Food poisoning from naturally occurring marine toxins is a worldwide public health issue, and it poses economic issues/concerns for the food industry. Marine toxins, such as saxitoxin (STX) and domoic acid (DA), are associated with algae blooms and can bioaccumulate in shellfish and herbivorous fishes causing food poisoning [1
]. The frequency of STX and DA producing algae blooms is on the rise [3
], possibly due to climate change, leading to increasing potential for adverse environmental, economic, and health implications.
STXs, produced by dinoflagellates, are the most well studied cause of paralytic shellfish poisoning (PSP) which is also the most common and lethal form of marine toxin poisoning [2
]. PSP results in paralysis of muscles throughout the body and at higher concentrations can cause death. STX, is a heat stable toxin whose primary target is the voltage-gated sodium channels in the nerve and muscle cells of the body affecting the gastrointestinal and neurological systems [4
]. Because of its potency, STX has been identified as a potential biothreat agent and is regulated as a select agent by the U.S. Center for Disease Control (CDC).
DA is a representative of toxins that cause amnesic shellfish poisoning (ASP). Besides typical food poisoning symptoms, short term memory loss, confusion, and disorientation are observed in ASP. Although higher levels of DA are needed for intoxication, it is a heat stable toxin that can cause kidney damage at levels several orders of magnitude lower than those that cause neurological symptoms [5
]. It acts by stimulating the glutamate receptors.
There is no known antidote for either STX or DA poisoning; therefore, the ability to detect the presence of toxin is vital for the early administration of the correct supportive care [7
]. Detection of these and related toxins are also of importance for environmental monitoring to limit use of shellfish from that region. Until the last decade, the mouse bioassay (MBA) was the only approved method for the detection of marine toxins. The MBA provides toxicity information about the food sample but not toxin identification. Moral and ethical issues with this method (animal usage) have led to the development of other methods. Liquid chromatography such as HPLC (with fluorometric detection for STX), LC-UV for DA, LC-MS and UPLC-MS/MS [2
], are methods that are able to detect and identify variants of the toxins. HPLC-FD and LC-UV have been validated for use in the EU. Surface Enhanced Raman Scattering (SERS) has also been demonstrated [9
]. While these methods have the specificity, they require extensive sample preparation, expensive equipment, highly trained personnel, and the availability of analytical standards. In addition, interferences from complex matrices have been observed.
Alternatively, biologically-based assays have been developed that can monitor toxicity similar to the MBA and, thereby, provide for rapid screening of samples. Cell-based sensors have been developed for PSP toxins, which utilize only cells instead of animals [11
]. In 2012, Van Dolan et al. described a receptor-based assay that was able to detect PSP toxins near the regulatory limits [4
]. Even though cells and receptors have been employed to identify marine toxins and their toxicity, most bio-based methods employ antibodies as a rapid screening tool. Several different immunoassays have successfully been applied to the detection of marine toxins that use either polyclonal (pAbs) or monoclonal antibodies (mAbs) against STX and DA [15
]. Antibodies have been used in many different detection formats such as: enzyme-linked immunosorbent assays (ELISAs) [20
], surface plasmon resonance (SPR) [22
], fluorescence/chemiluminescence-based microarray assays [26
] flow cytometry [28
], and lab-on-a-chip using fluorescence and magnetic particles [30
]. Recently, two groups have been developing lateral flow immunoassays (LFIs) for the detection of PSP toxins and DA as rapid, low tech screening assays [31
]. Campbell et al. reviews biological toxin binders in detail in their 2011 paper [34
In the last few years, concerns have grown over the reliability and reproducibility of standard antibodies. A. Bradbury wrote an article in Nature describing the issues with using traditional monoclonal and polyclonal antibodies [35
]. These issues include availability of antibody (polyclonals -batch-to-batch variability), use of animals, and for monoclonals, the death of hybridoma cell lines. A viable alternative is recombinant antibodies including single-chain variable fragment (scFv), and single domain antibodies (sdAb) [36
]. Recombinant constructs (scFv) that consist of linked heavy and light variable domains that make up the binding domains of conventional antibodies have been produced for the sensitive detection of DA, as well as a number of other biotoxins [37
]. Recombinantly produced binding elements serve to “preserve” mAbs as they can be easily recreated from their sequence information, thereby reducing variability and eliminating issues with loss of cell lines. An additional advantage of scFv over mAbs is the ability to produce reagent recombinantly in E. coli
and the potential to produce fusion constructs with enhanced utility that can potentially be tailored to particular sensor systems [38
]. While not universal, improvements in stability, affinity, and diversity have been observed in scFvs, for example, improvement in both stability and affinity was demonstrated by McConell et al. [42
Herein, we demonstrate recombinantly produced antibody recognition domains, scFvs, in a microsphere-based competitive immunoassay for the detection of STX and DA. This work utilized the previously described anti-DA binding fragment [37
] in conjunction with an anti-STX binding domain that was synthesized from the sequence of an anti-STX mAb [15
]. In addition to detection in buffer, we show the utility of the assay in shellfish matrices.
4. Experimental Section
MagPlex microspheres and MAGPIX Instrument were from Luminex Corporation (Austin, TX, USA). Streptavidin–R phycoerthryin (SA-PE) was purchased from Columbia Biosciences (Frederick, MD, USA) and the biotinylated anti-streptavidin (anti-SA) was obtained from Vector Laboratories (Burlingame, CA, USA). Phosphate-buffered saline with Tween and BSA (PBSTB) were prepared from phosphate-buffered saline with 0.05% Tween-20 (PBST) packets and 0.1% bovine serum albumin (BSA) from Sigma-Aldrich Chemical Co. (St. Louis, MO, USA). The saxitoxin dihydrochloride, 100 µg/mL in 20% ethanol-water solution (pH 3), was the kind gift of Dr. Sherwood Hall, US FDA, College Park, MD, USA. Domoic acid was purchased from Calbiochem EMD Biosciences (San Diego, CA, USA).
4.2. scFv Construction and Protein Production
The genes coding for the variable heavy and light domains from anti-STX mAbs 5F7 and 1E8 were synthesized with codons optimized for expression in E. coli (GenScript, Piscataway, NJ, USA). The variable heavy domain was PCR amplified from the plasmid provided by Genscript with primers that introduced flanking NcoI and NotI sites; the light chain was similarly amplified to introduce flanking BamH1 and XhoI sites. The PCR products were digested and gel purified prior to ligation into a pET22b derivative containing the linker sequence AAAGSGSGGGSSGGGSSGGGSGASGS, between the NotI (coded by AAA) and the Bam H1 (coded by the C-terminal GS) sites. Similarly, the sequence for the variable heavy and light chain of the anti-DA scFv DA24cB7, joined by a 15 amino acid linker, was synthesized with flanking NcoI and XhoI sites (GenScript) and cloned into the pET22b expression vector.
The anti-STX scFvs and the anti-DA scFv were grown and produced essentially as described previously [48
]. Cultures were grown at 25 or 30 °C in terrific broth (TB). Fifty mL overnight cultures were used to inoculate 500 mL of TB and were grown for 3 h before induction by addition of 1 mM isopropyl β-d
-1-thiogalactopyranoside (IPTG). After induction, cultures were grown 3 h before pelleting and subjected to an osmotic shock protocol [44
]. The scFv were purified from the shockate by immobilized metal affinity chromatography followed by size exclusion chromatography using a Superdex 75 10/300 GL column (GE Healthcare, Pittsburgh, PA, USA) and a BioLogic Duo-Flow Chromatography System (Bio-Rad, Hercules, CA, USA). The yield of the scFv was determined by UV spectroscopy, measuring the absorbance at 280 nm using a NanoDrop 2000 (Thermo Fisher, Waltham, MA, USA).
4.3. Food Matrices Preparation
A simple sample preparation protocol was used to extract the toxins as compared to the more stringent buffer conditions described in Campbell et al. and Szkola et al. [27
]. Bay scallops (live, frozen) and live oysters were purchased from a local grocery store. The oysters were placed at −20 °C overnight. The frozen bay scallops (200 g) were blended in a small Cuisinart food processor until smooth with no additional liquid. The puree was placed in 15 mL centrifuge tubes (VWR, Radnor, PA, USA) in 5 g aliquots and frozen until testing. For the frozen oysters, 120.5 g were blended with no additional liquid, aliquoted in 5 g sizes, and frozen until testing.
Just prior to testing, the 5 g samples were thawed and 10 mL PBSTB were added. To the thoroughly blended samples, either STX or DA were spiked to give 1000 ng/mL or 200 ng/mL, respectively. The samples were mixed and incubated for 2 h at room temperature, then spun to remove large particulates. The supernatant was used for analysis. Extraction efficiency was not determined.
4.4. Preparation and Biotinylation of mAbs and ScFv
The antibodies were purified from cell supernatants by MEP HyperCel hydrophobic charge induction chromatography (Pall, East Hills, NY, USA), as described previously [49
]. Antibodies and scFvs were biotinylated using NHS-LC-LC-Biotin (Thermo-Fischer, Waltham, MA, USA) dissolved in dimethyl sulfoxide (20 g/L). The antibodies were reacted with a 10:1 molar excess of the NHS-LC-LC-Biotin. To enhance the rate of reaction, the pH was increased by the addition of a half-volume of 100 mM sodium borate +100 mM sodium chloride (pH 9.1). After incubation for 1 h at room temperature, the biotinylated antibodies were separated from free biotin by gel filtration on a Bio-gel P10 column (Bio-Rad, Hercules, CA, USA) or by using Zeba spin 7 K desalting columns (Thermo Fisher, Waltham, MA, USA).
4.5. Surface Plasmon Resonance Evaluation of Anti-STX mAbs
Surface plasmon resonance (SPR) affinity and kinetics measurements were performed using the ProteOn XPR36 (Bio-Rad). Lanes of a general layer compact (GLC) chip were individually coated with STX covalently linked to an irrelevant HuIgG in 10 mM acetate buffer with pH 5.0. The covalent crosslinking of STX protocol was described previously [15
] and in Section 4.6
. The STX-HuIgG was attached to the chip following the standard 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC)/N-hydroxysulfosuccinimide (Sulfo-NHS) coupling chemistry available from the manufacturer. Binding kinetics of each antibody was tested at 25 °C by flowing six concentrations varying from 100 to 0 nM at 100 μL/min for 90 s over the antigen coated chip and then monitoring dissociation for 600 s. For comparison purposes, this data was analyzed using a global Langmuir fit.
4.6. Preparation Toxin-Coated MagPlex Microspheres
MagPlex microspheres were coated with DA by first reacting the surface carboxyls with ethylene diamine (EDA) using the standard two step protocol, where 30 µL of the microspheres were washed with 0.1 M sodium phosphate (pH 6.0) three times, and then activated using EDC and Sulfo-NHS 5 mg/mL each. After 20 min the microspheres were washed once with 0.1 M sodium phosphate (pH 6.0) and once with PBS, then microspheres were resuspended in EDA (1 mg/mL) in PBS and incubated overnight. The next day, the microspheres were washed three times with 0.1 M 2-(N-morpholino)ethanesulfonic acid buffer (MES) at pH 4.5. The microspheres were then coated with DA by adding a 2:1 EDC:DA ratio. The microspheres were incubated for 1 h and then washed three times with PBS and stored in the dark at 4 °C.
Use of a different chemistry to bind STX to the microspheres was indicated by the immunogen used to prepare the mAbs [15
]. MagPlex microspheres (50 µL) coated with STX were prepared by first coating two sets of microspheres with either a purified rabbit or human polyclonal IgG towards irrelevant targets, using the standard protocol described above except they were washed into PBS and then resuspended in 0.4 mL of PBS on the following day. The carbohydrates on the IgG were activated by the addition of 22 µL sodium periodate (46 mM). The microspheres were incubated in the dark for 1 h at room temperature, then washed three times with PBS. The PBS was removed and the microspheres resuspended in 40 µL of 0.1 M sodium bicarbonate (pH 8.5) to which 5 µL of STX (100 µg/mL) was added. The microspheres were incubated for 1 h and then 1 µL of sodium cyanoborohydride (5 M in 1 M NaOH) was added. The reaction was allowed to proceed for 30 min on ice. Then the microspheres were washed three times with PBS and then stored in 0.1 M sodium phosphate (pH 6.0).
Competitive immunoassay dose response curves were performed first in buffer and then with the spiked food samples. Briefly, added into a well of a 96-well microtiter plate was either a sample containing 1000 ng/mL STX or 200 ng/mL DA such that after serial dilutions (1:4 for STX and 1:3 for DA) 90 µL remained in each well. Next, 10 µL of the biotinylated scFvs (bt-anti-STX, 5F7, at 2.5 µg/mL final for STX and bt-anti-DA at 10 µg/mL final) were added, followed with 10 µL of toxin-coated beads. The foil-covered plate was placed on a FINEPCR micromixer MX4t (Gyeonggi-Do, Korea) for 30 min at room temperature. Using a 96-well flat magnetic plate (BioTek, Winooski, VT, USA), the supernatant was removed and the beads were washed with PBSTB. For signal generation, 50 µL of 2.5 µg/mL SA-PE were added to each well and plate was incubated on shaker for 15 min followed by a wash with PBSTB. Based on previous work which showed a ~5 fold increase in signal [50
], a second round of SA-PE was performed as follows. Fifty µL of biotinylated anti-SA (1 µg/mL) were added, incubated for 15 min, and washed with PBSTB. Lastly, another 50 µL of 2.5 µg/mL SA-PE was added and the mixture incubated for 15 min. The beads were washed 2× with PBSTB. PBSTB (100 µL) was added to each well and the plate was analyzed with Luminex MAGPIX. For the assay using the intact mAbs, only a single 30 min incubation with SA-PE (2.5 µg/mL) was performed. Percent inhibition was calculated using the following equation: