Biomedical Microdevices

, 20:60 | Cite as

A magnetic beads-based portable flow cytometry immunosensor for in-situ detection of marine biotoxin

  • Yuxiang Pan
  • Xinwei Wei
  • Tao Liang
  • Jie Zhou
  • Hao Wan
  • Ning Hu
  • Ping Wang


Okadaic acid (OA), a representative diarrhetic shellfish poisoning toxin, mainly produced by toxigenic dinoflagellates, has significant hazard to public health. Traditional methods for detection of OA can not give the consideration to the need of rapid, high sensitive, quantitative and in-situ detection at the same time. Herein, a new effective detection method of OA was developed based on fluorescence immunosensor and flow cytometry (FCM). In this assay, Streptavidin-coated magnetic beads (MBs) were used as the supporter to immobilize the biotinylated OA. Modified MBs competed with the free OA in the sample solution to bind with the anti-OA monoclonal antibody (OA-MAb). The R-phycoerythrin (R-PE) dye labeled IgG was served as a secondary antibody to perform fluorescence detection. A portable flow cytometry was applied for the in-situ fluorescence quantification. The results showed that the OA concentration was inversely proportional to the R-PE fluorescence intensity. The detection method took within 50 min with a limit of detection (LOD) was 0.05 μg/L and range from 0.2 to 20 μg/L for OA detection. Moreover, the matrix effect and the recovery rate were assessed during real sample measurement, showing a high recovery. Performance features such as high sensitivity, low LOD, speediness and simplicity of the analysis protocol, shows this biosensing-systems as a promising tool for routine use.


Fluorescence immunosensor Portable flow cytometry Magnetic beads-based flow cytometry In-situ detection Okadaic acid 

1 Introduction

Okadaic acid (OA), a common marine biotoxin produced by Dinophysis and Prorocentrum dinoflagellates, is the main representative diarrheic shellfish poisoning (DSP) toxin (Yasumoto and Murata 1993). It easily accumulates in digestive glands of shellfishes. When human consumes these contaminated seafood, a series of syndromes may occur such as gastrointestinal dysfunctions. Furthermore, studies carried out on animals suggested OA as a potential tumor promoter and immune toxin (EFSA 2008). In recent years, due to the deterioration of the marine environment, contaminated shellfishes attracte a common concern worldwide. In the European Union, the regulation (CE) No. 853/2004 established a maximum OA permitted level of 160 μg/kg in mussels. It is difficult to assess the existence of OA in real samples due to OA neither changes the odor or taste of contaminated seafood nor damages shellfishes. The standard method for OA toxicity screening is mouse bioassay, which suffers from low sensitivity, low specificity and additional ethical problems. Due to its drawbacks, the European Union introduced rapid, sensitive and accurate alternative methods, such as liquid chromatography coupled with mass spectrometry (LC-MS), PP2As inhibition assays and cytotoxicity assays. These techniques are well-proven and widely accepted, which have high specificity and sensitivity (Goto et al. 2001). Nevertheless, a rapid, high sensitive, in-situ and specific method is still urgently demanded for environmental monitoring and food security warning.

Biosensors are potential alternative approaches for rapid, sensitive and high specific detection of the OA (Campas et al. 2007). Cell-based biosensor, which is introduced to detect marine toxins years ago, has outstanding features in high throughput, low cost and good reproducibility (Zou et al. 2016). Enzyme inhibition-based biosensor presents high sensitivity, short analysis time without ethical considerations (Campàs and Marty 2007; Campàs et al. 2005). However, these methods are deficient in detecting specificity. Immunosensor has high specificity and sensitivity due to the antigen–antibody reaction. The strong specificity and affinity between antigen and antibody have lead a trend in the development of immunosensor. More and more micro-molecule toxin quantification are studied using indirect enzyme linked immunosorbent assay (id-ELISA) kit and electrochemical detection based on the interaction between antigen and antibody (Kreuzer et al. 1999; Hayat et al. 2011). However, the coating of antigen or antibody on plates and electrochemical sensor surface make them expensive. Nowadays, magnetic beads (MBs) have been increasingly applied in the development of immunoassays for food analysis (Llamas et al. 2007; Gessler et al. 2006) and clinical diagnosis (Haukanes and Kvam 1993; Bangs 1996) due to their large surface area for convenient functional groups attachment, efficient magnetic force, slow sedimentation rate during incubation, which can be easily manipulated by an external magnetic field (Centi et al. 2007; Hayat et al. 2012). These advantages of MBs can highly shorten washing time, decrease matrix effect, and simplify the fabrication processes. In the meantime, the noncovalent but exceptionally strong interaction of avidin-biotin has been used in various systems, such as gene map, selective absorption of cells and immobilization of antigen due to its high affinity and plentiful binding sites (Guesdon et al. 1979). To further improve the detection efficiency, sensitivity and reproducibility, Streptavidin coated MBs is a good choice of interaction supporter.

Moreover, various signal transduction systems have been used to improve sensitivity of immunosensors such as quartz-crystal microbalance, surface plasma resonanceand screen-printed electrode (Tang et al. 2002; Prieto-Simón et al. 2010). These signal transduction systems are high sensitive and specific, but these methods are facing severe challenges of in-situ analysis due to the huge equipment and complicated operations. Compared to other methods, portable flow cytometry (FCM), as a precise fluorescent detection instrument is widely used in cell classification, disease diagnoses, DNA analysis and chromatin association because of its excellent optical properties and high precision (Riccardi and Nicoletti 2006; Jung et al. 1993; Siena et al. 1991; Forment and Jackson 2015). In contrast to microplate readers that directly analyse the total fluorescence intensity in one well, FCM detects the fluorescence of each MB in the fluid sample one by one, which guarantees higher accuracy and sensitivity. The analysis time of FCM is as short as 10 s, which is faster than microplate readers and Bionic e-Eye (Su et al. 2016). Consequently, we adopted a portable FCM to detect the fluorescence intensity, which can not only achieve high sensitivity but also demonstrate good portability and ease of handling, making in-situ and precise OA detection achievable.

Herein, a practical bioanalysis method for rapid and in-situ detection of OA was developed by combination of an MB-based fluorescence immunosensor and a portable FCM for the first time. This immunosensor combined the advantages of antigen–antibody system, avidin-biotin system and MBs that can show high specificity, excellent sensitivity and fast response time. Cooperating with the portable FCM, this method realized the rapid in-situ analysis of OA. In addition, the flow cytometry fluorescence immunosensor (FCMFI) was compared with other enzyme-linked immunosorbent assay for OA detection to validate its performances. All the details will be discussed in the following sections.

2 Experimental and methods

2.1 Reagents and materials

Okadaic acid (OA) potassium salt was purchased from Algal Science Inc. (Taiwan) and dissolved in ethanol. Streptavidin-coated magnetic beads Dynabeads MyOne C1 (Invitrogen, USA) were used as the support to immobilize the OA. EZ-link aminePEO3-biotin was from Pierce (France). N-hydroxysuccinimide (NHS), N-(3-dimethylaminopropyle)-N′-ethyle-carbodiimide hydrochloride (EDC), hydrogen peroxide (H2O2), buffer components, Tween 20, 3,3′,5,5′-tetramethylbenzidine (TMB) liquid substrate and bovine serum albumin (BSA) were purchased from Aladdin (China). Anti-OA antibody (developed in mouse) was supplied by Abcam (UK). R-Phycoerythrin (R-PE) labeled goat anti-mouse IgG and HRP labeled goat anti-mouse IgG were purchased from Jackson Immuno Research Laboratories, Inc. (USA). Eppendorf tubes were obtained from Eppendorf (German). Microtiter plates were obtained from BD Falcon (USA).

2.2 Appratus

Colorimetric measurements were performed with SpectraMax Paradigm (Molecular Devices, USA). A Moxi Flow portable flow cytometry (Orflo, USA) was used as an in-situ fluorescence immunoassay platform. Moxi Flow only weights 1250 g with an overall dimension of 8 × 4.5 × 4.5 in. IKA horizontal shaker Vortex 4 was used to stir (IKA, German). A magnet (Aladdin, China) was used to separate the magnetic beads from the sample solution. A centrifuge (Organomation Association, Inc., Berlin/USA) was used for real sample preparation.

2.3 Synthesis of biotinylated OA

OA is a micro molecule toxin (MW 805) that is unable to adopt sandwich assay schemes. Therefore, we adopted the immunoassays based on the competition strategy. OA was conjugated to the biotin using EZ-Link amine-PEO3-biotin as a biotinylating agent. 60 μL mixture of NHS and EDC (both 26 mM) was added to 100 μL of OA sample (0.5 g/L in 0.1 M MES buffer, pH 5.5). The mixture was stirred for 1 h at room temperature. Then, 40 μL of 40 mM biotin-LC-PEO3-amine in MES buffer were added into the mixture and stirred for 24 h at room temperature for reaction. 600 μL of PBS (1×) were added to the reaction mixture. At last, the reaction mixture was transferred into a dialysis membrane (MW = 10,000) and dialyzed against 0.01 M PBS buffer for 4 h, in which dialysate was replaced every 8 h. The final prepared biotinylated-OA was stored at 4 °C.

2.4 MBs modification

100 μL of MBs (10 g/L) were injected in a eppendorf tube (EP tube), and the beads were collected using a magnet and subsequently washed for 3 times PHwith 0.1 M MES buffer (pH 5.0), and then the supernatant was removed after magnetic collection. 50 μL biotinylated OA were added into the MB solution to incubate 1 h at room temperature; and then the modified MBs were washed 3 times with 50 μL PBS (1×). At last, the biotinylated OA modified MBs (OA-MBs) were suspended in 100 mL PBS with 0.1% Tween®-20 and stored at 4 °C.

2.5 Flow cytometry fluorescence immunosensor

Firstly, 100 μL OA-MBs at dilution of 1/100 in PBS (1×) and 20 μL OA standard solution at different concentrations were added into the reaction EP tubes. After shaking the tubes for 30 s, 80 μL anti OA-MAb solution at dilution of 1/1000 in PBS (1×) were added into the tubes and incubated for 20 min on a roller. After removing the waste supernate by magnet, 100 μL R-PE labeled goat anti-mouse IgG secondary antibody solution at dilution of 1/200 in PBS (1×) were added and incubated for 20 min on a roller. All the processes were carried out at room temperature and washing steps were performed by adding PBS (1×) between each steps (Fig. 1). With the magnetic separation of magnets, the experimental efficiency has been greatly improved compared to the traditional centrifuge separation. The use of MBs make a break though for eliminating tedious washing steps, which is a typical drawback of a competitive raction-based detection platform.
Fig. 1

The construction process of fluorescence immunosensor system for the competitve OA detection

The employment of this portable FCM enables rapid and in-situ detection of OA with high sensitivity and precision based on the fluorescence immunosensor. Additionally, no pre-treatment or cleaning step was required before measurement. The R-PE labelled secondary antibody was used to amplify the fluorescence signals. The portable FCM system used 580 nm detection filter to quantify the fluorescence of R-PE labeled IgG that bounds to the MBs surface one by one. Then, the total fluorescence intensity of functionalized MBs was calculated by the software of FCM. With the special designed microfluid chip, the amount of sample solution required is only 75 μL (Fig. 2a). X axis of the scatter diagram represnts the particle diameter, Y axis of the scatter diagram represents particle surface complexity. In order to eliminate the interference of impurity particles in the sample solution, the relative particle diameters were used to gate the detection MBs (Fig. 2b). The more OA is contained in the sample solution, and the less OA-MAb will combine with the OA-MBs, so the weaker total counting fluorescence intensity of the OA-MBs is detected by FCM. Consequently, the R-PE fluorescence intensity is inversely proportional to the OA concentration in sample solution (Fig. 2c). All the fluorescence detection processes can be performed on the portable flow cytometry (Fig. 2d).
Fig. 2

a Working principle of FCM for fluorescence detection. b The gate process for the detection MBs and measuring the fluorescence intensity of MBs. c Typical fluorescence intensity curves of FCMFI under different OA concentration. d Image of portable flow cytometry

2.6 Colorimetric id-ELISA protocol

In the id-ELISA, 300 μL PBS (1×) containing 2% BSA was used to block the 96-well microtiter plates for 1 h at room temperature. 100 μL OA-MBs at dilution of 1/100 were added into the wells, and the buffer was removed using the magnet. Then 20 μL OA samples and 80 μL of anti OA-MAb solution with a dilution ratio of 1/1000 in PBS (1×) were mixed and incubated for 20 min on a roller for competitive reaction. Then 100 μL HRP labeled goat anti-mouse IgG solution at dilution of 1/200 in PBS (1×) were added and incubated 20 min on a roller. All the processes were carried out at room temperature, and washing steps were performed using PBS (1×). For the colorimetric detection step, 100 μL of TMB solution was incubated for 15 min and the absorbance values were measured at 620 nm.

2.7 Preparation of real samples

The real samples of OA were extracted from the shellfish according to Hayat’s method. 2 g mussels (Hangzhou, Zhejiang) were blended with a hand held homogenizer and extracted with methanol: water (80:20; 0.6 g mL−1) for 5 min at 4000 rpm. Crude extracts were centrifuged for 5 min at 2000 rpm. The supernatant was filtrated through 0.2 μm filters and the sediment was evaporated in a centrifuge; the residue was resuspended in 1 mL of PBS (1×). The filtrated samples were spiked with the stock solution of OA in the range from 2 to 10 μg/L to study matrix effect and OA recovery.

2.8 Calibration curve of immunoassay and fluorescence immunosensor

OA standard solution was diluted to the desired concentrations between 0 and 50 μg/L to establish the calibration curve. Blank sample without OA was used to assess the matrix effect, and recovery was calculated by spiking sample. The specificity of the FCMFI was also analyzed by introducing OA analogues dinophysistoxin-1 (DTX-1), dinophysistoxin-1 (DTX-3) toxins and shellfish extract as the interferences.

3 Results and discussion

In the past years, sandwich assay schemes are widely used for specific protein detection. However, due to the small molecule weight (MW 805), it is impossible for OA detection. Fortunately, immunoassays based on the competition strategy are highly matched to these cases (Bilitewski 2000). It was proved that colorimetric id-ELISA analysis had good performance in biochemical assays. In our work, the ability of FCMFI to assess low levels of OA in real samples was demonstrated and compared with colorimetric id-ELISA. In the meantime, experiment parameters were optimized for further routine use. The aim of study was to reduce the analysis time, improve sensitivity and portability with respect to the previously reported methods.

3.1 Optimization of MBs modification and reaction parameters

Firstly, experiments were performed to verify the modification of OA-MBs and optimize parameters. The amout of biotinylated OA for the MBs modification was optimized ranging from 10 to 50 μL. In order to eliminate impurity particle interference, the gating MBs process ran through each parameters optimization. The fluorescence signal intensity increased in the range of 10 μL to 50 μL, but no significant difference was observed between 50 and 100 μL (Fig. 3a). Considering the modification efficiency and sample cost, 50 μL was selected as the amount of biotinylated OA for modification. Meanwhile, studies showed that the antibody concentration was related to the fluorescence signal intensity, the sensitivity and the background noise of immunoassays. High antibody concentration increased the signal intensity but decreased the sensitivity of immunoassays, and led to background noise (Andrew 2001). Therefore, experiments with different dilution ratios of anti OA-MAb were also performed to determine appropriate antibody concentration. With the increase of antibody concentration, the fluorescence signal intensity increased accrodingly. Negligible intensity difference was observed between 1/500 and 1/1000, so the dilution ratio of 1/1000 was selected as the optimized concentration of anti OA-MAb (Fig. 3b). The incubation time during the detection was another crucial factor, especially for in-situ tests. We compared the incubation time during competitive reaction, and concluded that 20 min can be applied for competitive reaction since the absorbance values were similar to longer incubation time (Fig. 3c).
Fig. 3

a Optimization of biotinylated OA amount for MBs modification. b Optimization of OA-MAb dilution for OA detection. c Optimization of antigen-antibody reaction incubation time. All the data are represented by means ± SD (standard deviation), n = 3

3.2 FCMFI for OA detection

To calibrate the immunosensor for OA detection, OA with the concentration ranging from 0 to 20 μg/L was detected by id-ELISA and the immunosensor using the optimized parameters. Competition assays were performed using free OA. In principle, the higher concentration of OA presented in the sample induces the less fluorescence signal intensity. As shown in Fig. 4a, FCMFI (slope was 48.16, r2 was 0.9922) had higher sensitivity and linearity than id-ELISA (slope was 34.35, r2 was 0.9886) at present (Fig. 4b). The limit of detection (LOD) and working range were analyzed for further applications. The triple standard deviation (STD) was used to calculate the LODs of FCMFI and id-ELISA. The LOD of FCMFI was 0.05 μg/L, and that of id-ELISA was 0.5 μg/L (Fig. 4c). According to the results in Fig. 4a, FCMFI has the good linearity between 0.5 and 20 μg/L. Consequently, the effective detection range was 0.5–20 μg/L for FCMFI. The sampling of the microplate reader was the average fluorescence value in each well of the microtiter plate. As a comparison, FCM analyzed fluorescence value of each magnetic bead and generated a scatter diagram of MBs fluorescence. The accurate and precise analysis of FCM resulte in higher sensitivity, linearity and lower LOD of FCMFI.
Fig. 4

a Standard curve obtained with FCMFI and id-ELISA for OA detection. b Sensitivity of FCMFI and id-ELISA for OA detection. c LOD of FCMFI and id-ELISA for OA detection. All the data are represented by means ± SD (standard deviation), n = 3

In order to investigate the anti-interference ability of the FCMFI from other DSP toxins that may exist in real samples, we chose OA analogues DTX-1, DTX-3 and shellfish extract as interferents to evaluate the specificity of the immunosensor. The concentration of each toxin performed in this specificity assay was 20 μg/L. The PBS solution without toxins acted as a control group, and the results showed that this immunosensor has a good specificity in Fig. 5a. As an analytical method for in-situ applications, high efficient detection is also very important. Due to the employment of MBs, the washing and separation steps became convenient and efficient. The matrix effect was also minimized and total reaction time cost decreased to 50 min. We test the stability of FCMFI on the initial day, and 30th day. The sensitivity of calibration curves changes from initiate day (slope = 48.16) to (slope = 47.04) on 30th day indicating the good stability for OA detection in the long term (Fig. 5b). It demonstrated that the FCMFI has the suitable stability for OA detection. Fig. 5c showed the results obtained by the proposed FCMFI were highly correlated with those obtained by the id-ELISA.
Fig. 5

a The selectivity test of the FCMFI to OA, DTX-1, DTX-3 and PBS solution. b Comparison of calibration curves on 0 and 30th days (n = 3 for each results). c Correlation between the results of the analysis of spiked samples with the FCMFI and id-ELISA

3.3 Detection of OA in spiked mussel extracts

Considering the vital importance of practical application, the recovery, the extraction efficiency and the matrix effect of real sample were studied using spiked mussel samples. To assess the matrix effect of the FCMFI, we compared the data obtained from spiked mussels extract and standard OA solutions. No significant difference was observed between results obtained from two solutions, which proved that this method featured outstanding matrix effect control. The low matrix effect was owing to the specificity of antigen-antibody reaction, large surface area and efficient magnetic collection of MBs. This prominent advantage validated that our immunosensor was more suitable for real sample analysis than other biosensors such as cell-based biosensor and PP2A biosensor that have higher matrix effect. Then, mussel extracts were spiked with the stock solution of OA in the range from 1 to 20 μg/L after filtration. The recovery results of both two methods were showed in the Table 1, which demonstrated that the two methods had good recovery and met the standards of European Commission (160 μg kg−1). In the meantime, the precision and accuracy were calculated using the triplicate measurements. All the results obtained from the FCMFI indicated that the FCMFI can effectively decrease the demand of pre-treatment for real sample and can be used for real sample analysis.
Table 1

Results for the determination of OA in shellfish extracts using the FCMFI


OA added (μg/L)

OA Detected (μg/L)

% Recovery



0.96 ± 0.11




1.85 ± 0.35




4.82 ± 0.80




9.40 ± 1.10




21.60 ± 1.90


In this study, we focused on establishing a portable and highly sensitive biosensor for OA in-situ test. Compared with previos reported immunosensors for OA detection, FCMFI combined the advantages of rapid, high sensitivity and in-situ quantitative analysis (Table 2). Considering the preparation of real samples, the OA in real sample is diluted to 1/20. Comapared to the LOD of LC-MS (0.026 μg/g), the LOD of FCMFI is 0.001 μg/g in real sample (Louppis et al. 2010). All the above results demonstrated that the FCMFI was a potential platform for OA in-situ test.
Table 2

Comparison between FCMFI and previous reported immunosensors for OA detection


LOD (μg/L)

Required time


In-situ detection



<1 h





2 h





<1 h



Test strip


25 min



4 Conclusion

For the first time, we proposed a portable fluorescence immunosensor based on MBs and FCM for the in-situ detection of OA, which combined the advantages of MBs with FCM and resulted in a rapid, high performance and in-situ analytical method. The use of MBs made a break though for eliminating tedious washing steps, improved the sensitivity of immunosensor and eliminated the matrix effect from mussel extract. Moreover, the combination of the portable flow cytometry miniaturized the analysis equipment and made the OA detection system portable. The LOD and analysis time of the FCMFI obtained was lower than those obtained from previous reported immunosensors (Llamas et al. 2007; Centi et al. 2007; Su et al. 2016), which can not give the consideration to the need of rapid, high sensitive, quantitative and in-situ detection at the same time. Compared with previous methods, the new method simultaneously achieved rapid, in-situ and high sensitive quantitative analysis for OA detection. Meanwhile, this assay only needs to adjust the type of antigen-antibody for other small molecule marin biotoxin detection. As demonstrated by the above results, our developed FCMFI will provide a very promising portable instrument for the detection of marine biotoxin in-situ.



This work was supported by a key project of the Natural Science Foundation of China (No. 31627801), International Cooperation Project of Natural Science Foundation of China (No. 61320106002,31661143030,), National 973 Project of China (No.2015CB352101), and Natural Science Foundation of China (No. 31571004).


  1. R. Andrew, Clinical measurement of steroid metabolism. Best Pract. Res. Clin. Endocrinol. Metab. 15, 1–16 (2001)CrossRefGoogle Scholar
  2. L.B. Bangs, New developments in particle-based immunoassays: Introduction. Pure Appl. Chem. 68, 1873–1879 (1996)CrossRefGoogle Scholar
  3. U. Bilitewski, Peer reviewed: Can affinity sensors be used to detect food contaminants? Anal. Chem. 72, 692 A–701 A (2000)CrossRefGoogle Scholar
  4. M. Campàs, J.-L. Marty, Enzyme sensor for the electrochemical detection of the marine toxin okadaic acid. Anal. Chim. Acta 605, 87–93 (2007)CrossRefGoogle Scholar
  5. M. Campàs, D. Szydlowska, M. Trojanowicz, J.-L. Marty, Towards the protein phosphatase-based biosensor for microcystin detection. Biosens. Bioelectron. 20, 1520–1530 (2005)CrossRefGoogle Scholar
  6. M. Campas, B. Prieto-Simón, J.-L. Marty, Biosensors to detect marine toxins: Assessing seafood safety. Talanta 72, 884–895 (2007)CrossRefGoogle Scholar
  7. S. Centi, S. Laschi, M. Mascini, Improvement of analytical performances of a disposable electrochemical immunosensor by using magnetic beads. Talanta 73, 394–399 (2007)CrossRefGoogle Scholar
  8. EFSA, Opinion of the scientific panel on contaminants in the food chain on a request from the European Commission on marine biotoxins in shellfish—Okadaic acid and analogues. EFSA J. 589, 1–62 (2008)Google Scholar
  9. J.V. Forment, S.P. Jackson, A flow cytometry-based method to simplify the analysis and quantification of protein association to chromatin in mammalian cells. Nat. Protoc. 10, 1297–1307 (2015)CrossRefGoogle Scholar
  10. F. Gessler, K. Hampe, M. Schmidt, H. Böhnel, Immunomagnetic beads assay for the detection of botulinum neurotoxin types C and D. Diagn. Microbiol. Infect. Dis. 56, 225–232 (2006)CrossRefGoogle Scholar
  11. H. Goto, T. Igarashi, M. Yamamoto, M. Yasuda, R. Sekiguchi, M. Watai, et al., Quantitative determination of marine toxins associated with diarrhetic shellfish poisoning by liquid chromatography coupled with mass spectrometry. J. Chromatogr. A 907, 181–189 (2001)CrossRefGoogle Scholar
  12. J.L. Guesdon, T. Ternynck, S. Avrameas, J.L. Guesdon, J. Ternyck, S. Avrameas, The use of avidin-biotin interaction in immunoenzymatic techniques. J. Histochem. Cytochem. 27, 1131–1139 (1979)CrossRefGoogle Scholar
  13. B.I. Haukanes, C. Kvam, Application of magnetic beads in bioassays. Bio/Technology 11, 60–63 (1993)Google Scholar
  14. A. Hayat, L. Barthelmebs, J.-L. Marty, Enzyme-linked immunosensor based on super paramagnetic nanobeads for easy and rapid detection of okadaic acid. Anal. Chim. Acta 690, 248–252 (2011)CrossRefGoogle Scholar
  15. A. Hayat, L. Barthelmebs, A. Sassolas, J.-L. Marty, Development of a novel label-free amperometric immunosensor for the detection of okadaic acid. Anal. Chim. Acta 724, 92–97 (2012)CrossRefGoogle Scholar
  16. T. Jung, U. Schauer, C. Heusser, C. Neumann, C. Rieger, Detection of intracellular cytokines by flow cytometry. J. Immunol. Methods 159, 197–207 (1993)CrossRefGoogle Scholar
  17. M.P. Kreuzer, C.K. O'Sulliva, G.G. Guilbault, Development of an ultrasensitive immunoassay for rapid measurement of okadaic acid and its isomers. Anal. Chem. 71, 4198–4202 (1999)CrossRefGoogle Scholar
  18. N.M. Llamas, L. Stewart, T. Fodey, H.C. Higgins, M.L.R. Velasco, L.M. Botana, et al., Development of a novel immunobiosensor method for the rapid detection of okadaic acid contamination in shellfish extracts. Anal. Bioanal. Chem. 389, 581–587 (2007)CrossRefGoogle Scholar
  19. A.P. Louppis, A.V. Badeka, P. Katikou, et al., Determination of okadaic acid, dinophysistoxin-1 and related esters in Greek mussels using HPLC with fluorometric detection, LC-MS/MS and mouse bioassay. Toxicon 55(4), 724–733 (2010)CrossRefGoogle Scholar
  20. B. Prieto-Simón, H. Miyachi, I. Karube, H. Saiki, High-sensitive flow-based kinetic exclusion assay for okadaic acid assessment in shellfish samples. Biosens. Bioelectron. 25, 1395–1401 (2010)CrossRefGoogle Scholar
  21. C. Riccardi, I. Nicoletti, Analysis of apoptosis by propidium iodide staining and flow cytometry. Nat. Protoc. 1, 1458–1461 (2006)CrossRefGoogle Scholar
  22. S. Siena, M. Bregni, B. Brando, N. Belli, F. Ravagnani, L. Gandola, et al., Flow cytometry for clinical estimation of circulating hematopoietic progenitors for autologous transplantation in cancer patients. Blood 77, 400–409 (1991)Google Scholar
  23. K. Su, X. Qiu, J. Fang, Q. Zou, P. Wang, An improved efficient biochemical detection method to marine toxins with a smartphone-based portable system − bionic e-eye. Sensors Actuators B Chem. 238, 1165–1172 (2016)CrossRefGoogle Scholar
  24. A.X. Tang, M. Pravda, G.G. Guilbault, S. Piletsky, A.P. Turner, Immunosensor for okadaic acid using quartz crystal microbalance. Anal. Chim. Acta 471, 33–40 (2002)CrossRefGoogle Scholar
  25. T. Yasumoto, M. Murata, Marine toxins. Chem. Rev. 93, 1897–1909 (1993)CrossRefGoogle Scholar
  26. L. Zou, Q. Wang, M. Tong, H. Li, J. Wang, N. Hu, et al., Detection of diarrhetic shellfish poisoning toxins using high-sensitivity human cancer cell-based impedance biosensor. Sensors Actuators B Chem. 222(, 205–212 (2016)CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Biosensor National Special Laboratory, Key Laboratory for Biomedical Engineering of Education Ministry, Department of Biomedical EngineeringZhejiang UniversityHangzhouChina
  2. 2.State Key Laboratory of Transducer TechnologyChinese Academy of SciencesShanghaiChina
  3. 3.Key Laboratory of Optoelectronic Devices and Systems of Ministry of Education and Guangdong Province, College of Optoelectronic EngineeringShenzhen UniversityShenzhenChina
  4. 4.Department of Medicine, Biomaterials Innovation Research Center, Center for Biomedical EngineeringBrigham and Women’s Hospital, Harvard Medical SchoolCambridgeUSA

Personalised recommendations