Anal Bioanal Chem (2010) 398:155–166 DOI 10.1007/s00216-010-3959-2
REVIEW
Circular dichroism in drug discovery and development: an abridged review Carlo Bertucci & Marco Pistolozzi & Angela De Simone
Received: 27 April 2010 / Revised: 15 June 2010 / Accepted: 21 June 2010 / Published online: 25 July 2010 # Springer-Verlag 2010
Abstract Chirality plays a fundamental role in determining the pharmacodynamic and pharmacokinetic properties of drugs, and contributes significantly to our understanding of the mechanisms that lie behind biorecognition phenomena. Circular dichroism spectroscopy is the technique of choice for determining the stereochemistry of chiral drugs and proteins, and for monitoring and characterizing molecular recognition phenomena in solution. The role of chirality in our understanding of recognition phenomena at the molecular level is discussed here via several selected systems of interest in the drug discovery and development area. The examples were selected in order to underline the utility of circular dichroism in emerging studies of protein– protein interactions in biological context. In particular, the following aspects are discussed here: the relationship between stereochemistry and pharmacological activity—stereochemical characterization of new leads and drugs; stereoselective binding of leads and drugs to target proteins—the binding of drugs to serum albumins; conformational transitions of peptides and proteins of physiological relevance, and the stereochemical characterization of therapeutic peptides. Keywords Circular dichroism . Chiral analysis . Protein–protein interactions . Drug–protein interactions . Bioanalytical methods . Molecular recognition mechanism
C. Bertucci (*) : M. Pistolozzi : A. De Simone Department of Pharmaceutical Sciences, University of Bologna, via Belmeloro 6, 40126 Bologna, Italy e-mail:
[email protected]
Introduction Very little happens in biological systems unless two or more molecules interact to form a stable complex. Thus, it is essential to study the molecular-recognition phenomena behind a biological process in order to discover new pharmacological targets and thus to discover and develop potent and selective drugs. In particular, it is vital to locate and validate new targets, and then to rapidly investigate the bioavailability of each potential drug by determining its ADMET (absorption, distribution, metabolism, excretion, and toxicity) parameters. However, complex networks of interactions between proteins likely regulate biological processes. Life science is currently focused on studying the full complement of proteins involved in a specific biological process more than describing each protein individually. Changing the focus of cellular studies created a paradigm shift in modern life science research towards the system biology approach. Modulating protein–protein interactions and protein conformations are now recognised to be valuable strategies in drug discovery. Conformational changes quite often trigger the process of interest. Monitoring and modulating them can lead to a mechanistic understanding of the pathophysiology of a disease, and to the identification of treatable targets. Chirality often plays a fundamental role in determining the pharmacodynamics and pharmacokinetics of a drug [1, 2], and chirality contributes significantly to our understanding of the mechanisms behind biorecognition phenomena. Circular dichroism (CD) spectroscopy (both electronic and vibrational) is the technique of choice for
156
elucidating chirality, and, in particular, for monitoring and characterizing molecular recognition phenomena in solution [3–6]. CD is simply a variation of normal absorption spectroscopy that uses right- and left-handed circularly polarized light to investigate a sample. Differential absorption of the two components occurs for chiral compounds with chromophores that either possess intrinsic chirality or are located in a chiral environment. CD is mainly applied for the determination of the absolute configuration of chiral compounds in solution. One nonempirical method for determining the absolute configuration is X-ray crystallography method using the heavy atom effect [7, 8]. The results obtained are unambiguous and reliable; the main problem is obtaining the crystal. CD has the advantages that it can be applied to solutions and that it can monitor the dynamics of processes and stereochemically unstable compounds. A limitation could be the reliability of the configurational assignment if the compound analyzed or the systems under investigation are not ideal targets for this type of analysis. However, these days nonempirical methods are widely and successfully used to determine the absolute configurations of chiral drugs, drug metabolites and natural compounds [3, 5, 9–12]. Alternatively, the absolute configuration can be obtained through the application of relative and/or empirical methods, like X-ray crystallography using a chiral internal reference, and proton nuclear magnetic resonance (1H-NMR) spectroscopy using chiral anisotropic reagents [13]. In these cases, the relative configuration at the position of interest is determined with respect to a reference compound or a substituent with known absolute configuration. CD also allows the signal associated with the formation of a drug–protein or protein– protein complex to be monitored, thus providing direct information on the binding interaction. Information on a drug–protein complex or molecular recognition processes can be obtained using several techniques, including X-ray crystallography, mass spectrometry, fluorescence and NMR spectroscopy, and surface plasmon resonance-based optical biosensors. The use of independent techniques allows better characterization of the structure and yields a more accurate picture of the dynamics of the recognition process. CD can provide deeper insights into the stereochemistry of a protein-bound drug and protein folding, thus giving detailed information on the mechanism of binding. Spectroscopic investigations of drug–protein complexes, and in particular CD investigations, are invaluable for not only obtaining a complete picture of the binding process and mechanism, but also for quantitatively characterizing the binding process at equilibrium (i.e., binding parameters) without interfering with the equilibrium process. In this review we will present several cases that underline the role of chirality in our understanding of recognition
C. Bertucci et al.
phenomena at the molecular level and give information on the structural features that modulate these processes. The examples were selected to highlight the utility of CD during the different steps of drug discovery and development, and its potential in emerging studies of protein–protein interactions in the biological context. In particular, the following aspects are discussed: the relationship between stereochemistry and pharmacological activity—the stereochemical characterization of new leads and drugs; the stereoselective binding of leads and drugs to target proteins—the binding of drugs to serum albumins; conformational transitions of peptides and proteins of physiological relevance, and the stereochemical characterization of therapeutic peptides.
Relationship between stereochemistry and pharmacological activity: stereochemical characterization of chiral leads and drugs The study of the relationship between stereochemistry and pharmacological activity is the most common type of investigation among those reported in this short review. Numerous studies have been reported in the literature [1–6], starting with the fundamental papers of E.J. Ariens [2]. Reliable determination of the stereochemistry–activity relationship requires full stereochemical characterization of the possible stereoisomers. As a representative example of the use of circular dichroism in this type of investigation, the enantioselective inhibition of the aromatase system by a small series of new compounds is discussed here [9]. The inhibition of aromatase is one of the approaches employed to reduce the concentration of estrogens in the blood [14, 15] in the treatment of post-menopausal estrogen-dependent breast cancer. Fradrozole and the azole family are important aromatase inhibitors. A few compounds were selected to investigate the 3D QSARs of azoles, taking the stereochemistry into account in the computational molecular field analysis (CoMFA) [15, 16]. Different activities were predicted for the individual enantiomers of these compounds (Chart 1), with calculations suggesting that the (R)enantiomer is always more potent than the (S)-enantiomer. Thus, the individual enantiomers of compounds 1–3 were synthesized and their biological activities were determined to verify the predicted enantioselectivities of their aromatase inhibition activities. The main challenge in this type of study is to obtain a reliable relationship between stereochemistry and pharmacological activity. To this end, each compound, and then each enantiomeric sample, should be characterized for its stereochemistry (i.e., the enantiomeric excess and the absolute configuration of the prevailing enantiomer should be determined). The enantiomeric samples were obtained from the synthesized racemic mixtures of 1–3 by preparative HPLC upon chiral stationary phases. The
Circular dichroism in drug discovery and development: an abridged review Chart 1 Structures of the aromatase inhibitors
157
20
N O
N 10
R
O 1-3 Compound
R
1
NO2
2
Br
3
CN
Δε (M-1cm-1)
*
0 -10
-15 200
250
300
350
Wavelength (nm) Fig. 1 CD spectra of the first-eluted enantiomers of 1 (red), 2 (black) and 3 (blue), recorded in 2-propanol
carried out at the molecular mechanics level using MMFF94 [34] of Spartan02 [35]. Fourteen different conformers were selected in the range of 3 kcal/mol. Optimization at the DFT level led to the selection of just four conformers in the range of 2 kcal/mol. Accurately evaluating the conformer populations (obtained by calculating the free energy values) is essential for achieving a reliable determination of the contribution of each conformation to the CD spectrum of 3. Using the calculated CD spectra for the four conformations and their populations, the final result obtained was then the weighted-average CD spectrum of (S)-3, as reported in Fig. 2, together with the experimental CD spectra for the two enantiomeric fractions. This comparison allowed the absolute configurations (R)-
12 8 Δε (M-1cm-1)
same enantioselective HPLC methods were successfully applied to determine the enantiomeric excess of each sample. The absolute configuration of each enantiomeric fraction must then be established. The best ways to obtain this information are to use the chiroptic properties, optical rotation and circular dichroism [1, 3], as demonstrated in this successful multidisciplinary medicinal chemistry approach [9]. Here the absolute configurations of the enantiomeric fractions of 1–3 were determined by a combination of CD measurements and time-dependent density functional theory (TD-DFT) calculations. The method involves calculating the CD spectrum of one of the two enantiomers, and then the absolute configuration of the enantiomeric fraction is determined by comparing experimental CD spectra with this calculated spectrum. The CD spectra of the less-retained enantiomeric fractions obtained by enantioselective preparative HPLC showed almost the same numbers of bands and the same sequence of signs—at least in the 280–210 nm spectral region—even though significant differences were observed for the band energies, due to the varying nature of the substituent on the phenyl group in compounds 1–3 (Fig. 1). Thus, the same elution order was suggested for all of the compounds, irrespective of the substituent on the phenyl ring. Much progress has been made recently in correctly interpreting chiroptic data using coupled oscillator methods [17–22] and ab initio calculations, thus allowing reliable assignment of the absolute molecular configuration; some different programs are now available [23–25] for the ab initio calculation of the optical rotation [26–29] and CD spectra [30–33]. Quantum mechanical methods of calculating CD data were applied to the CD calculations for 3, and the (S) absolute configuration was arbitrarily assumed when performing these CD calculations. A detailed conformational analysis was performed to determine the conformer populations of 3. An initial conformational search was
4 0 -4 -8 -12 200
240
280
320
360
Wavelength (nm) Fig. 2 Comparison of the experimental CD spectra of (+)-3 (black) and (−)-3 (blue) to the calculated weighted-average CD spectrum of (S)-3 (red)
158
C. Bertucci et al.
and (S)- to be assigned to the first- and second-eluted enantiomers of 3, respectively, based on the sequence of CD band signs at 242 and 222 nm. As shown in Fig. 2, the calculated CD spectrum for (S)-3 exhibits the same CD spectral behavior as the second-eluted enantiomer of 3, which is the mirror-image of the first-eluted enantiomer. Both of the enantiomers of the new inhibitors were finally tested for their activity against human P450 19 enzyme, and significant enantioselectivity was observed, in agreement with the 3D QSAR predictions. Thus, stereochemistry proved fundamental to both the design of new inhibitors with higher activities and to gaining a deeper understanding of the mechanism involved in enzyme inhibition. It is worth noting that these CD calculation methods are so reliable and powerful that they were recently employed to show that the absolute configuration determined by X-ray diffraction analysis was incorrect, and thus the structural determination needed to be revised [36]. Applications of ab initio methods to the structural elucidation of natural compounds obtained by hyphenated HPLC/MS/CD techniques are reported almost routinely these days [10]. Quantum mechanical calculations of CD spectra are also useful for determining the absolute configurations of drug metabolites, and thus obtaining information on metabolic pathways [12]. Other nonempirical methods for determining the absolute configuration, such as exciton chirality and DeVoe methods, have been successfully applied to elucidate the absolute stereochemistries of drugs and natural compounds in a nonempirical manner [37–40].
Stereoselective binding of leads and drugs to target proteins: the example of serum albumins No matter how active or specific a chemical is, it needs good bioavailability parameters if it is to be turned into a
Detection of drug–protein binding phenomena: binding stereochemistry Drug distributions are widely studied by determining the drug fraction that is bound to serum proteins in vitro [42–45]. Again, stereochemistry is a fundamental aspect of the drug–protein interaction process, as it determines enantioselectivity phenomena during drug–protein binding [46, 47]. CD spectroscopy yields information on the stereochemistry of the bound drug, thus aiding in the elucidation of the enantioselection mechanism. The conformation of the bound drug can be determined by monitoring the CD signal induced by the drug when binding to the target protein. As an example, the HSA binding of a nonchiral drug like diazepam, which does not give any CD signal when it is free in solution, leads to a well-defined induced CD (ICD) spectrum (Fig. 3a). Given that there are two mirror-image conformations of the drug, M and P, that are interconverting in solution, the signs of the CD bands allow M to be reliably determined as the prevailing conformation (Fig. 3b). The assignment of M as the prevailing conformation is based on a comparison of the ICD spectrum of the bound diazepam with the almost superimposable CD spectra of 3-substituted benzodiazepines with the (S) absolute configuration at C3 of the
a
5 ΔA = AL– AR (x 10-4)
Fig. 3 a Induced CD (ICD) spectrum of the complex [diazepam]/[HSA] 1:1 (solid line) obtained by subtracting the CD spectrum of HSA (dotted line) from the CD spectrum of the equimolar mixture of diazepam and HSA (dashed line). [HSA] = [diazepam] = 15 μM, pathlength =10 mm;. b Plus (P) and minus (M) conformational equilibrium of diazepam. The M conformation prevails for the HSA-bound diazepam
drug. In particular, it should be well absorbed, distributed throughout the body, not too rapidly or slowly metabolized, and eliminated completely. Furthermore, the potential drug and its metabolites should not be toxic to the body. Thus, it is clear how the rapid determination of ADMET parameters during the early stages of drug discovery will save time and money by rapidly and cheaply identifying the most suitable compounds [41]; it is worth noting that about half of all drugs in development fail to reach the market because of ADME deficiencies, and that 50% of all registered drugs still have some ADME or toxicity problems [41].
2.5
b Cl
Hax CH3
Cl
CH3
Heq
N
0
N
N O
O
N Heq
-2.5
Hax
M -5 260
280
300
320
Wavelength (nm)
340
P
Circular dichroism in drug discovery and development: an abridged review
heptaatomic ring [46], for which the M conformation is largely prevalent. Difference CD spectra obtained by subtracting the protein’s CD contribution from the CD spectrum of the drug–protein complex (Fig. 3a) provides evidence of the stereoselective interaction. The titration of ΔCD and ΔUV at a constant protein concentration and varying drug concentrations allowed a stereoselective binding site for diazepam to HSA to be distinguished, as well as additional nonstereoselective binding sites, where diazepam binds without a single conformation being prevalent. The ΔCD values increase up to a stoichiometry of [HSA]/[diazepam] 1/1, while the ΔUV values also increase for higher stoichiometries [48], thus supporting the existence of several other low-affinity binding areas. Determination of binding parameters Circular dichroism investigations of drug–protein complexes are invaluable since they not only help to build up a complete picture of the binding process and mechanism but they also allow the quantitative characterization of the binding process at equilibrium without interfering with the equilibrium process, thus facilitating the derivation of affinity constants. Indeed, the induced CD (ICD) arising from the binding of a drug to a protein allows the selective detection of the drug when bound to the protein without interference from the free fraction of the drug in solution (which is “transparent;” i.e., only the bound fraction of the drug gives the ICD signal). Since the intensity of the ICD is proportional to the concentration of the drug–protein complex, it is possible to use CD to measure the binding constant of a drug to a protein in a very elegant way. This approach consists of monitoring the effect of sequential dilutions on the ICD signal of the drug–protein complex at fixed stoichiometry while increasing the pathlength in proportion to the dilution (e.g., increasing the pathlength tenfold while diluting the solution of the complex tenfold) [48, 49]. The decrease in the ICD signal detected by this method is correlated with the association constant of the complex, which is directly derivable from the experimental data. Using the Lambert–Beer law, the following linear equation was derived for a 1/1 drug–protein stoichiometry [48, 49]: p 1 qffiffiffiffiffiffi ¼ Δ" CD l
rffiffiffiffiffiffiffiffi CD 1 þ pffiffiffiffiffiffiffiffiffiffiffiffiffiffi ; l K Δ"
ð1Þ
where p is the protein concentration, CD is the measured signal, l is the cell pathlength, K is the association constant, and Δε is the differential molar extinction coefficient. Measuring the ICD while changing p and l yields K and Δε from the slope and the intercept of a
159
linear fit to the data. Obviously this approach requires the binding phenomenon to give a detectable ICD; furthermore, if the drug is chiral, its intrinsic CD signal must be much lower than the ICD to get reliable affinity data. On the other hand, this approach can obtain data on the highest-affinity binding site of a drug, since usually only the binding of the drug to that site gives an ICD [47]. This is particularly important when the binding of drugs to serum albumins is studied, since (at therapeutic concentrations) only the main binding site of the drug has pharmacokinetic relevance [47]. As an example, although this approach is not applicable to all drug–protein systems, it was successfully employed to measure the binding constants of ketoprofen to human and rat serum albumins [50], and it gave binding constants that agreed with those obtained by ultrafiltration [51]. Determination of binding sites and detection of drug–drug interactions Information on binding sites can also be obtained by carrying out competition experiments with ligands that are known to bind to specific binding sites on the protein [46, 49]. Depending on the change in the ICD spectrum upon the addition of increasing concentrations of the competitors, different co-binding situations can be observed. Independent binding does not cause any change in the ICD signal, while a decrease in the CD signal may be due to direct competition at the same binding site, or an allosteric interaction upon the binding of drug and competitor at different binding sites. In the case of allosteric interactions, it is possible to observe an increase in the CD signal due to a cooperative effect that leads to an increase in the affinities of both the ligands. As an example, the competitive binding of diazepam and ibuprofen to the site II binding site (according to the Sudlow model [52]) was carried out by monitoring the decrease in the ICD signal of the diazepam bound to HSA in the presence of increasing concentrations of ibuprofen, its competitor (Fig. 4). Competition experiments utilizing CD spectroscopy also allow the affinity constant of the displacer to be determined, yielding a value that is strictly limited to the binding site responsible for the ICD signal [49]. Competition between two drugs and/or between drugs and metabolites in binding to serum proteins may strongly affect the disposition of the drug, and this phenomenon can be clinically relevant. This is particularly true of drugs that show a high protein-bound fraction, because of the significant impact of even small changes in the bound fraction on the circulating free fraction. As an example, lowering the bound fraction by 2% from 99.9 and 97.9% corresponds to a 21-fold increase in the free fraction, while decreasing the bound fraction by 2% from
160
C. Bertucci et al.
ΔA = AL–AR (x 10-4)
6
4
2
0 -1 260
280
300
320
340
Wavelength (nm) Fig. 4 Displacement of diazepam from the complex [HSA]/[diazepam] 1:1, performed by the stepwise addition of rac-ibuprofen, and monitored by CD. [HSA] = [diazepam] = 15 μM; [rac-ibuprofen] = 15 μM (solid line), 30 μM (dashed line), 60 μM (dotted line); pathlength=10 mm
90 to 88% leads to only a 1.2-fold increase in the free fraction of the drug. Detection of protein binding differences among mammals Animal models are often used in protein binding and pharmacokinetic studies, and the results obtained are then extrapolated to humans. Each animal albumin has different binding and structural characteristics [50, 53–55]. This variation must therefore be considered when results of animal experimentation are used as the basis for investigations on humans. To this end, it is very important to collect serum carrier-binding data using albumin from different species, to ensure that distribution data can be reliably extrapolated between species. In one study [50], CD provided a useful and reliable method for screening drugs based on their binding to albumins from different mammals [human serum albumin (HSA), bovine serum albumin (BSA), rat serum albumin (RSA), and dog serum albumin (DSA)]. The CD signals induced by the binding of the drug to the albumins of different animal species were measured and analyzed in order to get information on the stereoselectivity of the binding process and to determine the binding parameters. Each ligand [phenylbutazone, diazepam, rac-ketoprofen, (S)-ketoprofen, and a metabolite (bilirubin)] binds to a specific (highest-affinity) binding site on HSA [42, 43]. Phenylbutazone binds in the large hydrophobic cavity of subdomain IIA, at site I, while diazepam, rac-ketoprofen, and (S)-ketoprofen bind in the cavity of subdomain IIIA, at site II [42, 43, 52]. Bilirubin also binds in the large hydrophobic cavity of subdomain
IIA, at the bilirubin or site III binding site [42, 43, 56]. All of the ligands investigated are nonchiral or racemic, with the exception of (S)-ketoprofen. If the ligand is nonchiral or racemic, the observed ICD signal is completely related to the bound drug or metabolite, thus reflecting the stereoselectivity of the binding process [46, 47]. Information on the ligand–protein complexes can be obtained in relation to both the stereoselectivities and the affinity constants. Interestingly, significant differences were observed in the ICD spectra for the ligands bound to the different albumins [50]. As far as the phenylbutazone was concerned, similar behavior was observed in its binding to the different albumins, at least for the lowest-energy ICD spectrum, which was positive at about 290 nm for the drug when bound to any of the proteins [50]. BSA yielded the highestintensity ICD signal, and high affinity constants were also determined for RSA, HSA, and DSA. This behavior is in agreement with affinity chromatography data [57, 58]. It is worth mentioning that the intensity of the ICD spectrum does not necessarily reflect the degree of affinity; the sign and the intensity of the ICD are mainly related to the conformation of the bound drug. Furthermore, the association constants (KA) obtained by CD analysis can differ from those obtained using other affinity-based techniques. This is due to the selective monitoring of stereoselective binding processes in the case of CD spectroscopy [46, 47]. Diazepam, a selective HSA site II marker, yields an almost identical ICD spectrum, in terms of the signs and the energies of the ICD bands, when binding to any of the four proteins. This behavior suggests that the M conformation of the bound diazepam prevails, although the intensities of the ICD spectra are significantly different due to differences in the populations of the mirrorimage M and P conformations for the bound drug [50]. The KA values—as determined by analyzing the change in the ICD signal upon changing the drug–albumin concentrations —were quite different; they were much higher for diazepam bound to HSA and DSA than diazepam bound to BSA and RSA. The stereoselectivity was impressive in the binding of the HSA site II marker ketoprofen to the different albumins. The ICD spectra were positive at about 340 nm for the binding of both rac-ketoprofen [50, 51] and (S)-ketoprofen [50] to RSA, while the opposite sign was observed in the case of HSA binding [50, 51]. A negligible signal was obtained in the binding of the drug to BSA [59] and DSA. The intensities of the ICD spectra were comparable for both the rac-ketoprofen and (S)-ketoprofen binding, suggesting high stereoselectivity during the binding process. The intensity of the CD due to the bound (S)-ketoprofen was much higher than that of the CD spectrum of the free drug [50, 59, 60]. Thus, a prevalent conformation and a much higher conformational homogeneity can be hypothesized for the benzophenone moiety in a chiral arrangement when
Circular dichroism in drug discovery and development: an abridged review
bound to the protein [48]. The KA values were quite high (on the order of 105–106) [50, 51] for the binding of rac-ketoprofen to both HSA and RSA. No reliable measurements were possible when attempting to determine the KA values for binding to BSA and DSA because of the extremely low intensities of the corresponding ICD spectra, even though high KA values are reported for the binding of ketoprofen to both of these albumins [51]. Very intense ICD spectra were obtained for the binding of bilirubin to all of the albumins (Fig. 5). Bilirubin is nonchiral, and the observed CD is due to the binding of this metabolite to protein, during which it adopts a prevailing conformation that is reflected in the sign of the CD couplet in the 550– 350 nm spectral region (Fig. 5) [61–63]. A P helicity conformation gives a CD couplet with a positive band at lower energy and a negative band at higher energy. Interestingly, a P helicity conformation is adopted in the binding of bilirubin to HSA, RSA and DSA, while an M helicity conformation prevails in the case of BSA binding [50]. A higher KA value was obtained for the binding of bilirubin to BSA than for the binding of RSA, HSA, or DSA [50]. The ICD intensities do not reflect the affinities, as the CD signal is mainly related to the contributions of the relative populations of the different chiral conformations of the bound bilirubin. Again, significant species dependencies were observed for both the stereoselectivity of the recognition process and the binding parameters. These results support the utility of the CD screening of new compounds for their binding to albumins from different mammalians in order to investigate the reliability of extrapolating distribution data between different species.
12
ΔA = AL – AR (x 10-4)
10
5
0
-5 -8 350
400
450
500
550
Wavelength (nm) Fig. 5 Induced CD (ICD) spectra of bilirubin bound to different mammalian albumins in 1:1 molar ratio complexes ([HSA] = [bilirubin] = 100 μM, pathlength=1 mm): RSA (black), DSA (red), HSA (green) and BSA (blue). The CD couplet sign in the 350–550 nm range reflects the stereochemistry of the HSA-bound bilirubin
161
In addition, CD spectroscopy can be used to monitor changes in the binding of drugs (and in their free fractions) to albumins when they are covalently modified by reactions that occur under physiological conditions, like the binding of aspirin [64] or ethacrynic acid [65]. Finally, it is worth mentioning that CD is very important for retrieving information on the conformation of the protein by analyzing its secondary structure, in order to check its stability in the presence of physiologic concentrations of drugs [66]. The methods applied here to drug–albumin interactions can be extended to other protein systems, and they exploit the multitude of information that arises from changes elicited in both ligand and protein spectra upon binding [67, 68].
Conformational transitions of proteins of physiological relevance Protein interactions are fundamental components of physiological and pathological processes. Knowledge of the protein interactions in a cell provides information on the structure and function of cell components. Impressive progress in different techniques has allowed the rapid acquisition of data relating to the description of the interactome (i.e, all possible interactions between biological molecules in a cell) [69]. In proteomics, the interactome refers to the protein–protein interaction network. The interactome can be studied by the two-hybrid technique [70], by immuno-coprecipitation [71], and by physicochemical studies of the binding between two or three proteins [72–76]. The latter approach is mainly achieved in the laboratory via surface plasmon resonance measurements [72, 73], affinity chromatography, and mass spectrometry [76]. Furthermore, there is a great effort to characterize the interactome through bioinformatic studies [69, 77–79]. Quite often, molecular recognition and complex stability are driven by conformational transitions. Thus, CD is becoming increasingly useful in this area, particularly due to the possibility of monitoring conformational transitions when the secondary structural changes are functional in the physiological or pathological process. Recently, the study of conformational transitions of peptides and proteins has gained considerable attention because of their importance as a molecular key event in a variety of degenerative diseases, such as Alzheimer’s disease, Parkinson’s disease, Huntington’s disease, amyotrophic lateral sclerosis, and prion diseases [80]. These neurodegenerative diseases are increasingly being found to have common cellular and molecular mechanisms, including protein aggregation. The aggregates usually consist of fibers containing misfolded protein with a beta-sheet conformation, termed amyloid.
162
C. Bertucci et al.
Among these processes, amyloid peptide fibril formation during Alzheimer’s disease has been proven to start with the adoption of an amyloidogenic conformation, which acts as a trigger for peptide aggregation and then fibril formation. Circular dichroism was extensively used to monitor peptide conformational changes [81–85]. As an example, this spectroscopy was successfully used to get information on the self-aggregation process of the βamyloid (1–42) peptide when concentrations, solvents and the temperature were varied [86], and on the aggregation of the β-amyloid (1–40) induced by the interaction of the peptide with a protein, as in the case of the oligomerization process induced by human recombinant acetylcholinesterase (AChE) [87]. CD has the advantage of being able to characterize the kinetics of the secondary structural change from the beginning of the peptide aggregation process and before fibril formation. The aggregation process is studied by several techniques, including light scattering, [88], turbidimetry [89, 90], X-ray diffraction [91, 92], atomic force microscopy [93], and imaging techniques such as electron microscopy [94, 95]. Therefore, CD can easily be used to follow the conformation transition associated with fibril formation, and to screen for new inhibitors of the selfaggregation process. Screening can also be performed for new AChE inhibitors, such as drugs that are capable of not only reducing the symptoms of AD but also preventing or delaying the degeneration of cholinergic neurons. Furthermore, these studies yield information on the mechanism of the inhibition process by discriminating between inhibitors that act in the lag phase and those that act in the exponential phase (Fig. 6). During the lag phase, the peptide maintains a non-amyloidogenic conformation, so an inhibitor that prolongs this phase will slow the aggregation process significantly. The inhibitor can also act on the exponential phase, thus reducing the rate of conformational transition to the beta-sheet form and insoluble fibril formation [86, 87].
Absolute CD signal
Plateau
Exponential growth phase
Lag phase
Time Fig. 6 The three phases of the secondary structural switch of amyloidogenic peptides
Another interesting example of the application of CD for monitoring conformational changes is the study of homo- and heterodimer formation, processes that are fundamental for modulating physiologic and pathologic processes. CD was used to monitor the complexation of two mimetic peptides. The entry of herpes simplex virus 1 (HSV-1) into cells, and cell–cell fusion mediated by HSV-1 glycoproteins, require four glycoproteins: gD, gB, gH, and gL. Among these, gH is the only one that has so far exhibited structural–functional features typical of viral fusion glycoproteins: a candidate fusion peptide, and— downstream of this—a heptad repeat (HR) segment that is able to form a coiled coil, named HR-1 [96]. gH was proven to carry a functional HR-2 that is capable of physical interaction with HR-1. Specifically, mutational analysis of gH aimed at increasing or decreasing the ability of HR-2 to form a coiled coil resulted in an increase or decrease in fusion activity, respectively. A mimetic peptide with the HR-2 sequence inhibited HSV-1 infection in a specific and dose-dependent manner [97, 98]. CD spectroscopy showed that both HR-2 and HR-1 mimetic peptides adopt mainly random conformations in aqueous solution, while a decrease in peptide environmental polarity causes a conformational change, with a significant increase in the α-helical conformation content, particularly for the HR-1 peptide. A less polar environment was obtained by adding up to 15% TFE to the buffer solution. A significant conformational change was observed for HR-1–25 upon adding TFE. The shape of the spectrum— clearly that of an unordered structure in aqueous buffer— changed, and two negative bands appeared at 208 nm and 222 nm (characteristic of an α-helix structure) at a TFE concentration of 15% [98]. In contrast, HR-2–25 underwent minor conformational changes during titration, and even at 15% TFE the spectrum indicated a mostly disordered structure. These results were in agreement with the results of a bioinformatic study on the N-terminal domain of the protein. Once the propensities of the isolated peptides to assume ordered structures were elucidated, the interaction between the two peptides was studied via difference CD spectra. A non-negligible signal was observed (about 12% of the signal from the mixture at 15% TFE), which was less than usually detected in analogous models [99, 100]. Interestingly, superimposing the spectra recorded at increasing concentrations of TFE indicates—for each peptide and for the mixture—a clean, conserved isodichroic point at 203 nm, which points to a two-state transition from a random coil (disordered) conformation to one with a high α-helical content. Furthermore, HR-1 and HR2 mimetic peptides formed a stable complex, as revealed in nondenaturing electrophoresis and by CD [97].
Circular dichroism in drug discovery and development: an abridged review
The utility of CD spectroscopy in the development of therapeutic peptides is widely documented. Since the end of the 1960s, when the peptide market emerged, CD has been used to study molecules of great pharmacological interest, such as the antibiotic gramicidin [101] and oxytocin [102]. The use of proteins or peptides as pharmaceutical drugs is rapidly increasing due to their proven efficacy for the treatment of different diseases. Peptides and proteins are mediators of very important biological functions, and their high biological activities are associated with low toxicity because of their very selective binding to functional molecular targets. However, the instability of these drugs can compromise their efficacy. Thus, CD represents an excellent analytical tool for studying pharmaceutical peptide stability under different conditions and shedding light on mechanisms of binding to cellular membrane or targets. Insulin (Humulin®, Regular Iletin I®, and Regular Iletin II®) and exenatide (Byetta®) are two peptides involved in the treatment of diabetes: insulin works by lowering levels of glucose in the blood, while exenatide is used to treat type II (non-insulin-dependent) diabetes. These drugs were studied by CD in order to evaluate the the dependence of their conformation and aggregation state on the medium. CD allowed the effect of agitation on the secondary structure of insulin in solution and in the liquid crystalline cubic phase gel of glyceryl monooleate (GMO, used as a chemical stability enhancer) to be investigated. The optical density value obtained suggested that insulin did not aggregate in the gel phase, while it showed a total loss of native conformation in solution. Therefore, the gel phase has the ability to inhibit insulin aggregation and subsequent precipitation [103]. For exenatide, CD studies established the high medium dependence of its structural propensities. In fact, CD and NMR data indicated that higher concentrations led to helix bundle formation, and highlighted the residues involved in the helix/helix interaction [104]. CD was also used for therapeutic antibody characterization. CD studies pointed out the relationship between the structural change induced by the freeze-drying process and a loss of the in vivo immunogenic properties of MMA 383. This is an antiidiotypic antibody that was designed to be an immunogenic surrogate for the Lewis Y carbohydrate, a potential cancer antigen whose specificity is related to its expression on the surfaces of most cancer cells of epithelial origin (its expression is limited in normal tissues). CD spectra recorded from lyophilized and nonlyophilized MMA 383 antibodies were indistinguishable when recorded between room temperature and 50 °C, while a significant difference was observed upon lowering the temperature to 11 °C. In particular, the CD spectral changes were more pronounced for the nonlyophilized antibodies than for the lyophilized antibodies. These results suggest that lyophilization induces a decrease in the flexibility of MMA 383 [105].
163
Another interesting CD application concerns the study of how the structural flexibility of Polymyxin B relates to its high selectivity. The structural changes observed for this potent and selective antimicrobial agent (Polytrim®) at the interfaces of anionic phospholipid vesicles were consistent with those registered in aqueous Tris buffer solution upon the addition of TFE. Polymyxin B CD spectra recorded under both conditions suggested the adoption of a more ordered structure, in agreement with its hypothesized membrane-mimetic behavior. CD and NMR studies suggested that the topological flexibility of Polymyxin B may represent a critical element that allows different functions to be performed depending on the local environment of the bacterial cell surface [106]. Lepirudin is a recombinant hirudin that is used as an anticoagulant when heparins are contraindicated due to heparin-induced thrombocytopenia. Hirudin is a potent and specific inhibitor of thrombin that was first obtained from the salivary gland of the European leech [107]. Information on the mechanism of its recognition process was obtained by measuring the CD spectra of the individual peptides and their complex. In particular, the CD spectrum of the hirudin–thrombin complex was not additive with respect to the individual spectra of thrombin and hirudin. Dissociation of hirudin aggregates was considered to drive the conformational changes monitored during the complex formation. One interesting example of the use of CD spectroscopy to monitor the modulation of peptide–protein interactions is a study of the binding mechanism of enfuvirtide (Fuzeon®)—a potent inhibitor of the entry of HIV-1 into its host cell—with gp41, a transmembrane envelope glycoprotein that modulates fusion between the viral and cellular membranes. In its active state, the core of gp41 consists of a trimer of heterodimers comprising a leucine/isoleucine zipper sequence. CD spectra suggested that enfuvirtide influences the leucine/isoleucine zipper-like sequence without perturbing the trimer of heterodimers conformation. Therefore, it was possible to determine that enfuvirtide acts by binding gp41 in an intermediate state, thus inhibiting the glycoprotein’s transition to the fusionactive conformation [108].
Conclusions and perspectives Circular dichroism is now widely used in all phases of drug discovery and development. Specific information can be obtained from it to assess the stereochemistries of chiral drugs; their absolute configurations can be accurately and reliably determined using software for ab initio calculations. Furthermore, analysis of ICD spectra makes it easy to stereochemically characterize the drug when bound to the target protein. Finally, monitoring the
164
conformational transitions of peptides and proteins yields information on oligomerization processes and molecular recognition phenomena, both of which are often key steps in physiological and pathological events. Circular dichroism will surely find extensive use in the relatively new research field of the interactome. The simultaneous detection of circular dichroism, linear dichroism, absorption, turbidity, light scattering and fluorescence [109] should place new emphasis on using polarization modulation spectrometry to better characterize biological structures. As an example, it should be possible to analyze aggregates [110], membrane proteins and dynamic processes while taking into account anisotropic artefacts. Even more importantly, the technical development of CD instrumentation that can be reliably applied to miniaturized systems would allow the application of this spectroscopy to the nano world. This would provide a valuable complementary technique to those based on the immobilization of target proteins, as shown by the combined use of CD and surface plasmon resonance in a study of the folding of protein monolayers [111]. Acknowledgments The work was supported by a grant from MIUR, Italy (PRIN 2008 National Program), and by the University of Bologna.
References 1. Aboul-Enein HY, Wainer IW (eds) (1997) The impact of stereochemistry on drug development and use (Chemical Analysis Series). Wiley, New York 2. Ariens EJ (1984) Stereochemistry, a basis for sophisticated nonsense in pharmacokinetics and clinical pharmacology. Eur J Clin Pharmacol 26:663–666 3. Nakanishi K, Berova N, Woody RW (eds)(1994) Circular dichroism: principles and applications. VCH, New York 4. Ascoli GA, Domenici E, Bertucci C (2006) Drug binding to human serum albumin: abridged review of results obtained with high-performance liquid chromatography and circular dichroism. Chirality 18(9):667–679 5. Stephens PJ, Devlin FJ, Pan JJ (2008) The determination of the absolute configuration of chiral molecules using vibrational circular dichroism (VCD) spectroscopy. Chirality 20(5):643–663 6. Urbanova M (2009) Bioinspired interactions studied by vibrational circular dichroism. Chirality 21:E215–E230 7. Bijvoet JM, Peerdeman AF, van Bommel AJ (1951) Determination of the absolute configuration of optically active compounds by means of X-rays. Nature 168:271–272 8. Flack HD (1983) On enantiomorph-polarity estimation. Acta Crystallogr A 39:876–881 9. Cavalli A, Bisi A, Bertucci C, Rosini C, Paluszcak A, Gobbi S, Giorgio E, Rampa A, Belluti F, Piazzi L, Valenti PW, Hartmann R, Recanatini M (2005) Enantioselective nonsteroidal aromatase inhibitors identified through a multidisciplinary medicinal chemistry approach. J Med Chem 48:7282–7289 10. Bringmann G, Gulder TAM, Reichert M, Gulder T (2008) The online assignment of the absolute configuration of natural products: HPLC-CD in combination with quantum chemical CD calculations. Chirality 20:628–642
C. Bertucci et al. 11. Iwasa K, Takahashi T, Nishiyama Y, Moriyasu M, Sugiura M, Takeuchi A, Tode C, Tpkuda H, Takeda K (2008) Online structural elucidation of alkaloids and other constituents in crude extracts and cultured cells of Nandina domestica by combination of LC-MS/ MS, LC-NMR. and LC-CD analyses. J Nat Prod 71:1376–1385 12. Bertucci C, Bonato GLF, PS BKB, Okano LT, Mazzeo G, Rosini C (2010) Assignment of the absolute configuration at the sulfur atom of thioridazine metabolites by the analysis of their chiroptical properties: the case of thioridazine 2-sulfoxide. J Pharm Biomed Anal 52(5):796–801 13. Harada N (2008) Determinationof absolute configurations by X-ray crystallography and 1H NMR anisotropy. Chirality 20:691–723 14. Owen CP, Nicholls PJ, Smith HJ, Whomsley R (1999) Inhibition of aromatase (P450Arom) by some 1-(benzofuran-2-ylmethyl) imidazoles. J Pharm Pharmacol 51:427–433 15. Recanatini M, Cavalli A, Valenti P (2002) Nonsteroidal aromatase inhibitors: recent advances. Med Res Rev 22:282–304 16. Recanatini M (1996) Comparative molecular field analysis of non-steroidal aromatase inhibitors related to fadrozole. J Comput Aided Mol Des 10:74–82 17. Mason SF (1962) Optical rotatory power. Quart Rev 17–20 18. Mason SF (1967) Theory II. Optical rotatory dispersion and circular dichroism in organic chemistry (Chap 4). Heyden and Son, London, p 71 19. Gottarelli G, Mason SF, Torre G (1971) The circular dichroism and absolute configuration of (+)-trans-stilbene oxide. J Chem Soc B 7:1349–1353 20. Harada N, Nakanishi K (1972) The exciton chirality method and its application to configurational and conformational studies of natural products. Acc Chem Res 5:257–263 21. Mason SF (1982) Molecular optical activity and the chiral discriminations. Cambridge University Press, Cambridge 22. Harada N, Nakanishi K (1983) Circular dichroic spectroscopy: exciton coupling in organic stereochemistry. University Science Books, Mill Valley 23. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA et al (2003) Gaussian 03. Gaussian, Inc., Pittsburgh 24. Helgaker T, Jensen HJA, Joergensen P, Olsen J, Ruud K et al. (2001) DALTON, a molecular electronic structure program, v.1.2. http://www.daltonprogram.org/ 25. Ahlrichs R, Bar M, Baron HP, Bauernschmitt R, Bocker S et al (2002) Turbomole v.5.6. Universitat Karlsruhe, Karlsruhe 26. Polavarapu PL (1997) Ab initio molecular optical rotations and absolute configurations. Mol Phys 91:551–554 27. Kondru RK, Wipf P, Beratan DN (1998) Theory-assisted determination of absolute stereochemistry for complex natural products via computation of molar rotation angles. J Am Chem Soc 120:2204–2205 28. Stephens PJ, Devlin FJ, Cheeseman JR, Frisch MJ (2001) Calculation of optical rotation using density functional theory. J Phys Chem A 105:5356–5371 29. Polavarapu PL (2002) Optical rotation: recent advances in determining the absolute configuration. Chirality 14:768–781 30. Diedrich C, Grimme S (2003) Systematic investigation of modern quantum chemical methods to predict electronic circular dichroism spectra. J Phys Chem A 107:2524–2539 31. Pedersen TB, Koch H (2000) Theoretical electronic absorption and natural circular dichroism spectra of (−)-trans-cycloctene. J Chem Phys 112:2139 32. Furche F, Ahlrichs R, Wachsmann C, Weber E, Sobanski A et al (2000) Circular dichroism of helicenes investigated by timedependent density functional theory. J Am Chem Soc 122:1717–1724 33. Braun M, Hohmann A, Rahematpura J, Buehne C, Grimme S (2004) Synthesis and determination of the absolute configuration of fugomycin and desoxyfugomycin: CD spectroscopy and fungicidal activity of butenolides. Chem Eur J 10:4584–4593
Circular dichroism in drug discovery and development: an abridged review 34. Halgren TA (1996) Merck molecular force field. I. Basis, form, scope, parameterization, and performance of MMFF94. J Comput Chem 17:490–519 35. Wavefunction Inc. (2002) SPARTAN ’02. Wavefunction Inc., Irvine 36. Devlin FJ, Stephens PJ, Besse P (2005) Are the absolute configurations of 2-(1-hydroxyethyl)-chromen-4-one and its 6bromo derivative determined by X-ray crystallography correct? A vibrational circular dichroism study of their acetate derivatives. Tetrahedron Asymmetr 16:1557–1566 37. Mason SF (1962) The absolute configuration of calycanthine. Proc Chem Soc 362–363 38. Harada N, Nakanishi K (1983) Circular dichroic spectroscopy— exciton coupling in organic stereochemistry. University Science Books, Mill Valley 39. Harada N, Nakanishi K (1972) Exciton chirality method and its application to configurational and conformational studies of natural products. Account Chem Res 5:257–263 40. Salvadori P, Bertucci C, Rosini C, Zandomeneghi M, Gallo G, Martinelli E, Ferraris P (1972) Circular dichroism of rifamycin S. J Am Chem Soc 103:5553–5557 41. Hodgson J (2001) ADMET— turning chemicals into drugs. Nat Biotech 19:722–726 42. Peters T Jr (1996) All about albumin, biochemistry, genetics, and medical applications. Academic, New York 43. Kragh-Hansen U, Chuang VTG, Otagiri M (2002) Practical aspects of the ligand-binding and enzymatic properties of human serum albumin. Biol Pharm Bull 25:695–704 44. Wainer IW (1994) Enantioselective high-performance liquid affinity chromatography as a probe of ligand–biopolymer interactions: an overview of a different use for highperformance liquid chromatographic chiral stationary phases. J Chromatogr A 666:221–234 45. Hage DS (2002) High-performance affinity chromatography: a powerful tool for studying serum protein binding. J Chromatogr B 768:3–30 46. Bertucci C, Domenici E (2002) Reversible and covalent binding of drugs to human serum albumin: methodological approaches and physiological relevance. Curr Med Chem 9:1463–1481 47. Ascoli GA, Domenici E, Bertucci C (2006) Drug binding to human serum albumin: abridged review of results obtained with high-performance liquid chromatography and circular dichroism. Chirality 18:667–679 48. Bertucci C, Domenici E, Salvadori P (1990) Stereochemical features of 1,4-benzodiazepin-2-ones bound to human serum albumin: difference circular dichroism and UV studies. Chirality 2:167–174 49. Ascoli G, Bertucci C, Salvadori P (1995) Stereospecific and competitive binding of drugs to human serum albumin: a difference circular dichroism approach. J Pharm Sci 84:737–741 50. Pistolozzi M, Bertucci C (2008) Species-dependent stereoselective drug binding to albumin: a circular dichroism study. Chirality 20:552–558 51. Kaneko K, Fukuda H, Chuang VT, Yamasaki K, Kawahara K, Nakayama H, Suenaga A, Maruyama T, Otagiri M (2008) Subdomain IIIA of dog albumin contains a binding site similar to site II of human albumin. Drug Metab Dispos 36 (1):81–86 52. Sudlow G, Birkett DJ, Wade DN (1975) The characterization of two specific drug binding sites on human serum albumin. Mol Pharmacol 11:824–832 53. Robertson A, Karp W, Brodersen R (1990) Comparison of the binding characteristic of serum albumins from various animal species. Dev Pharmacol Ther 15:106–111 54. Kosa T, Maruyama T, Sakai N, Yonemura N, Yahara S, Otagiri M (1998) Species difference of serum albumins: III. Analysis of structural characteristics and ligand binding properties during NB transitions. Pharm Res 15:592–598
165 55. Fitos I, Visy J, Simonyi M (2002) Species-dependency in chiraldrug recognition of serum albumin studied by chromatographic methods. J Biochem Biophys Methods 54:71–84 56. Dockal M, Carter DC, Ruker F (1999) The three recombinant domains of human serum albumin. J Biol Chem 41:29303– 29310 57. Massolini G, De Lorenzi E, Ponci MC, Caccialanza G (1996) Comparison of drug binding sites on rat and human serum albumins using immobilized-protein stationary phases as a tool for the selection of suitable animal models in pharmacological studies. Boll Chim Farm 135:382–386 58. Massolini G, Aubry AF, McGann A, Wainer IW (1993) Determination of the magnitude and enantioselectivity of ligand binding to rat and rabbit serum albumins using immobilizedprotein high performance liquid chromatography stationary phases. Biochem Pharmacol 46:1285–1293 59. Zandomeneghi M (1995) Circular dichroism of ketoprofen complexed to serum albumins: conformational selection by the protein: a novel optical purity determination technique. Chirality 7:446–451 60. Ghislandi V, La Manna A, Azzolina O, Gazzaniga A, Vercesi D (1982) Configurational relationships in antiphogistic hydratropic acids. Farmaco 37:81–93 61. Lightner DA, Wijekoon WM, Zhang MH (1988) Understanding bilirubin conformation and binding. Circular dichroism of human serum albumin complexes with bilirubin and its esters. J Biol Chem 263:16669–16676 62. Petersen CE, Ha C-E, Harohalli K, Feix JB, Bhagavan NV (2000) A dynamic model for bilirubin binding to human serum albumin. J Biol Chem 275:20985–20995 63. Boiadjiev SE, Lightner DA (1997) Exciton chirality of bilirubin homologs. Chirality 9:604–615 64. Bertucci C, Viegi A, Ascoli G, Salvadori P (1995) Protein binding investigation by difference circular dichroism: native and acetylated human serum albumin. Chirality 7:57–61 65. Bertucci C, Nanni B, Salvadori P (1999) Reversible binding of ethacrynic acid to human serum albumin: difference circular dichroism study. Chirality 11:33–38 66. Kelly SM, Price NC (2000) The use of circular dichroism in the investigation of protein structure and function. Curr Protein Pept Sci 1(4):349–384 67. Siligardi G, Hussain R (1998) Biomolecules interactions and competitions by non-immobilised ligand interaction assay by circular dichroism. Enantiomer 3(2):77–87 68. Rodi DJ, DJ JRW, Sanganee HJ, Holton RA, Fallace BA, Makowski L (1999) Screening of a library of phage-displayed peptides identifies human bcl-2 as a taxol-binding protein. J Mol Biol 285:197–203 69. Chua HN, Wong L (2008) Increasing the reliability of protein interactomes. Drug Discov Today 13(15/16):652–658 70. Fields S, Song O (1989) A novel genetic system to detect protein–protein interactions. Nature 340:245–246 71. El-Samalouti VT, Schletter J, Brade H, Brade L, Kusumoto S, Rietschel ET, Flad HD, Ulmer AJ (1997) Detection of lipopolysaccharide (LPS)binding membrane proteins by immuno-coprecipitation with LPS and anti-LPS antibodies. Eur J Biochem 250:418–424 72. Roman I, Figys J, Steurs G, Zizi M (2006) Direct measurement of VDAC–actin interaction by surface plasmon resonance. Biochim Biophys Acta 1758:479–486 73. Royer-Zemmour B, Ponsole-Lenfant M, Gara H, Roll P, Leveque C, Massacrier A, Ferracci G, Cillario J, RobagliaSchlupp A, Vincentelli R, Cau P, Szepetowski P (2008) Epileptic and developmental disorders of the speech cortex: ligand/receptor interaction of wild-type and mutant SRPX2 with the plasminogen activator receptor uPAR. Hum Mol Genet 17(23):3617–3630
166 74. Ortega-Roldan JL, Ringkjøbing Jensen M, Brutscher B, Azuaga AI, Blackledge M, van Nuland NAJ (2009) Accurate characterization of weak macromolecular interactions by titration of NMR residual dipolar couplings: application to the CD2AP SH3-C:ubiquitin complex. Nucleic Acids Res 37(9):1–12 75. Park S, Lim BBC, Perez-Terzic C, Mer G, Terzic A (2008) Interaction of asymmetric ABCC9-encoded nucleotide binding domains determines KATP channel SUR2A catalytic activity. J Proteome Res 7(4):1721–1728 76. Downard KM (2006) Ions of the interactome: the role of MS in the study of protein interactions in proteomics and structural biology. Proteomics 6:5374–5384 77. Martin S, Roe D, Faulon JL (2005) Predicting protein–protein interactions using signature products. Bioinformatics 21:218– 226 78. Saito R, Suzuki H, Hayashizaki Y (2003) Construction of reliable protein–protein interaction networks with a new interaction generality measure. Bioinformatics 19:756–763 79. Chen J, Hsu W, Lee ML, Ng SK (2006) Increasing confidence of protein interactomes using network topological metrics. Bioinformatics 22:1998–2004 80. Ross CA, Poirier MA (2004) Protein aggregation and neurodegenerative disease. Nat Med 10(Suppl):S10–S17 81. Barrow CJ, Zagorski MG (1991) Solution structures of beta peptide and its constituent fragments: relation to amyloid deposition. Science 253:179–182 82. Barrow CJ, Yasuda A, Kenny PT, Zagorski MG (1992) Solution conformations and aggregational properties of synthetic amyloid β-peptides of Alzheimer’s disease: analysis of circular dichroism spectra. J Mol Biol 225:1075–1093 83. Otvos L, Szendrei GI, Lee VM, Mantsch HH (1993) Human and rodent Alzheimer beta-amyloid peptides acquire distinct conformations in membrane-mimicking solvents. Eur J Biochem 211:249–257 84. Tomaselli S, Esposito V, Vangone P, van Nuland NAJ, Bonvin AMJJ, Guerrini R, Tancredi T, Temessi PA, Picone D (2006) The α to β conformational transition of Alzheimer’s Aβ-(1-42) peptide in aqueous media is reversible: a step by step conformational analysis suggests the location of β conformation seeding. Chembiochem 7:257–267 85. Fezoui Y, Teplow DB (2002) Kinetic studies of amyloid βprotein fibril assembly: differential effects of α-helix stabilization. J Biol Chem 277:36948–36954 86. Bartolini M, Bertucci C, Bolognesi ML, Cavalli A, Melchiorre C, Andrisano V (2007) Insight into the kinetics of amyloid β (1– 42) peptide self-aggregation: elucidation of inhibitors’ mechanism of action. Chembiochem 8:2152–2161 87. Bartolini M, Bertucci C, Cavrini V, Andrisano V (2003) Betaamyloid aggregation induced by human acetylcholinesterase: inhibition studies. Biochem Pharmacol 65:407–416 88. Lomakin A, Chung DS, Benedek GB, Kirschner DA, Teplow DB (1996) On the nucleation and growth of amyloid b-protein fibrils: detection of nuclei and quantitation of rate constants. Proc Natl Acad Sci USA 93:1125–1129 89. Pai A, Rubinstein I, Onyuksel H (2006) PEGylated phospholipid nanomicelles interact with beta-amyloid(1–42) and mitigate its beta-sheet formation, aggregation and neurotoxicity in vitro. Peptides 27:2858–2866 90. Wang SS, Chen YT, Chou SW (2005) Inhibition of amyloid fibril formation of beta-amyloid peptides via the amphiphilic surfactants. Biochim Biophys Acta Mol Basis Dis 1741:307– 313 91. Inouye H, Fraser PE, Kirschner DA (1993) Structure of betacrystallite assemblies formed by Alzheimer beta-amyloid protein analogues: analysis by X-ray diffraction. Biophys J 64:502–519
C. Bertucci et al. 92. Fraser PE, McLachlan DP, Surewicz WK, Mizzen CA, Snow AD, Nguyen JT, Kirschner DA (1994) Conformation and fibrillogenesis of Alzheimer A beta peptides with selected substitution of charged residues. J Mol Biol 244:64–73 93. Harper JD, Wong SS, Lieber CM, Lansbury PT Jr (1997) Observation of metastable Abeta amyloid protofibrils by atomic force microscopy. Chem Biol 4:119–125 94. Teplow DB (1998) Structural and kinetic features of amyloid beta-protein fibrillogenesis. Amyloid 5:121–142 95. Sunde M, Blake C (1997) The structure of amyloid fibrils by electron microscopy and X-ray diffraction. Adv Protein Chem 50:123–159 96. Campadelli-Fiume G, Amasio M, Avitabile E, Cerretani A, Forghieri C, Gianni T, Menotti L (2007) The multipartite system that mediates entry of herpes simplex virus into the cell. Rev Med Virol 17(5):313–326 97. Gianni T, Piccoli A, Bertucci C, Campadelli-Fiume G (2006) Heptad repeat 2 in herpes simplex virus 1 gH interacts with heptad repeat 1 and is critical for virus entry and fusion. J Virol 80:2216–2224 98. Sanavio B, Piccoli A, Gianni T, Bertucci C (2007) Helicity propensity and interaction of synthetic peptides from heptadrepeat domains of herpes simplex virus 1 glycoprotein H: a circular dichroism study. Biochim Biophys Acta 1774(7):781– 791 99. Liu S, Xiao G, Chen Y, He Y, Niu J, Escalante CR, Xiong H, Farmar J, Debnath AK, Tien P, Jiang S (2004) Interaction between heptad repeat 1 and 2 regions in spike protein of SARS-associated coronavirus: implications for virus fusogenic mechanism and identification of fusion inhibitors. Lancet 363:938–947 100. Tripet B, Howard MW, Jobling M, Holmes RK, Holmes KV, Hodges RS (2004) Structural characterization of the SARScoronavirus spike S fusion protein core. J Biol Chem 279:20836–20849 101. Quadrifoglio F, Urry DW (1967) Circular dichroism and optical rotatory dispersion of gramicidin S in aqueous solution. Biochem Biophys Res Commun 29(6):785–791 102. Beychok S, Breslov E (1968) Circular dichroism of oxytocin and several oxytocin analogues. J Biol Chem 243(1):151–154 103. Sadhale Y, Shah JC (1999) Stabilization of insulin against agitation-induced aggregation by the GMO cubic phase gel. Int J Pharm 191:51–64 104. Hudson FM, Handersen NH (2004) Exenatide: NMR/CD evaluation of the medium dependence of conformation and aggregation state. Biopolymers 76:298–308 105. Taschner N, Müller SA, Alumella VR, Goldie KN, Drake AF, Aebi U, Arvinte T (2001) Modulation of antigenicity related to changes in antibody flexibility upon lyophilization. J Mol Biol 310:169–179 106. Bruch MD, Cajal Y, Koh JT, Jain MK (1999) Higher-order structure of Polymixin B: the functional significance of topological flexibility. J Am Chem Soc 121:11993–12004 107. Konno S, Fenton JW, Villanueva GB (1988) Analysis of the secondary structure of hirudin and the mechanism of its interaction with thrombin. Arch Biochem Biophys 267:158– 166 108. K liger Y, Shai Y (2000) Inhibition of HIV-1 entry before gp41 folds into its fusion-active conformation. J Mol Biol 295:163–168 109. Arvinte T, Bui TTT, Dahab AA, Demeule B, Drake AF, Elhag D, King P (2004) The multi-mode polarization modulation spectrometer. Part 1: simultaneous detection of absorption, turbidity, and optical activity. Anal Biochem 332:46–57 110. Demeule B, Lawrence MJ, Drake AF, Gurny R, Arvinte T (2006) Characterization of protein aggregation: the case of a therapeutic immunoglobulin. Biochim Biophys Acta 1774(1):146–153 111. Keegan N, Wright NG, Lakey JH (2005) Circular dichroism spectroscopy of folding in a protein monolayer. Angew Chem 117:4879–4882