Nicotine, Cotinine, and Myosmine ... - American Chemical Society

Feb 22, 2012 - and Francis D'Souza*. ,§. †. Institute of Physical Chemistry, Polish Academy of Science, Kasprzaka 44/52, 01-224 Warsaw, Poland. ‡...
2 downloads 0 Views 1MB Size
Article pubs.acs.org/ac

Nicotine, Cotinine, and Myosmine Determination Using Polymer Films of Tailor-Designed Zinc Porphyrins as Recognition Units for Piezoelectric Microgravimetry Chemosensors Krzysztof Noworyta,*,† Wlodzimierz Kutner,†,‡ Channa A. Wijesinghe,§ Serge G. Srour,§ and Francis D’Souza*,§ †

Institute of Physical Chemistry, Polish Academy of Science, Kasprzaka 44/52, 01-224 Warsaw, Poland Faculty of Mathematics and Natural Sciences, School of Science, Cardinal Stefan Wyszynski University, Dewajtis 5, 01-815 Warsaw, Poland § Department of Chemistry, University of North Texas, 1155 Union Circle, No. 305070, Denton, Texas 76203-5017, United States ‡

S Supporting Information *

ABSTRACT: Two electropolymerizable zinc porphyrins with receptor sites tailor-designed for selective recognition of the nicotine, cotinine, or myosmine alkaloids were synthesized. These were 5-(2-phenoxyacetamide)10,15,20-tris(triphenylamino)porphyrinato zinc(II) 1 and 5-(2,5-phenylenebis(oxy)diacetamide)-10,15,20-tris(triphenylamino)porphyrinato zinc(II) 2 featuring one and two pendant amide side “pincers”, respectively, and three triphenylamine substituents at the meso positions of the porphyrin macrocycles capable of electrochemical polymerization. Thin polymer films of these porphyrins served for recognition and the piezoelectric microgravimetry (PM) for analytical signal transduction of a new chemical sensor devised for determination of these alkaloids. The films were deposited by potentiodynamic electropolymerization on the 10 MHz quartz resonators of the electrochemical quartz crystal microbalance (EQCM) without affecting the electronic structure of the porphyrin macrocycles. Under favorable flow injection analysis (FIA) conditions, the alkaloid analytes were determined at the concentration level of 0.1 mM with high sensitivity and selectivity. Affinity toward the analytes of the polymer of 2 was higher than that of 1 due to the higher binding ability offered by two pendant pincers of the former. Because of the selective receptors and PM applied under FIA conditions, the developed procedure offered an alternative to the time-consuming and relatively expensive high-performance liquid chromatography (HPLC), capillary electrophoresis (CE), and gas chromatography mass spectrometry (GC-MS) methods of detection and quantification of these alkaloids. Alkaloid 4 is the main (70%) human metabolite of 3.6 Alkaloid 5 is another tobacco genotoxic alkaloid, present in the tobacco leaves at lower concentrations. It is also known of damaging human DNA.7,8 In contrast to 3, the alkaloid 5 is also present in many common food products, fruits, and vegetables.9 Therefore, there is a need to monitor its concentration in fresh produce. Concentration of 4 and 5 in urine, saliva, as well as organs and body parts is used in medical research, along with the concentration of 3, to follow patient’s exposure to 3 or tobacco smoke.10−13 Over the past 2 decades, several analytical procedures have been developed for determination of 3−5. These include highperformance liquid chromatography (HPLC),14−17 gas chromatography (GC),13,18 capillary electrophoresis (CE),19−21 spectrophotometry22−24 [also coupled to solid-phase extraction (SPE)25 using molecularly imprinted polymers, MIPs] and radioimmunoassay.26,27 Moreover, enzyme-based biosensors

N

icotine, 3 (Scheme 1a), is the primary neuroactive alkaloid in tobacco leaves used to produce tobacco products.1 This alkaloid is easily absorbed in humans leading to variety of negative physiological effects. These include several illnesses, like cancer as well as pulmonary and cardiovascular disease.2,3 Despite of its high toxicity, 3 is said to have some therapeutic effects4 in neurodegenerative diseases, like those of Alzheimer’s and Parkinson’s.5 Therefore, it is important to monitor the presence and determine concentration of 3 both in the environment and in the human body. Apparently, determination of 3 in tobacco as well as in pharmaceutical formulations (transdermal patches and gums) is important in the tobacco and pharmaceutical industry. Besides, determination of 3 in living organisms, after their exposure to 3, is also of great interest in medicine and toxicology allowing evaluation of passive and active smoking effects on human health and correlating concentrations of 3 in the organism with the occurrence of certain serious diseases. For the latter, it is also important to monitor concentration of analogues of 3, such as cotinine 4 and myosmine 5 (Scheme 1a). © 2012 American Chemical Society

Received: August 24, 2011 Accepted: February 1, 2012 Published: February 22, 2012 2154

dx.doi.org/10.1021/ac2021344 | Anal. Chem. 2012, 84, 2154−2163

Analytical Chemistry

Article

Scheme 1. (a) Structural Formula of 5-(2-Phenoxyacetamide)-10,15,20-tris(triphenylamine)porphyrinato zinc(II) 1, 5-[2,2′(2,6-Phenylenebis(oxy)diacetamide)]-10,15,20-tris(triphenylamine)porphyrinato zinc(II) 2, Nicotine 3, Cotinine 4, and Myosmine 5; (b) Structural Formula proposed for the Complex of 1 and 3

have recently been devised for determination of 328−30 and 4.31,32 An approach employing piezoelectric microgravimetry (PM) signal transduction combined with the MIP recognition of 3 has also been tested, however, with limited selectivity.33 Furthermore, several electrochemical chemosensors for 3 and 4 were devised using amperometric34−42 or capacitive43 signal transduction methods. All of the above procedures suffer from certain disadvantages. Spectrophotometric methods typically require time-consuming isolation of the analyte from its complex matrix. The GC, CE, and HPLC procedures are time-consuming and require costly equipment and trained operators. Enzymatic procedures of 3 and 4 determination prevailingly experience low enzyme stability. Moreover, the required enzymes are relatively expensive. These deficiencies prompted the attempts of devising electrochemical sensors capable of detection of 3. Those sensors required neither expensive equipment nor trained operators being relatively easy to miniaturize. Unfortunately, 3 and 4 exhibit irreversible electrode kinetics, which strongly depends on the electrode material used.36 This behavior leads to low signal reproducibility. Attempts of

overcoming this problem included application of an electrode material, such as the boron-doped diamond,35 multiwall carbon nanotubes,37,39,40 or basal-plane pyrolytic graphite.36 The resulting sensors exhibited limit of detection (LOD) in the range of 1−10 μM as well as appreciable sensitivity and stability. Nevertheless, electrochemical sensors are hard to recommend for determination of 4 because its electrode kinetics is even more sluggish than that of 3. Further, these sensors typically lack selectivity. Only few examples of integration of 3-selective recognition films in electrochemical sensors are known.38,42 They use either direct38 or indirect42 amperometric detection. The former requires a complex, multistep preparation of a composite recognition film. The resulting sensor is prone to aging with its detection signal decreasing by 50% within 3 days. Indirect detection requires addition of an auxiliary reversible redox probe to the test solution. Unfortunately, this procedure is sensitive to the electrode-fouling impurities. In view of the present state of development of procedures for determination of 3, it seems desirable to devise a chemosensor of improved selectivity to this alkaloid, introduced by a 2155

dx.doi.org/10.1021/ac2021344 | Anal. Chem. 2012, 84, 2154−2163

Analytical Chemistry

Article

microbalance allowed for simultaneous batch measurements of changes of current, resonant frequency, and dynamic resistance of a 10 MHz, AT-cut plano−plano quartz resonator during potential scanning as the function of potential or time. The resonators featured 5 mm diameter Au-over-Ti film circular electrodes. The experiments were conducted using a dedicated three-electrode 15 mL electrochemical glass cell. However, for simultaneous potentiodynamic and PM polymer film deposition by electropolymerization, the EQCM 5710 holder was horizontally mounted in such a way that its resonator cavity formed a small-volume ∼0.2 mL electrochemical cell. This cell was filled with a 150 μL sample of the solution for electropolymerization, and both the spiraled Pt wire counter electrode and the Ag wire pseudoreference electrode were then immersed in this solution. Solutions used for deposition of the polymer films were not deaerated, while those used for simultaneous cyclic voltammetry (CV) and PM characterization of the deposited films were deaerated with the 15 min Ar purge before measurements. The FIA experiments used an experimental setup consisting of the NE-500 syringe pump driven by the WinPumpTerm software of New Era Pump Systems, Inc., the 9725 rotary sample injector of Rheodyne, and the EQCM 5610 electrochemical quartz crystal microbalance with a flow-through quartz crystal holder of IPC PAS. The inlet capillary-toresonator distance was set at 50 μm. The experiments were driven and data acquired using the EQCM 5710-S2 software of the same manufacturer. The polymer films were imaged with atomic force microscopy (AFM) in the tapping mode using the Multimode NS3D microscope with the 10 × 10 or 125 × 125 μm2 scanner of the Digital Instruments/Veeco metrology group. The operating either in the transmission or reflection mode UV2501 PC UV−vis spectrophotometer of Shimadzu served for the UV−vis spectroscopy measurements. The Raman spectra were recorded using the Almega Raman spectrometer of Nicolet equipped with a confocal microscope and a laser of the 532 nm excitation beam.

tailor-designed recognition unit. So far, little has been done toward this end.33,38,42,43 For example, selective to 3 MIP films were implemented as recognition units in chemosensors. These devices revealed appreciable sensitivity and the detection limit in the range of 0.2−5 μM.38,42 However, all these sensors suffer from disadvantages of inadequate stability, cumbersome preparation methods, or an indirect detection procedure adopted. In the present work, we designed and synthesized electrochemically polymerizable Zn porphyrin molecular receptors for alkaloids, 3, 4, and 5. These receptors feature three triphenylamine meso substituents and one or two phenoxyacetamide “pincers”. Structural formulas of the respective porphyrins viz. 5-(2-phenoxyacetamide)-10,15,20-tris(triphenylamino)porphyrinato zinc(II) 1, and 5-(2,5-phenylenebis-(oxy)-diacetamide)-10,15,20-tris(triphenylamino)porphyrinato zinc(II) 2, are shown in Scheme 1a. The triphenylamine substituents are capable of oxidative electrochemical polymerization resulting in thin polymer films.44 Each pincer facilitates analyte binding due to a “two-point” complexation. This complexation involves interaction of the aromatic nitrogen atom of the pyridine ring of 3 and the Zn atom of the porphyrin, on the one hand,45 and the nitrogen atom of the alicyclic ring of 3 and the side amide pincer on the other (Scheme 1b).46 Using films of the polymerized receptors, we fabricated PM chemosensors for selective determination of 3, 4, and 5 under flow injection analysis (FIA) conditions. The evaluated figures of merit unanimously suggest that the present method is a viable, competitive procedure for alkaloid detection and quantification compared to the currently used expensive and time-consuming methods.



MATERIALS AND METHODS Chemicals. (−)-Nicotine, (−)-cotinine, myosmine, and anhydrous 1,2-dichlorobenzene, ODCB, as well as reagent chemicals were purchased from Sigma-Aldrich. The HPLCgrade bulk solvents were from Fischer Scientific. Electrochemical grade solvents and supporting electrolyte salts were from Fluka. Analytical grade acetic acid (80%), absolute ethanol, 2-propanol, acetone, and toluene were from Chempur while analytical grade CH3COONa·3H2O was from POCh. Deionized water (18.2 MΩ cm) was Milli-Q purified. Argon was from Multax. Synthetic details of 1 and 2 (Scheme S-1) are given in the Supporting Information. Instrumentation and Procedures. Electrochemical experiments were conducted with the use of a conventional glass V-shaped three-electrode one-compartment 0.5 mL minicell with the Pt wire counter electrode and the Ag wire pseudoreference electrode. Either a 1 mm diameter Pt disk or an Au/Cr-coated glass slide was used as the working electrode. All electrodes were prepared according to the procedure described in the Supporting Information. The AUTOLAB electrochemical system of Eco Chemie, equipped with the PGSTAT13 potentiostat expansion card and driven by the GPES 4.9 software of the same manufacturer, was used for all voltammetry experiments. All polymer films were deposited by multicyclic potentiodynamic electropolymerization and monitored by PM using the EP-21 potentiostat of Elpan connected to the EQCM 5710 electrochemical quartz crystal microbalance (EQCM) of the Institute of Physical Chemistry (IPC PAS) under control of the EQCM 5710-S2 software of the same manufacturer. This



RESULTS AND DISCUSSION Deposition of the Poly1 and Poly2 Films. Devising of chemosensors for 3, 4, and 5 necessitates deposition on electrodes of recognition polymer films of Zn porphyrins featuring amide pincers. For that, the herein synthesized Zn porphyrins bearing three triphenylamine moieties (Scheme 1a) were used. These ternary amines eagerly polymerize under electrooxidative conditions.47 Therefore, stable porphyrin polymers were formed, indeed, via electrochemical oxidation of 1 and 2.44 That is, films of both the polymer of 1, poly1, and polymer of 2, poly2, were readily deposited on different electrodes from respective monomer solutions of 0.1 M (TBA)ClO4 in ODCB.44 Figure 1 shows the potential dependence of current, i, resonant frequency changes, Δf, and dynamic resistance changes, ΔR, simultaneously recorded during electropolymerization of 1 (Figure 1a) and 2 (Figure 1b), which resulted in respective polymer films. For 1, there are two small anodic peaks at 0.57 and 0.82 V, followed by a larger one at 1.14 V, in the initial positive potential excursion. In the subsequent negative scan of the first potential cycle, there is a broad cathodic peak at 0.75 V and a smaller one at 0.53 V. At the potential of onset of the anodic peak at 1.14 V, Δf rapidly decreases, which indicates the increase of the resonator mass 2156

dx.doi.org/10.1021/ac2021344 | Anal. Chem. 2012, 84, 2154−2163

Analytical Chemistry

Article

Figure 1. Simultaneously EQCM recorded potential dependence of (1 and 1′) current, (2 and 2′) resonance frequency change, and (3 and 3′) dynamic resistance change during deposition by multicyclic potentiodynamic electropolymerization, at 0.1 V s−1, of the polymer films from the solution of 0.1 M (TBA)ClO4 in ODCB (a) 0.48 mM 1 and (b) 0.70 mM 2.

amount of the entrapped species in poly1. That is, the molecular weight of the deposited species was determined as 1404.6 g mol−1, whereas it was calculated as 1254.9 g mol−1 for the porphyrin monomer. The difference of 149.7 g mol−1 may be ascribed to ingress in the film of 1 mol of ClO4− and ∼0.34 mol of the ODCB solvent accompanying electropolymerization of 1 mol of the monomer. Moreover, a concomitant ingress of TBA+ most plausibly contributes to the mass increase during electroreduction of the poly1 film at ∼0.7 V after reversal of the potential scan direction. Calculation of the Mw/z value for the species involved in this electroreduction resulted in Mw/z = 375 g mol−1. This value does not correspond exactly to any possible species present in the film or the solution, which points to a rather complex process of the ion and solvent exchange. Qualitatively, simultaneously recorded potentiodynamic and PM curves for electropolymerization of the monomer with two amide pincers 2 (Figure 1b) are similar to those for the monomer with only one pincers 1 (Figure 1a). Besides, fivepotential-cycle deposition of poly2 was accompanied by the ΔR increase over 4 times higher (>25 Ω) than that of poly1 indicating higher rigidity of the latter film. This behavior confirms that the poly1 film is more compact. Notably, the ΔR changes for poly2 are still small compared to the change of ∼200 Ω characteristic for deposition of viscous films.50 Interestingly, the Mw/z value determined for poly2 is merely equal to 201.5 g mol−1 being significantly lower than that of 442.6 g mol−1 calculated. This large discrepancy may indicate higher solubility of the electrooxidation products of poly2 than those of poly1. That way, some products formed during electrooxidation of 2 may remain in solution while participating in the electron exchange, i.e., contributing to the current flow.

corresponding to deposition of the polymer film. Oxidative electropolymerization of both monomers positively charges the resulting polymer film. The ClO4− ion of the supporting electrolyte enters the film for neutralization of this positive charge. Interestingly, the broad cathodic peak at 0.75 V is also accompanied by some frequency decrease. This is surprising because a frequency increase due to the counterion (anion) removal from the film during its reduction would rather be expected. Therefore, this decrease can be attributed to the TBA+ cation ingress to the film for maintaining its electroneutrality. Apparently, ClO4− remains trapped in the film. In subsequent potential cycles, a broad anodic peak at ∼0.94 V appears and then grows accompanied by the current increase of the cathodic peak at 0.75 V. Moreover, current of the anodic peak at 0.57 V and that of the cathodic peak at 0.53 V increase, but to a lesser extent. This behavior is typical for deposition of a conducting polymer film.48 This inference is additionally confirmed by the concomitant Δf decrease. The dynamic resistance increase accompanied deposition of the poly1 film. However, the overall ΔR change was rather small (