ionization time-of-flight mass spectrometry and hydrogen exchange combined with enzymatic digestion for the structural characterization of antimalaric Spf66 peptide

Talanta 72 (2007) 1192–1198

Application of matrix-assisted laser desorption/ionization time-of-flight mass spectrometry and hydrogen exchange combined with enzymatic digestion for the structural characterization of antimalaric Spf66 peptide Alexis Oliva ∗ , Mat´ıas Llabr´es, Jos´e B. Fari˜na Departamento de Ingenier´ıa Qu´ımica y Tecnolog´ıa Farmac´eutica, Facultad de Farmacia, Universidad de La Laguna, 38200, La Laguna, Tenerife, Spain Received 21 September 2006; received in revised form 29 December 2006; accepted 9 January 2007 Available online 14 January 2007

Abstract Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry was used in hydrogen exchange studies, exchanging deuteron (H/D) or proton (D/H), to determine the structure and conformational changes of antimalarial Spf66 synthetic peptide in its monomeric and dimeric forms. The accuracy of both analytical methods was assessed along with their suitability to study structural aspects. The results via these two approaches were in agreement, indicating that the dimer presents segments of secondary structure. In this last case, the combination of both methods with enzymatic digestion with pepsin was used in their identification. Although 100% coverage of Spf66 dimer was not observed, the higher levels of deuteration were observed for fragments located in the chain terminal where the structure may be more flexible, while the fragments near the disulfide bonds, which is, in theory, the more rigid region of the molecule, were not detected. This strategy is significantly time saving and allows rapid screening and help to characterize a protein, especially, when no prior structural information is available. However, a single spectrum is not certainly sufficient to obtain structural data; it is just an experimental limitation. Also, changes in peptide structure after storage at different temperatures and time were observed, which lead to a loss in the secondary structure as determined by circular dicroism measurements and an increase in aggregation products, since the trimer and tetramer species were detected by mass spectrometry. © 2007 Elsevier B.V. All rights reserved. Keywords: MALDI-TOF MS; Synthetic peptide; Secondary structure; Hydrogen exchange; Conformational stability

1. Introduction Protein and peptides production through recombinant techniques has benefited from numerous years of successful research, and is widely accepted as the approach of choice. Totally chemical synthesis is another mode of polypeptide production, since considerable progress has been made in terms of the lengths of polypeptides, which can be manufactured entirely by chemical means. Synthesis techniques have been developed and optimized that allow the routine production of polypeptides of up to 130 residues in milligram quantities. In addition, with the advance of processes permitting efficient formation of disulfide bonds, these polypeptides can be produced and isolated as folded, native-like structures [1].

∗

Corresponding author. Tel.: +34 922 318 451; fax: +34 922 318 514. E-mail address: [email protected] (A. Oliva).

0039-9140/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2007.01.012

In recent years, there has been increasing interest in de novo design and construction of novel synthetic peptides that mimic protein secondary structures, i.e., turns, helixes and sheets. The unique structural influences of unsubstituted, non-coded, non-chiral-␤-amino acids, e.g., ␤-alanine on the peptide backbone, when inserted into a peptide chain comprising ␣-amino acids, offer an excellent opportunity to design and build diverse well-defined three-dimensional structures [2]. Linear peptides are highly flexible molecules that can adopt a multiplicity of conformations in solution, but only a few are responsible for their own immunoreactivity. Cyclic peptides mimic native secondary structures better than linear ones, as they induce spatial arrangements that reproduce bioactive conformations, resulting in enhanced binding and improved immunological properties [3,4]. Many methods for structural analysis of peptides and proteins have been reported. Traditional techniques like X-ray crystallography, nuclear magnetic resonance spectroscopy or differential

A. Oliva et al. / Talanta 72 (2007) 1192–1198

scanning calorimetry often consume large amounts of materials and time. Other limitations are the inability to study large proteins and difficulties in producing crystalline proteins. Circular dicroism (CD) is a valuable tool for evaluation of protein and peptide conformation in solution or monitoring conformational changes [5,6]. Today, several methods for secondary structure calculations from far-UV CD spectra exist. These methods are based on different mathematical algorithms and use various sets of reference proteins with known crystallographic structure. The accuracy of these approaches must be confirmed by other techniques, such as NMR, X-ray, IR, etc. [6]. Matrix-assisted laser desorption/ionization (MALDI) coupled with time-of-flight (TOF) mass analysis is now widely used in analysis of biological molecules, and recently, to study protein folding and conformational changes. Hydrogen exchange of amide backbone protons has already been used for many years to study different aspects of protein structure [7]. The method is based on the fact that the hydrogen/deuterium (H/D) exchange rate of amide protons located in the peptide backbone depends on whether they are participant in intramolecular hydrogen bonding, and on the extent to which they are shielded from the solvent [8]. Two features make MALDI-TOF MS well suited for the measurements described here: high resolution, to easily resolve the multiple isotopic peaks resulting from amide H/D exchange experiments, and high mass accuracy, to aid in identifying peptides resulting from the specific cleavage by pepsin or trypsin [7,8]. As a result of these advances, several peptides and proteins have been identified for specific therapeutic targets; however, the development of these macromolecules into stable formulations still remains a great challenge. Formulation problems in solution arise from the complex native structure of protein molecules and often manifest as physical instability, such as unfolding, aggregation, and/or precipitation [9–11]. Physical instability problems result mainly from perturbation of the native structure of the protein molecule, which can easily be caused by alterations in solution conditions, such as pH, temperature, and ionic strength, presence of co-solvents, and other additives. This physical degradation of the protein can occur during production and purification but also result from improper formulation or storage or handling conditions. Therefore, the choice of formulation and dosage form are crucial to guarantee the physical stability of the protein [12]. The antimalarial Spf66 synthetic peptide was the first chemically synthesized vaccine to induce a partial protective immune response to malaria in people older than 1 year of age [13–16]. Several trials have shown that this molecule is safe, immuno-

1193

genic and induces protection, although, in general, the protection level and incidence of immune responses are low and variable. Peptide structure, formulation and physical degradation are some of the factors that could affect the immune response and may explain the high variability in protection levels observed [13–16]. The goal of the present investigation was to evaluate the hydrogen/deuterium exchange monitored by MALDI-TOF mass spectrometry, combined with enzymatic digestion for determining synthetic peptide structure, in particular the antimalarial Spf66 peptide, which is a good example of a peptide with a repeating amino acid sequence. For this, the uncertainty of different analytical methods was calculated, especially, in the case of enzymatic digestion, a method for its statistical evaluation is proposed. The possible conformational changes as a function of temperature were also evaluated using both circular dicroism and exchange hydrogen monitored by MALDI-TOF mass spectrometry, since these can most affect their overall long-term stability and biological activity, where a small change at one site may result in a major change in overall properties. 2. Experimental 2.1. Chemical and reagents All chemical reagents used in the present study were of highest grade purity from commercial sources. Pepsin and all proteins used as standard were purchased from Sigma (St. Louis, MO). The monomer and dimer species from SPf66 peptide were synthesized at the Instituto de Inmunolog´ıa, San Juan de Dios, Bogot´a, Colombia, under GMP conditions in accordance with the t-Bock technique [17]. The amino acid sequence of the SPf66 monomer species is: GANKKNAPPNANPLVMKEKQFLSYPNANPAAYVNQTEAELEDG. The SPf66 dimer species was synthesized with a cysteine residue introduced at the amino-terminal end, which enables polymerization of the molecule through an oxidation mechanism, obtaining a final product composed of two individual peptides joined by disulfide bridges (Fig. 1). Both species were characterized by HPLC using UV–vis and multi-angle laser light-scattering detection and MALDI-TOF mass spectrometry [18]. 2.2. Circular dicroism studies CD measurements were carried out using a Jasco J-600 spectropolarimeter, in a 10 mm path-length cell using a peptide concentration of 20 ␮g/mL and a scan speed of 20 nm/min from

Fig. 1. Amino acid sequence of SPf66 dimer species, indicating peptides with mass over 500 Da were obtained by pepsin digestion at pH > 2.

1194

A. Oliva et al. / Talanta 72 (2007) 1192–1198

190 to 300 nm. Each scan was the result of two accumulations. All spectra were recorded at room temperature, with a bandwidth of 1.0 nm and a step resolution of 1.0 nm, a sensitivity of 20 mdeg, and a time constant of 1 s. The K2d program [19] was used for the analysis of peptide secondary structure using circular dicroism data. 2.3. MALDI-TOF MS MALDI-TOF mass spectra were acquired on a Bruker Reflex III mass spectrometer equipped with a nitrogen laser with an emission wavelength of 337 nm, operating in the reflectronmode with an accelerating voltage of 20 kV. An external calibration was performed for each measurement using an appropriate standard protein. Each mass determination was averaged from three sample-standard pairs. All reported masses are monoisotopic [M + H]+ unless otherwise noted. The matrix used was 2,5-dihydroxybenzoic acid (DHB; Sigma) in a saturated solution containing water (0.1% TFA)– acetonitrile 1:1 by volume. 2.3.1. H/D experiments All experiments were carried out at room temperature in 0.5mL microcentrifuge capped tubes. Values given in the text for pH and pD were taken directly from the pHmeter, uncorrected for isotopic effect [20]. Lyophilized samples of both peptides were dissolved in D2 O (pD 5.5) to a concentration of 0.5 mg/mL. Aliquots were taken at different times, diluted 1:1 by volume in the matrix solution, and 10–20 pmol of analyte placed in the target, which was immediately placed in desiccators under a moderate vacuum so that the spots dried in 2–3 min, then finally, transferred as quickly as possible to the mass spectrometer for analysis. The matrix used was 2,5-dihydroxybenzoic acid in a saturated solution containing D2 O (0.1% TFA)–acetonitrile 1:1 (v/v). H/D exchange experiments for each time point were performed in triplicate. 2.3.2. D/H experiments Lyophilized aliquots of the peptide were dissolved in D2 O to a concentration of 0.5 mg/mL and incubated at 25 ◦ C until a complete exchange of the labile deuterons was attained. The native deuterated peptides were then allowed to exchange by dilution with nine volumes of deionized water. The rest of the conditions were as described earlier. 2.3.3. Enzyme digestion Lyophilized aliquots of the peptide were dissolved in D2 O to a concentration of 0.5 mg/mL and incubated at 25 ◦ C for 15 min to complete the exchange of labile deuterons, and the sample was then diluted with nine volumes of deionized water. Pepsin solutions were prepared as directed by the supplier and added to the peptide in a molar ratio of 1:1, at pH 2–3, and the resulting solution was incubated at 0 ◦ C for 5–10 min. MALDI-TOF mass spectrometry analysis was performed by taking 10 ␮L of the digested sample and diluting it with 10 ␮L of 1:1 ␣-cyano-4hydroxycinnamic acid matrix solution. The rest of conditions were as above. Samples were externally calibrated.

The digested fragments were compared with those of the peptides theoretically obtainable by cleavage at any two positions of the peptide sequence, in accordance with the cleavage rules of pepsin at pH > 2 (cleavage C-terminal side of F, L, W,Y, A, E, Q). All peptides with a mass above 500 Da were selected for further consideration, since many peptides with lower mass (<500 Da) cannot be identified by MALDI-TOF mass spectrometry, due to overlapping signals from the matrix [21], although several approaches to avoid or minimize interference via sample preparation and matrix selection as well as coupling of MALDI to liquid and planar chromatographic techniques have been proposed to extend its range of applicability [22]. A mass tolerance of ±2 Da was assumed. 3. Results and discussion In order to develop a method for monitoring hydrogen/deuteron exchange by MALDI-TOF MS, a series of trials were initially performed to optimize the sample preparation procedure in order to overcome back-exchange problems and attain reproducible results. For this, the SPf66 monomer species was chosen as a test compound for full exchange since all the labile protons (i.e., peptide backbone amide protons except proline, all side-chains and amino/carboxyl termini protons) are expected to be accessible to the solvent, due to its flexible and unordered structure [18]. This peptide, with a calculated molecular mass of 4643.5 Da, contains a total of 74 labile hydrogens. The measured average molecular mass of the exchanged species presents an increase of 71.5 Da over the average molecular mass of the unexchanged species, an exchange level of 96.6% after 1 h of incubation with D2 O, remaining constant up to the 24 h (Table 1). This data confirmed the previous results [18]. After 48 h, the result indicated a decreased deuteration percentage in the 1–4% range, whereas the analysis by MALDI-TOF mass spectrometry revealed the presence of aggregation product (Fig. 2). This result shows that the exchange reaction is considerable rapid, although the exchange time is not adequate and, possible more exchange time points in a much short time scale (i.e., 20 s, 1 min, etc.) should be used at kinetic studies [8,23], but this methodology was developed to rapid screening and help to characterize a protein or peptide when no prior structural information is available. To determine the loss of deuterons through the sample preparation and analysis, and back exchange, the uncertainty in deuteron incorporation was estimated. It was first expressed as a standard deviation, the so-called standard uncertainty and abbreviated as u(x). The expanded uncertainty U(x) defines an interval around the result of a measurement, x ± U(x), with U(x) = ku(x). The constant k is called the coverage factor, and for k = 2, the expanded uncertainty is roughly equivalent to half the length of a 95% confidence interval. Thus, the probability that the mean value is included in the expanded uncertainty is about 95% [24]. The standard uncertainty of the deuteron incorporation was 1.16 Da (n = 6), corresponding to the 15 min of incubation since the monomer is storage-temperature and time sensitive. The expanded uncertainty was calculated by multiplying u(x) by a coverage factor of 2 to give U(x) = 2.32 Da, the relative

A. Oliva et al. / Talanta 72 (2007) 1192–1198

1195

Table 1 Comparison of the percentage of labile hydrogen exchange at different time for the Spf66 monomer and dimer species after their incubation in D2 O Time (h)

0 0.25 1 4 24 48 72 a b c d e

Monomer

Dimer

Mw ± S.D.a

(n = 3)

% Deuteration

Mw ± S.D. (n = 3)

± ± ± ± ± ± ±

0.045 1.087 0.825 0.939 0.898 1.028c 0.837e

Reference 96.9 ± 1.47 96.6 ± 1.12 95.9 ± 1.27 95.7 ± 1.21 94.5 ± 1.39 93.5 ± 1.13

9490.506 9611.543 9607.432 9610.174 9612.979 9616.325 9618.671

4643.437 4715.162 4714.941 4714.385 4714.254 4713.336 4712.606

± ± ± ± ± ± ±

0.125 6.194 9.320 9.421 11.52b 7.242d 10.42d

% Deuteration Reference 82.9 ± 4.24 80.1 ± 6.38 82.0 ± 6.45 83.9 ± 7.88 86.2 ± 4.96 87.8 ± 7.15

Molecular weight ± standard deviation. Aggregation products detected by MALDI-TOF mass spectrometry: trimer species. Aggregation products detected by MALDI-TOF mass spectrometry: dimer species. Aggregation products detected by MALDI-TOF mass spectrometry: trimer and tetramer species. Aggregation products detected by MALDI-TOF mass spectrometry: dimer and trimer species.

error being 3.24%. This result indicates that a minimal loss of deuterons was achieved. Alternatively, hydrogen exchange has been measured as D/H exchange starting from the fully deuterated peptide. For this,

all labile protons are first exchanged by deuterons and after subsequent dilution of the peptide in a proton containing solvent, the reverse exchange to proton is monitored. At this point, the peptide solution was incubated in D2 O during 15 min and,

Fig. 2. MALDI-TOF mass spectra of the Spf66 monomer. The labels on the peaks indicate the measured average molecular mass. (A) Undeuterated sample; (B) deuterated sample stored at 25 ◦ C for 72 h, with a principal peak at m/z 4712.47, indicating a hydrogen exchange of 93.2%, although the dimer and trimer species were also detected; (C) sample after D/H exchange for 10-fold dilution, a deuterium exchange of 11.1%.

1196

A. Oliva et al. / Talanta 72 (2007) 1192–1198

after 5 min of off-exchange by 10-fold dilution in H2 O, the residual 8.02 deuteron number, corresponding to 11.1 ± 0.19% exchange (n = 3), was consistent with the 10% residual deuteration expected (Fig. 2). The expanded uncertainty, using a coverage factor k = 2, is then: U(x) = 0.27, so the mean number of exchanged protons is 8.02 ± 0.27 Da, which implies an error of 3.39%. This result was lower than those obtained by Ehring [7]. The Spf66 dimer, with a calculated molecular mass of 9490.5 Da containing a total of 146 labile hydrogen, shows a partial exchange of 82.9% after 15 min of incubation in D2 O at room temperature, remaining constant up to 24 h, similar values to those previously obtained [18]. The expanded uncertainty is clearly greater than the estimates in the previous cases (8.28%). This might be due to some uncertainty sources being overlooked and/or that the ones considered were underestimated or simply the nature of the molecule itself, since as expected for a folded conformation, the exchange level was not completed. The potato carboxypeptidase inhibitor (PCI), a small disulfide-bridged protein (Mw = 4295 Da) containing a total of 65 labile protons, showed similar behavior, since a partial hydrogen exchange of 61% was observed after 15 min of incubation in a deuterated solvent [25]. This exchange level seems to suggest that the Spf66 dimer presents two subpopulations of labile hydrogen: a group belonging to either solvent-exposed surface residues or relatively flexible regions that exchange rapidly and a second group involved in regular ordered structure as a determined by CD measurements [18]. To confirm these results, the D/H exchange experiment was also monitored by MALDI-TOF mass spectrometry, but this approach can only be applied to proteins or peptides with reversible denaturation [25]. Changing pH or adding denaturant may be a good choice for this purpose. Instead of chemical denaturation, thermal denaturation was used since the presence of chemical such as buffers, denaturants, etc., could seriously influence in the validation procedure, and could be necessary re-evaluate the analytical method; so the Spf66 dimer sample was stored at different temperatures in order to establish the best conditions for this. However, an irreversible thermal denaturation was observed in all conditions, for example, the percentage of deuteration rose to 87.8 ± 7.15% at 25 ◦ C and 72 h of storage, but the trimer and tetramer aggregation products were detected by MALDI-TOF mass spectrometry. These results point to a conformational change as a function of temperature. To verify this, CD spectroscopy was used. The changes in the UV-far CD spectra were found to be gradual with increasing temperature (Fig. 3), and even at lower temperatures, e.g., storage at 37 ◦ C for 1 h, a loss in secondary structure was observed, since the ␤sheet content decreased up to 42% against a 45% initial value. This was more evident as temperature increased reaching a value of 35% when stored at 60 ◦ C for 30 min (Table 2). This was confirmed by the H/D experiments, where an exchange of 94.2% of the total of 146 labile protons was observed. This result could be due to the amide hydrogen bonds being substantially weakened, increasing accessibility of the polypeptide backbone to hydrogen exchange. Moreover, these changes in peptide structure also

Fig. 3. Far-UV circular dicroism spectra of the Spf66 dimer as a function of temperature.

lead to increase aggregation since the trimer and tetramer species were detected, especially at high temperatures. Even though the percentage of exchanged protons is high for the Spf66 dimer in its natural state, a number of protons still resist exchange under normal conditions, which, in the most rigid part could be formed by tightly packed segments of secondary structure that could collapse through various factors such as temperature. However, as H/D experiments do not provide the accurate location of the protected hydrogen, protein fragmentation is a useful tool for identifying those regions of a protein that have little contact with the solvent or where there is a stable secondary structure [7,8,26,27]. Several approaches for obtaining high sequence coverage in protein (or mixture) have been reported; including the type of matrix used, the protein digestion method, and the use of fractionation for peptide digest to prior spectrometry mass analysis [21,26,28]. Based on these data, pepsin was then used for digestion of the Spf66 dimer because it has maximal activity in the pH range 2–3, where the H/D exchange rate is lowest, and cleaves the protein at many point to yield small peptides that more closely define the regions involved in the secondary structure. In this case, the fragments higher than 500 Da were only used since their size is the more appropriate for determining the D/H exchange level with an acceptable accuracy and precision. As a first step, peptide fragments obtained from digestion without H/D exchange were identified. For this, mass measurements in combination with sequence analysis were compared to identify the peptide fragments with an error set to ±2 Da [28], Table 2 Results obtained from circular dicroism data analysis using the K2d program Storage conditions

% Alpha % ␤-sheet % Random coil Distance (d2 )a Maximum error a

25 ◦ C, 15 min

37 ◦ C, 1 h

60 ◦ C, 30 min

8 45 47 89.4 0.112

8 42 50 101.7 0.182

9 35 56 21.7 0.100

Parameter used by Andrade et al. as an accuracy measurement.

A. Oliva et al. / Talanta 72 (2007) 1192–1198

1197

Table 3 Identification of peaks in the MALDI-TOF spectrum of an Spf66 dimer sample after pepsin digestion Theoretical massa

Residues position

Peptide sequence

Observed massb

D/H exchangec

1054.65 953.52 830.44 762.42 722.38 590.28 583.28 574.33 551.25 523.25 506.26 500.51

1-8+45-47 4-12/48-56 13-19/57-63 16-21/60-65 9-15/53-59 34-38/78-82 26-31/70-75 4-8/48-52 24-28/68-72 33-36/77-80 16-19/60-63 1-3+45-47

AGCCGANKKNA NKKNAPPNA NPLVMKE VMKEKQ PPNANPL VNQTE PNANPA NKKNA SYPNA YVNQ VMKE AGCCGA

n.d.d n.d. n.d. n.d. n.d. 590.15 n.d. n.d. 551.21 523.19 506.42 n.d.

– – – – – 6.78% (0.61/9) – – 9.43% (0.66/7) 9.00% (0.72/8) 9.17% (0.55/6) –

a

Theoretical masses [M + H]+ for identified peptides. Undeuterated observed peak mass. c Deuteration increase expressed as mean percentage D/H exchange for a 10-fold dilution, in brackets, the mean number of exchanges divided by the number of exchangeable amide protons. d Peak not detected. b

even though the error in mass determination is less than 300 ppm. The molecular mass and sequence of identified peptides are shown in Table 3. In theory, the MALDI spectrum of the entire peptic digest of Spf66 dimer should show 12 peaks in accordance with its primary structure, 10 peaks are present in each chain, plus 2 additional peaks containing the disulfide bridge (see Fig. 1). However, were only four identified, corresponding to 33% of the amino-acid coverage (Fig. 4A). Similar results were obtained in the complex protein mixture analysis, where the number of digest fragments covered between 10 and 40% of the entire amino-acid sequence [26,27]. Higher values (>65%) were obtained by Russell et al. [27] in mixed organic-aqueous solvent systems. However, under these conditions, physical instability problems were observed (data not shown). The peak at m/z 569.29 could be matched to residues 4-8 (or 48-52) with a theoretical m/z 574.31, but the mass tolerance is higher than the established limit (±2 Da). In this case, the mass tolerance could be increased up to 4 Da [29], but modern MALDI mass spectrometry equipment is capable of single decimal point mass resolution, the error being better than 50 ppm [21]. This peak could be a matrix adduct rather than an error in calibration. After determining that various fragments were observed in a single spectrum of digested, unlabeled Spf66 dimer, it was important to verify that deuterated Spf66 dimer could be digested and subsequently the deuteration of each fragment was determined from the mass spectrum of the complex mixture (Fig. 4B). As can be seen in Table 3, none of the fragments detected are protected from the D/H exchange, since the residual deuteron number matched the residual deuteration level expected from the 10-fold dilution, although a slightly lower value was observed for the 34-38 (or 78-82) fragment. In this case, the standard uncertainty was calculated as a standard deviation of the mean difference u(Diff ), where (Diff ) is defined as the average difference of the expected value for 10-fold dilution and the observed value (Table 3). A significance test was used to establish whether the measurements differ significantly from the expected value.

The test statistic t was carried out using Eq. (1). t=

|Diff | u(Diff )

(1)

This value was compared with the two-tailed critical value tcritical , for n − 1 degrees of freedom at 95% confidence. In the

Fig. 4. MALDI-TOF mass spectrum of an Spf66 dimer after pepsin digestion. (A) Undeuterated sample and (B) sample after D/H exchange for 10-fold dilution.

1198

A. Oliva et al. / Talanta 72 (2007) 1192–1198

case of fragment 34-38 (or 78-82), the bias was found to be significant (tcal > tcritical ). In spite of this, the results indicate that most of these segments are not involved in rigid structure since they are located at the end of the chain, where the structure may be more flexible and therefore more accessible to the solvent. In contrast, segments near the disulfide bridges, which in theory could be involved in forming rigid secondary structures, were not detected. Ehring [7] used this approach to study structural features of insulin-like growth factor I (IGF-1), and reported that the amount of H/D exchange was very close to 0% in all segments arranged around the disulfide bridge, which form the hydrophobic core of the protein. These findings were consistent with the known structure of IGF-1. 4. Conclusions In this work, several strategies to study structural features of synthetic peptides were used and validated. The H/D and D/H exchange experiments require only a small amount of sample, are relatively easy to use and fast since the information on the structure can be obtained from a few spectra in a matter of hours. All these aspects make these approaches attractive and useful in previous structural analysis of proteins and peptides. The validation of the hydrogen exchange methods provided enough evidence that they fulfill the requirements and purpose. The results obtained by both methods are in agreement and comparable, their uncertainties being very close. In contrast, the dimeric form is structured, but this structure is storage-sensitive (denaturing after time and temperature) since subtle changes were observed at high temperature and after a long time, which resulted in loss of secondary structure, as determined by CD measurements and increased aggregation products. As consequence, the D/H exchange cannot be applied to confirm the above results. Further combination of these two procedures with enzymatic digestion provided useful information on peptide regions involved in secondary structure, showing that all identified fragments belong to either solvent-exposed surface residues or regions located mainly in terminal chains where the structure is more flexible and accessible to solvent. However, the fragments involved in regular secondary structure, possibly those near disulfide bridges, were not detected. Although this method cannot be taken as definite proof that a specific region is involved or not in the structure, in combination with techniques, such as NMR spectroscopy and X-ray crystallography, an unambiguous decision could certainly be taken. The methods employed are of interest to those who want to do a quick structural analysis of synthetic peptides, especially, when no prior structural information is available, although can also be easily extended to other macromolecules.

Acknowledgements The authors sincerely thank Dr. M.E. Patarroyo and Dra. F. Guzman of the Instituto de Immunolog´ıa de Colombia for kindly providing the Spf66 peptide samples, Dr. J. Trujillo from Universidad de La Laguna, Spain, for his assistance in the CD analysis and Dra. V. Bonetto from Istituto di Ricerche Farmacologiche Mario Negri, Milan, Italy, for her assistance with the MALDITOF mass spectrometry analysis and helpful comments. References [1] S. Demotz, C. Moulon, M.A. Roggero, N. Fasel, S. Masina, Mol. Immunol. 38 (2001) 413. [2] R. Kishore, Curr. Protein Pept. Sci. 5 (2004) 435. [3] J.S. Davies, J. Pept. Sci. 9 (2003) 471. [4] I. Haro, M.J. G´omara, Curr. Protein Pept. Sci. 5 (2004) 425. [5] S.Y. Venyaminov, J.T. Yang, Determination of protein secondary structure, in: G. Fasman (Ed.), Circular Dicroism and the Conformational Analysis of Biomolecules, Plenum Press, New York, 1996. [6] S.Y. Tetin, F.G. Prendergast, S.Y. Venyaminov, Anal. Biochem. 321 (2003) 183. [7] H. Ehring, Anal. Biochem. 267 (1999) 252. [8] J.G. Mandell, A.M. Falick, E.A. Komives, Anal. Chem. 70 (1998) 3987. [9] T. Arakawa, S. Prestrelski, W.C. Kenney, J.F. Carpenter, Adv. Drug Deliv. Rev. 10 (1993) 1. [10] W. Wang, Int. J. Pharm. 185 (1999) 129. [11] V.K. Sharma, D.S. Kalonia, Pharm. Res. 20 (2003) 1721. [12] S. Hermeling, D.A.J. Crommelin, H. Schellekens, W. Jiskoot, Pharm. Res. 21 (2004) 897. [13] M.E. Patarroyo, P. Romero, M.L. Torres, P. Clavijo, A. Moreno, A. Mart´ınez, R. Rodr´ıguez, F. Guzman, E. Cabezas, Nature 328 (1987) 629. [14] M.V. Valero, R. Amador, J.J. Aponte, A. Narv´aez, C. Galindo, Y. Silva, J. Rosas, F. Guzman, M.E. Patarroyo, Vaccine 14 (1996) 1466. [15] H.P. Beck, I. Felger, W. Huber, S. Steiger, T. Smith, N. Weiss, P. Alonso, M. Tanner, J. Infect. Dis. 175 (1997) 921. [16] P. Graves, H. Gelband, P. Garner, Parasitol. Today 14 (1998) 218. [17] R.A. Houghten, Proc. Natl. Acad. Sci. U.S.A. 82 (1985) 5131. [18] A. Oliva, M.J. Dorta, A. Santove˜na, V. Bonetto, M. Salmona, J.B. Fari˜na, Peptides 23 (2002) 1527. [19] M. Andrade, J. Chac´on, J. Merelo, F. Mor´an, Protein Eng. 6 (1993) 383. [20] R. Li, C. Woodward, Protein Sci. 8 (1999) 1571. [21] C. Wa, R. Cerny, D.S. Hage, Anal. Biochem. 349 (2006) 229. [22] L.H. Cohen, A.I. Gusev, Anal. Bioanal. Chem. 373 (2002) 571. [23] K. Dharmasiri, D.L. Smith, Anal. Chem. 68 (1996) 2340. [24] E. Hund, D.L. Massart, J. Smeyers-Verbeke, Trends Anal. Chem. 20 (2001) 394. [25] J. Villanueva, F. Canals, V. Vilegas, E. Querol, F.X. Avil´es, FEBS Lett. 472 (2000) 27. [26] Z. Park, D.H. Russell, Anal. Chem. 73 (2001) 2558. [27] W.K. Russell, Z. Park, D.H. Russell, Anal. Chem. 73 (2001) 2682. [28] L.M. Jungbauer, S. Cavagnero, Anal. Chem. 78 (2006) 2841. [29] M.R. Wilkins, I. Lindskog, E. Gasteiger, A. Bairoch, J.C. Sanchez, D.F. Hochstrasser, R.D. Appel, Electrophoresis 18 (1997) 403.