Simple and fast determination of tetrodotoxin in human plasma based on hydrophilic-interaction/ion-exchange mixed-mode solid phase extraction combined with liquid chromatography-ta

In this study, we developed and validated a simple, fast and sensitive LC-MS/MS method for the measurement of tetrodotoxin (TTX) in human plasma. Three HILIC-type solid phase extraction (SPE) carriers (PSA, silica, Siphila i HILIX) with different stationary phase functional groups were compared.

Journal of Chromatography A 1684 (2022) 463567 Contents lists available at ScienceDirect Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma Simple and fast determination of tetrodotoxin in human plasma based on hydrophilic-interaction/ion-exchange mixed-mode solid phase extraction combined with liquid chromatography-tandem mass spectroscopy Liang Xin a,b,c,1 , Yan Liang a,b,c,1 , Shuangshuang Yang a,b,c,1 , Fengli Jiang a,b,c , Fan Yu a,b,c , Meiwei Zhang a,b,c , Wei Chang d , Wei Wang d , Chen Yu a,b,c , Gangyi Liu a,b,c,∗ , Youli Lu a,b,c,∗ a Central Laboratory, Shanghai Xuhui Central Hospital/Zhongshan-Xuhui Hospital, Fudan University, Shanghai, China Shanghai Engineering Research Center of Phase I Clinical Research & Quality Consistency Evaluation for Drugs, Shanghai, China c Shanghai Institute of Clinical Mass Spectrometry, Shanghai, China d Department of Emergency Medicine, Shanghai Xuhui Central Hospital/Zhongshan - Xuhui Hospital, Fudan University, Shanghai, China b a r t i c l e i n f o Article history: Received 26 August 2022 Revised 11 October 2022 Accepted 12 October 2022 Available online 18 October 2022 Keywords: Tetrodotoxin Hydrophilic-interaction/ion-exchange Mixed-mode solid phase extraction LC-MS/MS Human plasma a b s t r a c t In this study, we developed and validated a simple, fast and sensitive LC-MS/MS method for the measurement of tetrodotoxin (TTX) in human plasma Three HILIC-type solid phase extraction (SPE) carriers (PSA, silica, Siphila i HILIX) with different stationary phase functional groups were compared The Siphila i HILIX SPE plate containing multi-carboxyl groups was finally selected due to obviously better extraction recovery of TTX (about 80% of recovery from plasma samples) than the other two and no significant matrix effects were observed, which was speculated to have mixed-mode synergistic effects of hydrophilic interaction and ion exchange 100μL plasma sample was precipitated rapidly with acetonitrile containing 1% trichloroacetic acid, and filtrates were loaded onto Siphila i HILIX 96 well SPE plate After washed with 95% acetonitrile, TTX was eluted with 200μL of 50% acetonitrile containing 1% trichloroacetic acid 2μL of elution solution was directly injected into LC-MS/MS and the total run time on a BEH amide column was 4.5 The method avoids the evaporation and ultrafiltration processes which is simple and timesaving ( 10) of 0.1 ng/mL; the intra- and inter-assay accuracies were in the range of 98.5%-99.8% (relative standard deviations, RSDs ≤ 5.92%) and 98.8–99.5% (RSDs ≤ 6.23%), respectively Biases of spiking analysis were ranged from -7.00% to 7.43% for healthy human plasma samples (RSDs ≤ 8.83%) and from -5.00% to 3.93% for hemolytic, high triglyceride, high cholesterol and high bilirubin plasma samples (RSDs ≤ 6.40%), which proved the good anti-interference property of the method The results showed that the method is sensitive, accurate, specific, reliable, and can be used to monitor the concentration of TTX in plasma to meet the needs of clinical research and poisoning screening © 2022 The Authors Published by Elsevier B.V This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) Introduction Tetrodotoxin (TTX) poses a serious threat to human health in coastal areas where marine food is the main food source [1–2] Incidents of food poisoning caused by TTX were reported almost every year [3–7] Compared to other marine toxins, TTX intoxication ∗ Corresponding authors E-mail addresses: gyliu@shxh-centerlab.com (G Liu), yllu@shxh-centerlab.com (Y Lu) These authors are considered as co-first authors leads to more severe symptoms and higher mortality [8] The grade of TTX poisoning depends on the ingested amount, the time since ingestion, and the health status before poisoning [9–10] There is a strong correlation between the symptoms and the levels of TTX in the blood of victims [11] As a potent toxin, TTX selectively blocks voltage-gated sodium (NaV ) channels, such as NaV 1.1, NaV 1.2, NaV 1.3, NaV 1.6, and NaV 1.7 [12–13], which are widely distributed in the human central and peripheral nervous system [14], making it a very promising drug for the treatment of pain [15–16] The therapeutic doses reported in current clinical trials were 15–45 μg with Cmax of approximately https://doi.org/10.1016/j.chroma.2022.463567 0021-9673/© 2022 The Authors Published by Elsevier B.V This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) L Xin, Y Liang, S Yang et al Journal of Chromatography A 1684 (2022) 463567 ng/ml in plasma [17] While, the lethal dose of TTX in human is about 1.5–2.0 mg, with a blood concentration of ng/ml [11,18] Therefore, to ensure drug safety and study the relationship between drug concentration and efficacy, a highly sensitive and accurate assay is needed to monitor TTX blood or plasma levels in human Although various methods used for the determination of TTX have been published, such as mouse bioassay [19], immunoassay [20], nuclear magnetic resonance (NMR) [21], and liquid chromatography with fluorescence detection (LC-FL) [22], most of them are for shellfish or fish tissue samples Dealing with human biological samples is much more challenging because the amount of TTX in the urine and plasma is generally extremely low The liquid chromatography-tandem mass spectroscopy (LC-MS/MS) method was considered the most popular method for the identification and quantification of TTX because of its high sensitivity, selectivity, and specificity [23–24] Since TTX is a highly polar substance, there would be poor retention on the conventional reverse-phase column and low ionization efficiency in mass spectrometry [25] Hydrophilic interaction liquid chromatography (HILIC) could help to address these issues and showed good retention capability and improved ionization efficiency in LC-MS [26–27] While ion suppression (one kind of matrix effect) was another challenge in analyzing TTX in biological samples Ion suppression usually results from the coelution of nondetected interferences [28–29] In general, ion suppression can be reduced by sample preparation, chromatographic separation, and MS analysis [30–32] The most daunting analysis bottleneck is sample preparation Because its effective extraction and enrichment from complicated human biological samples are challenging [33] The most popular technique for the cleanup of TTX samples is solid phase extraction (SPE), usually using cation exchange extraction [34–35] The positively charged guanidinium group of TTX strongly interacts with the sulfonic acid group in the stationary phase, allowing TTX to be strongly retained on SPE cartridges Only strong acids such as hydrochloric acid can elute it, requiring additional solvent evaporation and residue redissolution steps, which are both timeconsuming and labor-intensive Therefore, strong cation exchange extraction techniques are rarely used for the detection of TTX at present Another SPE technique applied to TTX detection is the use of graphitized carbon carriers However, the TTX recoveries of urine, serum, and plasma samples were 43%, 23%, and 20%, respectively [36] Low recoveries can be attributed to toxin loss and/or ion suppression during SPE cleanup The limits of quantification for serum and plasma samples were 21.8 and 25.4 ng/mL, respectively, which were not sensitive enough for pharmacokinetic studies Ochi [37] also used graphitized carbon carriers and optimized the method by ion-pair solid-phase extraction The difference was that the use of ion-pair solvent in sample loading condition improve the retention of TTX on graphitized carbon While the method was only applied to the determination of toxins in bivalve extracts, there was no data to prove its recovery and matrix effect in human biological samples Hydrophilic interaction liquid chromatography-solid phase extraction (HILIC-SPE) can be a third option due to the hydrophilic nature of TTX However, few attempts have been made to extract TTX from plasma using HILIC-SPE technology One reason is the limited impurity removal capability of a single HILIC-SPE Considering this, double SPE using C18-HILIC had been used to improve the efficiency of removing ion suppression [38] In this study, we attempted to use three HILIC-type SPE carriers for sample preparation to compare their extraction recoveries and influence on matrix effects A simple and fast SPE protocol based on hydrophilic-interaction/ion-exchange mixed-mode was developed to extract TTX from human plasma without evaporation and ultrafiltration The pretreatment method combined with LC-MS/MS method was used for the quantitation of TTX in human plasma with satisfactory sensitivity, recovery, and matrix effect, which greatly simplified the sample preparation steps and shortened the total analytical time Material and methods 1.1 Reagents and materials TTX (99.5%) was purchased from Standardpharm Co., Ltd (New York, USA) Arginine-15 N4 (90.4%) was purchased from TLC Pharmaceutical Standards Ltd (Ontario, Canada) Voglibose (98%) was purchased from Alta Scientific Ltd (Tianjin, China) Siphila i HILIX 96 well plates (30 mg/2 mL) were obtained from Siphila Technologies Co., Ltd (Tianjin, China) PSA 96 well plates (30 mg/2 mL) and silica 96 well plates (30 mg/2 mL) were obtained from TupLabs Co., Ltd (Tianjin, China) The links for the three SPE plates were shown in the supplementary materials Protein precipitation plates (2 mL/well) were obtained from TupLabs Co., Ltd (Tianjin, China) Trichloroacetic acid (TCA, ≥99%) was supplied by Sinopharm Chemical Reagent Co., Ltd (Shanghai, China) HPLCgrade acetonitrile (ACN) and methanol were supplied by Merck KGaA (Darmstadt, Germany) HPLC-grade formic acid (FA, 99%) and ammonium formate (98%) were supplied by ACS (Wilmington, USA) Water was purified using Simplicity system (Merck KGaA, Darmstadt, Germany) Blank human plasma was obtained from healthy donors who signed the Informed Consent Forms The standard stock solutions of TTX (100 μg/mL) were prepared in deionized water and stored at −80 °C until further use Working standard solutions were prepared by serial dilution of standard stock solutions using deionized water at different points (1, 2, 10, 50, 200 ng/mL) before they were added to human blank plasma to create calibration curves with ranges from 0.1 to 20 ng/mL The same way was used to make quality control samples (QCs) at different concentration levels (0.1, 0.5, 2, 10 ng/mL) 1.2 Sample pretreatment One hundred microliters of human plasma sample (calibration curves, QCs or real samples) and 10μL arginine-15 N4 (internal standard, IS, 100 ng/mL) working solution were mixed and transferred to 96 well protein precipitation plate, then 700μL of 1%TCAACN was added for protein precipitation After filtering with a positive pressure device, all filtrates were transferred to Siphila i HILIX 96 well SPE plate, which was previously activated with mL of methanol and equilibrated with mL ACN Wash once with 200μL of 95% ACN and elute with 200μL of 1%TCA-50% ACN 2μL of elution solution was directly injected into LC-MS/MS 1.3 LC-MS/MS analysis The LC-MS/MS system consisted of a triple quadrupole mass spectrometer 6500 system (SCIEX, USA) and an ExionLC system (SCIEX, USA) Data was acquired and statistically calculated by Analyst 1.6.3 (SCIEX, USA) TTX and IS were monitored using multiple reaction monitoring (MRM) in positive mode with the transitions of m/z 320.3→162.2 (quantification transition for TTX), 320.3→284.1 (confirmation transition for TTX) and 179.2→63.0 (transition for IS), respectively Ion source parameters were also optimized: 550 °C source temperature, 60 psi ion source gas1, 70 psi ion source gas2, 30 psi curtain gas, 40 0 V ion spray voltage, 120/43 V declustering potential, and 49/27 eV collision energy L Xin, Y Liang, S Yang et al Journal of Chromatography A 1684 (2022) 463567 Fig Effects of ammonium formate (a) and formic acid (b) concentration in mobile phase for TTX Chromatographic separation was performed on an ACQUITY UPLC BEH Amide column (50×2.1 mm, 1.7 μm, Waters, USA) at room temperature The mobile phase consisted of (A) deionized water and (B) ACN, both containing 10μmol/L ammonium formate and 0.01% FA The following gradient program was used [mobile phase B]: 85–55% at 0–2 min; held 55% at 2–3 min; 55–85% at 3– 3.01 min; held 85% at 3.01–4.5 with a flow rate of 0.5 mL/min 2.2 Selection and optimization of SPE In the beginning, MCX and WCX cation exchange SPE plates (Waters, USA) were tested to clean up the plasma samples and purify TTX It was found that the WCX plate can hardly retain TTX during loading, and the MCX plate can hardly elute TTX using organic solvent containing moderate acids like fomic acid or acetic acid Strong acid like hydrochloric acid wasn’t tried on MCX plate due to its damage to liquid chromatography and mass spectrometry, even if it can be dried up, poor peak shapes and severe matrix effect might not be avoided In turn, three different HILIC-type SPE plates were attempted to separate TTX from the plasma matrix PSA plate containing polyamide material, silica plate containing silica gel material, and Siphila i HILIX plate containing multi-carboxyl material were compared The flow charts of comparison and optimization for three different HILIC-SPE were shown in Fig All the loading, washing, and elution samples were collected and analyzed to calculate the loss or recovery of TTX HILIC-type SPE usually requires high concentrations of organic solvents as loading solutions and water is used as a strong eluent [40] In this study, pure ACN and ACN containing 1%TCA, 5%TCA, and 2%FA were investigated respectively as loading solutions, which also acted as protein precipitating agents, denaturing plasma proteins and releasing TTX 100μL of standard solution (prepared by deionized water) and plasma samples were precipitated with 700μL of loading solutions in 96-well protein precipitation plates respectively and then all filtrates were loaded onto three different SPE plates, which were previously activated with methanol and equilibrated with ACN The results showed TTX was most strongly retained on the Siphila i HILIX SPE plate during sample loading ACN was the optimal loading solvent and facilitated the retention of TTX on SPE plates, followed by 2%FA-ACN, 1%TCAACN, and 5%TCA-ACN (Fig 3) The loss of TTX in plasma samples were less than that in standard solution samples during loading, which might be due to the poor buffering capacity of standard solution samples compared with plasma To further investigate the loss and recovery of TTX during washing and elution steps, 200μL of 95%ACN was used for washing followed by 200μL of 1%TCA-50%ACN for elution In the 2%FA-ACN and ACN loading groups, due to the low recoveries after elution, another 200μL of 5%TCA-ACN was used for further elution to improve the recovery The results were shown in Fig Different retention and elution capabilities of TTX were observed on the same SPE plate under different loading conditions during washing and elution Siphila i HILIX was the most distinct and optimal one (Fig 4a) In the 1%TCA-ACN loading group, washing with 95% ACN resulted in a loss of 18% TTX, while elution with 1% TCA-50% ACN resulted in the recovery of 80% TTX in plasma samples While during the same loading and washing steps, the corresponding standard solution samples would lose even more, with less than 30% TTX recovered by elution In the 5%TCA-ACN loading group, both the loading and washing steps resulted in 1.4 Method validation The lower limit of quantification (LLOQ), linearity, inter/intrabatch accuracy and precision, matrix effect, extraction recovery, specificity, and interference of the established method were investigated according to the U.S food and drug administration (FDA) bioanalytical method validation guidelines [39] Results and discussion 2.1 Development of chromatography and mass spectrometry method TTX is difficult to retain on reversed-phase columns [26], and most recent reports used an amide column with a length of 10 cm to improve the retention of 10 TTX analogs [37] Here, an ACQUITY UPLC BEH C18 Amide column (50×2.1 mm, 1.7 μm, Waters, USA) was used to separate TTX from its interferences, which performed better compared with an ACQUITY UPLC BEH C18 HILIC column (50×2.1 mm, 1.7 μm, Waters, USA) The retention times for TTX and IS on the amide column were 2.41 and 2.31 min, respectively, with a total analysis time of 4.5 The pH value and the concentration of ammonium formate in the mobile phase were considered to have significant effects on retention time and peak area [37] Therefore, different FA and ammonium formate concentrations were tested under the same separation conditions It was observed that the retention time and peak area of TTX increased with the decrease of FA (from 0.05% to 0.01%, v: v) (see Fig 1a) and ammonium formate minimized the retention time shift between pure solvent and matrix samples but decreased the peak area of TTX (Fig 1b) In this study, 10 μmol per liter of ammonium formate and 0.01%FA (v: v) were used in both mobile phases to acquire better sensitivity and make the peaks more stable The variation of 56 consecutive injections prepared by plasma samples was tested (see Fig S1) and the mean retention time was 2.42±0.1 (2.37–2.47 min), which indicated that the chromatographic separation conditions performed well with excellent reproducibility MS/MS parameters for TTX and IS were optimized by direct infusion of individual standard solutions The quantifier and confirmation product ion for TTX were m/z 162.2 and 284.1, respectively, although the most abundant product ion was observed as [M + H–H2 O]+ ion (m/z 302.2) The latter has higher baseline and stronger interference in plasma samples L Xin, Y Liang, S Yang et al Journal of Chromatography A 1684 (2022) 463567 Fig The flow charts of comparison and optimization for three different HILIC-SPE more loss of TTX, about 60% for plasma samples and even higher for standard solution samples (>80%) In the 2%FA-ACN loading group, no significant TTX loss was observed for plasma samples after washing with 95%ACN, and the recovery was about 22% after elution with 1%TCA-ACN Compared to the two TCA-ACN loading groups, the total recovery of TTX in plasma samples was reduced throughout the SPE, so an additional 5%TCA-ACN elution was performed to look for the lost TTX and the total recovery was increased to 60% after the additional elution This was quite different from the total recovery of standard solution samples which was close to 100% The total recovery of TTX in plasma samples was significantly lower in the ACN loading group, no matter in the loading and washing solutions, or the eluted and twice-eluted samples, while the recovery of standard solution samples was nearly 100% In the 2%FA-ACN and ACN loading groups, the lower recovery of TTX from plasma samples compared to standard solution samples may be due to the incomplete release of TTX from plasma proteins upon precipitation of these two loading solutions Similar results were observed when using PSA and silica plates (Fig 4b and 4c) Strong acids such as TCA in the precipitating agent may be necessary for the release of TTX from plasma proteins Excessive TCA resulted in weak retention of TTX on SPE, which may be related to the decreased polarity of the TTX-TCA conjugates in the presence of ion pairs solvents, thus reducing the retention of TTX on HILIC-type SPE plates The results showed that the retention capacity among Siphila i HILIX, PSA, and silica SPE were quite different Compared to Siphila i HILIX, PSA and silica had weaker retention of TTX under almost all loading conditions It was speculated that both Siphila i HILIX and silica may have two interaction modes: hydrophilic interaction and ion exchange Some SiOH-containing silica materials had been shown to have cation-exchange properties [41] While in our study, we found that the ion exchange effect of silica SPE might be very weak which could be easily inhibited by TCA or FA In addition to hydrophilic interaction, Siphila i HILIX may also retain polar basic compounds through cation-exchange interaction which was similar or even stronger to that of weak cation-exchange material with carboxylic acid moieties [42] It seemed that PSA had only a pure hydrophilic effect on TTX, and retention was the weakest Among the three HILIC materials, Siphila i HILIX was considered to have better synergistic effect of hydrophilic interaction and ion exchange, which was suitable for the retention of TTX After optimizing the loading conditions, the washing solvents were selected among 95% ACN, 98% ACN, and ACN No significant difference was observed, so 95%ACN was chosen because it had the strongest cleaning capability The elution solvent remained 1%TCA50%ACN, which had no obvious ion suppression or enhancement effects on TTX Since stable isotope labeled TTX is very expensive and difficult to obtain, we used the reported voglibose and the available arginine-15 N4 as candidate internal standards in this L Xin, Y Liang, S Yang et al Journal of Chromatography A 1684 (2022) 463567 Fig The loss of TTX under different sample loading conditions on three HILIC-SPE plates Table The major validation parameters for TTX and IS in human plasma TTX Spiked conc (ng/mL) Intra-assay (n = 6) Inter-assay (n = × 3) Extraction recovery (n = 6) IS-normalized MF (n = 6) Mean accuracy (%) RSD (%) Mean accuracy (%) RSD (%) Mean (%) RSD (%) Mean (%) RSD (%) 0.1 0.5 10 99.8 98.7 98.5 99.2 5.92 5.75 2.29 4.37 99.5 99.4 99.1 98.8 5.47 6.23 4.16 5.26 81.5 80.8 81.2 5.53 3.58 6.78 0.979 3.41 0.968 4.29 Spiked conc (ng/mL) arginine-15 N4 Extraction recovery (n = 6) voglibose Extraction recovery (n = 6) Mean (%) 74.3 Mean (%) 71.5 10 RSD (%) 4.26 study, both of which showed similar chromatographic behavior to TTX on the amide column The performance of the two candidate internal standards was further compared in plasma matrix Post-column injection experiment was used to investigate the different matrix effects of TTX and the two candidate internal standards in pure solvent and plasma samples It was found that TTX was barely enhanced or inhibited by plasma matrix, arginine-15 N4 was slightly enhanced, while voglibose was severely enhanced (Fig 5) By comparing the recoveries, it can be found that both arginine-15 N4 and voglibose had similar extraction recoveries as TTX Therefore, arginine-15 N4 was finally selected as IS in this study (see Table 1) RSD (%) 3.17 a 200-fold linear dynamic range (0.1–20 ng/mL) with the LLOQ (signal-to-noise ratio, S/N>10) of about 0.1 ng/mL Moreover, spikerecovery studies assessed for four different QCs (at 0.1, 0.5, 2, and 10 ng/mL, n = 6) at three consecutive days The intra- and interassay accuracies (%) were 98.5%−99.8% and 98.8%−99.5%, respectively, calculated by comparing measured concentrations to theoretically spiked concentrations The intra- and inter-assay precision investigated in terms of relative standard deviations (RSDs%), were ≤ 5.92% and ≤ 6.23%, respectively The results were summarized in Table 1, which showed acceptable accuracy and precision for plasma TTX in the investigated concentration range The validation of the method selection proved that samples were prepared well by Siphila i HILIX The representative chromatograms of blank samples and LLOQ samples at concentration of 0.1 ng/mL of TTX and 10 ng/mL of IS were shown in Fig No obvious interferences coeluting with the TTX and the IS was observed and the baselines of noise were acceptable The LLOQ was about 0.3 ng/mL for confirmation transition (m/z 320.3→284.1) according to the criterion of S/N>10, indicating that the method in this study would be suit- 2.3 Method validation After determining the optimization of SPE conditions, validation was attempted to investigate the practicability of the present method The method displayed excellent linearity of TTX with 1/x as weighting factor, and R value was higher than 0.999 over L Xin, Y Liang, S Yang et al Journal of Chromatography A 1684 (2022) 463567 Fig The recoveries of TTX in each step on three HILIC-SPE plates under different sample loading conditions a Siphila i HILIX SPE, b Silica, c PSA L Xin, Y Liang, S Yang et al Journal of Chromatography A 1684 (2022) 463567 Fig Plasma matrix effect of a TTX, b arginine-15 N4 and c voglibose Fig The representative chromatograms of blank, LLOQ (0.1 ng/mL) of a TTX and b IS (10 ng/mL) L Xin, Y Liang, S Yang et al Journal of Chromatography A 1684 (2022) 463567 Table Bias and precision of spiked samples from different sources sources Plasma1 Plasma2 Plasma3 Plasma4 Plasma5 Plasma6 hemolytic plasma high triglyceride plasma high cholesterol plasma high bilirubin plasma 0.1 ng/mL (n = 3) 10 ng/mL (n = 3) Mean (bias%) RSD (%) Mean (bias%) RSD (%) 3.67 −3.37 1.73 −6.00 −1.43 −2.67 −3.53 0.90 −3.67 −5.00 8.83 6.16 8.59 2.50 5.90 1.49 4.28 4.80 3.10 4.95 4.23 0.97 −0.37 3.20 −7.00 7.43 −2.50 −4.67 −2.50 3.93 0.32 1.73 2.99 2.51 4.07 1.04 3.86 2.92 6.40 2.37 alency factors [46] Although their SPE and LC-MS behavior may be similar to that of TTX, it is likely that they share the same or similar parent or product ions with TTX, which will interfere the detection of TTX and require further chromatographic separation More work is needed to prove whether interference exists from TTX analogs or whether TTX analogs can also be detected by our method or an improved method Conclusions An LC-MS/MS method had been developed and validated for the quantitation of TTX in human plasma This method used multicarboxyl group SPE materials and did not require evaporation and ultrafiltration steps Compared with previous reports, the method was highly sensitive, simple, and rapid in sample pre-treatment and had short chromatographic time The method presented in this paper showed wide linearity, low LLOQ, good precision and accuracy, and high recovery characteristics which well satisfied the requirements of pharmacokinetic study and clinical diagnosis of TTX able for the identification and quantification of TTX at poisoned concentration levels The extraction recovery was assessed by comparing the peak areas of spiked plasma samples with those of post-extraction spiked samples at three QC levels The quantitative matrix factors were evaluated by comparing the peak areas of post-extraction spiked samples from six sources at two QC levels with those of neat standard solutions The method demonstrated excellent extraction recovery for TTX, ranging from 80.8% to 81.5% at three different concentrations (RSDs ≤ 6.78%) The mean IS normalized matrix factor of samples from six sources at two levels were 0.979 and 0.968 respectively, which indicated no significant ion suppression or enhancement existed by the optimized method (Table 1) The lost extraction recovery of TTX was mainly in the washing step, as shown in Fig 4a, which was speculated to be improved by increasing the amount of SPE packed material or reducing the washing volume Good specificity means that the analytical method can differentiate between the target analyte and interfering substances for samples from different sources and the bias and precision were acceptable Ten blank human plasma samples including samples from six healthy volunteers and pooled blank human plasma samples mixing with high concentrations of hemoglobin, triglyceride, cholesterol, and bilirubin were prepared and investigated by spiking analysis In detail, working standard solutions of TTX were spiked to the above 10 blank plasma samples to make QCs at LLOQ and High concentration levels (0.1, 10 ng/mL) The QCs were pretreated and analyzed The biases and RSDs were listed in Table The biases of target value were ranged from −7.00% to 7.43% for healthy human plasma samples (RSDs ≤ 8.83%) and from −5.00% to 3.93% for hemolytic, high triglyceride, high cholesterol and high bilirubin plasma samples (RSDs ≤ 6.40%), which proved the good anti-interference property of the method Saxitoxin and its approximately 60 natural analogues, which are also potent and specific voltage-gated sodium channel blockers and share the same binding site as that of TTX [43, 44], also cause paralysis and even death in extreme cases In our supplemental study, gonyautoxin5 (GTX5) was purchased for interference test since its retention time was closest to TTX on amide column [45, 37] Under the optimized LC and SPE conditions described in this study, the retention time of GTX5 was 2.53 which was close to TTX (see Fig S2) No interferences were observed between the TTX and GTX5 MRM transitions Therefore, we believe that saxitoxin and its analogues may not interfere the detection of TTX Besides, in a preliminary study, the total recovery of GTX5 from plasma samples was about 70% according to the peak area ratio of spiked plasma samples and neat standard solutions which indicated that the described method in our study may be suitable for quantitation of these compounds after optimization In addition, the analogs of TTX like 4,9-anhydro TTX, 4–epi TTX and so on, usually have similar chemical structures and lower toxic equiv- Credit authorship contribution statement Liang Xin: Conceptualization, Investigation, Methodology, Writing - Original Draft Yan Liang: Data Curation, Investigation, Writing - Original Draft Shuangshuang Yang: Methodology, Data Curation, Visualization Fengli Jiang: Investigation, Methodology Fan Yu: Investigation, Validation Meiwei Zhang: Data curation, Methodology Wei Chang: Funding acquisition Wei Wang: Data curation, Validation Chen Yu: Supervision Gangyi Liu: Formal analysis, Project administration, Writing - Review & Editing Youli Lu: Funding acquisition, Formal analysis, Project administration, Writing - Review & Editing Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper Data Availability Data will be made available on request Acknowledgments This research was funded by the Innovation Fund of Science and Technology Commission of Shanghai Municipality (20Y1190 080 0) and the General Project of Medical Research Fund of Xuhui District, Shanghai (SHXH202121) Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.chroma.2022.463567 References [1] L Biessy, M.J Boundy, K.F Smith, D.T Harwood, I Hawes, S.A Wood, Tetrodotoxin in marine bivalves and edible gastropods: a mini-review, Chemosphere 236 (Dec 01, 2019), doi:10.1016/j.chemosphere.2019.124404 [2] P Gouel, C.M iti Gatti, L de Haro, A Liautaud, J Langrand, D Boucaud-Maitre, Tetrodotoxin poisoning in Mainland France and French overseas territories: a review of published and unpublished cases, Toxins (Basel) 14 (5) (May 2022) 351, doi:10.3390/toxins14050351 [3] D.F Hwang, T Noguchi, Tetrodotoxin Poisoning, Adv Food Nutr Res 52 (2007) 141–236, doi:10.1016/S1043-4526(06)52004-2 [4] J.F Fernndez-Ortega, et al., Seafood intoxication by tetrodotoxin: first case in europe, J Emerg Med 39 (5) (Nov 2010) 612–617, doi:10.1016/j.jemermed 2008.09.024 L Xin, Y Liang, S Yang et al Journal of Chromatography A 1684 (2022) 463567 [5] J You, Y.J Yue, F Xing, W Xia, S.Y Lai, F.L Zhang, Tetrodotoxin poisoning caused by goby fish consumption in southeast China: a retrospective case series analysis, Clinics 70 (1) (2015) 24–29, doi:10.6061/clinics/2015(01)05 [6] P Almeida, R Diaz, F Hernandez, G Ferrer, Blow: a case of pufferfish intoxication in South Florida, BMJ Case Rep 12 (6) (Jun 2019), doi:10.1136/ bcr- 2019- 229272 [7] B Alhatali, et al., A cluster of tetrodotoxin poisoning in Oman, Clin Toxicol 60 (2) (2022) 262–266, doi:10.1080/15563650.2021.1917595 [8] M.C.K Geoffrey K Isbister, Neurotoxic marine poisoning, Lancet Neurol (2005) 219–228, doi:10.1016/S1474- 4422(05)70041- [9] Y Bentur, et al., Lessepsian migration and tetrodotoxin poisoning due to Lagocephalus sceleratus in the eastern Mediterranean, Toxicon 52 (8) (Dec 15, 2008) 964–968, doi:10.1016/j.toxicon.2008.10.001 [10] V Bane, M Lehane, M Dikshit, A O’Riordan, A Furey, Tetrodotoxin: chemistry, toxicity, source, distribution and detection, Toxins (Basel) (2) (2014) 693– 755, doi:10.3390/toxins6020693 [11] Q.T Islam, et al., Puffer fish poisoning in Bangladesh: clinical and toxicological results from large outbreaks in 2008, Trans R Soc Trop Med Hyg 105 (2) (Feb 2011) 74–80, doi:10.1016/j.trstmh.2010.10.002 [12] N Ogata and Y Ohishi, “Molecular diversity of structure and function of the voltage-gated Na+ channels,” 2002 doi: 10.1254/jjp.88.365 [13] H.L Chong, P.C Ruben, Interaction between voltage-gated sodium channels and the neurotoxin, tetrodotoxin, Channels (6) (2008) 407–412, doi:10.4161/ chan.2.6.7429 [14] W.A Catterall, A.L Goldin, S.G Waxman, International Union of Pharmacology XLVII Nomenclature and structure-function relationships of voltage-gated sodium channels, Pharmacol Rev 57 (4) (Dec 2005) 397–409, doi:10.1124/pr 57.4.4 [15] F.R Nieto, E.J Cobos, M.Á Tejada, C Sánchez-Fernández, R González-Cano, C.M Cendán, Tetrodotoxin (TTX) as a therapeutic agent for pain, Mar Drugs 10 (2) (2012) 281–305, doi:10.3390/md10020281 [16] R González-Cano, M.C Ruiz-Cantero, M Santos-Caballero, C Gómez-Navas, M Tejada, F.R Nieto, Tetrodotoxin, a potential drug for neuropathic and cancer pain relief? Toxins (Basel) 13 (7) (2021), doi:10.3390/toxins13070483 [17] M Kavoosi, et al., Safety, tolerability, pharmacokinetics, and concentration-QTc analysis of tetrodotoxin: a randomized, dose escalation study in healthy adults, Toxins (Basel) 12 (8) (Aug 2020), doi:10.3390/toxins12080511 [18] G.M Bucciarelli, M Lechner, A Fontes, L.B Kats, H.L Eisthen, H.B Shaffer, From poison to promise: the evolution of tetrodotoxin and its potential as a therapeutic, Toxins (Basel) 13 (8) (2021) MDPI AG, Aug 01, doi:10.3390/ toxins13080517 [19] G.S Shephard, “Committee on natural toxins and food allergens mycotoxins,” 2007 doi: 10.1093/jaoac/90.1.1b [20] D Neagu, L Micheli, G Palleschi, Study of a toxin-alkaline phosphatase conjugate for the development of an immunosensor for tetrodotoxin determination, Anal Bioanal Chem 385 (6) (Jul 2006) 1068–1074, doi:10.1007/ s0 0216-0 06-0522-2 [21] R Watanabe, et al., Quantitation of Tetrodotoxin and Its Analogues with a Combination of Liquid Chromatography-Tandem Mass Spectrometry and Quantitative 1H-NMR Spectroscopy, J Agric Food Chem 67 (46) (Nov 2019) 12911– 12917, doi:10.1021/acs.jafc.9b06380 [22] M.A O’Leary, J.J Schneider, G.K Isbister, Use of high performance liquid chromatography to measure tetrodotoxin in serum and urine of poisoned patients, Toxicon 44 (5) (Oct 2004) 549–553, doi:10.1016/j.toxicon.20 04.07.0 08 [23] K.S.Y Leung, B.M.W Fong, Y.K Tsoi, Analytical challenges: determination of tetrodotoxin in human urine and plasma by LC-MS/MS, Mar Drugs (11) (2011) 2291–2303 MDPI AG, doi:10.3390/md9112291 [24] H.K Knutsen, et al., Risks for public health related to the presence of tetrodotoxin (TTX) and TTX analogues in marine bivalves and gastropods, EFSA J 15 (4) (Apr 2017), doi:10.2903/j.efsa.2017.4752 [25] F Long, M Zhang, J Yuan, J Du, A Ma, J Pan, A simple, versatile, and automated pulse-diffusion-focusing strategy for sensitive milliliter-level-injection HILIC-MS/MS analysis of hydrophilic toxins, J Hazard Mater 392 (Jun 2020), doi:10.1016/j.jhazmat.2020.122318 [26] A.L.N van Nuijs, I Tarcomnicu, A Covaci, Application of hydrophilic interaction chromatography for the analysis of polar contaminants in food and environmental samples, J Chromatogr A 1218 (35) (Sep 02, 2011) 5964–5974, doi:10.1016/j.chroma.2011.01.075 [27] D Salas, F Borrull, N Fontanals, R.M Marcé, Hydrophilic interaction liquid chromatography coupled to mass spectrometry-based detection to determine emerging organic contaminants in environmental samples, TrAC - Trends Analyt Chem 94 (Sep 01, 2017) 141–149, doi:10.1016/j.trac.2017.07.017 [28] H Mei, et al., Investigation of matrix effects in bioanalytical high-performance liquid chromatography/tandem mass spectrometric assays: application to drug [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] discovery, Rapid Commun Mass Spectrom 17 (1) (2003) 97–103, doi:10.1002/ rcm.876 T.M Annesley, “Ion suppression in mass spectrometry,” 2003 doi: 10.1373/49.7.1041 B.K Matuszewski, M.L Constanzer, C.M Chavez-Eng, Strategies for the assessment of matrix effect in quantitative bioanalytical methods based on HPLCMS/MS, Anal Chem 75 (13) (Jul 2003) 3019–3030, doi:10.1021/ac020361s H Trufelli, P Palma, G Famiglini, A Cappiello, An overview of matrix effects in liquid chromatography-mass spectrometry, Mass Spectrom Rev 30 (3) (May 2011) 491–509, doi:10.1002/mas.20298 F Gosetti, E Mazzucco, D Zampieri, M.C Gennaro, Signal suppression/enhancement in high-performance liquid chromatography tandem mass spectrometry, J Chromatogr A 1217 (25) (Jun 2010) 3929–3937, doi:10.1016/j chroma.2009.11.060 M.M Khamis, D.J Adamko, A El-Aneed, Strategies and Challenges in Method Development and Validation for the Absolute Quantification of Endogenous Biomarker Metabolites Using Liquid Chromatography-Tandem Mass Spectrometry, Mass Spectrom Rev (2019) John Wiley and Sons Inc., doi:10.1002/mas 21607 H.C Jen, S.J Lin, Y.H Tsai, C.H Chen, Z.C Lin, D.F Hwang, Tetrodotoxin poisoning evidenced by solid-phase extraction combining with liquid chromatography-tandem mass spectrometry, J Chromatogr B Anal Technol Biomed Life Sci 871 (1) (Aug 2008) 95–100, doi:10.1016/j.jchromb.2008.06 030 H.E Cho, et al., Determination and validation of tetrodotoxin in human whole blood using hydrophilic interaction liquid chromatography-tandem mass spectroscopy and its application, Forensic Sci Int 217 (1–3) (Apr 2012) 76–80, doi:10.1016/j.forsciint.2011.10.026 M Rambla-Alegre, et al., Rapid screening and multi-toxin profile confirmation of tetrodotoxins and analogues in human body fluids derived from a puffer fish poisoning incident in New Caledonia, Food Chem Toxicol 112 (Feb 2018) 188–193, doi:10.1016/j.fct.2017.12.039 N Ochi, Simultaneous determination of ten paralytic shellfish toxins and tetrodotoxin in scallop and short-necked clam by ion-pair solid-phase extraction and hydrophilic interaction chromatography with tandem mass spectrometry, J Chromatogr A 1651 (Aug 2021), doi:10.1016/j.chroma.2021.462328 B.M.W Fong, S Tam, S.H Tsui, K.S.Y Leung, Development and validation of a high-throughput double solid phase extraction-liquid chromatography-tandem mass spectrometry method for the determination of tetrodotoxin in human urine and plasma, Talanta 83 (3) (Jan 2011) 1030–1036, doi:10.1016/j.talanta 2010.11.014 U.S Food and Drug Administration, Bioanalytical Method Validation Guidance for Industry, https://www.fda.gov/files/drugs/published/ Bioanalytical- Method- Validation- Guidance- for- Industry.pdf#:∼:text= Bioanalytical%20Method%20Validation%20Guidance%20for%20Industry%201% 20This,is%20not%20binding%20on%20FDA%20or%20the%20public, 2018 R Li, W Sun, X Xiao, B Chen, Y Wei, Retention of stevioside polar compounds on a sulfonic acid-functionalized stationary phase, J Chromatogr A (1620) 2020, doi:10.1016/j.chroma.2020.460978 T Ikegami, A Taniguchi, T Okada, K Horie, S Arase, Y Ikegami, Functionalization using polymer or silane? A practical test method to characterize hydrophilic interaction chromatography phases in terms of their functionalization method, J Chromatogr A 1638 (2021) 461850, doi:10.1016/j.chroma.2020 461850 D Bratkowska, R.M Marcé, P.A.G Cormack, D.C Sherrington, F Borrull, N Fontanals, Synthesis and application of hypercrosslinked polymers with weak cation-exchange character for the selective extraction of basic pharmaceuticals from complex environmental water samples, J Chromatogr A 1217 (10) (2010) 1575–1582, doi:10.1016/j.chroma.2010.01.037 M Wiese, P.M D’Agostino, T.K Mihali, M.C Moffitt, B.A Neilan, Neurotoxic alkaloids: saxitoxin and its analogs, Mar Drugs (7) (2010) 2185–2211, doi:10 3390/md8072185 T.W Soong, B Venkatesh, Adaptive evolution of tetrodotoxin resistance in animals, Trends Genet 22 (11) (2006) 621–626, doi:10.1016/j.tig.2006.08.010 S Numano, Y Kudo, Y Cho, K Konoki, M Yotsu-Yamashita, Temporal variation of the profile and concentrations of paralytic shellfish toxins and tetrodotoxin in the scallop, Patinopecten yessoensis, cultured in a bay of East Japan, Mar Drugs 17 (12) (2019), doi:10.3390/md17120653 I.J Tamele, M Silva, V Vasconcelos, The incidence of tetrodotoxin and its analogs in the Indian Ocean and the Red Sea, Mar Drugs 17 (1) (2019) 28– 38, doi:10.3390/md17010028  ... Simultaneous determination of ten paralytic shellfish toxins and tetrodotoxin in scallop and short-necked clam by ion-pair solid- phase extraction and hydrophilic interaction chromatography with tandem... John Wiley and Sons Inc., doi:10.1002/mas 21607 H.C Jen, S.J Lin, Y.H Tsai, C.H Chen, Z.C Lin, D.F Hwang, Tetrodotoxin poisoning evidenced by solid- phase extraction combining with liquid chromatography-tandem... blank human plasma samples mixing with high concentrations of hemoglobin, triglyceride, cholesterol, and bilirubin were prepared and investigated by spiking analysis In detail, working standard