
	Version PDF

Biomimetic study of the Ca ²⁺-Mg ²⁺ and K ⁺-Li ⁺ antagonism

Magnesium Research. Volume 22, Numéro 1, 10-20, March 2009, Original article

DOI : 10.1684/mrh.2009.0159

Summary

Auteur(s) : Beata Paczosa-Bator, Milena Stepien, Magdalena Maj-Zurawska, Andrzej Lewenstam , Faculty of Material Science and Ceramics, AGH-University of Science and Technology, Cracow, Poland, Center for Process Analytical Chemistry and Sensor Technology (ProSens), c/o Process Chemistry Center of Excellence, Åbo Akademi University, FIN-20500 Turku-Åbo, Finland, Department of Chemistry, University of Warsaw, Warsaw, Poland.

Illustrations

ARTICLE

Auteur(s) : Beata Paczosa-Bator¹, Milena Stepien^1,2, Magdalena Maj-Zurawska³, Andrzej Lewenstam^1,2

¹Faculty of Material Science and Ceramics, AGH-University of Science and Technology, Cracow, Poland
²Center for Process Analytical Chemistry and Sensor Technology (ProSens), c/o Process Chemistry Center of Excellence, Åbo Akademi University, FIN-20500 Turku-Åbo, Finland
³Department of Chemistry, University of Warsaw, Warsaw, Poland

Membrane potential is rudimentary in biological life and sensor technology. Recently, we proposed taking advantage of processes common for those areas which lead to membrane potential formation [1-4]. Namely, this approach employed the redistribution of ions in the vicinity of the membrane and coupled dynamic potential change. From the biological point of view, we are interested in the mechanism of voltage-dependent channel block and related ionic antagonism, such as divalent calcium-magnesium ion competition in the case of NMDA [4-7], or monovalent ion effects such as potassium-sodium/lithium/TEA in the case of potassium and sodium channels [8]. From the electrochemical perspective it means that we are concerned with the time-dependent, characteristics and those resulting from a competitive ion-exchange of ion-selective sensors such as magnesium [9] or any other ion-selective electrode [10, 11].

Our deliberate biomimetic concept allows for studying of a competitive (“antagonistic”) ion exchange and its coupling to membrane potential formation on biologically active sites. The novelty is in using conductive polymers as the membranes in which the sites are dispersed.

In this report, recent results on calcium-magnesium antagonism, and for the first time, the data on potassium-lithium/sodium are shown.

Experimental

Reagents

Pyrrole, N-methylpyrrole (Merck) was purified by double distillation under argon and then stored under argon at low temperature and protected from light. The L-asparagine, L-glutamine, adenosine 5′- triphosphate sodium salts were obtained from Fluka. The other reagents were also purchased from Fluka. All the chemicals used were of analytical grade. Doubly distilled and freshly deionized ELGA water (resistivity 18.2 MΩ cm) was used throughout the work.

Apparatus

The potentiostatic synthesis of polymer films on GC and ITO electrodes was carried out using an Autolab general Purpose System (AUT20.Fra2-Autolab, Eco Chemie, B.V., Utrecht, The Netherlands) connected to a conventional, three-electrode cell. The working electrode was a glassy carbon (GC) disk with an area of 0.07 cm² or conducting glass pieces with an area of about 1 cm² (ITO, Lohja Electronics, Lohja, Finland, used for the EDAX, XPS and LA-ICP-MS experiments). The reference electrode was an Ag/AgCl/3M KCl electrode connected to the cell via a bridge filled with supporting electrolyte solution, and a glassy carbon (GC) rod was used as the auxiliary electrode. The solutions used for polymerization contained 0.1 M monomer of polymer and an electrolyte that provided the doping ion. Electropolymerization was performed in solutions saturated with argon at room temperature.

The chrono-amperometric measurements were carried out also using an Autolab general Purpose System (AUT20.Fra2-Autolab. Eco Chemie. B.V.) connected to a conventional, three-electrode cell where a glassy carbon electrode prepared with PMPy-ATP-Na polymer was used as the working electrode, and as the reference electrode - an Ag/AgCl/3M KCl electrode, and the auxiliary electrode - a glassy carbon electrode.

The potentiometric measurements were made with a homemade multi-channel set-up. The reference electrode was an Ag/AgCl/3M KCl electrode. All experiments were performed at room temperature.

The elemental analysis of the polypyrrole films was performed using X-ray photoelectron spectroscopy (XPS). The XPS analysis was performed with a Physical Electronics Quantum 2000 XPS-spectrometer equipped with a monochromatized Al-X-ray source to assess qualitatively the influence of soaking on the composition of these films. The size of the analyzed area was 100 μm in diameter and the analysis depth was about 2-5 nm depending on the investigated element.

The Energy Dispersive Analysis of X-ray (EDAX) measurements were performed using a Scanning Electron Microscope, SEM model LEO 1530 from LEO Electron Microscopy Ltd, which was connected to an Image and X-ray analysis system – model Vantage from ThermoNoran.

The LA-ICP-MS measurements were performed using a model 6100 Elan DRC Plus of ICP-MS from Perkin Elmer SCIEX (Waltham, USA) and UP-213 of Laser Ablation from “New wave Research” Merchantek Products (Fremont, USA). The LA-ICP-MS measurements for the PMPy-ATP sample were investigated by vaporization in laser plasma with a spot size of 15 μm and a scan speed of 10 μm per second. The repetition rate was 10 Hz and output 40% in every case during the measurement.

Procedures of CP-BL-Me electrode preparation

Poly(N-methylpyrrole) and poly(pyrrole) deposition

The electrodeposition of the poly(N-methylpyrrole) and poly(pyrrole) films was carried out potentiostatically from a solution that contained 0.1 M ATP as dopant and 0.1 M N-methylpyrrole or pyrrole. Potentiostatic growth was achieved by holding a potential at + 0.8 V vs Ag/AgCl/3M KCl for different times in order to obtain different amounts of charge density (510-750 mC cm^-2).

The potentiostatic method was also used to grow poly(pyrrole) films doped with amino acids. PPy-Asn(Gln) films were grown on the working electrode at a potential of +1.0 V vs Ag/AgCl/3M KCl and charge density of 240 mC cm^-2. Besides 0.1 M pyrrole, the solution contained 0.1 M Gln (or 0.1 M Asn).

The process of making cation-sensitive CP-BL films

After synthesis, the polymer membranes were washed with deionized water and then the electrodes were soaked and stored in a mixture of 0.1 M MeCl_n and Me(OH)_n (pH of about 9-10) where Me was a main ion. As a rule, a cationic response with a linear range within the K⁺, Na⁺, Li⁺ activities from 10^-1 M to 10^-4 M or Ca²⁺, Mg²⁺ activities from 10^-1 M to 10^-5 M and with a close-to-Nernstian slope was observed for the CP-BL films after 1 week of soaking. Only conditioning in the alkaline solution of main ions was effective. The cation complexes with BL were formed after PPy-BL film deprotonation in alkaline solutions (protons were substituted with other cations) [1, 2].

Results and discussion

The results described below reflect a methodological pattern used to explore the biomimetic properties of CP-BL membranes and the mechanism of their membrane potential formation. In the first stage, the films were electrosynthesized and doped with respective biological ligands. Next, a potentiometric sensitivity for the main ion of interest was induced. After this, the influence of other ions, i.e. the selectivity of interest was studied.

Only properly behaving membranes, characterized with a time-independent response for both the main and other than main ions, were used in the studies of a dynamic (time-dependent) response for changes to the electrochemical membrane potential driving forces, i.e. changes in bulk ion concentrations in the bathing solutions and influence of the external electric potential [1-4].

Potentiometric response of CP-BL membranes

The potentiometric response of all membranes was studied in a solution of chloride salt of main ions in the concentration range from 10^-6 M to 10^-1 M.

The first calibration was performed before conditioning and then after a different time of soaking. Exemplary calibration and dynamic response curves are shown in figure 1: for a membrane sensitive towards monovalent potassium cation PMPy-ATP-K (figure 1A) and a membrane sensitive to divalent magnesium ion PMPy-ATP-Mg (figure 1B). Figure 1 illustrates a need for conditioning of the films to achieve a close-to-Nernstian potentiometric response, i.e. the slope close to 59.2 mV/pNa, (pLi or pK) and 29.6 mV/pMg or pCa in the range of 10^-1 M to 10^-4 M for Na⁺, Li⁺, K⁺ and 10^-1 M to 10^-5 M for Ca²⁺ and Mg²⁺ activities. It is shown that, after one week of conditioning in the alkaline solution of the main ion, both potentiometric sensitivity and a short potential response time (less than 10 s) were obtained. This applies to all the membranes studied.

Potentiometric selectivity of CP-BL membranes

After inducing a proper sensitivity, the selectivity of the films to other ions was studied by a separate solution method (SSM) [12]. This method allows determining a potentiometric selectivity coefficient () which is a quantitative measure of the ability of the ion-sensitive film to discriminate the main ion from the other ions. The coefficient is dictated by the thermodynamic properties of ions in the phases submitted for ion exchange and should correlate with a corresponding ion-exchange equilibrium constant at the solution-membrane interface [11].

The values of the potentiometric selectivity coefficients obtained were in excellent agreement with the stability constants of BL complexes with ions, as shown in figure 2 for two chosen membranes, PMPy-ATP-Na and PMPy-ATP-Mg. The potentiometric selectivity coefficients follow the same sequence as the stability constants of ATP complexes with cations.

As expected, in the case of membranes sensitive towards monovalent cations, strong interferences of divalent cations were observed. By subsequently soaking the CP-BL-Na(Li)(K) films in the solution of Ca²⁺, or Mg²⁺, forming a stronger complex with BL, it was possible to obtain calcium- or magnesium-sensitive films by competitive ion exchange.

The calcium- and magnesium-sensitive CP-BL films were insensitive towards sodium, potassium or lithium. Importantly, the selectivity coefficient values for the membranes sensitive to divalent metal ions and monovalent ions were close to = 1. This manifestation of similar thermodynamic properties of ions (in the groups) studied, makes any dissimilarity of the response attributable to the kinetic properties of these ions in the membrane systems studied.

Chemical analysis of CP-BL membranes

The elemental analysis of poly(N-methylpyrrole) and polypyrrole films was performed using three different methods, X-ray photoelectron spectroscopy for membranes doped with amino acids, energy dispersive analysis of X-ray for CP-ATP films sensitive to divalent ions, and laser ablation inductively coupled plasma mass spectrometry for CP-ATP films sensitive to monovalent ions. For the chemical and morphological analysis, two kinds of samples were prepared, namely: CP-BL without soaking and CP-BL after 2 weeks’ soaking in the solution of main ions.

The presence of the phosphorus signal in the case of PPy-ATP and PMPy-ATP films in the EDAX and LA-ICP-MS spectra, as shown in figure 3 and 4, proves that counter-ions dope the films formed during electrodeposition.

The EDAX analysis of PPy-ATP films showed that, after the conditioning process, calcium or magnesium peaks had also appeared on the spectrum. The LA-ICP-MS measurements for the PMPy-ATP sample sensitive to monovalent ions proved that, after the conditioning process, ions that were present in the solution also appeared in the membranes.

Changing the composition of the conditioning solution from sodium to lithium (or potassium) gave membranes sensitive to these ions and the signal from the new ions appeared on the spectrum as shown in figure 3.

Depending on the composition of the solution used during conditioning CP-ATP membranes, different potentiometric sensitivity was observed. After conditioning in alkaline lithium solution, the potentiometric sensitivity to these ions had been induced and the LA-ICP-MS spectrum showed a signal from Li (which was not observed before the conditioning process) as presented in figure 3A. The same behaviour was observed for calcium, magnesium and potassium ions. The chemical composition of the PPy-Asn(Gln) films was investigated by XPS. Figure 3B (curve i) shows on overview spectrum of a freshly deposited PPy-Asn sample on which the following peaks were observed at the given binding energies: O1s at 532.8 eV, N1s at 400 eV and C1s at 284.8 eV. The peaks in the region of 88.8 eV binding energy, as shown in figure 3B (curve ii), are identified as an Mg2s signal, which was observed only after conditioning in alkaline magnesium solution. Similarly, the XPS analysis taken after soaking of the PPy-Asn(Gln) films in an alkaline calcium solution proved to peak Ca2p3 at 347.2 eV (figure 3B, curve iii).

In the case of sodium-sensitive membranes, the signal from Na was observed after soaking with EDAX, providing good evidence for introducing sodium to the membrane during soaking, as shown in figure 4.

The chemical analysis provides the evidence that the sensitivity for a particular ion is always associated with the presence of this ion in the membrane.

Ion competition during open-circuit response

In spite of similar sensitivity and selectivity of both groups of polymer films (namely CP-BL-Ca(Mg) or CP-BL-K(Li)(Na)) towards divalent (calcium and magnesium) ions or monovalent (sodium, potassium and lithium) ions, the transitory potential provoked by the changes in bulk concentrations of these groups of ions was strikingly different.

The representative plots for the measurements made for monovalent ion-sensitive membranes are shown in figure 5A-C, and the membranes sensitive towards divalent ions are shown in figure 5D.

As can be seen from figure 5, potential-time (E-t) response strongly depends on the kind of ion that was involved in the competitive ion-exchange equilibration process. Lithium ion-exchange with sodium-rich CP-ATP-Na membranes results in a monotonic response (figure 5A), while if potassium ions are engaged in the ion exchange, instead of lithium, a non-monotonic (overshoot-type) response is observed (figure 5B). If a sodium-rich membrane is converted to a potassium-rich one, then both the lithium and sodium responses, as expected, are monotonic (figure 5C). A similar pattern in observed for the CP-ATP-Mg membrane (figure 5D). Changes in bulk concentrations of magnesium ion are always associated with monotonic potential changes, while changes in the concentration of calcium are associated with overshoot-type responses. These characteristic differences between potassium, sodium and lithium, as well as magnesium and calcium can be called “ionic antagonism”. Interestingly, and most probably not coincidentally, the same pairs of ions, i.e. Ca²⁺-Mg²⁺, Na⁺/Li⁺-K⁺, are indeed considered as antagonistic in real biological membrane systems, and specialized voltage and/or ligand-gated ion channels engage these ions, e.g. NMDA.

As stated above, when discussing the selectivity of the membranes used in our study, thermodynamically, the differences do not allow prediction of a striking difference in membrane responses. However, this fact lends credence to kinetic aspects in signal formation. A different rate in the transport of ions to (the source of ion exchange) the interface of the solution and membrane could be a primary source of the “antagonism” observed [4]. The hypothesis is that faster ions, (Ca and K characterized by the mobility 6.17 and 7.62 10^-8 m² s^-1 V^-1, respectively [13]) coming from the solution bulk and substituting via ion exchange slower ions from the film (Mg, Na and Li, characterized by the mobility 5.49, 5.19 and 4.01 10^-8 m² s^-1 V^-1, respectively [13]) sites, allow for local accumulation of slower ions in the solution bathing the film. And vice-versa, if slower ions come from the solution to the film containing faster ions, a deficit of this ion can be observed. This mechanism is schematically illustrated for Mg²⁺ - Ca²⁺ pair in figure 6

Figure 6 clearly shows the limited value of our approach when confronted with realistic biological cell membranes, which contain a lipid environment and complicated 3D protein-channel structures. Undoubtedly, our CP membrane serves only one aspect of many relevant to real potential gated channels. It allows addressing and controlling electrical potential on the site, and observing resulting ion redistributions and fluxes.

Ion competition during amperometric response

As shown in figure 5, the changes in bulk concentration of ions result in characteristic changes of potential vs. time, and are attributed, as shown in figure 6, to local redistributions of ions in the vicinity of the membrane-solution interface. It is of great interest to convert the problem and ask whether one could observe any manifestation of this process in an experiment where the membrane redistribution is provoked by an external electric potential impulse. In this respect, in the absence of a method for direct visualization of the ionic concentration changes in the vicinity of a membrane interface, a chrono-amperometric method was used. In this method, the external potential (+ 5/-5 and + 10/-10 mV from the open-circuit potential) is applied to provoke ion fluxes to and from the membrane, and the fluxes are characterized by (ionic) current over time. It would be expected that faster ions (potassium or calcium), in response to the potential applied, would produce currents that come to a base-line faster than ions of lower mobility (sodium/lithium or magnesium) do. The current response of the PMPy-ATP-Na electrode with time was measured in solutions of chloride salts of sodium, lithium and potassium with concentrations equal to 0.0001M under different values of potential. In figure 7, the current-time (I-t) responses for a sodium sensitive PMPy-ATP membrane are shown. The plots indeed prove the interrelation between the size of ions (resp. mobility of ions) and the I-t signal measured.

For the faster potassium ion (resp. calcium ion), after bigger initial cathodic or anodic current values, a fast current drop was observed, while for slower sodium and lithium (resp. magnesium) ions, the initial currents were smaller and followed by a slower drop in current. This amperometric behaviour can be attributed to different mobility “antagonistic” ions. It can be concluded that both in the open circuit and under potential, the electrochemical behaviour of the biomimetic membrane is dictated by different transport numbers of the ions. The kinetic difference is thus a prerequisite of the “ionic antagonism”.

Theoretical interpretation and implications

The change in membrane potential over time provoked by bulk concentration changes is attributed to local redistributions of ions at the membrane-solution interface and transport to and from this interface. If the membrane potential is changed by an external source, resulting ionic fluxes could be observed and the ionic currents depend on the physicochemical properties of the ions engaged. These observations allow the development of a general interpretation of E-t and I-t for this biomimetic system using the Nernst-Planck-Poisson model (NPP) [11, 14, 15] or a simpler diffusion-layer model (DLM) [2, 4, 11, 16]). Both these theoretical models support a fundamental idea of the biomimetic membrane concept presented and show that, when a biologically active site is allowed for a competitive ion-exchange, the extent of the competition is regulated by the electric potential “applied” to this site and the transport of ions to and from the site. Different E-t and I-t patterns have to be observed for faster ions in comparison to slower, which is known as “ionic antagonism”. Here is a key for applying our biomimetic approach for responses that follow electric potential incoming to the site (e.g. as excitation potential) and to observe ionic distributions on the sites accumulated in 3D (i.e. three dimensional) structures, e.g. nanotubes or 3D architectures in a form of artificial channels, and/or on 2D nanofilms. Our study shows that a local e.g. magnesium ion concentration increase is be expected when positive vs. equilibrium (rest) potential is applied. In other words, it means that magnesium ions leave the coordinating sites and the smaller calcium ion is admitted. This is exactly what happens at the neck of a magnesium blocked NMDA channel where this ion is attracted by Asn and Gln. When excited by action potential, the channel gets unblocked and allows the faster calcium ions to pass through [5, 17]. Obviously, a deficiency of magnesium in external compartments can facilitate calcium influx and modulation of intracellular calcium concentration. Interestingly, this mechanism and the magnesium-calcium antagonism in relation to the NMDA receptor channel were recently considered as one possible reason for inflammatory responses and the metabolic syndrome [18, 19]. The importance of the effects of Ca²⁺, Mg²⁺, ATP and other phosphorylated species on cardiac action potentials was recently emphasized as well [20]. A similar case of a competitive ion mechanism could be in the interaction of the exogenous lithium ion with negatively-charged inositol phospholipids, which is considered to be relevant in the treatment of bipolar disorders [21, 22].

As presented here, potential-dependent local concentration redistribution of ions at the membrane binding sites undoubtedly adds a new dimension in the interpretation of the above effects.We address these issues in our present research.

Conclusion

The biomimetic membrane methodology allows visualization and inspection of competitive and voltage-dependent ion exchange on biologically active sites. By using selected biological membranes and their channels (e.g. ATP, Asn, Gln), relevant to real ionic sites, it is possible to access the ionic redistribution in the sites in function of the bulk concentration of ions, external potential and time. In other words, the concept presented provides a tool to study the role of ions and the influence of ion supplementation, ion deficiencies and ion antagonism on membrane potential. It is, as well, a tool for investigating the nature of bias between voltage effects (long-term potentiation (LTP), cardiac arrhythmias, low-frequency signals in brain) on local (at/on site) ionic or transmembrane redistributions.

Acknowledgments

This work is supported by the Polish Ministry of Science and Higher Education, Grant No. DWM/232/MATERA/2006 and is a part of MASTRA MATERA ERA-NET project funded under the 6^th FP EU. Financial support from the Polish State Committee for Scientific Research (KBN), project number R15 005 03, is gratefully acknowledged.

References

1 Migdalski J, Błaż T, Paczosa B, Lewenstam A. Magnesium and calcium-dependent membrane potential of poly(pyrrole) films doped with adenosine triphosphate. Microchim Acta 2003; 143: 177-85.

2 Paczosa-Bator B, Migdalski J, Lewenstam A. Conducting polymer films as model biological membranes. Electrochemical and ion-exchange properties of poly(pyrrole) films doped with asparagine and glutamine. Electrochim Acta 2006; 51: 2173-81.

3 Paczosa-Bator B, Peltonen J, Bobacka J, Lewenstam A. Influence of morphology and topography on potentiometric response of magnesium and calcium sensitive PEDOT films doped with adenosine triphosphate (ATP). Anal Chim Acta 2006; 555: 118-27.

4 Paczosa-Bator B, Blaz T, Migdalski J, Lewenstam A. Conducting polymers in modelling transient potential of biological membranes. Bioelectrochemistry 2007; 71: 66-74.

5 Nowak L, Bregestovski P, Ascher P, Herbet A, Prochiantz A. Magnesium gates glutamate-activated channels in mouse central neurons. Nature 1984; 307: 462-5.

6 McBain CJ, Mayer ML. N-methyl-D-aspartic acid receptor structure and function. Physiol Rev 1994; 74: 723-60.

7 Saris NE, Mervaala E, Karppanen H, Khawaja JA, Lewenstam A. Magnesium - an update on physiological, pathological and analytical aspects. Clin Chim Acta 2000; 294: 1-26.

8 Hille B. Ionic Channels of excitable membanes. Sinauer Associates Inc., 1992.

9 Maj-Zurawska M, Lewenstam A. Fully automated potentiometric determination of ionized magnesium in blood serum. Anal Chim Acta 1990; 236: 331-5.

10 Lewenstam A. Design and pitfalls of ion-selective electrodes. Scand J Clin Lab Invest 1994; 54: 11-20.

11 Bobacka J, Ivaska A, Lewenstam A. Potentiometric ion sensors. Chem Rev 2008; 108: 329-51.

12 Umezawa Y, Bühlmann P, Umezawa K, Tohda K, Amemiya S. Potentiometric Selectivity Coefficients of Ion-Selective Electrodes. Part I. Inorganic Cations. Pure Appl Chem 2000; 72: 1851-2082.

13 Fraústoda Silva JJR, Williams RJP. The Biological Chemistry of the Elements. Oxford: University Press, 2001.

14 Sokalski T, Lewenstam A. Application of Nernst-Planck and Poisson equations for interpretation of liquid-junction and membrane potentials in real-time and space domains. Electrochem Comm 2001; 3: 107-12.

15 Sokalski T, Lingenfelter P, Lewenstam A. Numerical solution of the coupled Nernst-Planck and Poisson equations for liquid-junction and ion-selective membrane potentials. J Phys Chem B 2003; 117: 2443-52.

16 Lewenstam A, Hulanicki A, Sokalski T. Response mechanism of solid-state ion-selective electrodes in the presence of interfering ions. Anal Chem 1987; 59: 1539-44.

17 Vargas-Caballero M, Robinson HPC. Fast and slow voltage-dependent dynamics of magnesium block in the NMDA receptor: The asymmetric trapping block model. J Neurosci 2004; 24: 6171-80.

18 Rayssiguier Y, Gueux E, Nowacki W, Rock E, Mazur A. High fructose consumption combined with low dietary magnesium intake may increase the incidence of the metabolic syndrome by inducing inflammation. Magnes Res 2006; 19: 237-43.

19 Mazur A, Maier JAM, Rock E, Gueux E, Nowacki W, Rayssiguier Y. Magnesium and the infammatory response: Potential physiopathological implications. Archiv Biochem Biophys 2007; 458: 48-56.

20 Michailova A, McCulloch AD. Effects of Mg²⁺, pH and PCr on cardiac excitation-metabolic coupling. Magnes Res 2008; 21: 16-28.

21 Atack JR, Broughton HB, Pollack SJ. Inositol monophosphatase – a putative target for Li⁺ in the treatment of bipolar disorder. Trends Neurosci 1995; 18: 343-9.

22 Gibbons CE, Maldonado-Pérez D, Shah AN, Riccardi D, Ward DT. Calcium-sensing receptor antagonism or lithium treatment ameliorates aminoglycoside-induced cell death in renal epithelial cells. Biochim Biophys Acta 2008; 1782: 188-95.

Biomimetic study of the Ca 2+-Mg 2+ and K +-Li + antagonism