Fluorescence spectra decomposition by asymmetric functions: Laurdan spectrum revisited Mihaela Bacalum a,b,c , Bogdan Zorila ˘ a , Mihai Radu a,d,⇑ a Department of Life and Environmental Physics, Horia Hulubei National Institute of Physics and Nuclear Engineering, 077125 Magurele, Romania b Biomedical Research Institute, Hasselt University, B-3590 Diepenbeek, Belgium c Department of Electricity, Solid State and Biophysics, Faculty of Physics, University of Bucharest, 077125 Magurele–Ilfov, Romania d Department of Neurological, Neuropsychological, Morphological, and Movement Sciences, Section of Anatomy and Histology, University of Verona, Verona 37134, Italy article info Article history: Received 8 April 2013 Received in revised form 17 May 2013 Accepted 22 May 2013 Available online 6 June 2013 Keywords: Complex emission spectrum Log-normal function Large unilamellar vesicles Laurdan Generalized polarization Bilayer hydration abstract Due to their asymmetric nature, complex fluorescence spectra of molecules can be analyzed much better by log-normal distributions than by Gaussian ones. So far, the log-normal function has been used for deconvolution of emission spectra of different fluorescent molecules, such as Tryptophan and Prodan, but to our knowledge it is far less used for Laurdan (2-dimethylamino-6-lauroylnaphthalene). In this arti- cle, we present the decomposition of Laurdan emission spectra in large unilamellar vesicles using a pro- cedure that relies on the log-normal asymmetric function. The procedure was calibrated using Laurdan spectra in homogeneous solutions of various solvents. Comparing our results with the ones obtained from a Gaussian fit, we show that (i) the position of the elementary peaks (440 and 490 nm) is preserved in a large range of temperatures that include the main phase transition of lipid bilayer and (ii) the bilayer hydration, as reported by Laurdan, increases approximately 8 times from the gel phase to the liquid crys- talline one, a result that fits with other reports, providing a more realistic description. In addition, we pro- pose a new parameter to globally evaluate Laurdan emission spectra with the prospect of acquiring a larger range of values than the classical ‘‘generalized polarization’’. Ó 2013 Elsevier Inc. All rights reserved. The complex emission spectrum decomposition is a very impor- tant issue in analytical fluorescence spectroscopy. The structure of a spectrum is produced by the presence of either several different fluorophores or different excited states of the same fluorophore in the investigated solution. Most fluorophores produce asymmetric emission spectra even in a homogeneous solution where only one emitting state is presumed to be present. The main reason is given by the differences in vibronic level distribution between ground and excited states [1]. The solution for analysis of such complex spectra is the decomposition in elementary emission peaks. This is not a trivial task, with a valid mathematical model for an analyt- ical description of the elementary peak shape being needed. Very often, the first choice is oriented to the popular Gaussian and Lorentzian functions in spite of their symmetry. Few alternatives for complex spectrum decomposition have been proposed in the lit- erature, from which one requires the use of asymmetric functions. Siano and Metzler proposed a four-parameter log-normal (LN) 1 function to describe band shape in absorption spectra [2]. Burstein and coworkers developed a procedure based on the mirror symmetric form of LN to analyze the emission spectra of a larger class of fluoro- phores [3]. In particular, they developed and validated a robust algo- rithm for analysis of the tryptophan emission spectrum in proteins [4]. The procedure was extended by the same group for Prodan and Acrylodan [5]. Kalauzi and coworkers presented a comparison among their own asymmetric model and the LN and Gaussian models, con- cluding that LN and their new asymmetric function provide very sim- ilar results with a better fit than Gaussian-based models [6]. LN has been successfully used as the base of an algorithm to analyze the spectra of a class of fluorophores presenting two coupled ground and excited states (3-hydroxychromone derivatives) [7]. Another sensitive issue of the complex spectra decomposition is the meaning of the parameters resulting from a fit procedure, par- ticularly the parameters related to the shape function (e.g., full width at half-maximum, FWHM). If only v 2 is considered as a un- ique criterion to establish the optimal parameter values, it is pos- sible that some of these values are far from a physical meaning. In the particular case of complex spectrum produced by several pop- ulations of the same fluorophore, Burstein and coworkers proposed a procedure to calibrate the fitting algorithm, imposing some empirically derived constraints on the LN shape parameters [3]. Laurdan (2-dimethylamino-6-lauroylnaphthalene) is a deriva- tive of Prodan (6-propionyl-2-(dimethy-1-amino)-naphthalene) 0003-2697/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ab.2013.05.031 ⇑ Corresponding author at: Department of Neurological, Neuropsychological, Morphological, and Movement Sciences, Section of Anatomy and Histology, University of Verona, Verona 37134, Italy. Fax: +39 045 8027163. E-mail address: mihai.radu@univr.it (M. Radu). 1 Abbreviations used: LN, log-normal; FWHM, full width at half-maximum; GP, generalized polarization; LUV, large unilamellar vesicle; PBS, phosphate-buffered saline; DMPC, 1,2-dimyristoyl-sn-glycero-3-phosphocholine; MLV, multilamellar vesicle. Analytical Biochemistry 440 (2013) 123–129 Contents lists available at SciVerse ScienceDirect Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio