The quality safety issue of traditional Chinese medicine (TCM) caused by the mycotoxin contamination is of great concern and urgently needs to be addressed. Mycotoxins are the secondary metabolites produced by the fungi under suitable conditions [1], widely distributing in the agricultural products, feeds, dairy products, and TCM samples, and possess the teratogenicity, carcinogenicity and immunosuppression effects to the humans and wildlife [2]. As previously reported, more than 400 mycotoxins have been identified [3], including zearalenone (ZEN), aflatoxins (AFs), deoxynivalenol (DON), ochratoxins (OTs), and fumonisins (FBs), as the most common ones [4].

ZEN is naturally produced by various Fusarium fungi, and mainly contaminates the medicinal materials and cereal crops in trace amounts [5], [6], [7]. In view of the huge economic losses caused by ZEN to agriculture and the threat of ZEN to the health of animals and humans, countries around the world have successively formulated and continuously revised the ZEN limit standards in the foods and drugs, with a limit range of 20–1000 μg kg−1 [8], [9], [10], [11]. To date, several works have reported the ZEN contamination of different samples, involving the feed [5], breakfast cereal [6], and corn [7], in different countries and regions around the world. Unfortunately, the ZEN content in some samples even exceeds the limit standard of the code. In recent years, with the increasing attention paid to the safety of TCMs at home and abroad, the reports of ZEN contamination in TCMs have emerged endlessly [12,13]. According to the literature, ZEN contamination widely exists in TCMs, such as the root and rhizome rich in the starch (0.04–4.53 μg kg−1), as well as the fruit and seed rich in the oil (325.00 μg kg−1) [14]. Specifically, the ZEN-contaminated TCMs, including Coix lacryma-jobi L. var. ma-yuen (Roman.) Stapf, Prunus armeniaca L. var. ansu Maxim., Alpinia oxyphylla Miq., Panax notoginseng (Burk.) F. H. Chen, Polygonum multiflorum Thunb., Scutellaria baicalensis Georgi, Salvia miltiorrhiza Bge., Glycyrrhiza uralensis Fisch., Tribulus terrestris L., and Lygodium japonicum (Thunb.) Sw., etc. [12], [13], [14], are commonly used for the prevention and treatment of various diseases. Moreover, ZEN has the estrogenic effects on various organisms even at a very low concentration level [15]. Considering the low concentration and high toxicity of ZEN as well as the complexity of the TCM matrices, it is essential and urgent to develop the highly selective, sensitive, and simple sample pretreatment methods for monitoring ZEN.

Up to now, sample pretreatment methods, such as the solid-phase extraction (SPE) [16], QuEChERS [17], dispersive liquid-liquid microextraction [18], and magnetic solid phase extraction (MSPE) [19,20], have been explored to enrich ZEN from various matrices. Among them, MSPE is regarded as an efficient, low-cost, and practical sample pretreatment technology for the extraction of ZEN. In general, the magnetic adsorbents can be efficiently separated from the sample solution with the aid of an external magnetic field, avoiding the cumbersome and time-consuming filtration or centrifugation procedure during the SPE operation. In recent years, MSPE has been widely used in the pretreatment of mycotoxins by developing diverse porous magnetic adsorbents [19,20]. Since the structures and properties of the magnetic adsorbents can largely determine the selectivity and sensitivity of the developed MSPE method, the rational design and preparation of efficient magnetic adsorbents for the enrichment of trace ZEN from the complex TCM samples are of great significance and challenging [21].

Microporous organic networks (MONs) are an emerging category of porous materials prepared via the Sonogashira-Hagihara coupling reaction [22], which are widely explored in the adsorption [23], [24], [25], sample pretreatment [22,[26], [27], [28], [29]], and chromatographic separation [22,30]. Owing to the ultrafast separation performance of magnetic units and the advantage of MONs (such as the extremely large specific surface area, excellent solvent and thermal stability, convenient functionalization, and easy combination with other matrix materials), the magnetic MONs have been receiving the increasing attention in sample pretreatment [22], [23], [24], [25], [26], [27], [28], [29], [30]. In addition, the Fe3O4@UiO-66-NH2@MON composite has been utilized as an advanced adsorbent for MSPE of AFs in the food [31]. ZEN is a resorcylic acid lactone derivative with diverse function groups. Generally, the benzene ring, double bond, and alkyls containing structure leads to the good hydrophobicity of ZEN, which can be reasonably extracted onto MONs with highly conjugated networks via the possible hydrophobic and π-π interactions. Additionally, some hydrophilic (e.g. hydroxyl, ester, and carbonyl) groups are involved in ZEN. Therefore, introduction of proper hydrogen bond donor/receptor (-NH2 and -OH) groups into the MONs’ networks can be a feasible way to improve the extraction performance of MONs for ZEN through the synergistic hydrogen bonding, hydrophobic, and π-π interactions.

In the current work, we reported the fabrication of a -NH2 and -OH groups dual-functionalized magnetic MON (Fe3O4@MON-NH2-OH) for the efficient MSPE of trace ZEN from the complex TCM samples prior to the high-performance liquid chromatography (HPLC) determination. Extraction conditions such as the desorption solvent, adsorbent dosage, extraction time, pH, and salt effect affecting the extraction of ZEN were optimized in detail. Fe3O4@MON-NH2-OH provides numerous hydrophobic, π-π, and hydrogen bonding interaction sites, allowing the good extraction for ZEN. By coupling with the HPLC-DAD analysis, the Fe3O4@MON-NH2-OH based MSPE approach was completely validated and used for the detection of ZEN from 10 kinds of TCM samples. This work can demonstrate the great potential of the magnetic Fe3O4@MON-NH2-OH for ZEN enrichment and presents a new strategy to design the novel functionalized magnetic MONs for the extraction of AFs in the complex TCM matrices.