Influence of response factors on determining equilibrium association constants of non-covalent complexes by electrospray ionization mass spectrometry

Investigation of the first shot phenomenon in MALDI mass spectrometry of protein complexes

Matrix-assisted laser desorption/ionization-mass spectrometry (MALDI-MS) of noncovalent protein complexes is difficult, due to the disruptive nature of processes occurring during MALDI sample preparation and ion formation. Sometimes the observation of intact noncovalent protein complexes with MALDI is only possible if data are acquired from the first laser shot fired at a fresh sample; this is called the 'first shot phenomenon'. To study the origin of the first shot phenomenon, we used MALDI-MS and confocal laser scanning microscopy (CLSM) to examine typical MALDI sample preparations with embedded protein complexes, labeled with fluorophores. Fluorescence energy transfer techniques allowed the differentiation between intact and dissociated protein complexes with CLSM. In cases where a first shot behavior was observed by MALDI-MS, it was found to be accompanied by localization of protein complexes at the exterior of the sample crystals. Segregation of the large protein complexes to the exterior and dissociation of the complexes in the crystal interior during sample crystallization can rationalize this observation.

Quantitative determination of lysozyme-ligand binding in the solution and gas phases by electrospray ionisation mass spectrometry

Affinity constants for the binding of a range of substrate and non-substrate oligosaccharides to hen egg white lysozyme were determined by direct observation of the protein.ligand complexes using electrospray ionisation mass spectrometry (ESI-MS) with a chip-based nano-ESI source. The values obtained for a series of beta-1,4-N-acetylglucosamine oligomers (NAGn) were found to be in good agreement with those determined by fluorescence measurement. Oligomers of alpha-1,4-glucose (Glcn), which are believed to bind to lysozyme non-specifically, exhibited a 10(6)- to 10(8)-fold lower affinity for the enzyme. Lysozyme.NAGn complexes displayed an increase in Ka from n=2 to n=4, but then reached a plateau. In contrast non-specific lysozyme.Glcn complexes showed no such trend. Determination of gas-phase complex stability was achieved by quantitative collision-induced dissociation (CID) and infrared multiphoton dissociation (IRMPD) measurements. The collision energy (Ec50) or laser power (IRMPD50) required to dissociate precursor ions to 50% of their original intensity was determined for lysozyme.NAGn and Glcn complexes using the [M+8H]8+ charge state. An excellent correlation between trends in Ka and gas-phase stability was seen for NAGn oligomers bound to lysozyme, whereas no such relationship was observed with the non-specific, weaker lysozyme.Glcn complexes. These results illustrate that ESI-MS can be used to quantify the interactions between lysozyme and oligosaccharides in both the solution and gas phase and that measurement of gas-phase complex stability by CID or IRMPD can provide information about specific solution binding events.

DNA stability in the gas versus solution phases: a systematic study of thirty-one duplexes with varying length, sequence, and charge level

We report herein a systematic mass spectrometric study of a series of thirty-one non-self-complementary, matched, DNA duplexes ranging in size from 5- to 12-mers. The purpose of this work is threefold: (1) to establish the viability of using mass spectrometry as a tool for examining solution phase stabilities of DNA duplexes; (2) to systematically assess gas-phase stabilities of DNA duplexes; and (3) to compare gas and solution phase stabilities in an effort to understand how media affects DNA stability. These fundamental issues are of importance both on their own, and also for harnessing the potential of mass spectrometry for biological applications. We have found that ion abundances do not always track with solution phase stability; GC content must be taken into account. Two duplexes with the same Tm yet with differing GC content can yield different ion abundances. That is, if two duplexes have the exact same melting temperature, yet one has a higher GC content, the duplex with the higher GC content yields a higher ion abundance. It thus appears that not only is a GC base pair stronger than an AT base pair, but the relative strengths of each differ in the gas phase versus in solution, such that the electrospray process can differentiate between them. We also characterize the gas-phase stabilities of the duplexes, using collision-induced dissociation (CID) as a method to assess stability. We focus on two aspects of this CID experiment. One, we examine what factors appear to control whether the duplexes dissociate into single strands or covalently fragment; we are able to utilize a charge state normalization we coin "charge level" to compare our results with others' and establish generalities regarding dissociation versus fragmentation patterns. Two, we examine those duplexes that primarily dissociate and use CID to assess the gas-phase stabilities. We find that correlation of gas-phase to solution-phase stabilities is more likely to occur when duplexes of varying GC content are examined. Duplexes with the same GC content tend to have stabilities that do not parallel those in solution. We discuss these results in light of the different roles that hydrogen bonding and base stacking play in solution versus the gas phase. Ultimately, we apply what we learn to lend insight into the biological problem of how the carcinogenic, damaged nucleobase O6-methylguanine causes mutations.

Electrospray mass spectrometry (ESI-MS) in the study of metal-ligand solution equilibria

In the 20 years, since the introduction of electrospray mass spectrometry (ESI-MS), the use of this technique in various fields of inorganic, organometallic, and analytical chemistry has been steadily increasing. In this study, the application of ESI-MS to the study of metal-ligand solution equilibria is reviewed (till 2004 included). In a first section, advantages and drawbacks of ESI-MS in this type of application are described. Subsequently, a list of ca. 300 studies is reported, in which ESI-MS was used to give number and stoichiometry of the species at equilibrium, or also to estimate their stability constants. All studies are classified according to the metal ions under examination. Other related applications, such as host-guest interactions and metal ion-protein binding studies, are briefly reviewed as well.

Evaluation of binding of perylene diimide and benzannulated perylene diimide ligands to DNA by electrospray ionization mass spectrometry

Electrospray ionization mass spectrometry (ESI-MS) and spectroscopic studies in solution were used to evaluate the self-association, G-quadruplex DNA binding, and selectivity of a series of perylene diimides (PDIs) (PIPER, Tel01, Tel11, Tel12, and Tel18) or benzannulated perylene diimide ligands (Tel34 and Tel32). Fluorescence and resonance light scattering spectra of Tel01, Tel12, Tel32, and Tel34 reveal that these analogs undergo self-association in solution. UV-Vis and fluorescence titrations with G-quadruplex, duplex, or single-stranded DNA demonstrate that all the analogs, with the exception of Tel32, bind to G-quadruplex DNA, with those PDIs that are self-associated in solution showing the highest degree of selectivity for binding G-quadruplex DNA. Parallel ESI-MS analysis of the stoichiometries demonstrates the ability of the ligands, with the exception of Tel32, to bind to G-quadruplex DNA. While most ligands show major 1:1 and 2:1 binding stoichiometries as expected in the case of end-stacking, interestingly, three of the most quadruplex-selective ligands show a different behavior. Tel01 forms 3:1 complexes, while Tel12 and Tel32 only form 1:1 complexes. Collisional activation dissociation patterns are compatible with ligand binding to G-quadruplex DNA via stacking on the ends of the terminal G-tetrads. Experiments with duplex and single strand DNA were performed to assess the binding selectivities of the ligands. PIPER, Tel11, and Tel18 demonstrated extensive complexation with duplex DNA, while Tel11 and Tel18 bound to single strand DNA, confirming the lack of selectivity of these two ligands. Our results indicate that Tel01, Tel12, and Tel34 are the most selective for G-quadruplex DNA.

Positive effect of natural and negatively charged cyclodextrins on the stabilization of penicillins towards β-lactamase degradation due to inclusion and external guest–host association. An NMR and MS study

The complexation of penicillin (1a-c) and cephalosporin (2a,b) antibiotics with cyclodextrins (CDs), both natural [beta-CD (3b) and gamma-CD (3c)] and carboxylated [heptakis(6-oxycarbonylethylthio-6-deoxy)-beta-CD sodium salt (4b) and octakis(6-oxycarbonylethylthio-6-deoxy)-gamma-CD (4c) sodium salt], has been studied at neutral pH. Penicillins [ampicillin (1a), amoxicillin (1b) and dicloxacillin (1c) form inclusion complexes with the above CDs, as was shown by extensive NMR spectroscopic studies, whereas cephalosporins (cephalexin, cefadroxil) do not. Inclusion of the penicillins into either 3b or 4b was not accompanied by significant chemical shift changes in the 1H NMR spectra. On the contrary, with the wider 3c and its derivative 4c inclusion was evidenced by both chemical shift displacements of the cavity protons and intermolecular interactions, indicating the formation of primarily 1:1 guest-host inclusion complexes. The binding constants for 1a/3c, 1a/4c and 1c/3c were calculated as 19 +/- 4, 17 +/- 0.9 and 622 +/- 200 M(-1), respectively. With 4c, a 1:2 stoichiometry was also found. In addition, simultaneous formation of aggregates by external association takes place in solution, as shown by the ESI-mass spectrometric data. Studies on the hydrolysis of ampicillin under pseudo-first order conditions using an excess of 3c, 4c and of linear maltoheptaose at pH 7 showed that the drug hydrolysed at a similar rate in all cases. In the presence, however, of beta-lactamase enzyme and the carboxylated host 4c, ampicillin degraded twice as slowly (0.008 h(-1)) as in the presence of beta-lactamase alone (0.017 h(-1)). This was explained by the effective protection provided by both inclusion and external association of the host. The interaction, therefore, of penicillins with carboxylated CDs may present a means to lessen the chemical instability of these drugs in the presence of beta-lactamase enzymes.

Secondary organic aerosol formation by glyoxal hydration and oligomer formation: humidity effects and equilibrium shifts during analysis

Glyoxal is a significant atmospheric aldehyde formed from both anthropogenic aromatic compounds and biogenic isoprene emissions. The chemical behavior of glyoxal relevant to secondary organic aerosol (SOA) formation and analysis is examined in GC-MS, electrospray ionization (ESI)-MS, and particle chamber experiments. Glyoxal oligomers are shown to rapidly decompose to glyoxal in GC injection ports at temperatures > or = 120 degrees C. Glyoxal dihydrate monomer is dehydrated at temperatures > or = 140 degrees C during GC analysis but shows only oligomers (n < or = 7) upon ESI-MS analysis. Thus both of these analytical techniques will cause artifacts in speciation of glyoxal in SOA. In particle chamber experiments, glyoxal (at -0.1 Torr) condensed via particle-phase reactions when relative humidity levels exceeded a threshold of -26%. Both the threshold humidity and particle growth rates (-0.1 nm/min) are consistent with a recent study performed at glyoxal concentrations 4 orders of magnitude below those used here. This consistency suggests a mechanism where the surface water layer of solid-phase aerosol becomes saturated with glyoxal dihydrate monomer, triggering polymerization and the establishment of an organic phase.

Interactions of cationic porphyrins with double-stranded oligodeoxynucleotides: a study by electrospray ionisation mass spectrometry

Electrospray ionisation mass spectrometry (ESI-MS), electrospray ionisation tandem mass spectrometry (ESI-MS/MS) and Ultraviolet-visible (UV-vis) spectroscopy were used to investigate the non-covalent interactions between small oligonucleotide duplexes with the GC motif and a group of cationic meso(N-methylpyridynium-4-yl)porphyrins (four free bases with one to four positive charges, and the zinc complex of the tetracationic free base). The results obtained point to outside binding of the porphyrins, with the binding strength increasing with the number of positive charges. Fragmentations involving losses from both chains were observed for the porphyrins with N-methylpyridinium-4-yl groups in opposite meso positions.

Determination of affinity constants and response factors of the noncovalent dimer of gramicidin by electrospray ionization mass spectrometry and mathematical modeling

The dimerization of gramicidin, a 15-residue membrane peptide, in solution can be viewed as a model for protein-protein interactions. We reported previously that the dimer can be observed when electrosprayed from organic solvents and that the abundances of the dimer depends on the dielectric constant of the solvent. Here, we report an effort to determine an affinity constant for the dimerization of gramicidin by using gas-phase abundance. Two issues affecting the determination are the electrospray-induced dissociation of the dimer and discrimination in the electrospray of the dimer compared with the monomer. Other methods developed for the purpose of determining affinity from mass spectral abundance do not address the dissociation of the complex in the gas phase or can not be applied for cases of low affinity constant, K(a). We present a mathematical model that uses the ratio of the signal intensities of the dimer and the monomer during a titration. The model also incorporates the dissociation and an electrospray ionization-response factor of the dimer for extracting the affinity constant for the dimerization of gramicidin. The dimerization constants from the new method agree within a factor of two with values reported in the literature.

Gas-Phase Ion Association Provides Increased Selectivity and Sensitivity for Measuring Perchlorate by Mass Spectrometry

Perchlorate (ClO4-) competitively inhibits the uptake of iodide by the thyroid gland. Trace quantities of perchlorate are being increasingly detected in food and environmental samples. There is great concern that perchlorate contamination may be far more widespread than believed until now. Increasingly sensitive and unambiguous methods are needed for measuring perchlorate. We report here an ion chromatography-ion association-electrospray ionization-mass spectrometry (IC/IA-ESI-MS) method of substantially greater selectivity and sensitivity than other available single-stage MS approaches. A long chain dipositive cationic agent (D2+) is added postcolumn in low concentration. This ion associates with perchlorate, even in the gas phase. Perchlorate is, thus, detected as DClO4+ in the positive ion mode at an m/z value between 300 and 400 (depending on the choice of D2+). This results in much better S/N and selectivity, as compared to detecting 35ClO4- at m/z 99, where H34SO4- also responds. We show results for various dicationic agents which vary in their selectivity and affinity for ClO4-, typically being at least 1 order of magnitude more selective for ClO4- over HSO4-. For a 100-microL injected standard, limits of detection (LOD, S/N = 3) are as good as 25 ng/L on a single quadrupole mass spectrometer. Calibration for concentrations up to 100 microg/L displays an r2 value of > or =0.9993. We show applicability to various real samples. A number of the studied reagents are suitable for such applications.

Determination of zinc to beta-peptide binding constants with electrospray ionization mass spectrometry

We present an improvement of the titration method for binding constant determination with electrospray ionization (ESI) mass spectrometry that is unaffected by differences in ESI response of measured species in solution. The method consists of a calibration and titration, both using an internal standard that allows relative quantitation. This avoids artifacts such as a decrease in overall signal intensity with increasing ligand concentrations, rendering this approach more reliable and meaningful than direct evaluation of ESI peak intensities. We demonstrate the de novo binding constant determination of novel zinc binding beta-peptides, which have been synthesized with the goal of creating secondary structures stabilized by metal complexation.

Measurement of Solution-Phase Chiral Molecular Recognition in the Gas Phase Using Electrospray Ionization-Mass Spectrometry

Development of chiral selectors (SOs) is important both for understanding chiral molecular recognition processes and for their use in the separation of chiral species (selectands). Their evaluation by chromatographic procedures (e.g., as chiral stationary phase) can, however, be time-consuming. In this respect, electrospray ionization-MS (ESI-MS) is tested here as a possible alternative for screening enantioselective binding by SOs. The set of well-characterized cinchona alkaloid SOs are investigated with respect to their enantioselective binding to a set of model enantiomers, dinitrobenzoyl-(R)- and dinitrobenzoyl-(S)-leucine. MS-based enantioselectivity values from normalized gas-phase ion abundances for the diastereomeric complexes are compared empirically to chromatographic (HPLC) enantioselectivity results and shown to be consistent. Investigations into the fundamentals of measuring unbiased enantioselectivity values in the limit of dilute solution by correlation between experimental and modeled theoretical data are shown. Titration experiments are used to extract binding constants and are compared with published calorimetric (ITC) data. Results show that while the magnitude of binding affinities determined for various diastereomeric complexes is attenuated, the relative ranking and stereochemical preference in binding are consistently reproduced. This work represents a fundamental study of solution- versus gas-phase correlation for enantioselective systems by ESI-MS and indicates that, although not all questions and assumptions can be clearly engaged, for these enthalpically driven binding systems, the relative degree of binding affinity and selectivity is preserved.

Affinity capillary electrophoresis on microchips

The present study shows that the application of the method of affinity capillary electrophoresis (ACE) to investigate interactions between ligands and their substrates can be realized on microchips. With ACE it is possible to characterize non-covalent molecular interactions (complexation and partition equilibria). Binding constants (K(B)) provide a measured value of the affinity of a ligand molecule to a substrate, which is basic information for the understanding of hormones, drugs and their targets, e.g. receptors in the human body. A microchip electrophoresis instrument equipped with a UV-detector and a home-built chip-station with electrochemical detection were used. ACE could be achieved with model solutions of neurotransmitters using sulfated beta-cyclodextrin (sCD) as substrate in a background buffer. This paper describes the advantages of microchip-ACE (MC-ACE) to traditional affinity capillary electrophoresis on a capillary. The results show that MC-ACE has great potential as a tool for fast scanning of interactions and to calculate binding constants of ligands with their substrates.

Ligand–Metal Ion Binding to Proteins: Investigation by ESI Mass Spectrometry

The objective of this chapter is to show the general mass spectrometry (MS)-based strategies that can be used to retrieve information regarding protein-metal and protein-ligand noncovalent complexes. Indeed, when using carefully controlled conditions in the atmospheric pressure-vacuum interface of the mass spectrometer, and when sample preparation is optimized, it is possible to preserve large specific multiprotein-metal-ligand noncovalent complexes during MS analysis. Examples describing the possibilities of electrospray ionization MS (ESI-MS) are shown. For instance, it can be used to probe cooperativity in the binding of a ligand or a metal to a protein or may constitute a new methodology for a more rational approach for drug discovery and for human genome annotation. Thanks to its ability to directly give information on stoichiometry or dynamics of the interactions formed in solution, MS offers new possibilities to tackle more and more various applications.

Mass spectrometric analysis of the interactions between CP12, a chloroplast protein, and metal ions: a possible regulatory role within a PRK/GAPDH/CP12 complex

The small chloroplast protein CP12 plays the role of a protein linker in the assembly process of a PRK/GAPDH/CP12 complex that is involved in CO2 assimilation in photosynthetic organisms. The redox state of CP12 regulates its role as a protein linker. Only the oxidized protein, with two disulfide bonds, is active in complex formation. Several observations indicating that CP12 might bind a metal ion led us to screen the binding of different metal ions on oxidized or reduced CP12 using non-covalent electrospray ionization mass spectrometry (ESI-MS) experiments. The oxidized protein bound specifically Cu2+ and Ni2+ (Kd of 26+/-1 microM and 11+/-1 microM, respectively); other cations such as Fe2+ and Zn2+ did not bind, while cations such as Cd2+ formed non-specific adducts to CP12. Similar results were obtained for metal ions on screening with the reduced CP12. Interestingly, the present results suggest that Cu2+ catalyzes the re-formation of the disulfide bonds of the reduced CP12, leading to recovery of the fully oxidized CP12 that is then able to bind a Cu2+ ion. Finally the high similarity between CP12 and copper chaperones from Arabidopsis thaliana, as judged by hydrophobic cluster analysis, provides additional evidence for the relevance of metal binding for the in vivo situation. The findings that CP12 is able to bind a metal ion, and that Cu2+ catalyzes the oxidation of the thiol groups of CP12, are new characteristics of this protein that may prove to be important in the regulation of the assembly process of the PRK/GAPDH/CP12 complex.

De Novo Design of a Copper(II)-Binding Helix−Turn−Helix Chimera: The Prion Octarepeat Motif in a New Context

A chimeric Cu-binding peptide has been designed on the basis of a turn substitution of the prion (PrP) octarepeat Cu-binding site into the engrailed homeodomain helix-turn-helix motif (HTH). This system is a model for the investigation of a single PrP Cu-binding site in a defined protein context. The 28-mer Cu-HTH peptide P7 spectroscopically mimics the PrP octarepeat (P7 = TERRRQQLSHGGGWGEAQIKIWFQNKRA). The Cu(II)-binding affinity of P7 was determined by ESI-MS and tryptophan fluorescence titrations to be K(d) = 2.5 +/- 0.7 microM at pH = 7.0. The quenching of fluorescence of the Trp within the binding loop (underlined above) is pH dependent and highly specific for Cu(II). No Trp quenching was observed in the presence of divalent Zn, Mn, Co, Ni, or Ca ions, and ESI-MS titrations confirmed that these divalent ions do not appreciably bind to P7. The EPR spectrum of Cu(II)-P7 shows that the Cu environment is axial and consistent with 6-coordinate N(3)O(H(2)O)(2) or N(4)(H(2)O)(2) coordination (A( parallel) = 172 x10(-)(4) cm(-)(1); g( parallel) = 2.27), very similar to that of the PrP octarepeat itself. Also like PrP, circular dichroism studies show that apo P7 is predominantly disordered in solution, and the structure is slightly enhanced by Cu binding. These data show the Cu-PrP HTH peptide reproduces the Cu-binding behavior of a single PrP octarepeat in a new context.

Evaluation of complexes of DNA duplexes and novel benzoxazoles or benzimidazoles by electrospray ionization mass spectrometry

Electrospray ionization mass spectrometry is used to compare the metal ion binding and metal-mediated DNA binding of benzoxazole (1, 2, 3, 4) and benzimidazole (5) compounds and to elucidate the putative binding modes and stoichiometries. The observed metal versus non-metal-mediated DNA binding, as well as the specificity of DNA binding, is correlated with the biological activities of the analogs. The ESI-MS spectra for the antibacterial benzoxazole and benzimidazole analogs 4 and 5 demonstrated non-specific and non-metal-mediated binding to DNA, with the appearance of DNA complexes containing multiple ligands. The anticancer analog 2 demonstrates a clear preference for metal-mediated DNA interactions, with an apparent selectivity for Ni2+ -mediated binding over the more physiologically relevant Mg2+ or Zn2+ cations. Complexation between DNA and the biologically inactive analog 1 was not observed, either in the absence or presence of metal cations.

Production and Properties of Nanoelectrospray Emitters Used in Fourier Transform Ion Cyclotron Resonance Mass Spectrometry: Implications for Determination of Association Constants for Noncovalent Complexes

Electrospray ionization (ESI) is extensively used in the analysis of biological compounds; yet some fundamental properties of this technique are not completely understood. It is widely recognized that care should be exercised when noncovalent complexes are being studied by ESI, since weak noncovalent binding can be broken or formed during the desolvation process. In the present work, spectra from the noncovalent complex, vancomycin/diacetyl-L-lysyl-D-alanyl-D-alanine, obtained from ESI and from nanoelectrospray ionization (nanoESI), have been compared. The results indicated that the milder desolvation conditions arising as a result of the smaller sizes of droplets produced in the nanoESI source attenuated effects upon weak bonds in the desolvation process. The association constant values calculated from the relative peak intensities suggest that, when using ESI, the analyzed noncovalent complex dissociated in the condensed phase during the spraying process. The influences of experimental parameters such as tip diameter and coating for nanoESI needles were investigated. Principal component analysis, a multivariate analysis method, was applied to achieve a better evaluation of the spectra obtained using different needle diameters and coatings for the analysis of the noncovalent complex vancomycin/diacetyl-L-lysyl-D-alanyl-D-alanine. It was found that 2-microm tip diameter resulted in more reproducible spectra than the larger tip diameters tested (6-20 microm).

Evidence for the 2:1 molecular recognition and inclusion behaviour between β- and γ-cyclodextrins and cinchonine

Cinchonine (Cin) is the primary drug of choice in the treatment of malaria, but its poor solubility has restricted its use via the oral route. Cyclodextrins (CDs) form inclusion complexes with cinchonine to form soluble complexes. This interaction was investigated by solubility studies, electrospray ionization mass spectrometry (ESI-MS), and molecular modeling. ESI-MS evaluated successfully the nature of the solution-phase inclusion complexes. The experimental results showed that not only 1:1, but also stable 2:1 inclusion complexes can be formed between CDs and Cin. Multi-component complexes of beta-CD-Cin-beta-CD (1:1:1), gamma-CD-Cin-gamma-CD (1:1:1), and beta-CD-Cin-gamma-CD (1:1:1) were found in equimolar beta- and gamma-CD mixtures with Cin. The formation of 2:1 and multi-component 1:1:1 non-covalent CD-Cin complexes indicates that beta- and gamma-CD are able to form sandwich-type inclusion complexes with Cin in high concentrations. The phase-solubility diagram showed non-linear type A(p) profile, indicating that more than one cyclodextrin molecule is involved in the complexation of one guest molecule. Molecular modeling calculations have been carried out to rationalize the experimental findings and predict the lowest energy molecular structure of inclusion complex.

Self-Association of Cyclodextrins and Cyclodextrin Complexes

Cyclodextrins are useful functional excipients, which are being used in an ever-increasing way to camouflage undesirable pharmaceutical characteristics, especially poor aqueous solubility. It has generally been assumed that the mechanism whereby cyclodextrins exert their effects, especially their augmentation of solubility, is via the formation of noncovalent, dynamic inclusion complexes. This is a model, which regards drug-cyclodextrin interactions as a discrete phenomenon and ignores the possible interaction of these complexes with one another. It is becoming increasingly apparent that such assumptions may not be universally applicable or all encompassing. Specifically, there is a growing body of evidence that supports the important contribution of non-inclusion-based aspects for drug solubilization by cyclodextrins including surfactant-like effects and molecular aggregation. This short review attempts to assess the available literature for areas in which such non-inclusion mechanisms are apparent and tries to interpret these in the context of a broader working theory as to how cyclodextrins exert their beneficial effects.