of sufficient perme- stretching of pyridine ring. These characteristic peaks in the ability and mechanical strength of the monolith, 10% was spectrum showed that the MIP monolith was synthesized chosen as the optimum content of toluene for further studies. through the polymerization of EDMA and 4-VP. The result Yang F. et al.

Fig. 5 The mechanical stability of poly(4-VP-co-EDMA) MIP monolith

of elemental analysis showed that the C, H, O, N content of aggregated to form functional clusters, and flowing paths the MIP monolith were 57.66, 7.01, 27.63, and 1.47%, respec- through the monolith were existed. Large through-pores of tively, which were close to the theoretical calculated value (C the organic monolith were clearly observed and could provide 62.86%, H 7.07%, O 28.51%, N 1.56%). These results dem- a decreased mass-transfer resistance and high permeability for onstrated that the poly(4-VP-co-EDMA) MIP monolith was further application. successfully synthesized by in situ polymerization. Pore size distribution of the monolith was characterized by SEM was used to characterize the morphological structure mercury intrusion porosimetry. Based on measurement, aver- of the MIP monolith prepared under optimized conditions. As age pore diameter 895 nm, total surface area 15.03 m2·g−1,and shown in Fig. 3, the obtained MIP monolith had the morphol- total porosity 59.6% were determined. It was shown in Fig. 4 ogy of an interconnected, homogeneous, and continuous skel- that the monolith had a unimodal pore size distribution around eton with uniform pore distribution. On the cross-section im- 1 μm. The homogeneous macroporous structure allowed mo- ages of monolith, it was found that globular particles were bile phase flowing through the monolithic stationary phase

Fig. 6 Breakthrough curves of PTA on the MIP and NIP monolith Preparation and application of a molecularly imprinted monolith for specific recognition of domoic acid

Table 3 The recoveries of DA on the MIP monolith

Concentration Recovery (%) RSD (%) of DA (mg·L−1)

191.33.43 289.52.76 589.33.01

Experimental conditions were same as in Fig. 7

measured. As shown in Fig. 5, the backpressure increased linearly with the increase of flow rate. It indicated that the structure of the monolith was not compressed under high flow rate and the resulted monolith possessed good mechanical stability. Also, a relatively low backpressure of 4.9 and 3.9 MPa were observed under a high flow rate of deionized water and acetonitrile at 4 mL·min−1, respectively. The permeability of the MIP monolith was evalu- 2 ated by Darcy’slaw[28]: B0 = ηFL/(πr ΔP),whereF is the solvent flow rate (m3·s−1), L is the column length (m), η is the viscosity of the mobile phase (Pa·s), r is the radius of the column (m), and ΔP isthepressuredropofthe column (Pa). The permeability of the resulted monolith under optimum conditions was calculated as 6.72 × 10−14 m2 for water (η = 1.005 cP, 20 °C) and 3.37 × 10−14 m2 for acetonitrile (η = 0.369 cP, 20 °C) at flow rate of 1 mL·min−1, respectively. The result indicated that MIP monolith possessed good permeability. Permeability decreased when the mobile phase changing from aqueous buffer to organic buffer. It was caused by slight solvent swelling or shrinkage of organic skeleton, but stability of the monolith was not affected. The monolith was stable Fig. 7 Chromatograms of domoic acid and benzoic acid on the NIP monolith (a) and MIP monolith (b). Experimental conditions: mobile enough for continuous usage of weeks with a constant phase, acetonitrile/0.05% aqueous acetic acid (7:3, v/v); detection backpressure. − wavelength, 242 nm; flow rate, 0.5 mL·min 1 As above, these results of characterization experiments indicated that the MIP monolithic column possessed a homoge- with low flow resistance, and high permeability could be neous and porous skeleton with high permeability and good expected. mechanical stability, which make it more specifically suitable Additionally, to evaluate the mechanical stability of the for the rapid and high throughput analysis on HPLC. resulted MIP monolith, deionized water and acetonitrile were used as mobile phase to equilibrate the monoliths with flow rate ranging from 0.1 to 4.0 mL·min−1, respec- Dynamic binding capacity tively, and the change trend of backpressure was Binding capacity is one of the most important properties of imprinting material. Frontal chromatography method Table 2 Chromatographic separation data of poly(4-VP-co-EDMA) was employed to determine and compare the specific MIP monolith binding capacity of MIP and NIP monolith. PTA −1 MIP monolith NIP monolith IF (1 mg·mL ) dissolved in mobile phase was pumped through the pre-equilibrated monolith. The breakthrough k α k α t (min) t (min) curves of the MIP and NIP monoliths were shown in DA 4.94 1.47 4.90 2.77 0.39 1.22 3.77 Fig. 6. It is obvious that the plot of MIP monolith was BA 2.60 0.30 2.63 0.32 different from the plot of NIP monolith, which indicated that analyte had different adsorption behavior on MIP and Experimental conditions were same as in Fig. 7 NIP monolith. PTA was eluted out near the void time on Yang F. et al.

Fig. 8 Chromatograms of mussel extract sample containing 5mg·L−1 DA on the MIP monolith. Experimental conditions were same with Fig. 7

the NIP monolith while was captured by the MIP mono- retention toward to DA on the NIP monolith. In contrast, lith until saturation was reached. The difference of reten- BA was still eluted out near the void time while DA was tion behavior can be attributed to the existence of the well retained by the MIP monolith. The retention factor specific imprinting sites on prepared MIP monolith. The (k) for DA on the MIP monolith was 1.47, much higher dynamic binding capacity of PTA on the MIP and NIP than that obtained on the NIP monolith, where the k was monolith were 39.91 and 10.73 μmol·g−1, respectively. 0.39. The imprinted factor (IF) of DA reached 3.77. The The result revealed that the recognition cavities and sites chromatographic separation data of the MIP and NIP were successfully formed by using dummy template, and monoliths were summarized in Table 2. The results indi- the specific recognition ability toward to DA was further cated that the MIP monolith had specific retention ability verified by HPLC in subsequent experiment. toward to DA, and the preparation of MIP monolith using The binding capacity of resultant poly(4-VP-co-EDMA) dummy template was efficient for selective recognition of MIP monolith and other DA-MIP materials were compared. the target analyte. In relevant literatures, DA-molecularly imprinted polymer The recovery of MIP monolith was measured by using particles were prepared by bulk polymerization [22] and emul- different concentration of DA as analytes, and the results were sion polymerization [24], and the maximum adsorption capac- listed in Table 3. With the increasing of DA concentration, the −1 ity Qmax were determined to be 32.21 μmol·g and 1229 μg· recoveries of DA ranged from 89.3 to 91.3%, which indicated g−1, respectively. It was found that Q of MIP monolith in this that the prepared MIP monolith showed good recoveries for work was higher than that of relevant researches. This result DA at different concentrations. Furthermore, a liner standard indicated that MIP monolith exhibited higher specific binding curve for DA analysis on MIP monolith was constructed (y = ability toward to target compounds, and it could be attributed 4.563x + 2.669, R = 0.996), and detection limit was deter- to the advantages of monolithic columns, such as large surface mined to be 0.076 mg·L−1, which was lower than the action area and rapid mass-transfer rate. limit of 20 μg·g−1 imposed by Canada, the European Union, and the USA. The recovery of MIP monolith and the detection Specific recognition of DA on the MIP monolith limit for DA in this work were comparable to that in relevant reports [22, 24]. These results suggested that the MIP mono- To evaluate the specific selectivity of the prepared MIP lith possessed good recognition properties and could be used monolith, DA and benzoic acid (BA) were used as target for specific recognition of DA. analyte and competitive analyte, respectively, and the sample concentration was 2 mg·L−1.Figure7 showed Application for biological sample analysis the different chromatographic retention behavior of DA and BA on the MIP and NIP monoliths. On the NIP The prepared MIP monolith was further applied for analysis of monolith, analytes were both eluted out almost at the biological samples. Mussel extract was prepared according to same time, which suggested that there was no specific the method proposed by Zhou et al. [22]. The extract was Preparation and application of a molecularly imprinted monolith for specific recognition of domoic acid diluted 20-fold with ACN and 5 mg·L−1 DA was added into References the sample. The chromatographic retention behavior of the biological sample was shown in Fig. 8. It was found that the 1. Lefebvre KA, Frame ER, Kendrick PS. Domoic acid and fish be- extract from mussel was eluted quickly near void time, while havior: a review. Harmful Algae. 2012;13:126–30. DA was retained by the MIP monolith and separated from the 2. He Y, Fekete A, Chen GN, Harir M, Zhang L, Tong P, et al. −1 Analytical approaches for an important shellfish poisoning agent: complex matrix. The recovery of 5 mg·L DA in mussel domoic acid. J Agric Food Chem. 2010;58:11525–33. extract was measured to be 88.6 ± 2.5% (n =3).Theresultof 3. de la Iglesia P, Gimenez G, Diogene J. Determination of dissolved the chromatographic analysis confirmed that the MIP mono- domoic acid in seawater with reversed-phase extraction disks and lith using dummy imprinting template had the ability of chro- rapid resolution liquid chromatography tandem mass spectrometry – matographic separation of DA from complex biological with head-column trapping. J Chromatogr A. 2008;1215:116 24. 4. Lefebvre KA, Robertson A. Domoic acid and human exposure samples. risks: a review. Toxicon. 2010;56:218–30. 5. Morgan KL, Larkin SL, Adams CM. Firm-level economic effects of HABS: a tool for business loss assessment. Harmful Algae. Reproducibility and stability of the MIP monolith 2009;8:212–8. 6. ZhouY,ZhangYY,ShenQF,LuSY,RenHL,LiYS,etal. Reproducibility is an important parameter to evaluate effec- Development of a novel antibody probe useful for domoic acid tiveness and practicability of the preparation method. detection. Biosens Bioelectron. 2009;24:3159–63. Reproducibility of MIP monolith separation was investigated 7. Yu FY, Liu BH, Wu TS, Chi TF, Su MC. Development of a sensitive enzyme-linked immunosorbent assay for the determination of by calculating the relative standard deviation (RSD) of the domoic acid in shellfish. J Agric Food Chem. 2004;52:5334–9. retention time of DA. The MIP monoliths were prepared un- 8. Kvasnička F, Ševčík R, Voldřich M. Determination of domoic acid der optimized conditions. After each HPLC analysis, mono- by on-line coupled capillary isotachophoresis with capillary zone liths were sufficiently eluted and equilibrated before a new electrophoresis. J Chromatogr A. 2006;1113:255–8. cycle of analysis. The run-to-run reproducibility was evaluat- 9. Zhang X, Zhang Z. Quantification of domoic acid in shellfish samples by capillary electrophoresis-based enzyme immunoassay with ed on a single MIP monolith, and the RSD value for retention electrochemical detection. Toxicon. 2012;59:626–32. times of the analyte was 3.9% (n =5).TheRSDofthe 10. Wu W, Wu X, Lin X, Xie Z, Giesy JP. Quantification of domoic column-to-column and batch-to-batch reproducibility of MIP acid in shellfish tissues by pressurized capillary monoliths was 5.1% (n = 3) and 7.8% (n = 3), respectively. electrochromatography. J Sep Sci. 2009;32:2117–22. The results suggested that the reproducibility and stability of 11. Wang Z, Maucher-Fuquay J, Fire SE, Mikulski CM, Haynes B, Doucette GJ, et al. Optimization of solid-phase extraction and liquid the MIP monolith was acceptable for routine analysis, and the chromatography-tandem mass spectrometry for the determination preparation method was reliable. of domoic acid in seawater, phytoplankton, and mammalian fluids and tissues. Anal Chim Acta. 2012;715:71–9. 12. Regueiro J, Álvarez G, Mauriz A, Blanco J. High throughput analysis of amnesic shellfish poisoning toxins in bivalvemolluscs by Conclusions dispersive solid-phase extraction and high-performance liquid chromatography using a monolithic column. Food Chem. 2011;127: 1884–91. In summary, a MIP monolith prepared via a facile approach 13. Beach DG, Liu H, Quilliam MA. Sensitive determination of was successfully developed for specific recognition of DA. domoic acid in mussel tissue using dansyl chloride derivatization It is the first time to introduce MIP monolith to the field of and liquid chromatography-mass spectrometry. Anal Methods. DA detection. By using PTA as dummy template, the ob- 2015;7:1000–7. tained MIP monolith exhibited excellent chromatographic 14. Zhang W, Yan Z, Gao J, Tong P, Liu W, Zhang L. Metal-organic framework UiO-66 modified magnetite@ silica core–shell magnet- performance and selective recognition ability toward to tar- ic microspheres for magnetic solid-phase extraction of domoic acid get molecule DA. Moreover, the monolith was successfully from shellfish samples. J Chromatogr A. 2015;1400:10–8. applied for chromatographic separation of DA from com- 15. Chen LX, Xu SF, Li JH. Recent advances in molecular imprinting plex biological sample. The easy preparation protocol and technology: current status, challenges and highlighted applications. Chem Soc Rev. 2011;40:2922–42. good recognition properties suggest that MIP monolith has 16. Chen L, Wang X, Lu W, Wu X, Li J. Molecular imprinting: per- great potential for selective removal of low-level contami- spectives and applications. Chem Soc Rev. 2016;45:2137–211. nants from complex samples. 17. Vasapollo G, Sole RD, Mergola L, Lazzoi MR, Scardino A, Scorrano S, et al. Molecularly imprinted polymers: present and Funding Information The authors are grateful to the support of National future prospective. Int J Mol Sci. 2011;12:5908–45. Natural Science Foundation of China (No 21405026). 18. Tse Sum Bui B, Haupt K. Molecularly imprinted polymers: syn- thetic receptors in bioanalysis. Anal Bioanal Chem. 2010;398: – Compliance with ethical standards 2481 92. 19. Lotierzo M, Henry OYF, Piletsky S, Tothill I, Cullen D, Kania M, et al. Surface plasmon resonance sensor for domoic acid based on Conflict of interest The authors declare that they have no conflict of grafted imprinted polymer. Biosens Bioelectron. 2004;20:145–52. interest. Yang F. et al.

20. Kubo T, Nomachi M, Nemoto K, Sano T, Hosoya K, Tanaka N, 24. Lin ZZ, Wang D, Peng AH, Huang ZY, Lin YH. Determination of et al. Chromatographic separation for domoic acid using a fragment domoic acid in shellfish extracted by molecularly imprinted poly- imprinted polymer. Anal Chim Acta. 2006;577:1–7. mers. J Sep Sci. 2016;39:3254–9. 21. Nemoto K, Kubo T, Nomachi M, Sano T, Matsumoto T, Hosoya K, 25. Svec F, Huber CG. Monolithic materials: promises, challenges, et al. Simple and effective 3D recognition of domoic acid using a achievements. Anal Chem. 2006;78:2100–7. molecularly imprinted polymer. J Am Chem Soc. 2007;129:13626– 26. Zheng C, Huang YP, Liu ZS. Recent developments and applica- 32. tions of molecularly imprinted monolithic column for HPLC and 22. Zhou WH, Guo XC, Zhao HQ, Wu SX, Yang HH, Wang XR. CEC. J Sep Sci. 2011;34:1988–2002. Molecularly imprinted polymer for selective extraction of domoic 27. Yang F, Mao J, He XW, Chen LX, Zhang YK. Preparation of a acid from seafood coupled with high-performance liquid chromato- boronate-functionalized affinity hybrid monolith for specific cap- graphic determination. Talanta. 2011;84:777–82. ture of glycoproteins. Anal Bioanal Chem. 2013;405:5321–31. 23. Dan L, Wang HF. Mn-doped ZnS quantum dot imbedded two- 28. Wang FJ, Dong J, Jiang XG, Ye ML, Zou HF. Capillary trap column fragment imprinting silica for enhanced room temperature phos- with strong cation-exchange monolith for automated shotgun pro- phorescence probing of domoic acid. Anal Chem. 2013;85:4844–8. teome analysis. Anal Chem. 2007;79:6599–606.