**3. Discussion**

In this work, we report the generation of new anti-AMA mAbs (AMA9G3 and AMA9C12) using our previously synthesized immunogen (PERI-AMA-KLH) based on periodate oxidation of α-AMA. Unlike early reports of generating AMA-conjugated immunogens that exhibit toxicity [21,22], this immunogen did not cause any death in both mice (this study) or rabbits [20], corroborating the low toxicity observed by other investigators [23].

These mAbs exhibited high selectivity for AMAs, but the selectivity within AMAs varied from what was observed from our previously generated pAbs using the same immunogen [20]. The pAbs detected α-AMA, β-AMA and γ-AMA within a narrow range of IC50 values (2–3 ng mL−1). Conversely, both mAbs distinguish the AMAs di fferently, such that AMA9G3 detected α-AMA and γ-AMA (IC50 ≈ 1.6 ng mL−1), and, to a lesser extent, β-AMA (IC50 = 24 ng mL−1), while AMA9C12 detected only α-AMA and γ-AMA (IC50 = 2.3 and 2.7 ng mL−1, respectively). Similar to mAb AMA9C12, the commercially available, pAb-based cELISA detected only α-AMA and γ-AMA, and not β-AMA [15]. Another mAb, generated using a di fferent coupling strategy to generate an α-AMA immunogen, produced mAbs with broad selectivity to α-AMA, β-AMA and γ-AMA (IC50 = 66, 97, and 163 ng mL−1, respectively) [16]. A single chain variable fragment of a mAb, produced by this same group, also exhibited broad selectivity to α-AMA, β-AMA and γ-AMA (IC50 = 77, 115, and 199 ng mL−1, respectively) [24].

Regarding total AMA extraction efficiency, the recovery obtained with all of these extraction methods is reputable. Previously reported α-AMA concentrations were approximately 1–2 mg of toxin per gram of dried mushroom [8,10,11], but this could vary depending on the location and age of the specimen [11]. Nonetheless, given our extraction ratio of 1 mL per 0.1 g of dried tissue, we would expect to recover 0.1–0.2 mg of α-AMA in 1 mL, as well as 0.1–0.2 mg of β-AMA and 0.05 mg of γ-AMA. This equates to extracts containing 10,000–45,000 ng mL−<sup>1</sup> for total AMAs, which greatly exceeds our assay's detection limit of 1 ng mL−<sup>1</sup> for total AMAs. Furthermore, the detection of AMAs in the 27,000-fold dilution samples of these AMA-containing sample extracts provided evidence that we were obtaining a high recovery of AMA. Other investigators performed a second extraction following a 1 h extraction (using an acetonitrile solution) and did not recover any detectable amounts of residual toxin in the second extract [14]. While most of the AMAs appear to be easily extracted, further work is needed to determine the recovery coefficient for the rapid extraction procedure.

Furthermore, none of these tissue samples were macerated prior to extraction for any of the solutions tested. Maceration of dried mushroom samples increases exposure of the researcher to the toxin-containing dust. In our previous work, and that of many others, the mushroom tissue was ground to a powder [11–14,19,20,25]. In this study, however, we did not grind the samples and still achieved sufficient toxin extraction, suitable for cELISA detection.

The sensitivity of our mAb-based cELISA permitted the detection of as little as 1 ng mL−<sup>1</sup> of total AMAs in simple, water-based extracts obtained from dried mushrooms known to contain AMAs. Most LC methods achieve a detection limit of approximately 10 ng mL−<sup>1</sup> for each individual AMA [8,14]. The amount of tissue required for an extraction can be as little as 50 mg for LC instrumental techniques, whereas the antibody-based detection methods could conceivably utilize sub-μg amounts of material using a simple, water-based extraction. The sensitivity, speed, and throughput of this mAb-based immunoassay for the detection of AMAs provides a simplified strategy for the evaluation of these toxins in wild mushrooms to characterize their occurrence and geographic distribution.

#### **4. Materials and Methods**

#### *4.1. Immunization and Antibody Production*

The Institutional Animal Care and Use Committee of the United States Department of Agriculture, Western Regional Research Center approved the experimental procedures used in these studies (protocol #16-1). Three 6-week-old female BALB/c mice were immunized by intraperitoneal injection (i.p.) of 100 μL of a 1:1 Sigma Adjuvant System (Sigma-Aldrich, St. Louis, MO, USA) containing 50 μg of PERI-AMA-KLH [20]. Two subsequent booster immunizations were administered i.p. at 2-week intervals using 20 μg of PERI-AMA-KLH in Sigma Adjuvant System. Serum were collected one week after the third immunization. Another two booster immunizations were performed four months later, two weeks apart, and serum was collected one week after this round of immunizations. After determining by indirect ELISA that the antibody response was still elevated to this target immunogen, a final booster immunization containing 10 μg of PERI-AMA-KLH in saline was administered i.p., four days prior to being euthanized and cell fusion.

## *4.2. ELISA Procedure*

For serum antibody screening, black 96-well microtiter plates (Nunc, Thermo Fisher Scientific, Waltham, MA, USA) were coated at 1 μg mL−<sup>1</sup> with PERI-AMA-BSA for 1 h at 37 ◦C in carbonate buffer (0.05 M carbonate-bicarbonate, pH 9.6). Then, the plates were blocked for 1 h at 37 ◦C with 3% non-fat dry milk in tris-buffered saline with 0.05% Tween-20 (TBST). After incubation for 1 h at 37 ◦C, TBST was removed and serum was loaded at a dilution of 1:100 in TBST and serially diluted. After another incubation for 1 h at 37 ◦C, plates were washed three times with TBST. Plates were then loaded with a secondary horse radish peroxidase labeled goat-anti-mouse antibody (Sigma) at 1:5000 in TBST and incubated for 1 h at 37 ◦C. Plates were washed and loaded with SuperSignal West Pico Chemiluminescent substrate (Fisher), incubated for 3 min, and then luminescent counts were recorded on a Victor<sup>3</sup> Multilabel Counter (PerkinElmer, Waltham, MA, USA).

#### *4.3. Monoclonal Antibody Production and Screening*

The cell fusion and expansion procedures were completed as previously described [26]. The screening of the cell culture plates following cell fusion, in particular the use of an indirect cELISA, was carried out as previously described with minor modifications [27]. The screening process was the same, but the reagents used were changed. Briefly, wells of clear-bottom microtiter plates coated with PERI-AMA-BSA were pre-loaded with 50 μL/well of either TBST for noncompetitive screening or α-AMA at 100 ng mL−<sup>1</sup> for competitive screening. Plates were incubated for 1 h at 37 ◦C, washed, and then incubated with a secondary antibody, as described for the direct screening. After incubation and washing, antibody activity was visualized using Enhanced K-Blue Substrate (Neogen, Lexington, KY, USA) and read on a VersaMax Microplate Reader (Molecular Devices, San Jose, CA, USA).

Hybridomas from wells exhibiting a significant reaction to the presence of α-AMA (i.e., a reduction in signal intensity) were selected for clonal expansion. Cells were cloned by limiting dilution, repeated until every well with cell growth presented positive activity via ELISA. MAbs were purified from the cell culture supernatant on a Protein G Sepharose a ffinity column (GE Healthcare Life Sciences, Pittsburgh, PA, USA), eluted with 0.1 M glycine-HCl, pH 2.7. Purified protein was extensively dialyzed against the phosphate bu ffered saline (PBS; 10 mM phosphate, 138 mM NaCl, 2.7 mM KCl, pH 7.4) and then stored at −20 ◦C until further use. Antibody protein concentrations were determined on a NanoDrop Lite Spectrophotometer (Thermo). Antibody isotyping was completed using an IsoStrip Mouse Monoclonal Antibody Isotyping Kit (Roche, Indianapolis, IN USA), following the manufacturer's protocol. Purified mAbs were titrated by indirect ELISA to determine the concentration of the antibody at half of the maximal signal. This determined concentration was used as the working concentration of the antibody for the cELISAs, to evaluate antibody cross-reactivity.

#### *4.4. Antibody Characterization: Cross-Reactivity*

Indirect cELISAs were completed using a panel of inhibitors to determine the selectivity of the mAbs. The cELISA procedure was nearly the same as that described for the serum screening, except for the addition of inhibitors (50 μL), which were mixed with 50 μL of the antibody solution during the primary antibody incubation step. The inhibitors tested were α-AMA (≥90%, Enzo Life Sciences, Farmingdale, NY, USA), β-AMA (≥90%, Enzo), γ-AMA (≥90%, Enzo), microcystin-LR (≥95%, Enzo), nodularin (≥95%, Enzo), phalloidin (>90%, Enzo), phallacidin (≥85%, Sigma), pysilocybin (>99%, Cerilliant, Round Rock, TX, USA), muscimol (>99%, Abcam, Cambridge, MA, USA), ibotenic acid (>98%, Abcam). Each analyte stock was dissolved in dH2O, then serially diluted into TBST, starting at the highest concentration of ,000 ng mL−1, and assessed in triplicate. Data were analyzed using a 4-parameter logistic equation (GraphPad Prism 7 Software, La Jolla, CA, USA) to determine the concentration of inhibition at half of the maximal signal (IC50). Cross-reactivity (%) was calculated as follows: (IC50 α-AMA)/(IC50 test inhibitor) × 100.

#### *4.5. Antibody Characterization: Kinetic Measurements*

All KinExA experiments were performed on a KinExA 3200 with Autosampler (Sapidyne Instruments, Boise, ID, USA) and data were analyzed using KinExA Pro software provided by Sapidyne. A ffinity values ( *K*d) utilized were their template protocol for an Equilibrium Experiment and kinetic parameters were determined using the Kinetics Injection method. Flow rates and volumes used were the default settings defined in the software.

Polymethylmethacrylate particles (aliquots of 200 mg, Syringa Labs, Boise, ID, USA) were adsorption coated with 30 μg of BSA-AMA-PERI in 1 mL of carbonate bu ffer for 1 h at room temperature with end-over-end rotation. The particles were blocked with a solution of 1% BSA (Sigma) in PBS for 1 h at room temperature with end-over-end rotation, and stored at 4 ◦C for no more than one week before use. The diluent for all reagents was PBS containing 1% BSA. Three antibodies were evaluated—two mouse mAbs (AMA9G3 and AMA9C12) generated from this study and one rabbit pAB #58 generated from the previous study [20]. The secondary antibody used for the mouse mAb experiments was DyLight650 labeled anti-mouse Ig (Fisher) (used at 0.5 μg mL−1) and the secondary antibody used for the rabbit pAb experiments was AlexaFluor647 labeled anti-rabbit Ig (Jackson Immunoresearch, West Grove, PA, USA) (used at 0.25 μg mL−1).

Signal test runs were completed on each antibody to determine the amount of antibody needed to generate the appropriate signal change (1 Δv). Then, for the Equilibrium experiments, antibody was prepared at 2× this concentration and then mixed with an equal volume of a solution containing α-AMA diluted 2-fold, ranging from 300 ng mL−<sup>1</sup> (326 nM) to 9.2 pg mL−<sup>1</sup> (10 pM) of final concentrations, including one sample with no α-AMA and one sample containing only diluent. For the Kinetics Injection experiments, the same 2× antibody concentration was used, along with solutions containing α-AMA diluted 2-fold, ranging from 920 ng mL−<sup>1</sup> (1000 nM) to 1.8 ng mL−<sup>1</sup> (2 nM). The Equilibrium and Kinetics Injection experiments were completed in duplicate.

## *4.6. Mushroom Extraction*

Whole mushroom specimens were identified, dried, and provided by expert mycologists. The specimens included two that were known to contain AMAs, *A. phalloides* and *A. ocreata*, and one that was known to not contain AMAs, but was from the same genus, *A. gemmata*. Small portions of the specimens were weighed (~100–200 mg) and then placed into a 15 mL Falcon tube containing one of the five extraction buffers: (1) methanol (methanol:water:0.01 N HCl, 5:4:4, v:v:v), (2) diH2O, (3) phosphate buffer (PB; 0.1 M, pH 7.6), (4) PB with Tween-20 (PBT), or (5) TBST at a ratio of 1 mL per 100 mg tissue. The samples that were extracted with the methanol buffer were shaken for 1 h at room temp and then centrifuged at 1000× *g* for 10 min. Aliquots of the supernatant were drawn <sup>o</sup>ff, diluted in TBST as necessary, and assessed by indirect cELISA. The samples in diH2O, PB, PBT, or TBST were briefly shaken by hand for 1 min. Immediately after shaking, a 50 μL aliquot of the liquid phase was drawn <sup>o</sup>ff, diluted in TBST as necessary, and assessed by indirect cELISA, as described earlier. At least two individual mushrooms from each species were extracted, and extractions for each extraction condition were completed in duplicate.

**Author Contributions:** Conceived and designed the experiments, C.S.B. and L.H.S.; performed the experiments, C.S.B.; analyzed the data, C.S.B., L.W.C., R.M.H. and L.H.S.; wrote the original draft, C.S.B. and L.H.S.; reviewed and edited, L.W.C. and R.M.H.

**Funding:** This research was funded by the United States Department of Agriculture, Agricultural Research Service, National Program project NP108, CRIS 2030-42000-049-00D. C.S.B. and L.H.S. were also funded by interagency agreemen<sup>t</sup> IAA #60-2030-5-004 with Department of Homeland Security.

**Acknowledgments:** The authors are grateful to Tom Bruns and Catharine Adams (University of California, Berkeley) for their generous donation of identified mushrooms.

**Conflicts of Interest:** The authors declare a competing interest. C.S.B., R.M.H., L.W.C. and L.H.S. are named as inventors on a provisional patent application filed by ARS-USDA related to the data presented in this work.
