*4.3. Saxitoxins*

As a group of carbamate alkaloid neurotoxins, saxitoxins (STXs) contain sixteen variants, which are commonly associated with "red tides," and found as a paralytic shellfish toxin. Australia has a drinking water guideline of 3 μg L−<sup>1</sup> of STX equivalence. Recently, Bratakou et al. constructed a miniaturized potentiometric STX immunosensor on graphene nanosheets with incorporated lipid films and anti-STX (the natural STX receptor) [168]. The potentiometric STX immunosensor can be easily miniaturized because graphene nanosheets have a high surface area and good conductivity, and incorporate well with the lipid bilayer membrane for immobilizing anti-STX antibody. The potentiometric STX immunosensor exhibits several advantages such as a rapid response time (ca. 5–20 min), low LOD (1 nmol <sup>L</sup>−1) with high sensitivity (ca. 60 mV/decade of toxin concentration), good reproducibility (maximum deviation only 6.8%), reusability, high selectivity and long shelf life (> 1 month). The practicability of the method was demonstrated by detecting STX in lake water and shellfish samples. This graphene nanosheets with incorporated lipid films could be used to develop biosensors for monitoring other toxins.

## *4.4. Brevetoxin B*

Brevetoxins (BTXs) are potent cyclic polyether neurotoxins, which are naturally produced by the marine "red tide" dinoflagellate, *Karenia brevis*. BTX exposure can cause neurological shellfish poisoning (NSP), which has increased in geographical distribution over the past decade [139]. As early as 2012, Tang et al. constructed a magneto-controlled electrochemical immunosensor for sensitive detection of brevetoxin B (BTX-2) in seafood by using guanine-assembled graphene nanoribbons (GGNRs) as molecular tags on a home-made magnetic carbon paste electrode [169]. In this case, the GGNRs were modified by bioconjugates of BSA with BTX-2 (BTX-2-BSA), while monoclonal mouse anti-BTX-2 antibodies were covalently immobilized on the surface of magnetic beads for the capture of BTX-2 through a competitive-type immunoassay format. The formed magnetic immunocomplex was integrated on the electrode with an external magnet, followed by determination in pH 6.5 phosphate-bu ffered solution containing 2 μmol L−<sup>1</sup> Ru(bpy)3Cl2. Compared with pure guanine-labeled molecular tags, the GNR-labeled electrochemical immunoassays show a much wider linear range and lower detection limit. Under optimal conditions, the electrochemical signals decreased by increasing concentration of BTX-2 in the sample. The magneto-controlled immunosensing platform has a wide dynamic range from 1.0 pg mL−<sup>1</sup> to 10 ng mL−<sup>1</sup> with a LOD of 1.0 pg mL−<sup>1</sup> BTX-2. The analytical reliability of the magneto-controlled electrochemical immunosensing platform is demonstrated by the detection of BTX-2 in 12 spiked samples including *S. constricta, M. senhousia and T. granosa*. The as-obtained results are consistent with those of traditional ELISA.

#### *4.5. Okadaic Acid*

The family of okadaic acid (OA) biotoxins consists of OA and its analogues dinophysistoxins 1, 2 and 3 (named as DTX-1, DTX-2 and DTX-3) [170]. As a by-product of harmful algal blooms (HABs), OA originates from the algal genera *Prorocentrum* and *Dynophysis.* Eissa and Zourob developed a direct competitive voltammetric immunosensor for the sensitive detection of OA based on carboxyphenyl-functionalized graphene-modified SPCEs (GSPCEs) [171]. The anti-OA antibodies were immobilized on the GSPE via carbodiimide chemistry, where OA and OA-ovalbumin (OA-OVA) in solution compete for their binding to the immobilized antibody. Benefitting from the unique electrochemical properties of graphene and the stability of the carboxyphenyl layer, the immunosensor exhibits a linear response up to 5000 ng L−<sup>1</sup> with a LOD of 19 pg mL−1. The immunosensor was successfully applied for detecting OA in the spiked shellfish extracts, showing good recovery. Very recently, Ramalingam et al. fabricated an electrochemical microfluidic biochip for detecting OA by using phosphorene-gold (BP-Au) nanocomposite-modified SPCE (as shown in Figure 11) [172]. The as-synthesized BP-Au nanocomposite not only serves as a backbone to the aptamer sequence, but also significantly enhances the electrochemical response of the aptasensor. DPV measurements

revealed a LOD of 8 pmol <sup>L</sup>−1, while a linear range was found between 10 nmol L−<sup>1</sup> to 250 nmol L−1. The electrochemical aptasensor has excellent selectivity and can be employed to detect OA in fresh mussel extracts. The results sugges<sup>t</sup> that the microfluidic electrochemical aptasensor can be served as an easy-to-use POC device for an on-field assay.

**Figure 11. A** microfluidic electrochemical aptasensor for the detection of okadaic acid: (**A**) graphic of the fabricated PDMS microfluidic chip, and (**B**) schematic representation of the process of aptamer-based sensing (adapted from Ramalingam et al. 2019 [172], Copyright 2019 Elsevier B.V. and reproduced with permission).

#### **5. Conclusions and Perspective**

This review has summarized the recent progress in electrochemical biosensing systems for the determination of various microbial toxins by using 2D nanomaterials and their nanocomposites (hereinafter referred to 2D nanomaterials). The literature results demonstrate that the integration of 2D nanomaterials into electrochemical biosensors has led to the significant enhancement of their analytical efficiency, including a high sensitivity (e.g., very low LODs) with a wide linearity range over several orders of magnitude, rapid assaying time, and simplified analytical procedures, and they are also suitable for on-site monitoring. During the determination processes, 2D nanomaterials mainly have two roles: as substrates for efficient immobilization of capturing biomolecules (e.g., anti-toxin antibodies and aptamers) and high active electrochemical probes for signal amplification. Some 2D nanomaterials have multifunctionality, and are capable of playing both of the above roles. Furthermore, the 2D nanomaterial-based electrochemical aptasensors have been proven as reusable platforms for detecting toxins.

Although the 2D nanomaterial-based electrochemical biosensors show grea<sup>t</sup> promise within laboratory investigations, such as the detection of toxins in buffer solutions and/or toxin-spiked samples, the technique remains relatively immature in development compared with standard toxin assaying tools (e.g., HPLC and ELISA), and several technical challenges are still awaiting further investigation. (1) The multiple electrode modification steps are normally required for increasing the recognition performance of the immobilized aptamer or antibody, and reducing background signals. This phenomenon requires manual and tedious work, which not only increases the preparation cost of biosensors, but also leads to poor reproducibility of the results among laboratories. In order to simplify the biosensor construction procedure, future research should increase the reaction efficiency of 2D nanomaterials with biomolecules (such as an antibody and apatmer) and decrease unreacted activity groups on the surface of 2D nanomaterials after biomolecule immobilization. Furthermore, development of automatic methods for modification of 2D nanomaterials on the electrode surface may help to increase the inter-laboratory reproducibility of biosensors. (2) The properties of 2D nanomaterials, including their electrical conductivity, PEC conversion capability and biomolecule immobilization capacity, are strongly dependent on their morphology, such as shape, size, purity, and defects. Therefore, 2D-nanomaterials

should be fully characterized before biosensor fabrication. In further research, researchers are strongly encouraged to establish the synthesis standard of 2D-nanomaterials in order to improve the reproducibility of 2D nanomaterial-based electrochemical biosensors. In addition, the as-proposed synthesis strategy should be easily employed to produce 2D-nanomaterials on a large-scale by simply adjusting the synthesis conditions, such as increasing the amount of reactants. This factor is very important for industrialization of the 2D nanomaterial-based electrochemical sensors. (3) To date, one kind of 2D nanomaterial-based electrochemical biosensor is merely confined to determine a single microbial toxin. Because of coexistence of various microbial toxins in nature, future research should focus on development of a universal biosensor production technology for enabling rapid analysis of various toxins. (4) In order to achieve large-scale application, in particular for on-site monitoring, further efforts should be directed toward the development of 2D nanomaterial-based electrochemical biosensors, which can be used to detect toxins in practical samples such as various agricultural, food stuff, body fluids, and environmental sectors (e.g., lake water and sea water). The practicability of 2D nanomaterial-based electrochemical biosensors could be improved through integration of the biosensor with other techniques such as microfluidic devices and microarrays because miniaturization will help to increase the detection throughput, e.g., recognize multiple elements simultaneously. (5) Currently, aptamers and antibodies are mainly used for recognition of the toxins. In order to obtain high selectivity, the key epitope residues of the aptamer and antibody should be unrestrained after immobilization on the 2D nanomaterials. In addition, the molecular structures of the aptamer and antibody are sensitive to the environmental conditions (such as temperature, ionic strength and interferences from sample matrices). The high apparent affinity of the aptamer and/or antibody could be achieved through immobilization of the aptamer and/or antibody by stereoselective reactions (e.g., chick chemistry, DNA hybridization, biotin-avidin recognition). In addition, future research should aim to increase the biocompatibility of 2D nanomaterials. Finally, we expect commercialization of 2D nanomaterial-based electrochemical biosensors into practical procedures for detecting multiple toxins in practical samples through efforts of researchers in different disciplines, which would give significant benefit to the public.

**Author Contributions:** Conceptualization, Z.L. and X.L. contributed equally for preparation of the draft of this article, M.J. prepared the figures, G.S.G. and Z.W. contributed equally to its further elaboration and discussion. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Natural Science Foundation of China (Grant no. 21775145) to Z.W.

**Conflicts of Interest:** The authors declare no conflict of interest.
