Liquid chromatographic method for toxic biogenic amines in foods using a chaotropic salt

Journal of Chromatography A, 1406 (2015) 331–336

Contents lists available at ScienceDirect

Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma

Short communication

Liquid chromatographic method for toxic biogenic amines in foods using a chaotropic salt Jian-Jun Zhong a , Ningbo Liao b , Tian Ding a , Xingqian Ye a , Dong-Hong Liu a,∗ a Fuli Institute of Food Science, Zhejiang Key Laboratory for Agro-Food Processing, Zhejiang R & D Center for Food Technology and Equipment, Department of Food Science and Nutrition, Zhejiang University, Hangzhou 310058, China b Department of Nutrition and Food Safety, Zhejiang Provincial Center for Disease Control and Prevention, Hangzhou 310006, China

a r t i c l e

i n f o

Article history: Received 30 April 2015 Received in revised form 6 June 2015 Accepted 14 June 2015 Available online 23 June 2015 Keywords: Biogenic amine Chaotropic salt Food analysis Ion-pair reagent Retention behavior Reversed-phase chromatography

a b s t r a c t Direct separation of biogenic amines by reversed-phase liquid chromatography (RPLC) is not an easy task because their basic and hydrophilic characteristics can lead to poor retention, column overloading, peak tailing, and hence low efficiency. Rather than routinely resorting to derivatization or using classical hydrophobic ion-pair reagents (IPR), this work proposes a new RPLC method making use of the chaotropic salt KPF6 as inorganic additive to an acidic acetonitrilic eluent to remedy the difficulties. Amine retention, overload behavior, peak shape, and column efficiency were significantly improved. The use of excess KPF6 led to a very slight decrease of amine retention. Depending on amine, the dependence of the logarithmic retention factor on the volume percent of acetonitrile could be reasonably linear or quite convex. Coupled with UV detection, the method was applied to trace analysis for six biogenic, aromatic or heterocyclic amines in three types of food after a sample cleanup, as necessary, by ion-pair extraction. The reliability of the whole analysis was demonstrated to be satisfactory. The proposed method outperforms existing methods in that it eliminates the need for long and cumbersome derivatization procedures without losing sensitivity; it also represents a good surrogate for classical ion-pair chromatography (IPC) because of the desirable hydrophilicity of chaotropic salts. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Biogenic amines are bioactive organic bases mainly arising from undesirable microbial decarboxylase activity during the processing, storage or spoilage of food [1,2]. Based on structure, they are classified as aliphatic, aromatic or heterocyclic [3]. Many aromatic or heterocyclic amines possess well-documented toxicity when ingested in sufficient amounts, scombroid fish poisoning caused by histamine and the “cheese reaction” by tyramine and 2-phenylethylamine being the most notorious. Aliphatic amines are generally far less toxic, albeit can potentiate the toxicity of other amines [3,4]. Besides being a public health problem, biogenic amines have been proposed as microbiological quality indicators of certain foods: the Codex Alimentarius standards for fish provide histamine concentrations as indicators for decomposition, hygiene, and handling [3].

∗ Corresponding author at: Department of Food Science and Nutrition, Zhejiang University, Hangzhou 310058, China. Tel.: +86 571 88982169; fax: +86 571 88982154. E-mail address: [email protected] (D.-H. Liu). http://dx.doi.org/10.1016/j.chroma.2015.06.048 0021-9673/© 2015 Elsevier B.V. All rights reserved.

RP-HPLC with pre- or post-column derivatization has been the most common technique for simultaneously analyzing several biogenic amines [1]. Due to their lacking chromophores and fluorophores, derivatization (chromophore- or fluorophore-tagging) is indispensable to the direct photometric detection of aliphatic amines. However, this may not be the case for aromatic and heterocyclic amines whose UV detection is straightforward [5–7] provided the sensitivity and selectivity prove adequate for trace analysis. Besides the consideration of detection, the aim of a precolumn derivatization strategy to ease the separation should be underscored [1,8], because biogenic amines are highly polar bases that often exhibit poor behavior in reversed-phase chromatography (RPC). For the same reason, a post-column derivatization strategy is commonly combined with the adoption of ion-pair chromatography (IPC) using hydrophobic ion-pair reagents (IPR) like alkyl sulfonates [7,9]. Chaotropic salts such as perchlorates, tetrafluoroborates, triflates, and hexafluorophosphates, are a new class of inorganic IPR for bases. Chaotropic salts mimic the role of hydrophobic IPR in improving the retention and peak shape of bases in RPC, while they are quite hydrophilic and easily dissolved in the mobile phase as opposed to hydrophobic IPR that tend to stick strongly to the

332

J.-J. Zhong et al. / J. Chromatogr. A 1406 (2015) 331–336

Table 1 Structures and chemical properties of biogenic amines. Name Histamine

Structure

N

Chemical formula and molecular weight

pKa

Log P

C5 H9 N3 = 111.15

5.9 (imidazole), 9.7 ( NH2 )

−0.70

C8 H11 N = 121.18

9.83

1.41

C8 H11 NO = 137.18

9.5 (phenol), 10.8 ( NH2 )

(−0.14, 0.86)a

C8 H11 NO2 = 153.18

8.81

−0.90

C10 H12 N2 = 160.22

10.2

1.55

C10 H12 N2 O = 176.22

10 ( NH2 ), 10.7 (phenol)

0.21

NH 2

HN 2-Phenylethylamine

NH 2 Tyramine

NH2

HO Octopamine

OH NH2

HO Tryptamine

H N NH2

Serotonin

H N HO

a

NH2

The two were predicted by ALOGPs and KOWWIN programs, respectively. All the other pKa and Log P data were retrieved from the PHYSPROP database.

stationary phase, leading to slow column-equilibration and difficulty in recovering the initial column properties [10–13]. Nowadays, chaotropic salts are usually used in the analysis of active ingredients and impurities in pharmaceutical formulations [11,14–16]. In the meanwhile, attempts to understand the fundamentals behind the applications are still ongoing (see Ref. [17] for a review). Using NaPF6 as mobile-phase additive, Flieger and ˙ Czajkowska-Zelazko investigated the chromatographic behavior of five neurotransmitter amines [18]. To the best of our knowledge, no approach has been proposed dealing with analyzing biogenic amines in foods, taking advantage of chaotropic salts. As Erim [1] correctly concluded, the recent innovations in biogenic amine analysis by HPLC mostly concern extraction and derivatization procedures rather than separation science. The focus of this work, in contrast, is on developing a new RP-HPLC method making use of KPF6 for analyzing six biogenic, aromatic or heterocyclic amines (Table 1) of toxicological relevance in food. For detection their natural chromophoric properties were straightforwardly exploited.

2. Experimental

other common chemicals (analytical grade) were from Sinopharm (Shanghai, China). 2.2. Chromatographic system and conditions The HPLC system consisted of a Waters e2695 separation module, a 2489 TUV detector, and an Agilent Zorbax SB-C8 column (150 mm × 4.6 mm I.D., 5-␮m particles). The system void volume was 1.64 mL, determined by NaNO3 injection with the test eluent 10 mM H3 PO4 –ACN 90:10. This value was used for all retention factor calculations. Peak tailing factors (TF) were calculated according to US Pharmacopeia recommendations. The binary mobile phase consisted of (A) 50 mM KPF6 in 10 mM H3 PO4 (pH 2.2) and (B) ACN. The flow rate was 1.5 mL min−1 and the column temperature maintained at 30 ◦ C. The gradient was programmed as follows: 11–11–26–26% B at 0–5–5–11 min. A 5-min column-wash (with 100% ACN) and a 10-min equilibration were employed between successive runs. Typical injection volumes were 10 ␮L. The TUV detector was operated at dual-wavelength mode, 210 nm for quantitation and a different wavelength for coelution check. Baseline was autozeroed at 7.5 min.

2.1. Chemicals

2.3. Sample preparation

Histamine, tyramine, and tryptamine were from Sigma–Aldrich (Shanghai, China); serotonin hydrochloride, 2-phenylethylamine, and octopamine hydrochloride were from Aladdin (Shanghai, China). Their purities were all over 98%. Bis (2-ethylhexyl) phosphate (BEHPA) 95% and KPF6 99.98% were from Aladdin and HPLC-grade acetonitrile (ACN) from Scharlau (Barcelona, Spain). All

One vinegar and one baijiu (a Chinese liquor at 56% alcohol by volume) sample were purchased from a local store. A sample of chub mackerel, spoiled at room temperature for 2 days, was kindly provided by C. Wu. The baijiu sample was diluted fivefold with deionized water and filtered before analysis.

J.-J. Zhong et al. / J. Chromatogr. A 1406 (2015) 331–336

333

The vinegar sample was subjected to an ion-pair extraction (IPE) cleanup as described in Ref. [19] with minor modifications: to a 0.5mL aliquot was added concentrated NaOH solution to adjust the pH to near neutrality. The solution was then buffered with 2.5 mL of 50 mM pH 7.4 potassium phosphate, extracted with 2 mL of 0.1 M BEHPA in chloroform, and centrifuged. The organic phase was gathered and back-extracted with 2 mL of 0.1 M HCl. After centrifugation, 1.5 mL of the aqueous phase was withdrawn, de-acidified, and filtered. Ten grams of deboned, finely minced mackerel was weighed and acid-extracted as detailed in Ref. [20], followed by the IPE procedure as described above. 2.4. Preparation of calibrators and calibration Stock solutions were prepared by dissolving each standard substance in 10% (v/v) ACN–water. Seven calibrators (0.1–10 ␮g mL−1 ) were prepared by diluting the stock solutions with deionized water. They were injected without any pretreatment to establish the basic calibration [21], which could also be used for quantification in the baijiu sample. Calibration standards for the vinegar sample were prepared separately from the above calibrators. Amines were dissolved in 35 g L−1 acetic acid to prepare more matrix-matched calibrators. Seven calibrators (0.5–50 ␮g mL−1 ) plus a blank were ion-pair extracted before analysis. Calibration curves were generated from the plots of peak height (␮AU) against analyte concentration (␮g mL−1 ) using linear leastsquares regression. The curves were forced through the origin when appropriate (y-intercept ≤ its standard error [22]). To assess linearity the Mandel’s fitting test [21] was applied and the coefficients of determination (r2 ) also calculated. The method of standard additions was employed to quantify biogenic amines in the mackerel sample. The spiking levels were 0, 200, and 500 ␮g g−1 with three replicates each.

Fig. 1. Dependence of amine retention on KPF6 concentration (calculated with respect to the aqueous portion of the mobile phase). Symbols and lines from top to bottom correspond to tryptamine, 2-phenylethylamine, serotonin, tyramine, histamine, and octopamine. Conditions: Zorbax SB-C8 , (100 mM KPF6 in 10 mM H3 PO4 )–ACN–10 mM H3 PO4 x:12: (88 − x), 1.5 mL min−1 , 30 ◦ C.

overloading of those ionized bases [24]. It was impossible to improve the chromatographic behavior by operating at alkaline pH where the amines are deprotonated, because of the problematic stability of common silica-based columns at those pHs. In further experiments KPF6 , ranked first in terms of chaotropicity in the Hofmeister series, was used as mobile-phase additive to remedy the difficulties. It is important that the mobile-phase pH remained low to transfer the amines into protonated form, which can then efficiently interact with the PF6 − ions. Here 10 mM H3 PO4 was invariably used as the aqueous portion of mobile phases.

2.5. Validation 3.1. Effect of KPF6 concentration on amine retention Method accuracy (trueness and intra-assay precision) was evaluated by recovery studies using the vinegar sample spiked at 2 and 5 ␮g mL−1 . At each spiking level three replicates were prepared and each replicate analyzed twice; the apparent recoveries (R* ) [23] and the RSD values were calculated. The retention time precision was assessed by repeatedly (n = 5) injecting a calibration standard solution. Detection (DL) and quantitation limits (QL) were estimated using the signal-to-noise ratio approach (S/N = 3 and 10 for DL and QL, respectively). The method detection limits (MDL) were estimated in the vinegar sample and, therefore, were distinguished from the instrument detection limits (IDL) estimated by injecting clean standard solutions. 3. Results and discussion From Table 1 it can be concluded that the six biogenic amines under analysis are small size, strong organic bases with considerable hydrophilicity. It has long been recognized that the separation of such compounds by RPC can be challenging, such as poor retention, low efficiency, detrimental silanol interactions, and column overload (the latter two both lead to poor peak shapes) [24]. Preliminary experiments showed that, with an acidic (to suppress silanol ionization), eluotropically weak eluent (10 mM H3 PO4 –ACN 95:5), octopamine, histamine, and tyramine had very poor retention (k = 0–1). Furthermore, at sample loads greater than ca. 0.5 ␮g, the peaks of 2-phenylethylamine and tryptamine clearly exhibited a right-sided “shark-fin” type appearance and the apex retention times decreased with increasing sample load, suggesting mass

As expected, each amine showed a clear retention increase upon adding KPF6 (Fig. 1). Good peak symmetry (TF < 1.2; 0.5 ␮g sample load) was obtained since 10 mM KPF6 addition, suggesting a significant improvement in their overload behavior. The role of PF6 − in enhancing retention was most apparent at low addition levels. As the PF6 − concentration became higher and higher, a saturation effect gradually appeared and even a very slight but reproducible decrease of retention was seen (e.g., when >50 mM for octopamine and histamine). A notable selectivity change was also observed: with no more than 2 mM PF6 − , histamine was eluted before octopamine, whereas using PF6 − concentrations higher than 5 mM, the elution order was reversed. Both stoichiometric and non-stoichiometric models have been proposed to rationalize the chaotrope-enhanced retention of ionized bases. The main weakness of all stoichiometric models has been indicated by Cecchi et al. [25,26] recently: they erroneously neglect the development of an electrostatic surface potential due to the adsorption of chaotropic anions onto the stationary phase. Our finding is clearly discordant with the important prediction of the stoichiometric “chaotropic model” that, with increasing chaotrope concentration in the mobile phase, base retention would finally approach an asymptotic upper limit [27]. According to Cecchi’s non-stoichiometric model (named the “extended thermodynamic approach”) [28], such retention decrease can be attributed to ion-pair formation in the mobile phase and/or a salting-in effect, provided adsorption competitions are negligible (usually the case).

334

J.-J. Zhong et al. / J. Chromatogr. A 1406 (2015) 331–336

Fig. 2. (a) Plots of amine retention (, serotonin; 䊐, tyramine; 䊏, octopamine; 䊉, histamine) against the volume fraction of ACN in the eluent (ϕ). (b) Peak splitting was observed when injecting 40 ␮L of 10 ␮g mL−1 histamine in water, but not when injecting 10 ␮L of 40 ␮g mL−1 , indicating a sample solvent effect (see text for details). (c) The “atypical” ϕ-dependence of histamine retention may help peak identity confirmation (see text for details). Conditions: Zorbax SB-C8 , (50 mM KPF6 in 10 mM H3 PO4 )–ACN, 1.5 mL min−1 , 30 ◦ C.

3.2. Effect of %ACN on amine retention Fig. 2a shows the dependence of histamine, octopamine, tyramine, and serotonin retention on the volume fraction of ACN in the mobile phase (ϕ = 0–0.24). The KPF6 concentration in 10 mM H3 PO4 was constant at 50 mM, around which concentration retention of the four amines approached their maxima and leveled off (see Fig. 1), so that the PF6 − concentration dependence of their retention was negligible compared to the ϕ-dependence. The retention of tyramine and serotonin could be approximated by the well-known empirical relationship satisfactorily in terms of r2 statistic (0.996 and 0.995, respectively), log k = a − bϕ

(1)

Here log k is the logarithmic retention and a and b are positive constants for a given compound. For octopamine a small downward deviation from the linearity (r2 = 0.996) occurred below 4% ACN. With respect to histamine, a strikingly different, convex plot of log k vs ϕ was observed with a retention maximum at 12% ACN. Wang and Carr [29] also found a “quite nonlinear” log k–ϕ relationship at low %ACN while analyzing basic drugs with perchlorate as additive, but they did not give any explanation for the nonlinearity. Actually, their study presumed a linear relationship. It was not surprising to us that deviations from Eq. (1) may be observed because previous research (Cecchi et al. [30]) has already demonstrated that the empirical relationship does not hold when an electrostatic surface potential is developed and acts as a determining factor of solute retention. A qualitative explanation for the initial rise in histamine retention with increasing %ACN may be obtained by considering the unusual retention behavior of PF6 − ions in RPC. As indicated in Refs. [31,32], with increasing %ACN, PF6 − retention first increases considerably to a maximum and then decays. This may lead to a similar change of the electrostatic surface potential resulted from PF6 − adsorption, and a corresponding change of the surface potential determined retention. Note that histamine hardly interacted hydrophobically with the stationary phase and, under the acidic pH, it was dicationic while all others were monocationic. This implies that histamine retention would be more influenced by the magnitude of the surface potential than would the others, which possibly explains why a significantly convex plot of log k against ϕ was found for histamine.

It is worth noting that significant peak distortion or splitting was observed when injecting medium volumes (>20–30 ␮L) of histamine in water. Fig. 2b shows the chromatograms obtained from two successive injections (volume in ␮L × concentration in ␮g mL−1 ), 40 × 10 and 10 × 40. Peak splitting was observed only in the former case, suggesting a sample solvent effect. The underlying causes of this phenomenon are the strong eluotropic strength of the sample solvent (water) compared to the mobile phase containing 12% ACN (recalling Fig. 2a) or viscosity mismatch [33] or a combination of both. Of course the problem can be alleviated, as we observed, by using the mobile phase as the sample solvent, but for convenience, a small injection volume was used. The “atypical” retention behavior of histamine can be exploited to help peak identity confirmation and to attain separation selectivity. Take for instance the analysis of the baijiu sample (Fig. 2c): with an eluent containing 10% ACN, a peak was eluted exactly at the elution time of histamine (3 min); when the ACN percentage was increased to 12%, the signal was seen earlier (2.8 min). Easily and confidently, it was determined as an interferent. 3.3. Separation optimization and food analyses The chromatographic conditions were finally optimized for analyzing the three food samples. A 50 mM concentration of KPF6 was used in the aqueous portion. Acetonitrile was initially chosen over methanol as the organic modifier because of its lower UV cutoff and because a more increase in solute retention is obtained by using a hydro-acetonitrile system [31,32]. Due to the excessive retention of 2-phenylethylamine and tryptamine relative to the others, a simple step gradient was used. The gradient was optimized by trial and error with these goals: adequate retention (k ≥ 1) for earlier-eluting analytes, appropriate overall selectivity, and a short run time. Standard and sample chromatograms acquired under the optimal conditions are presented in Fig. 3. Excellent peak symmetry was achieved at a sample load of 0.5 ␮g for each analyte, thereby the overload tailing problem was perfectly addressed. The plate number was 58,353 per meter determined from the histamine peak with the half height method, indicating excellent efficiency. Large and toxic quantities of histamine, tyramine, 2-phenylethylamine (103–487 mg kg−1 ), and tryptamine (36 mg kg−1 ) were found in the 2-day-spoiled mackerel sample, suggesting their potential role

J.-J. Zhong et al. / J. Chromatogr. A 1406 (2015) 331–336

335

Fig. 3. Chromatograms of a standard mixture of biogenic amine (50 ␮g mL−1 ) and food samples. Peaks 1–6 were identified as octopamine, histamine, tyramine, serotonin, 2-phenylethylamine, and tryptamine, respectively. Peaks labeled with an asterisk were potential interferents under sub-optimized conditions. Gradient shape indicated by (. . .). N, plate number per meter; TF, tailing factor. For HPLC conditions, see Section 2.2. Table 2 Analytical performance characteristics. Compound

Retention timea (min)

Regression equationb

2 ␮g mL Octopamine Histamine Tyramine Serotonin 2-Phenylethylamine Tryptamine

2.35 (0.3) 2.96 (0.2) 4.17 (0.2) 5.73 (0.2) 8.67 (0.1) 10.10 (0.1)

y = 3202x y = 633x + 40 y = 3941x y = 935x y = 2559x y = 608x + 61 y = 5209x y = 1231x − 337 y = 3986x y = 880x + 202 y = 7169x y = 1837x + 289

DLd (␮g mL−1 )

%R* c at two spiking levels −1

89.5 (5.5) 98.2 (4.9) 96.7 (3.8) 94.4 (4.4) 95.0 (5.6) 94.8 (4.9)

−1

5 ␮g mL

IDL

MDL

91.0 (5.2) 99.1 (4.0) 96.5 (3.7) 92.6 (5.2) 95.3 (4.9) 96.9 (2.1)

0.01

0.09

0.01

0.13

0.02

0.10

0.01

0.06

0.04

0.14

0.03

0.15

DL, detection limit; IDL, instrument detection limit; MDL, method detection limit in the vinegar sample; QL, quantitation limit; R* , apparent recovery. a Mean of five injections. Values in brackets denote %RSD. b Upper equation, the basic calibration; lower equation created for the vinegar sample; all r2 > 0.9999. c Mean of three replicate samples. Values in brackets denote %RSD. d QL = 3.3DL.

as freshness indicators. Trace amounts of histamine (0.96 mg L−1 ) and tyramine (1.44 mg L−1 ) were detected in the vinegar sample, slightly higher than the levels reported by Callejón et al. but still within recommended upper limits [2]. As for the baijiu (liquor) sample, none of the analytes could be detected. This result seems plausible since nonvolatile amines (excluding 2phenylethylamine) can hardly be recovered by distillation.

3.4. Method validation Linearity was demonstrated by the r2 values higher than 0.9999 and by the Mandel test which showed that a quadratic calibration function did not provide a significantly better fit. The method accuracy was satisfactory as indicated by the apparent recoveries close to unity (89.5–99.1%) and by the RSD values that were lower than

336

J.-J. Zhong et al. / J. Chromatogr. A 1406 (2015) 331–336

6% (Table 2). The method detection limits were in the low ␮g mL−1 range, similar to those of some HPLC–UV methods reported recently which require derivatization [2,34]. Reasonably good specificity was obtained by using several means in combination. First was the efficient ion-pair sample cleanup, as had been proven by Ref. [19] and as indicated by the clean sample chromatograms. Second, other than the excellent retention, peak shape, and column efficiency already demonstrated, optimum overall resolution was obtained with judiciously considering the interference from co-extracted sample impurities: analytes were eluted well after the early “junk” peaks and also successfully resolved from other components which could interfere under sub-optimized conditions (Fig. 3). Third, peak height quantitation was used to minimize the influence of any incomplete resolution.

[10]

[11]

[12] [13]

[14]

[15]

4. Conclusions [16]

A good RPLC separation of six biogenic, aromatic or heterocyclic amines was obtained upon using the chaotropic salt KPF6 , which, in the acetonitrilic mobile phase buffered at acidic pH, significantly improved the retention, overload behavior, peak symmetry, and separation efficiency of protonated amines. Coupled with UV detection, the method was successfully applied to trace analysis for those toxic compounds in different food samples. As a favorable alternative to classical IPC, the proposed method eliminates the need for derivatization while still offering detection limits well below 1 ␮g mL−1 . Further work is required for a good theoretical modeling [28,30] of (1) the slight decrease of amine retention caused by excessive addition of KPF6 and (2) the different log k–ϕ relationships among amines.

[17] [18]

[19]

[20]

[21] [22] [23]

Acknowledgements Financial support from the Ministry of Science and Technology of China (Grants # 2012BAD29B07-02 and # 2011BAD24B01-3) is acknowledged. JJZ is grateful to Dan Wu for research assistance and to JL Gao for help in preparing the manuscript. References [1] F.B. Erim, Recent analytical approaches to the analysis of biogenic amines in food samples, Trends Anal. Chem. 52 (2013) 239–247. ˜ [2] J.L. Ordónez, R.M. Callejón, M.L. Morales, M.C. García-Parrilla, A survey of biogenic amines in vinegars, Food Chem. 141 (2013) 2713–2719. [3] EFSA Panel on Biological Hazards (BIOHAZ), Scientific opinion on risk based control of biogenic amine formation in fermented foods, EFSA J. 9 (2011) 2393. [4] M. Glória, Bioactive amines, in: Handbook of Food Science, Technology, and Engineering, Taylor & Francis, Boca Raton, FL, 2006, 13-1–13-38 pp. [5] A. Dang, J.J. Pesek, M.T. Matyska, The use of aqueous normal phase chromatography as an analytical tool for food analysis: determination of histamine in a model system, Food Chem. 141 (2013) 4226–4230. [6] W.-C. Lin, C.-E. Lin, E.C. Lin, Capillary zone electrophoretic separation of biogenic amines: influence of organic modifier, J. Chromatogr. A 755 (1996) 142–146. [7] S.-F. Chang, J.W. Ayres, W.E. Sandine, Analysis of cheese for histamine, tyramine, tryptamine, histidine, tyrosine, and tryptophane, J. Dairy Sci. 68 (1985) 2840–2846. [8] S. Hernández-Cassou, J. Saurina, Derivatization strategies for the determination of biogenic amines in wines by chromatographic and electrophoretic techniques, J. Chromatogr. B 879 (2011) 1270–1281. [9] M.L. Latorre-Moratalla, J. Bosch-Fusté, T. Lavizzari, S. Bover-Cid, M.T. VecianaNogués, M.C. Vidal-Carou, Validation of an ultra high pressure liquid

[24]

[25] [26]

[27]

[28]

[29]

[30]

[31]

[32] [33]

[34]

chromatographic method for the determination of biologically active amines in food, J. Chromatogr. A 1216 (2009) 7715–7720. L. Pan, R. LoBrutto, Y.V. Kazakevich, R. Thompson, Influence of inorganic mobile phase additives on the retention, efficiency and peak symmetry of protonated basic compounds in reversed-phase liquid chromatography, J. Chromatogr. A 1049 (2004) 63–73. ´ A. Malenovic, ´ M. Medenica, The influence of inorganic salts with A. Vemic, chaotropic properties on the chromatographic behavior of ropinirole and its two impurities, Talanta 123 (2014) 122–127. T. Cecchi, Ion-Pair Chromatography and Related Techniques, Taylor & Francis, Boca Raton, FL, 2010. T. Cecchi, Retention mechanism for ion-pair chromatography with chaotropic reagents: from ion-pair chromatography toward a unified salt chromatography Advances in Chromatography, vol. 49, Taylor & Francis, Boca Raton, FL, 2011, pp. 1–36. ´ B.J. Stojanovic, ´ I. Stamenkovic, ´ A. Malenovic, ´ Chaotropic agents in A. Vemic, liquid chromatographic method development for the simultaneous analysis of levodopa, carbidopa, entacapone and their impurities, J. Pharm. Biomed. Anal. 77 (2013) 9–15. ´ J. Flieger, R. Swieboda, Application of chaotropic effect in reversed-phase liquid chromatography of structurally related phenothiazine and thioxanthene derivatives, J. Chromatogr. A 1192 (2008) 218–224. E. Loeser, S. Babiak, M. DelaCruz, P. Karpinski, Triflic acid and sodium triflate as chaotropic mobile phase additives in RP-LC, J. Chromatogr. Sci. 49 (2011) 57–62. T. Cecchi, Theoretical models of ion pair chromatography: a close up of recent literature production, J. Liq. Chromatogr. Relat. Technol. 38 (2015) 404–414. ˙ J. Flieger, A. Czajkowska-Zelazko, Comparison of chaotropic salt and ionic liquid as mobile phase additives in reversed-phase high-performance liquid chromatography of biogenic amines, J. Sep. Sci. 34 (2011) 733–739. J.O. Fernandes, M.A. Ferreira, Combined ion-pair extraction and gas chromatography–mass spectrometry for the simultaneous determination of diamines, polyamines and aromatic amines in Port wine and grape juice, J. Chromatogr. A 886 (2000) 183–195. M. Kˇríˇzek, T. Pelikánová, Determination of seven biogenic amines in foods by micellar electrokinetic capillary chromatography, J. Chromatogr. A 815 (1998) 243–250. M. Koch, Calibration, in: Quality Assurance in Analytical Chemistry: Training and Teaching, 2nd ed., Springer, Heidelberg, 2010, pp. 183–200. J.W. Dolan, Calibration curves. Part I. To b or not to b? LCGC N. Am. 27 (2009) 224–230. D.T. Burns, K. Danzer, A. Townshend, Use of the terms “recovery” and “apparent recovery” in analytical procedures (IUPAC Recommendations 2002), Pure Appl. Chem. 74 (2002) 2201–2205. D.V. McCalley, The challenges of the analysis of basic compounds by high performance liquid chromatography: some possible approaches for improved separations, J. Chromatogr. A 1217 (2010) 858–880. T. Cecchi, P. Passamonti, Retention mechanism for ion-pair chromatography with chaotropic reagents, J. Chromatogr. A 1216 (2009) 1789–1797. T. Cecchi, P. Passamonti, Erratum to “Retention mechanism for ion-pair chromatography with chaotropic reagents” [J. Chromatogr. A 1216 (2009) 1789–1797], J. Chromatogr. A 1216 (2009) 5164. R. LoBrutto, A. Jones, Y.V. Kazakevich, Effect of counter-anion concentration on retention in high-performance liquid chromatography of protonated basic analytes, J. Chromatogr. A 913 (2001) 189–196. T. Cecchi, F. Pucciarelli, P. Passamonti, Extended thermodynamic approach to ion interaction chromatography: a mono- and bivariate strategy to model the influence of ionic strength, J. Sep. Sci. 27 (2004) 1323–1332. X. Wang, P.W. Carr, An unexpected observation concerning the effect of anionic additives on the retention behavior of basic drugs and peptides in reversedphase liquid chromatography, J. Chromatogr. A 1154 (2007) 165–173. T. Cecchi, F. Pucciarelli, P. Passamonti, Extended thermodynamic approach to ion interaction chromatography: influence of organic modifier concentration, Chromatographia 58 (2003) 411–419. Y.V. Kazakevich, R. LoBrutto, R. Vivilecchia, Reversed-phase high-performance liquid chromatography behavior of chaotropic counteranions, J. Chromatogr. A 1064 (2005) 9–18. I.L. Kazakevich, N.H. Snow, Adsorption behavior of hexafluorophosphate on selected bonded phases, J. Chromatogr. A 1119 (2006) 43–50. S. Keunchkarian, M. Reta, L. Romero, C. Castells, Effect of sample solvent on the chromatographic peak shape of analytes eluted under reversedphase liquid chromatographic conditions, J. Chromatogr. A 1119 (2006) 20–28. M. Saaid, B. Saad, A.S.M. Ali, M.I. Saleh, C. Basheer, H.K. Lee, In situ derivatization hollow fibre liquid-phase microextraction for the determination of biogenic amines in food samples, J. Chromatogr. A 1216 (2009) 5165–5170.