- Email: [email protected]
Effects of Cyclodextrins on Chemically and Thermally Induced Unfolding and Aggregation of Lysozyme and Basic Fibroblast Growth Factor
Effects of Cyclodextrins on Chemically and Thermally Induced Unfolding and Aggregation of Lysozyme and Basic Fibroblast Growth Factor SUMITRA TAVORNVIPAS, FUMITOSHI HIRAYAMA, SEIKO TAKEDA, HIDETOSHI ARIMA, KANETO UEKAMA Graduate School of Pharmaceutical Sciences, Kumamoto University, 5-1 Oe-honmachi, Kumamoto 862-0973, Japan
Received 7 November 2005; revised 12 May 2006; accepted 18 May 2006 Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.20715
ABSTRACT: Effects of cyclodextrin (CyDs) on unfolding and aggregation of lysozyme and basic fibroblast growth factor (bFGF) were investigated. CyDs inhibited the chemically induced aggregation and its inhibition was generally in the order of g-CyDs < a-CyDs < b-CyDs. Among these CyDs, branched b-CyDs and dimethyl-b-CyD (DM-b-CyD) significantly reduced the aggregation of lysozyme. Hydrophilic CyDs reduced the thermally induced unfolding of lysozyme as shown by a decrease in the thermal unfolding temperature (Tm) value of lysozyme, suggesting that CyDs destabilize native lysozyme or stabilize the unfolded state of lysozyme. In the case of bFGF, branched b-CyDs showed greater effects on inhibition of the chemically and thermally induced denaturation. Interestingly, sulfobutyl ether b-CyD (SBE-b-CyD), which was not effective in case of lysozyme, provided the inhibitory effect for bFGF on the chemically, thermally and acid-induced denaturation, suggesting that both the inclusion and electrostatic interaction may be operative in the inhibition of aggregation of the positively charged protein. The results indicated that the use of CyDs for protein stabilization is dependent not only on the structure and property of CyDs but also on the nature of the denaturing stimuli, and the most appropriate CyD should be used for the stabilization of each protein. ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 95:2722–2729, 2006
Keywords: unfolding
lysozyme; basic fibroblast growth factor; cyclodextrins; aggregation;
INTRODUCTION Protein stability is a particular relevant issue in the pharmaceutical field and will continue to gain more importance as the number of therapeutic protein products increases.1,2 Of these, protein aggregation is the most common problem in
This article contains supplementary material, available at www.interscience.wiley.com/jpages/0022-3549/suppmat. Fumitoshi Hirayama and K. Uekama’s present address is Faculty of Pharmaceutical Sciences, Sojo University, 4-22-1 Ikeda, Kumamoto 860-0082, Japan. Correspondence to: Prof. Kaneto Uekama (Telephone: þ8196-326-4096; Fax: þ81-96-326-4096; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 95, 2722–2729 (2006) ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association
2722
protein instability, because the aggregation process is often irreversible and the aggregates sometimes contain high levels of nonnative, intermolecular b-sheet structures which have a direct impact on drug potency, immunogenicity and the unfolded protein response.3–5 Protein unfolding and aggregation are usually triggered by various environmental stimuli such as changes in pH, ionic strength, heating, agitation, light irradiation, adsorption on hydrophobic surface, and addition of chemicals and organic solvents. A common method for inhibition of protein denaturation is to add excipients/additives in the protein preparation.1,6 Cyclodextrin (CyD) is one of the excipients used for the solubilization or stabilization of proteins. CyDs are cyclic oligosaccharides consisting of 6–8 glucose units. The
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006
AGGREGATION OF LYSOZYME AND BASIC FIBROBLAST GROWTH FACTOR
characteristic of CyD molecules is the presence of a hydrophobic cavity and hydrophilic exterior. In this way, CyDs can bind to hydrophobic parts of the protein surface leading to increase the solubility and stability of proteins.7 Recently, hydrophilic CyDs such as branched b-CyDs (glucosyl, maltosyl, and glucuronylglucosyl forms), 2hydroxypropyl-b-CyD (HP-b-CyD) and sulfobutyl ether b-CyD (SBE-b-CyD) have been evaluated as a new class of parenteral drug carriers, because they are highly hydrophilic and have less hemolytic activity than the parent and other hydrophilic CyDs.8–10 The objective of this study is to compare the effects of these hydrophilic CyDs on unfolding and aggregation of two model proteins, hen egg lysozyme and heparin-binding protein, basic fibroblast growth factor (bFGF).
MATERIALS AND METHODS Materials a-, b-, and g-CyDs and 2-hydroxypropyl-b-CyD (HP-b-CyD, degree of substitution (DS) ¼ 4.8) were supplied from Nihon Shokuhin Kako Co. (Tokyo, Japan). Branched CyDs (Maltosyl-a-, -band -g-CyDs (G2-a-CyD, G2-b-CyD, and G2-gCyD) and glucuronylglucosyl-b-CyD (GUG-bCyD) were obtained from Bio Research Corporation of Yokohama (Yokohama, Japan). SBE4-bCyD (DS ¼ 3.9) and SBE7-b-CyD (DS ¼ 6.2) were supplied by CyDex, Inc. (Lawrence, KS). 2,6-Di-Omethyl-a- and -b-CyDs (DM-a-CyD and DM-bCyD) were purchased from Wako Pure Chemical Co. (Osaka, Japan). Sulfated b-CyD (S-b-CyD, DS ¼ 10.7) was supplied by Kokusan Chemical Works, Ltd. (Tokyo, Japan). Recombinant human bFGF (bFGF, MW 17.1 kDa) was kindly donated by Kaken Pharmaceutical Co. (Urayasu, Japan). Hen egg lysozyme (58,100 U/mg, MW 14.3 kDa) was purchased from Sigma (St. Louis, MO). Guanidine hydrochloride (GuHCl) and dithiothretiol (DTT) was purchased from Nacalai Tesque Co. (Kyoto, Japan). All other materials and solvents were of analytical grade. Deionized, double-distilled water was used throughout the study.
Chemically Induced Denaturation Experiments Native lysozyme was dissolved in 6 M GuHCl, 50 mM DTT in phosphate buffer (pH 6.5, I ¼ 0.03) DOI 10.1002/jps
2723
at 1.0 mM (final concentration). After denaturation for 24 h at 258C, the mixture was diluted rapidly 10-fold with phosphate buffer (pH 6.5, I ¼ 0.03) containing CyD (10–40 mM), the mixture was then allowed to stand for 1 h at 258C. All insoluble materials in the denaturation solution were removed by centrifugation for 20 min at 15,000 rpm and the absorbance of the supernatant solution was measured at 280 nm using a Hitachi U-2000 spectrophotometer. The percent aggregation was estimated by subtracting from the percent remaining protein determined by UV absorbance. bFGF (5.84 mM) was incubated for 1 h at room temperature in 4 M GuHCl containing 0.1 M phosphate buffer at pH 7.0. The solution was diluted 10-fold with pH 7.0 phosphate buffer containing CyDs (100 mM) to induce refolding. After 4 h, the extent of protein denaturation was determined by fluorescence spectroscopy at excitation wavelength of 277 nm as reported previously by Sluzky et al.11 In native bFGF, fluorescence arose from tyrosine residues (308 nm), while tryptophan emission (350 nm) was quenched; upon unfolding, tryptophan emission increased. Fluorescence index (F350/F308) represents the degree of unfolding of bFGF.
Thermally Induced Denaturation Experiment The differential scanning calorimetry (DSC) thermograms of lysozyme solutions (150 mM) in the absence and presence of CyDs or maltose (100 mM) in phosphate buffer (pH 4.0, I ¼ 0.05) were recorded by a MC-2 differential microcalorimeter (MicroCal, Inc., Northampton, MA) using the MicroCal Origin for data acquisition and analysis. bFGF (0.12 mM) was denatured at 608C for 10 min in the absence and presence of CyDs (100 mM), and then incubated at room temperature for 2 h before centrifuging at 15,000 rpm for 10 min. The soluble proteins in the supernatant were determined by bicinchoninic acid method (BCA Protein Assay Kit, (Pierce Co., IL)).
Acid-Induced Denaturation Experiments The acid inactivation of bFGF (5.84 mM) in the absence and presence of CyDs (50 mM) was determined at 258C in 0.1 M phosphate buffer (pH 2.0). The degree of unfolding was expressed as fluorescence index (F350/F308) at excitation wavelength at 277 nm as described above.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006
2724
TAVORNVIPAS ET AL.
RESULTS AND DISCUSSION Effects of CyDs on Unfolding and Aggregation of Lysozyme We investigated effects of CyDs on chemically and thermally induced unfolding and aggregations of lysozyme in aqueous solution. It is well-known that when native lysozyme is dissolved in 6 M guanidine hydrochloride (GuHCl)/50 mM dithiothretiol (DTT) solution, the protein is reduced and denatured to disrupt its tertiary structure and to form a molten globule species.12 However, when this denaturated protein solution is diluted to
lower concentrations of GuHCl/DTT, not only the refolding of the protein to a native state starts, but also the aggregation of the protein to insoluble aggregates occur in a competitive way. Therefore, we investigated the effect of CyDs on the aggregation of lysozyme occurring during the dilution process of the molten globule species formed in the GuHCl/DTT solution (chemical denaturation). Figure 1A shows percents of aggregated insoluble lysozyme formed after a 10-fold dilution of the protein solution (1 mM in phosphate buffer (pH 6.5, I ¼ 0.03)) with the same phosphate buffer containing CyDs (10 mM), in comparison with that in the absence of CyDs. The
Figure 1. (A) Effects of CyDs (10 mM) on the aggregation of lysozymea) in phosphate buffer (pH 6.5, I ¼ 0.03) containing GuHClb) and DTTc) at 258C. (B) Effects of CyD concentrations on the aggregation of lysozymea) in phosphate buffer (pH 6.5, I ¼ 0.03) containing GuHClb) and DTTc) at 258C. Each value represents the mean SE of 3–5 experiments. a), b), c) The initial concentration of lysozyme, GuHCl and DTT were 1 mM, 6 M, and 50 mM, respectively. *p < 0.05 versus lysozyme alone. JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006
DOI 10.1002/jps
AGGREGATION OF LYSOZYME AND BASIC FIBROBLAST GROWTH FACTOR
Figure 2. DSC Thermograms of lysozyme (150 mM) in the absence and presence of CyDs (100 mM) and maltose (1 M) in phosphate buffer (pH 4.0, I ¼ 0.05). (A) lysozyme alone, (B) with G2-b-CyD, (C) with GUG-b-CyD, (D) with maltose.
inhibitory effects of b-CyDs, except for HP-b-CyD and SBE-b-CyDs, at 10 mM were larger than those of a- and g-CyDs (b-CyDs > a-CyDs > gCyDs, see Supporting Information Figure S1 for the structures of CyDs). It is apparent that in bCyD systems, DM-b-CyD and branched b-CyDs (G2-b-CyD and GUG-b-CyD) showed large inhibitory effects on the aggregation, while HPb-CyD and SBE-b-CyDs had smaller effects and parent b-CyD had a moderate inhibiting effect. Figure 1B shows effects of CyD concentration on the aggregation of lysozyme occurring during the dilution process. The inhibitory effect of CyDs on the aggregation augmented and tended to saturate as the CyD concentration increased. It is apparent that the ability of DM-b-CyD, G2-bCyD, and GUG-b-CyD to inhibit the lysozyme aggregation was greater than those of a-CyD, g-CyD, their derivatives, HP-b-CyD and SBE-b-
2725
CyDs in 10–40 mM CyD concentrations. The large inhibitory effect of DM-b-CyD, G2-b-CyD, and GUG-b-CyD may be due to a better fit of their cavity to aromatic amino acids that are involved in the aggregation or in the protein folding. Of these CyDs, the higher antiaggregation property was obtained with G2-b-CyD, GUG-b-CyD and DM-b-CyD, whereas HP-b-CyD and SBE-b-CyDs had the weak inhibiting effect. These observations can be deduced that both the cavity size of CyDs and the presence of appropriate substituents on the CyD rim are important in the inhibition of the protein aggregation, that is, HP-b-CyD and SBE-b-CyDs have randomly substituted 2-hydroxypropyl and sulfobutyl groups, respectively, at both sides of the CyD rim, which may hinder the accessibility of hydrophobic amino acid side chains of the protein to the CyD cavity. The cavity size of g-CyD may be too large to form strong inclusion complexes with the amino acids of lysozyme. Our 1H-NMR spectroscopic studies indicated that b-CyDs, particularly branched b-CyDs, interact with the tryptophan residues located in the self-association site of lysozyme, by inserting the branched moieties of the b-CyDs in the site. These results suggest that the inhibition of insoluble aggregate formation by b-CyDs increases amounts of lysozyme monomer in a similar manner to the case of rhGH as reported previously.13 Next, the thermal unfolding of lysozyme was studied by DSC. Figure 2 shows the thermograms of lysozyme in phosphate buffer (pH 4.0, I ¼ 0.05). In the absence of any additives, lysozyme exhibited a single sharp endothermic transition peak, which had the mean unfolding temperature (Tm) of 76.78C. The thermal unfolding of lysozyme under the experimental conditions was essentially a twostate process between the native and unfolded states, because of DHv/DHcal & 1.14 As shown in Table 1, the Tm value of lysozyme was changed by
Table 1. Thermodynamic Parameters of Lysozyme (150 mM) in the Absence and Presence of CyDs (100 mM) and Maltose (1 M) in Phosphate Buffer (pH 4, I ¼ 0.05) System Lysozyme Alone With a-CyD With g-CyD With HP-b-CyD With G2-b-CyD With GUG-b-CyD With Maltose
Tm (8C)
DHcal (kcal/mol)
DHv (kcal/mol)
DHv/DHcal
76.7 0.0 73.9 0.1 76.4 0.1 73.3 0.1 71.5 0.1 71.1 0.2 82.6 0.1
123 1 98 1 102 2 108 3 103 2 107 0 126 5
123 1 105 1 104 1 107 2 109 1 108 1 139 3
1.0 0.0 1.1 0.0 1.0 0.0 1.0 0.0 1.1 0.0 1.0 0.0 1.1 0.1
Each value represents the mean SE of 3–4 experiments. DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006
2726
TAVORNVIPAS ET AL.
the addition of additives, where they gave two contrary effects: CyDs lowered the Tm value, but maltose made the Tm value higher (about 68C). The Tm value of lysozyme decreased in the order of g-CyD < a-CyD < b-CyDs. Of CyDs used, branched b-CyDs dramatically decreased the Tm value (about 58C), whereas g-CyD showed little effect which coincides with the results of the chemical aggregation studies described above. Unfortunately, DM-b-CyD could not be used in this thermal unfolding study, because it precipitated at higher temperature (>608C). As shown in Table 1, the decrease in the change of enthalpy upon thermal denaturation (DH) agreed with the change in Tm values, that is, the decrease of DH in the CyD system, while the increase of DH in the maltose system. However, the DHv/DHcal value was not changed by the addition of these CyDs and maltose, indicating no change in the two-state transition between the native and unfolded states of lysozyme and no involvement of additional intermediates in the transition. These results suggested that the binding of branched b-CyDs to the exposed hydrophobic side chains during the heating destabilizes the native conformations of lysozyme by shifting the equilibrium in favor of the unfolded state.15 On the other hand, maltose enhanced the thermal stability of lysozyme, probably by preferential hydration of the native state, as reported.16 A recent study on the effects of arginine on refolding of aggregated proteins suggested that arginine contributes the stability of proteins in a way different from other amino acids, that is, arginines preferentially bind to the native state of proteins and solubilize the unfolded state.17 Therefore, CyDs may stabilize the unfolded state of lysozyme rather than the native state, shifting the equilibrium to the unfolded state. However, the presence of CyDs on surface of protein molecules or around some amino acid moieties may inhibit an access of further protein molecules to form aggregates. We assume that the inhibition of the aggregate formation, in spite of the slight destabilization of the native state, can be ascribed to the steric hindrance of CyDs, that is, CyDs occupy the self-association site of the protein, by inserting the branched moieties of b-CyDs in the site, inhibiting the aggregation.
(308 nm), while tryptophan emission (350 nm) is quenched.11 Upon unfolding of bFGF molecule, however, the tryptophan emission increases. Therefore, we monitored the ratio of the emission at 350 nm to that at 308 nm to determine the extent of unfolding of bFGF, that is, the intensity ratio (F350/F308) increases upon unfolding, while the ratio decreases when the protein is in a native state. Chemically induced unfolding of bFGF was studied in 4 M GuHCl in a similar manner described in the lysozyme studies, and the results are shown in Figure 3. The F350/F308 indices of bFGF in the absence and presence of 4 M GuHCl were 0.17 0.01 and 1.62 0.03, respectively, which agreed with the previous report.11 The F350/F308 index was decreased by the dilution with solutions of DM-b-CyD, SBE7-b-CyD, G2-b-CyD, GUG-b-CD, g-CyD, and G2-g-CyD, when compared with that diluted with phosphate buffer without CyDs, whereas a-CyD and G2-a-CyD showed negligible or moderate effects on the index. These results suggested that the former CyDs inhibit further unfolding to form soluble aggregates, whereas the latter CyD had negligible effects. Contrary to the case of lysozyme, the refolding effect of g-CyDs on bFGF was larger than those of a-CyDs. These results suggested that b- and g-CyDs may bind to hydrophobic sites in bFGF molecule, inhibiting unfolding to form soluble aggregates. The negatively charged sulfonate groups of SBE7-b-CyD may interact with positively charged bFGF molecule, inhibiting the
Effects of CyDs on Unfolding and Aggregation of bFGF
Figure 3. Fluorescence indices (F350/F308) of bFGF (5.84 mM) in the absence and presence of CyDs (100 mM) in 0.1 M phosphate buffer (pH 7.0) containing 4 M GuHCl at 258C. The excitation wavelength was 277 nm. *p < 0.05 versus without CyDs.
When bFGF molecule is in a native state, fluorescence arises from its tyrosine residues
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006
DOI 10.1002/jps
AGGREGATION OF LYSOZYME AND BASIC FIBROBLAST GROWTH FACTOR
2727
unfolding, in a similar way reported previously that sulfated oligosaccharides have high affinity to bFGF and can protect it from heat, acid and proteolytic degradations.18,19 The effect of CyDs on the acid-induced unfolding of bFGF was carried out in phosphate buffer at 258C. The fluorescence indices (F350/F308) increased as the pH of bFGF solutions was reduced from 5.0 to 2.0, but did not changed when the pH was raised from 5.0 into the alkaline region up to pH 8.0 (data not shown). The increased fluorescence indices in the pH 2.0 solution are attributable to the unfolding of protein in acidic solutions. These results agreed with the previous studies that an exposure of bFGF to a pH lower than 3.0 results in the rapid inactivation of the protein.20 Therefore, we have chosen pH 2.0 to study the effect of CyDs (50 mM) on the acid-induced inactivation of bFGF. As shown in Figure 4, the (F350/F308) ratio showed a large value of about 1.55, when bFGF was dissolved in the acid solution (pH 2.0), indicating that a significant unfolding of the protein occurred. On the other hand, the ratio was decreased when the protein is dissolved in the bCyD solutions (pH 2.0). It is apparent that b-CyD sulfate (S-b-CyD) and sulfobutylether derivatives such as SBE4-b-CyD and SBE7-b-CyD significantly protected bFGF from the acid-induced inactivation. The branched b-CyDs showed superior effect to protect bFGF from the chemically induced aggregation as described above, but not to protect against the acid-induced denaturation, when compared to those of SBE4-b-CyD and
SBE7-b-CyD. The inhibitory effect of HP-b-CyD was also smaller than those of other b-CyDs employed in this study. Heat treatment has been shown in previous studies to strongly decrease the potency of bFGF.20,21 To determine whether CyDs could inhibit bFGF from thermally induced aggregation, bFGF was heated at 608C for 10 min in the absence and presence of CyDs. Then, the aggregates were separated by centrifugation and the amounts of native bFGF in the supernatant were determined. As shown in Figure 5, branched b-CyDs, S-b-CyD, SBE4-b-CyD, and SBE7-b-CyD inhibited the aggregation of bFGF, compared with that of bFGF alone, whereas HP-b-CyD did not. In the case of SBE-b-CyDs, the inhibitory effect on the thermal aggregation may be attributed to the inclusion of aromatic amino acids or highly exposed hydrophobic residues of the protein, because the inhibitory effect of SBE7-b-CyD having higher negative charges (DS 6.2) was lower than that of SBE4-bCyD (DS 3.9). On the other hand, S-b-CyD with higher negative charges (DS 10.7) gave the larger inhibitory effect, suggesting that the interaction site of bFGF and the inhibitory mechanism are different between SBE-b-CyD and S-b-CyD. The effects of degree of sulfation of sulfated ligands on the ability to stabilize bFGF have been reported. bFGF is stabilized by dextran with degree of sulfation of more than three sulfated per disaccharide unit.18,22,23 Such the electrostatic interaction may be operative in the bFGF/S-b-CyD system. Unfortunately, DM-b-CyD could not be studied because of the precipitation at higher temperature as described before.
Figure 4. Effects of b-CyDs (50 mM) on acid inactivation of bFGF (5.84 mM) in 0.1 M phosphate buffer (pH 2.0) at 258C. Each value represents the mean SE of three experiments. The excitation wavelength was 277 nm. *p < 0.05 versus bFGF alone.
Figure 5. Effects of CyDs (100 mM) on aggregation of bFGF (0.12 mM) in phosphate buffer (pH 7.0) after heating at 608C for 10 min. Each value represents the mean SE of 3–6 experiments. *p < 0.05 versus bFGF alone.
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006
2728
TAVORNVIPAS ET AL.
CONCLUSION Among various CyDs, branched b-CyDs and DMb-CyD have the greater ability to inhibit the aggregation of lysozyme and bFGF induced by various stresses. Branched b-CyDs may be preferable over DM-b-CyD for use as an antiaggregation agent in parenteral preparations of proteins, because of the good bioadaptability and the solubility property8–10 at higher temperatures. SBE-b-CyDs may be useful for stabilization of positively charged proteins because of the inclusion and the electrostatic interaction with their negatively charges, although attention should be paid in the salt effects such as changes in ionic strength, osmotic pressure etc. HP-b-CyD seems to be unsuitable for inhibition of chemically and thermally induced denaturation of proteins, although it protects the interfacial aggregation of rhGH, as reported.13 As described here, it can be concluded that what is effective as a stabilizer for one stress on protein solutions may not be applicable to other stresses. The use of CyDs for protein stabilization is dependent not only on the structure and property of CyD itself, but also on the nature of the denaturing stimuli, and therefore the most appropriate CyD should be selected for the stabilization of each protein.
SUPPORTING INFORMATION Supporting Information of the structures of the CyDs used is available free of charge via the Internet (http://www.interscience.wiley.com/).
REFERENCES 1. Wang W. 2005. Protein aggregation and its inhibition in biopharmaceutics. Int J Pharm 289:1–30. 2. Stefani M, Dobson CM. 2003. Protein aggregation and aggregate toxicity: News insight into protein folding, misfolding diseases and biological evolutaion. J Mol Med 81:689–699. 3. Dong AC, Prestrelski SJ, Allison SD, Carpenter JF. 1995. Infrared spectroscopic studies of lyophilization-induced and temperature-induced protein aggregation. J Pharm Sci 84:415–424. 4. Hermeling S, Crommelin DJ, Schellekens H, Jiskoot W. 2004. Structure-immunogenicity relationships of therapeutic proteins. Pharm Res 21: 897–903.
5. Gass JN, Gunn KE, Sriburiand R, Brewer JW. 2004. Stressed-out B cells? Plasma-cell differentiation and the unfolded protein response. Trends Immunol 25:17–24. 6. Frokjaer S, Otzen DE. 2005. Protein drug stability: A formulation challenge. Nat Rev Drug Discov 4:298–306. 7. Aachmann FL, Otzen DE, Larsen KL, Wimmer R. 2003. Structural background of cyclodextrin-protein interactions. Protein Eng 16:905–912. 8. Brewster ME, Simpkins JW, Hora MS, Sternand WC, Bodor N. 1989. The potential use of cyclodextrins in parenteral formulations. J Parenter Sci Technol 43:231–240. 9. Yamamoto M, Aritomi H, Irie T, Hirayama F, Uekama K. 1991. Biopharmaceutical evaluation of maltosyl-b-cyclodextrin as a parenteral drug carrier. STP Pharma Sci 1:397–402. 10. Tavornvipas S, Hirayama F, Arima H, Uekama K, Ishiguro T, Oka M, Hamayasu K, Hashimoto H. 2002. 6-O-a-(4-O-a-D-Glucuronyl)-D-glucosylb-cyclodextrin: Solubilizing ability and some cellular effects. Int J Pharm 249:199–209. 11. Sluzky V, Shahrokh Z, Stratton PA, Eberlein GA, Wang J. 1994. Chromatographic methods for quantitation of native, denatured and aggregated bFGF in solution formulations. Pharm Res 11:485– 490. 12. Dong X, Shi J, Sun Y. 2002. Cooperative effects of artificial chaperones and guanidium chloride on lysozyme renaturation at high concentrations. Biotechnol Prog 18:663–665. 13. Tavornvipas S, Tajiri S, Hirayama F, Arima H, Uekama K. 2004. Effects of hydrophilic cyclodextrins on aggregation of recombinant human growth hormone. Pharm Res 21:2369– 2376. 14. Jelesarov I, Bosshard HR. 1999. Isothermal titration calorimetry and differential scanning calorimetry as complementary tools to investigate the energetics of biomolecular recognition. J Mol Recognit 12:3–18. 15. Cooper A. 1992. Effect of cyclodextrins on the thermal stability of globular proteins. J Am Chem Soc 114:9208–9209. 16. Arakawa T, Timasheff SN. 1992. Stabilization of protein structure by Sugars. Biochemistry 21: 6536–6544. 17. Arakawa T, Tsumoto K. 2003. The effects of arginine on refolding of aggregated proteins: Not facilitate refolding, but suppress aggregation. Biochem Biophys Res Comm 304:148–152. 18. Kajio T, Kawahara K, Kato K. 1992. Stabilization of basic fibroblast growth factor with dextran sulfate. FEBS Lett 306:243–246. 19. Fukunaga K, Hijikata S, Ishimura K, Sonoda R, Irie T, Uekama K. 1994. Aluminium b-cyclodextrin sulphate as a stabilizer and sustained-release
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006
DOI 10.1002/jps
AGGREGATION OF LYSOZYME AND BASIC FIBROBLAST GROWTH FACTOR
carrier for basic fibroblast growth factor. J Pharm Pharmacol 46:168–171. 20. Vemuri S, Beylin I, Sluzky V, Stratton PA, Eberlein GA, Wang YJ. 1994. The stability of bFGF against thermal denaturation. J Pharm Pharmacol 46: 481–486. 21. Gospodarowicz D, Cheng J. 1986. Heparin protects basic and acidic FGF from inactivation. J Cell Physiol 128:475–484. 22. Wang YJ, Shahrokh A, Vemuri S, Eberlein G, Beylin I, Busch M. 1996. Characterization, stability
DOI 10.1002/jps
2729
and formulation of basic fibroblast growth factor. In: Pearlman R, Wang YJ, editors. Stability and characterization of protein and peptide drugs. New York: Plenum Press, pp 141–180. 23. Copeland RA, Ji H, Halfpenny AJ, Williams RW, Thompson KC, Herber WK, Thomas KA, Bruner MW, Ryan JA, Marquis-Omer D, Sanyal G, Sitrin RD, Yamazaki S, Middaugh CR. 1991. The structure of human acidic fibroblast growth factor and its interaction with heparin. Arch Biochem Biophys 239:53–61.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006
Received 7 November 2005; revised 12 May 2006; accepted 18 May 2006 Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.20715
ABSTRACT: Effects of cyclodextrin (CyDs) on unfolding and aggregation of lysozyme and basic fibroblast growth factor (bFGF) were investigated. CyDs inhibited the chemically induced aggregation and its inhibition was generally in the order of g-CyDs < a-CyDs < b-CyDs. Among these CyDs, branched b-CyDs and dimethyl-b-CyD (DM-b-CyD) significantly reduced the aggregation of lysozyme. Hydrophilic CyDs reduced the thermally induced unfolding of lysozyme as shown by a decrease in the thermal unfolding temperature (Tm) value of lysozyme, suggesting that CyDs destabilize native lysozyme or stabilize the unfolded state of lysozyme. In the case of bFGF, branched b-CyDs showed greater effects on inhibition of the chemically and thermally induced denaturation. Interestingly, sulfobutyl ether b-CyD (SBE-b-CyD), which was not effective in case of lysozyme, provided the inhibitory effect for bFGF on the chemically, thermally and acid-induced denaturation, suggesting that both the inclusion and electrostatic interaction may be operative in the inhibition of aggregation of the positively charged protein. The results indicated that the use of CyDs for protein stabilization is dependent not only on the structure and property of CyDs but also on the nature of the denaturing stimuli, and the most appropriate CyD should be used for the stabilization of each protein. ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 95:2722–2729, 2006
Keywords: unfolding
lysozyme; basic fibroblast growth factor; cyclodextrins; aggregation;
INTRODUCTION Protein stability is a particular relevant issue in the pharmaceutical field and will continue to gain more importance as the number of therapeutic protein products increases.1,2 Of these, protein aggregation is the most common problem in
This article contains supplementary material, available at www.interscience.wiley.com/jpages/0022-3549/suppmat. Fumitoshi Hirayama and K. Uekama’s present address is Faculty of Pharmaceutical Sciences, Sojo University, 4-22-1 Ikeda, Kumamoto 860-0082, Japan. Correspondence to: Prof. Kaneto Uekama (Telephone: þ8196-326-4096; Fax: þ81-96-326-4096; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 95, 2722–2729 (2006) ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association
2722
protein instability, because the aggregation process is often irreversible and the aggregates sometimes contain high levels of nonnative, intermolecular b-sheet structures which have a direct impact on drug potency, immunogenicity and the unfolded protein response.3–5 Protein unfolding and aggregation are usually triggered by various environmental stimuli such as changes in pH, ionic strength, heating, agitation, light irradiation, adsorption on hydrophobic surface, and addition of chemicals and organic solvents. A common method for inhibition of protein denaturation is to add excipients/additives in the protein preparation.1,6 Cyclodextrin (CyD) is one of the excipients used for the solubilization or stabilization of proteins. CyDs are cyclic oligosaccharides consisting of 6–8 glucose units. The
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006
AGGREGATION OF LYSOZYME AND BASIC FIBROBLAST GROWTH FACTOR
characteristic of CyD molecules is the presence of a hydrophobic cavity and hydrophilic exterior. In this way, CyDs can bind to hydrophobic parts of the protein surface leading to increase the solubility and stability of proteins.7 Recently, hydrophilic CyDs such as branched b-CyDs (glucosyl, maltosyl, and glucuronylglucosyl forms), 2hydroxypropyl-b-CyD (HP-b-CyD) and sulfobutyl ether b-CyD (SBE-b-CyD) have been evaluated as a new class of parenteral drug carriers, because they are highly hydrophilic and have less hemolytic activity than the parent and other hydrophilic CyDs.8–10 The objective of this study is to compare the effects of these hydrophilic CyDs on unfolding and aggregation of two model proteins, hen egg lysozyme and heparin-binding protein, basic fibroblast growth factor (bFGF).
MATERIALS AND METHODS Materials a-, b-, and g-CyDs and 2-hydroxypropyl-b-CyD (HP-b-CyD, degree of substitution (DS) ¼ 4.8) were supplied from Nihon Shokuhin Kako Co. (Tokyo, Japan). Branched CyDs (Maltosyl-a-, -band -g-CyDs (G2-a-CyD, G2-b-CyD, and G2-gCyD) and glucuronylglucosyl-b-CyD (GUG-bCyD) were obtained from Bio Research Corporation of Yokohama (Yokohama, Japan). SBE4-bCyD (DS ¼ 3.9) and SBE7-b-CyD (DS ¼ 6.2) were supplied by CyDex, Inc. (Lawrence, KS). 2,6-Di-Omethyl-a- and -b-CyDs (DM-a-CyD and DM-bCyD) were purchased from Wako Pure Chemical Co. (Osaka, Japan). Sulfated b-CyD (S-b-CyD, DS ¼ 10.7) was supplied by Kokusan Chemical Works, Ltd. (Tokyo, Japan). Recombinant human bFGF (bFGF, MW 17.1 kDa) was kindly donated by Kaken Pharmaceutical Co. (Urayasu, Japan). Hen egg lysozyme (58,100 U/mg, MW 14.3 kDa) was purchased from Sigma (St. Louis, MO). Guanidine hydrochloride (GuHCl) and dithiothretiol (DTT) was purchased from Nacalai Tesque Co. (Kyoto, Japan). All other materials and solvents were of analytical grade. Deionized, double-distilled water was used throughout the study.
Chemically Induced Denaturation Experiments Native lysozyme was dissolved in 6 M GuHCl, 50 mM DTT in phosphate buffer (pH 6.5, I ¼ 0.03) DOI 10.1002/jps
2723
at 1.0 mM (final concentration). After denaturation for 24 h at 258C, the mixture was diluted rapidly 10-fold with phosphate buffer (pH 6.5, I ¼ 0.03) containing CyD (10–40 mM), the mixture was then allowed to stand for 1 h at 258C. All insoluble materials in the denaturation solution were removed by centrifugation for 20 min at 15,000 rpm and the absorbance of the supernatant solution was measured at 280 nm using a Hitachi U-2000 spectrophotometer. The percent aggregation was estimated by subtracting from the percent remaining protein determined by UV absorbance. bFGF (5.84 mM) was incubated for 1 h at room temperature in 4 M GuHCl containing 0.1 M phosphate buffer at pH 7.0. The solution was diluted 10-fold with pH 7.0 phosphate buffer containing CyDs (100 mM) to induce refolding. After 4 h, the extent of protein denaturation was determined by fluorescence spectroscopy at excitation wavelength of 277 nm as reported previously by Sluzky et al.11 In native bFGF, fluorescence arose from tyrosine residues (308 nm), while tryptophan emission (350 nm) was quenched; upon unfolding, tryptophan emission increased. Fluorescence index (F350/F308) represents the degree of unfolding of bFGF.
Thermally Induced Denaturation Experiment The differential scanning calorimetry (DSC) thermograms of lysozyme solutions (150 mM) in the absence and presence of CyDs or maltose (100 mM) in phosphate buffer (pH 4.0, I ¼ 0.05) were recorded by a MC-2 differential microcalorimeter (MicroCal, Inc., Northampton, MA) using the MicroCal Origin for data acquisition and analysis. bFGF (0.12 mM) was denatured at 608C for 10 min in the absence and presence of CyDs (100 mM), and then incubated at room temperature for 2 h before centrifuging at 15,000 rpm for 10 min. The soluble proteins in the supernatant were determined by bicinchoninic acid method (BCA Protein Assay Kit, (Pierce Co., IL)).
Acid-Induced Denaturation Experiments The acid inactivation of bFGF (5.84 mM) in the absence and presence of CyDs (50 mM) was determined at 258C in 0.1 M phosphate buffer (pH 2.0). The degree of unfolding was expressed as fluorescence index (F350/F308) at excitation wavelength at 277 nm as described above.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006
2724
TAVORNVIPAS ET AL.
RESULTS AND DISCUSSION Effects of CyDs on Unfolding and Aggregation of Lysozyme We investigated effects of CyDs on chemically and thermally induced unfolding and aggregations of lysozyme in aqueous solution. It is well-known that when native lysozyme is dissolved in 6 M guanidine hydrochloride (GuHCl)/50 mM dithiothretiol (DTT) solution, the protein is reduced and denatured to disrupt its tertiary structure and to form a molten globule species.12 However, when this denaturated protein solution is diluted to
lower concentrations of GuHCl/DTT, not only the refolding of the protein to a native state starts, but also the aggregation of the protein to insoluble aggregates occur in a competitive way. Therefore, we investigated the effect of CyDs on the aggregation of lysozyme occurring during the dilution process of the molten globule species formed in the GuHCl/DTT solution (chemical denaturation). Figure 1A shows percents of aggregated insoluble lysozyme formed after a 10-fold dilution of the protein solution (1 mM in phosphate buffer (pH 6.5, I ¼ 0.03)) with the same phosphate buffer containing CyDs (10 mM), in comparison with that in the absence of CyDs. The
Figure 1. (A) Effects of CyDs (10 mM) on the aggregation of lysozymea) in phosphate buffer (pH 6.5, I ¼ 0.03) containing GuHClb) and DTTc) at 258C. (B) Effects of CyD concentrations on the aggregation of lysozymea) in phosphate buffer (pH 6.5, I ¼ 0.03) containing GuHClb) and DTTc) at 258C. Each value represents the mean SE of 3–5 experiments. a), b), c) The initial concentration of lysozyme, GuHCl and DTT were 1 mM, 6 M, and 50 mM, respectively. *p < 0.05 versus lysozyme alone. JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006
DOI 10.1002/jps
AGGREGATION OF LYSOZYME AND BASIC FIBROBLAST GROWTH FACTOR
Figure 2. DSC Thermograms of lysozyme (150 mM) in the absence and presence of CyDs (100 mM) and maltose (1 M) in phosphate buffer (pH 4.0, I ¼ 0.05). (A) lysozyme alone, (B) with G2-b-CyD, (C) with GUG-b-CyD, (D) with maltose.
inhibitory effects of b-CyDs, except for HP-b-CyD and SBE-b-CyDs, at 10 mM were larger than those of a- and g-CyDs (b-CyDs > a-CyDs > gCyDs, see Supporting Information Figure S1 for the structures of CyDs). It is apparent that in bCyD systems, DM-b-CyD and branched b-CyDs (G2-b-CyD and GUG-b-CyD) showed large inhibitory effects on the aggregation, while HPb-CyD and SBE-b-CyDs had smaller effects and parent b-CyD had a moderate inhibiting effect. Figure 1B shows effects of CyD concentration on the aggregation of lysozyme occurring during the dilution process. The inhibitory effect of CyDs on the aggregation augmented and tended to saturate as the CyD concentration increased. It is apparent that the ability of DM-b-CyD, G2-bCyD, and GUG-b-CyD to inhibit the lysozyme aggregation was greater than those of a-CyD, g-CyD, their derivatives, HP-b-CyD and SBE-b-
2725
CyDs in 10–40 mM CyD concentrations. The large inhibitory effect of DM-b-CyD, G2-b-CyD, and GUG-b-CyD may be due to a better fit of their cavity to aromatic amino acids that are involved in the aggregation or in the protein folding. Of these CyDs, the higher antiaggregation property was obtained with G2-b-CyD, GUG-b-CyD and DM-b-CyD, whereas HP-b-CyD and SBE-b-CyDs had the weak inhibiting effect. These observations can be deduced that both the cavity size of CyDs and the presence of appropriate substituents on the CyD rim are important in the inhibition of the protein aggregation, that is, HP-b-CyD and SBE-b-CyDs have randomly substituted 2-hydroxypropyl and sulfobutyl groups, respectively, at both sides of the CyD rim, which may hinder the accessibility of hydrophobic amino acid side chains of the protein to the CyD cavity. The cavity size of g-CyD may be too large to form strong inclusion complexes with the amino acids of lysozyme. Our 1H-NMR spectroscopic studies indicated that b-CyDs, particularly branched b-CyDs, interact with the tryptophan residues located in the self-association site of lysozyme, by inserting the branched moieties of the b-CyDs in the site. These results suggest that the inhibition of insoluble aggregate formation by b-CyDs increases amounts of lysozyme monomer in a similar manner to the case of rhGH as reported previously.13 Next, the thermal unfolding of lysozyme was studied by DSC. Figure 2 shows the thermograms of lysozyme in phosphate buffer (pH 4.0, I ¼ 0.05). In the absence of any additives, lysozyme exhibited a single sharp endothermic transition peak, which had the mean unfolding temperature (Tm) of 76.78C. The thermal unfolding of lysozyme under the experimental conditions was essentially a twostate process between the native and unfolded states, because of DHv/DHcal & 1.14 As shown in Table 1, the Tm value of lysozyme was changed by
Table 1. Thermodynamic Parameters of Lysozyme (150 mM) in the Absence and Presence of CyDs (100 mM) and Maltose (1 M) in Phosphate Buffer (pH 4, I ¼ 0.05) System Lysozyme Alone With a-CyD With g-CyD With HP-b-CyD With G2-b-CyD With GUG-b-CyD With Maltose
Tm (8C)
DHcal (kcal/mol)
DHv (kcal/mol)
DHv/DHcal
76.7 0.0 73.9 0.1 76.4 0.1 73.3 0.1 71.5 0.1 71.1 0.2 82.6 0.1
123 1 98 1 102 2 108 3 103 2 107 0 126 5
123 1 105 1 104 1 107 2 109 1 108 1 139 3
1.0 0.0 1.1 0.0 1.0 0.0 1.0 0.0 1.1 0.0 1.0 0.0 1.1 0.1
Each value represents the mean SE of 3–4 experiments. DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006
2726
TAVORNVIPAS ET AL.
the addition of additives, where they gave two contrary effects: CyDs lowered the Tm value, but maltose made the Tm value higher (about 68C). The Tm value of lysozyme decreased in the order of g-CyD < a-CyD < b-CyDs. Of CyDs used, branched b-CyDs dramatically decreased the Tm value (about 58C), whereas g-CyD showed little effect which coincides with the results of the chemical aggregation studies described above. Unfortunately, DM-b-CyD could not be used in this thermal unfolding study, because it precipitated at higher temperature (>608C). As shown in Table 1, the decrease in the change of enthalpy upon thermal denaturation (DH) agreed with the change in Tm values, that is, the decrease of DH in the CyD system, while the increase of DH in the maltose system. However, the DHv/DHcal value was not changed by the addition of these CyDs and maltose, indicating no change in the two-state transition between the native and unfolded states of lysozyme and no involvement of additional intermediates in the transition. These results suggested that the binding of branched b-CyDs to the exposed hydrophobic side chains during the heating destabilizes the native conformations of lysozyme by shifting the equilibrium in favor of the unfolded state.15 On the other hand, maltose enhanced the thermal stability of lysozyme, probably by preferential hydration of the native state, as reported.16 A recent study on the effects of arginine on refolding of aggregated proteins suggested that arginine contributes the stability of proteins in a way different from other amino acids, that is, arginines preferentially bind to the native state of proteins and solubilize the unfolded state.17 Therefore, CyDs may stabilize the unfolded state of lysozyme rather than the native state, shifting the equilibrium to the unfolded state. However, the presence of CyDs on surface of protein molecules or around some amino acid moieties may inhibit an access of further protein molecules to form aggregates. We assume that the inhibition of the aggregate formation, in spite of the slight destabilization of the native state, can be ascribed to the steric hindrance of CyDs, that is, CyDs occupy the self-association site of the protein, by inserting the branched moieties of b-CyDs in the site, inhibiting the aggregation.
(308 nm), while tryptophan emission (350 nm) is quenched.11 Upon unfolding of bFGF molecule, however, the tryptophan emission increases. Therefore, we monitored the ratio of the emission at 350 nm to that at 308 nm to determine the extent of unfolding of bFGF, that is, the intensity ratio (F350/F308) increases upon unfolding, while the ratio decreases when the protein is in a native state. Chemically induced unfolding of bFGF was studied in 4 M GuHCl in a similar manner described in the lysozyme studies, and the results are shown in Figure 3. The F350/F308 indices of bFGF in the absence and presence of 4 M GuHCl were 0.17 0.01 and 1.62 0.03, respectively, which agreed with the previous report.11 The F350/F308 index was decreased by the dilution with solutions of DM-b-CyD, SBE7-b-CyD, G2-b-CyD, GUG-b-CD, g-CyD, and G2-g-CyD, when compared with that diluted with phosphate buffer without CyDs, whereas a-CyD and G2-a-CyD showed negligible or moderate effects on the index. These results suggested that the former CyDs inhibit further unfolding to form soluble aggregates, whereas the latter CyD had negligible effects. Contrary to the case of lysozyme, the refolding effect of g-CyDs on bFGF was larger than those of a-CyDs. These results suggested that b- and g-CyDs may bind to hydrophobic sites in bFGF molecule, inhibiting unfolding to form soluble aggregates. The negatively charged sulfonate groups of SBE7-b-CyD may interact with positively charged bFGF molecule, inhibiting the
Effects of CyDs on Unfolding and Aggregation of bFGF
Figure 3. Fluorescence indices (F350/F308) of bFGF (5.84 mM) in the absence and presence of CyDs (100 mM) in 0.1 M phosphate buffer (pH 7.0) containing 4 M GuHCl at 258C. The excitation wavelength was 277 nm. *p < 0.05 versus without CyDs.
When bFGF molecule is in a native state, fluorescence arises from its tyrosine residues
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006
DOI 10.1002/jps
AGGREGATION OF LYSOZYME AND BASIC FIBROBLAST GROWTH FACTOR
2727
unfolding, in a similar way reported previously that sulfated oligosaccharides have high affinity to bFGF and can protect it from heat, acid and proteolytic degradations.18,19 The effect of CyDs on the acid-induced unfolding of bFGF was carried out in phosphate buffer at 258C. The fluorescence indices (F350/F308) increased as the pH of bFGF solutions was reduced from 5.0 to 2.0, but did not changed when the pH was raised from 5.0 into the alkaline region up to pH 8.0 (data not shown). The increased fluorescence indices in the pH 2.0 solution are attributable to the unfolding of protein in acidic solutions. These results agreed with the previous studies that an exposure of bFGF to a pH lower than 3.0 results in the rapid inactivation of the protein.20 Therefore, we have chosen pH 2.0 to study the effect of CyDs (50 mM) on the acid-induced inactivation of bFGF. As shown in Figure 4, the (F350/F308) ratio showed a large value of about 1.55, when bFGF was dissolved in the acid solution (pH 2.0), indicating that a significant unfolding of the protein occurred. On the other hand, the ratio was decreased when the protein is dissolved in the bCyD solutions (pH 2.0). It is apparent that b-CyD sulfate (S-b-CyD) and sulfobutylether derivatives such as SBE4-b-CyD and SBE7-b-CyD significantly protected bFGF from the acid-induced inactivation. The branched b-CyDs showed superior effect to protect bFGF from the chemically induced aggregation as described above, but not to protect against the acid-induced denaturation, when compared to those of SBE4-b-CyD and
SBE7-b-CyD. The inhibitory effect of HP-b-CyD was also smaller than those of other b-CyDs employed in this study. Heat treatment has been shown in previous studies to strongly decrease the potency of bFGF.20,21 To determine whether CyDs could inhibit bFGF from thermally induced aggregation, bFGF was heated at 608C for 10 min in the absence and presence of CyDs. Then, the aggregates were separated by centrifugation and the amounts of native bFGF in the supernatant were determined. As shown in Figure 5, branched b-CyDs, S-b-CyD, SBE4-b-CyD, and SBE7-b-CyD inhibited the aggregation of bFGF, compared with that of bFGF alone, whereas HP-b-CyD did not. In the case of SBE-b-CyDs, the inhibitory effect on the thermal aggregation may be attributed to the inclusion of aromatic amino acids or highly exposed hydrophobic residues of the protein, because the inhibitory effect of SBE7-b-CyD having higher negative charges (DS 6.2) was lower than that of SBE4-bCyD (DS 3.9). On the other hand, S-b-CyD with higher negative charges (DS 10.7) gave the larger inhibitory effect, suggesting that the interaction site of bFGF and the inhibitory mechanism are different between SBE-b-CyD and S-b-CyD. The effects of degree of sulfation of sulfated ligands on the ability to stabilize bFGF have been reported. bFGF is stabilized by dextran with degree of sulfation of more than three sulfated per disaccharide unit.18,22,23 Such the electrostatic interaction may be operative in the bFGF/S-b-CyD system. Unfortunately, DM-b-CyD could not be studied because of the precipitation at higher temperature as described before.
Figure 4. Effects of b-CyDs (50 mM) on acid inactivation of bFGF (5.84 mM) in 0.1 M phosphate buffer (pH 2.0) at 258C. Each value represents the mean SE of three experiments. The excitation wavelength was 277 nm. *p < 0.05 versus bFGF alone.
Figure 5. Effects of CyDs (100 mM) on aggregation of bFGF (0.12 mM) in phosphate buffer (pH 7.0) after heating at 608C for 10 min. Each value represents the mean SE of 3–6 experiments. *p < 0.05 versus bFGF alone.
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006
2728
TAVORNVIPAS ET AL.
CONCLUSION Among various CyDs, branched b-CyDs and DMb-CyD have the greater ability to inhibit the aggregation of lysozyme and bFGF induced by various stresses. Branched b-CyDs may be preferable over DM-b-CyD for use as an antiaggregation agent in parenteral preparations of proteins, because of the good bioadaptability and the solubility property8–10 at higher temperatures. SBE-b-CyDs may be useful for stabilization of positively charged proteins because of the inclusion and the electrostatic interaction with their negatively charges, although attention should be paid in the salt effects such as changes in ionic strength, osmotic pressure etc. HP-b-CyD seems to be unsuitable for inhibition of chemically and thermally induced denaturation of proteins, although it protects the interfacial aggregation of rhGH, as reported.13 As described here, it can be concluded that what is effective as a stabilizer for one stress on protein solutions may not be applicable to other stresses. The use of CyDs for protein stabilization is dependent not only on the structure and property of CyD itself, but also on the nature of the denaturing stimuli, and therefore the most appropriate CyD should be selected for the stabilization of each protein.
SUPPORTING INFORMATION Supporting Information of the structures of the CyDs used is available free of charge via the Internet (http://www.interscience.wiley.com/).
REFERENCES 1. Wang W. 2005. Protein aggregation and its inhibition in biopharmaceutics. Int J Pharm 289:1–30. 2. Stefani M, Dobson CM. 2003. Protein aggregation and aggregate toxicity: News insight into protein folding, misfolding diseases and biological evolutaion. J Mol Med 81:689–699. 3. Dong AC, Prestrelski SJ, Allison SD, Carpenter JF. 1995. Infrared spectroscopic studies of lyophilization-induced and temperature-induced protein aggregation. J Pharm Sci 84:415–424. 4. Hermeling S, Crommelin DJ, Schellekens H, Jiskoot W. 2004. Structure-immunogenicity relationships of therapeutic proteins. Pharm Res 21: 897–903.
5. Gass JN, Gunn KE, Sriburiand R, Brewer JW. 2004. Stressed-out B cells? Plasma-cell differentiation and the unfolded protein response. Trends Immunol 25:17–24. 6. Frokjaer S, Otzen DE. 2005. Protein drug stability: A formulation challenge. Nat Rev Drug Discov 4:298–306. 7. Aachmann FL, Otzen DE, Larsen KL, Wimmer R. 2003. Structural background of cyclodextrin-protein interactions. Protein Eng 16:905–912. 8. Brewster ME, Simpkins JW, Hora MS, Sternand WC, Bodor N. 1989. The potential use of cyclodextrins in parenteral formulations. J Parenter Sci Technol 43:231–240. 9. Yamamoto M, Aritomi H, Irie T, Hirayama F, Uekama K. 1991. Biopharmaceutical evaluation of maltosyl-b-cyclodextrin as a parenteral drug carrier. STP Pharma Sci 1:397–402. 10. Tavornvipas S, Hirayama F, Arima H, Uekama K, Ishiguro T, Oka M, Hamayasu K, Hashimoto H. 2002. 6-O-a-(4-O-a-D-Glucuronyl)-D-glucosylb-cyclodextrin: Solubilizing ability and some cellular effects. Int J Pharm 249:199–209. 11. Sluzky V, Shahrokh Z, Stratton PA, Eberlein GA, Wang J. 1994. Chromatographic methods for quantitation of native, denatured and aggregated bFGF in solution formulations. Pharm Res 11:485– 490. 12. Dong X, Shi J, Sun Y. 2002. Cooperative effects of artificial chaperones and guanidium chloride on lysozyme renaturation at high concentrations. Biotechnol Prog 18:663–665. 13. Tavornvipas S, Tajiri S, Hirayama F, Arima H, Uekama K. 2004. Effects of hydrophilic cyclodextrins on aggregation of recombinant human growth hormone. Pharm Res 21:2369– 2376. 14. Jelesarov I, Bosshard HR. 1999. Isothermal titration calorimetry and differential scanning calorimetry as complementary tools to investigate the energetics of biomolecular recognition. J Mol Recognit 12:3–18. 15. Cooper A. 1992. Effect of cyclodextrins on the thermal stability of globular proteins. J Am Chem Soc 114:9208–9209. 16. Arakawa T, Timasheff SN. 1992. Stabilization of protein structure by Sugars. Biochemistry 21: 6536–6544. 17. Arakawa T, Tsumoto K. 2003. The effects of arginine on refolding of aggregated proteins: Not facilitate refolding, but suppress aggregation. Biochem Biophys Res Comm 304:148–152. 18. Kajio T, Kawahara K, Kato K. 1992. Stabilization of basic fibroblast growth factor with dextran sulfate. FEBS Lett 306:243–246. 19. Fukunaga K, Hijikata S, Ishimura K, Sonoda R, Irie T, Uekama K. 1994. Aluminium b-cyclodextrin sulphate as a stabilizer and sustained-release
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006
DOI 10.1002/jps
AGGREGATION OF LYSOZYME AND BASIC FIBROBLAST GROWTH FACTOR
carrier for basic fibroblast growth factor. J Pharm Pharmacol 46:168–171. 20. Vemuri S, Beylin I, Sluzky V, Stratton PA, Eberlein GA, Wang YJ. 1994. The stability of bFGF against thermal denaturation. J Pharm Pharmacol 46: 481–486. 21. Gospodarowicz D, Cheng J. 1986. Heparin protects basic and acidic FGF from inactivation. J Cell Physiol 128:475–484. 22. Wang YJ, Shahrokh A, Vemuri S, Eberlein G, Beylin I, Busch M. 1996. Characterization, stability
DOI 10.1002/jps
2729
and formulation of basic fibroblast growth factor. In: Pearlman R, Wang YJ, editors. Stability and characterization of protein and peptide drugs. New York: Plenum Press, pp 141–180. 23. Copeland RA, Ji H, Halfpenny AJ, Williams RW, Thompson KC, Herber WK, Thomas KA, Bruner MW, Ryan JA, Marquis-Omer D, Sanyal G, Sitrin RD, Yamazaki S, Middaugh CR. 1991. The structure of human acidic fibroblast growth factor and its interaction with heparin. Arch Biochem Biophys 239:53–61.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 12, DECEMBER 2006