' glargine and maltosyl-R-cyclodextrin

Description

The invention relates to a pharmaceutical formulation comprising insulin glargine and maltosyl- β -cyciodextrin. Insulin glargine is the first long-acting basal insulin analogue used for subcutaneous administration once daily in patients with type 1 or type 2 diabetes mellitus. To obtain the further desirable blood glucose lowering effect of insulin glargine, in the present study, we investigated the effect of maltosyl- β -cyciodextrin (G ₂ - β -CyD) on pharmaceutical properties of insulin glargine and the release of insulin glargine after subcutaneous injection to rats. G ₂ - β -CyD increased the solubility and suppressed aggregation of insulin glargine in phosphate buffer at pH 9.5, probably due to the interaction of G ₂ - -CyD with aromatic amino acid residues such as tyrosine of insulin glargine. In addition, G ₂ - β -CyD accelerated the dissolution rate of insulin glargine from its precipitates, compared to that of insulin glargine alone. Furthermore, we revealed that subcutaneous administration of an insulin glargine solution with G ₂ - β -CyD to rats gradually decreased a blood glucose level and provided a sustained-blood glucose lowering effect, possibly due to the conformational change of insulin glargine and the inhibitory effects of G ₂ - β -CyD on the enzymatic degradation of insulin glargine at the injection site. These results suggest that G ₂ - β -CyD can be a useful excipient for sustained release and a truly peak-less formulation of insulin glargine.

Diabetes is a chronic disease that the pancreas does not produce enough insulin (type 1 diabetes) or the body does not respond correctly to insulin and relative insulin deficiency (type 2 diabetes). It can be a life-threatening disease and also lead to serious complications such as cardiovascular disease, kidney failure, blindness and nerve damage (Blickle et al., 2007, Patterson et al., 2009, Simo et al., 2006). The global prevalence of diabetes has been increasing in recent decades, reaching near-epidemic proportions, and is projected to more than double by 2030 (Horton, 2008). The global diabetes epidemic has devastated on not only patients and their families but also national economies.

Human insulin is a major backbone for the treatment of diabetes. Although human insulin has attributed much in clinical treatment of diabetes for long time, there are still some difficulties and challenges in hypoglycemia and short half-life. In order to overcome these drawbacks, insulin glargine (Lantus ^® ), an insulin analogue (C267H404N72O73S6, MW=6,063) was developed by replacing the asparagine at the position of 21 of the A chain with glycine, and two arginines were added to the C-terminus of the B chain in human insulin (Fig. 1 ). It has a prolonged duration of action after subcutaneous injection and therefore can provide a basal insulin level of 24 hours by once daily injection (Rolla, 2008). This alteration resulted in low aqueous solubility at neutral pH (Wang et al., 2003). Insulin glargine is supplied in an acidic solution, which becomes neutralized at the injection site, leading to a formation of microprecipitates from which insulin glargine is slowly released into the circulation (Wang et al., 2003).

Cyclodextrins (CyDs) are known to form inclusion complexes with various guest molecules (Szente and Szejtli, 1999, Uekama et al., 1998). However, the low aqueous solubility of natural CyDs, especially β -CyD, has restricted their range of applications. To improve their solubility, alkylated, hydroxy! alkylated, sulfobutyl alkylated and branched CyDs have been used (Stella and Rajewski, 1997, Uekama, 2004, Uekama and Otagiri, 1987). Of these hydrophilic CyDs, maltosyl- β -CyD (G ₂ - β -CyD), 2-hydroxypropyl- β -CyD (HP- β -CyD) and sulfobutyl ether- β -CyD (SBE- β -CyD) have higher solubility in water and relatively low hemolytic activity, and thus have potential as pharmaceutical excipients for parenteral preparation (Uekama et al., 1998). In fact, natural β-CyD has a toxic effect on kidney, which is the main organ for removal of CyDs from the systemic circulation and for concentrating CyDs in the proximal convoluted tubule after glomerular filtration (Irie and Uekama, 1997). On the other hand, highly water-soluble β -CyD derivatives such as G ₂ - β -CyD, HP- β -CyD and SBE- β -CyD have very low systemic toxicity, compared with β-CyD.

We previously reported the effects of hydrophilic β -CyDs on the aggregation of bovine insulin in aqueous solution and its adsorption onto hydrophilic surfaces (Tokihiro et aL 1996, Tokihiro et ai., 1995, Tokihiro et al., 1997). Of the CyDs tested, G ₂ - β -CyD potently inhibited insulin aggregation in a neutral solution and its adsorption onto the surfaces of glass and polypropylene tubes. Furthermore, we reported that subcutaneous administration of insulin solution with SBE4- β -CyD to rats rapidly increased plasma insulin level and maintained higher plasma insulin levels for at least 8 h, possibly due to the inhibitory effects of SBE4- β -CyD on the enzymatic degradation and/or the adsorption of insulin onto the subcutaneous tissue at the injection site (Tokihiro et al., 2000). However, it is still unknown whether G ₂ - β -CyD shows the sustained glucose lowering effects for insulin analogues. Of various insulin analogues, only a few experiments on pharmaceutical application of insulin glargine were performed. Therefore, in the present study, to evaluate the potential use of G ₂ - β -CyD on not only bioavailability of insulin glargine but also the sustained-glucose lowering effect, we examined the effects of G ₂ - β -CyD on physicochemical properties and pharmacokinetics/pharmacodynamics of insulin glargine. Surprisingly, we revealed that G ₂ - β -CyD provided a sustained-blood- glucose lowering effect of insulin glargine after subcutaneous injection to rats. These findings indicate that G ₂ - -CyD can be a useful excipient for sustained release and a truly peak-less profile of insulin glargine. Therefore, an embodiment of the invention is a pharmaceutical formulation comprising insulin glargine and maltosyl- β -cyclodextrin.

A further embodiment of the invention is a pharmaceutical formulation as described above, additionally comprising one or more ingredients selected from a group comprising m-cresoL zinc, glycerol and polysorbate 20.

A further embodiment of the invention is a pharmaceutical formulation as described above, wherein the zinc concentration is 0 to 40 pg/ml, preferably 30 pg /ml.

A further embodiment of the invention is a pharmaceutical formulation as described above, wherein the glycerol content per 1 ml is 10 to 30 mg/ml, preferably 20 mg/ml of a 85% glycerol solution.

A further embodiment of the invention is a pharmaceutical formulation as described above, wherein the polysorbate 20 concentration is 10 to 30 pg /ml, preferable 20 pg /ml.

A further embodiment of the invention is a pharmaceutical formulation as described above, wherein the m-cresol concentration is 2,4 to 3,0 mg/ml, preferable 2,7 mg/ml. A further embodiment of the invention is a pharmaceutical formulation as described above, wherein the maltosyl- β -cyclodextrin concentration is 10 mM to 800 mM. ^'

A further embodiment of the invention is a pharmaceutical formulation as described above, wherein the maltosyl- β -cyclodextrin concentration is 150 to 250 mM, preferably 200 mM.

A further embodiment of the invention is a pharmaceutical formulation as described above, wherein the maltosyl- β -cyclodextrin concentration is selected from a group comprising 10 mM, 100 mM and 200 mM.

A further embodiment of the invention is a pharmaceutical formulation as described above, which additionally comprises a glucagon-like peptide- 1 (GLP1 ) or an analogue or derivative thereof, or exendin-3 or -4 or an analogue or derivative thereof.

A further embodiment of the invention is a pharmaceutical as described above, which additionally comprises exendin-4 or an analogue thereof, wherein the analogue is selected from a group comprising lixisenatide, exenatide and liraglutide, H-desPro ³⁶ -exendin-4-Lys ₆ -NH ₂ , H-des(Pro ^36,37 )- exendin-4-Lys ₄ -NH ₂ and H-des(Pro ³⁶ ' ³⁷ )-exendin-4-Lys ₅ -NH ₂ , or a pharmacologically tolerable salt thereof.

A further embodiment of the invention is the use of a pharmaceutical formulation as described above for the treatment of Type 1 or Type 2 Diabetes mellitus. A further embodiment of the invention is the preparation of a formulation as described above by adding insulin glargine, maltosyl- β -cyclodextrin and the excipients to an aqueous solution.

The invention is exemplified in the following by working examples which are not indended to be limiting.

MATERIALS

Insulin glargine was a gift from Sanofi-Aventis (Paris, France). G ₂ - β -CyD was obtained from Ensuiko Sugar Refining Co. Ltd (Yokohama, Japan). Recombinant trypsin (EC 3.4.21 .4) of proteomics grade was purchased from Roche Diagnostics (Tokyo, Japan). All other materials were of analytical reagent grade, and deionized double-distilled water was used. METHODS

Spectroscopic studies

Fluorescence and circular dichroism (CD) spectra were measured at 25°C using a HITACHI fluorescence spectrophotometer F-2500 (Tokyo, Japan) and a JASCO J-720 polarimeter (Tokyo, Japan), respectively. For preparation of the phosphate buffer (pH 9.5, 1=0.2), 0.1 mol/L phosphoric acid solution and 0.1 mol/L sodium hydroxide solution were mixed, which followed by addition of Sodium chloride.

Solubility studies

Excess amounts of insulin glargine were shaken in phosphate buffer (pH 9.5, /=0.2) in the absence and presence of G ₂ - β -CyDs at 25°C. After equilibrium was attained, the solutions were filtered with Millex® GV filter 0.22 μιη and insulin glargine dissolved was determined by the high performance liquid chromatography (HPLC) with Agilent 1 100 series (Tokyo, Japan) under the following conditions: Merck Superspher® 100 RP-18 column (4 pm, 3 mm x 250 mm, Tokyo, Japan), a mobile phase of phosphate buffer (pH 2.5) and acetonitrile and a gradient flow, increasing the ratio of the acetonitrile (25- 40%) over 30 min, a flow rate of 0.55 mL/min, a detection of UV at 214 nm.

Ultrafiltration studies

Ultrafiltration studies were performed using stirred ultrafiltration cells model 8010 (Millipore, Tokyo, Japan) applied with YM30 ultrafiltration discs (MWCO=30,000) in phosphate buffer (pH 9.5, 1=0.2) in the absence and presence of G ₂ - β -CyD at 25°C under nitrogen current. Insulin glargine levels in filtrates were determined by HPLC as described above.

Particle size determination

Particle sizes of insulin glargine (0.1 mM) with or without G ₂ - β -CyD (10 mM) in phosphate buffer (pH 9.5, /=0.2) were measured by Zetasizer Nano (Malvern Instruments, Worcestershire, UK).

Dissolution study of insulin glargine

Insulin glargine (0.1 mM) dissolved in phosphate buffer (pH 9.5, /=0.2) in the absence and presence of G ₂ - β -CyD (10 mM) was precipitated by a pH shift to 7.4. After centrifugation (2,500 rpm, 10 min), the supernatant was discarded and then phosphate buffer (pH 7.4, /=0.2) was newly added to the precipitate at 25°C. At appropriate intervals, an aliquot of the dissolution medium was withdrawn, centrifuged at 2,500 rpm for 10 min, and analyzed for the insulin glargine by HPLC as described in the paragraph 2.2.2.

Stability of insulin glargine against tryptic cleavage

Insulin glargine (0.1 m ) in phosphate buffer (pH 9.5, /=0.2) was incubated with recombinant trypsin (0.02 mg/mL) in the absence and presence of G ₂ - β -CyD at 37°C. At appropriate intervals, 5 μΙ_ of sample solution was withdrawn and determined intact insulin glargine level by HPLC. The rate constants (k _c ) and stability constants (K _c ) of 1 : 1 complexes of insulin glargine/G ₂ - β -CyD under the tryptic cleavage were determined by quantitative analysis according to the following equation, where k ₀ and [CyD] _t stands for the rate constants without CyD and the total concentration of CyD, respectively (Ikeda et al. , 1975).

Subcutaneous administration of insulin glargine/G2- β -CyD solution to rats Serum insulin glargine and glucose levels of rats were measured by the enzyme immunoassay and the mutarotase-glucose oxidase method. The solution (0.582 mL/kg) of the insulin glargine (2 l U/kg) in phosphate buffer (pH 9.5, /=0.2) in the absence and presence of G ₂ - β -CyD (200 mM) was subcutaneously injected in male Wistar rats (200-250 g), and at appropriate intervals blood samples were taken from the jugular veins. Serum insulin glargine and glucose were determined by Glyzyme Insulin-EIA Test Wako (Wako Pure Chemicals, Osaka, Japan) and Glucose-CI I-Test Wako (Wako Pure Chemicals Ind., Osaka, Japan), respectively. Serum glucose levels after the administration of insulin glargine/G ₂ - β -CyD solutions were expressed as a percentage of the initial glucose level before injection. Statistical Analysis

Data are given as the mean ± S.E. M. Statistical significance of means for the studies was determined by analysis of variance followed by Scheffe s test. p-Values for significance were set at 0.05.

Examplel : Spectroscopic studies

CyDs have been claimed to interact with hydrophobic residues exposed on protein surfaces and thereby to decrease aggregation of proteins (Brewster et al., 1991 , Tavornvipas et al., 2006). We previously reported that G ₂ - - CyD inhibited the insulin aggregation in neutral solution, possibly due to the inclusion of hydrophobic side chains of insulin within the CyD cavity, and hence perturbs the intermolecular hydrophobic contacts between aromatic side chains across the monomer-monomer interfaces (Tokihiro et al. , 1997). In the present study, to reveal whether G ₂ - β -CyD interacts with insulin glargine, we investigated the effects of G ₂ - β -CyD (10 or 100 mM) on the fluorescence and CD spectrum of insulin glargine (0.1 mM) (Fig. 2). To obtain the clear solution of insulin glargine (0.1 mM) in the present study, insulin glargine with G ₂ - β -CyD was dissolved in phosphate buffer (pH 9.5, /=0.2) at 25°C. The fluorescence intensity of tyrosine of insulin glargine at 306 nm was slightly enhanced by the addition of G ₂ - β -CyD (10 mM) (Fig. 2A). As tyrosine is a hydrophobic amino acid having a phenyl group in the molecule, G ₂ - β -CyD interacts with those aromatic amino acid residues of insulin glargine. The apparent 1 : 1 stability constant (K _c ) of the insulin glargine/G ₂ - β -CyD complexes was determined by the titration curves of the fluorescence intensity against a concentration of G ₂ - β -CyD with the Scott ^' s equation (Ikeda et al., 1975). The stability constant of insulin glargine/G ₂ - β - CyD complex in phosphate buffer (pH 9.5, /=0.2) at 25°C were calculated to be 27 ± 2 M ^"1 . The CD spectrum of insulin glargine (0.1 mM) showed negative bands at 210 and 220 nm in phosphate buffer (pH 9.5, /=θ.2) (Fig. 2B). The two negative bands assigned to alpha-helical (a characteristic feature of the monomer) and β -structure (a predominant feature of dimer) (Goldman and Carpenter, 1974). Furthermore, another negative band was observed at 273 nm in the CD spectrum of insulin glargine (Fig. 2C). This band is assigned to aromatic amino residues (tyrosine and phenylalanine) which exhibit optical activity as a function of aggregation of insulin molecule (Goldman and Carpenter, 974). In the presence of G ₂ - β -CyD (100 mM), the both negative bands at 210 and 220 nm in CD spectrum of insulin glargine were significantly increased, while the negative band at 275 nm was decreased. These results suggest that 1 ) G ₂ -(3-CyD increased monomer and dimer of insulin glargine and decreased aggregation of insulin glargine in the phosphate buffer (pH 9.5, /=0.2), and 2) G ₂ -(3-CyD changed conformation of insulin glargine by the complexation between aromatic amino residues of insulin glargine and G ₂ -(3-CyD in the phosphate buffer (pH 9.5, /=0.2).

Example 2: Solubility studies

Currently subcutaneous injection of clear solution is the main stream for administration of insulin and its analogues. However, insulin or insulin glargine is poorly soluble in aqueous solutions, in particular around the isoelectic point (pi), approximately pH 6.7, close to the physiological pH (Brange et al., 1997). Then, the effect of G ₂ - β -CyD on the solubility of insulin glargine was examined. As shown in Fig. 3, the solubility of insulin glargine in phosphate buffer at pH 9.5 was significantly increased by the addition of G ₂ - β -CyD. It is estimated that the increase in the solubility of insulin glargine was caused by the complexation between the G ₂ - β -CyD and aromatic amino acid residues of insulin glargine such as tyrosine. This solubilizing effect of G ₂ - β -CyD was also confirmed in phosphate buffer at pH 7.4 (data not shown). These results suggest that G ₂ - β -CyD potentially enhances the solubility of insulin glargine in phosphate buffer. Example 3: Ultrafiltration studies

The aggregation of insulin and its analogue is elicited by many kinds of factors such as the concentration of insulin, pH, temperature, shaking and so on (Rolla, 2008, Wang et al., 2003). Insulin glargine forms dimer, tetramer, hexamer and further soluble multimer by non-covalent interaction as proceeding in self-association (Havelund et al., 2004, Kurtzhals, 2004). Therefore, we performed ultrafiltration studies to estimate the effects of G ₂ - β -CyD on aggregation of insulin glargine using the membrane YM30 (MWCO=30,000) in phosphate buffer (pH 9.5, /=0.2). As shown in Fig. 4, insulin glargine permeated the ultrafiltration membrane by 48%. G ₂ - β -CyD significantly enhanced the permeation of insulin glargine up to 55%. These results suggest that G ₂ - β -CyD leads to dissociation of soluble multimer of insulin glargine. Following ultrafiltration experiment, particle sizes of insulin glargine were determined in the absence and presence of the G ₂ - β -CyD (Table 1 ). There were no significant difference in particle sizes of insulin glargine in the absence and presence of G ₂ - β -CyD in phosphate buffer (pH 9.5, /=0.2). These results suggest the potential use of G ₂ - -CyD as an aggregation-inhibitor for insulin glargine without remarkable influence on the particle size of insulin glargine.

Example 4: Dissolution study of insulin glargine Insulin glargine is believed to precipitate at the physiological pH after subcutaneous injection of the solution due to pi (about pH 6.7), which is followed by a sustained release of insulin glargine over 24 h from injection site because of an extremely low solubility in aqueous solution at pH of around pi (Wang et al., 2003). In order to investigate the effects of G ₂ - β - CyD on the sustained release of insulin glargine, the dissolution rate of insulin glargine from isoelectic precipitates formed in the absence and presence of G ₂ ~ β -CyD was determined (Fig. 5). Insulin glargine (0.1 mM) was dissolved in the phosphate buffer (pH 9.5) in the presence and absence of G ₂ - β -CyD (10 mM), and then isoelectric precipitation of insulin glargine was obtained after pH shift from 9.5 to 7.4. Then, the release of insulin glargine was determined in the pH 7.4 phosphate buffer in the absence of G ₂ - β -CyD. G ₂ - β -CyD significantly increased the dissolution rate of insulin glargine after 24 h, compare to insulin glargine alone. This enhancing effect of G ₂ - β -CyD is consistent with the solubilizing effect as shown in Fig. 3. These results suggest that G ₂ - β -CyD increases dissolution of insulin glargine from its precipitate.

Example 5: Stability of insulin glargine against tryptic cleavage

Insulin and its analogues are digested by proteinase such as trypsin, which cleaves insulin at the carboxyl side of residues B29-Lysine and B22-Arginine, at injection site and systemic circulation (Schilling and Mitra, 1991 ). Therefore, a resistance toward enzymatic degradation is required for insulin or its analogues formulation to improve their bioavailability. Next, we investigated the effects of the G ₂ - β -CyD on stability of insulin glargine against trypsin digestion. In this study, insulin glargine was digested by trypsin at 2 ID of the initial concentration at pH 9.5 at 37°C in the absence and presence of G ₂ - β -CyD. As shown in Fig. 6A, the apparent degradation rate constant of insulin glargine in the absence of the G ₂ - β -CyD (ko) was 0.357 ± 0.004 h ^"1 . Furthermore, the apparent rate constant (k _obs ) in the presence of the G ₂ - β -CyD decreased with the increase in the concentration of G ₂ - β -CyD. The rate constants (k _c ) and stability constants (K _c ) of 1 : 1 complex calculated with the regression lines shown in the Fig. 6B were 0.207 ± 0.023 h ^~1 and 563 ± 139 IVT ¹ , respectively. These results suggest that the inhibition of tryptic cleavage of insulin glargine by G ₂ - β -CyD was caused by a formation of complex with insulin glargine, resulting from decreasing the free insulin glargine to be easily digested by trypsin. We previously reported G ₂ - β -CyD inhibited hydrolysis of the buserelin acetate, agonist of luteinizing hormone-releasing hormone, by alpha-chymotrypsin. Deceleration of the β - chymotrypsin-catalyzed hydrolysis by G ₂ - β -CyD could be explained solely by a non-productive encounter between a complex of the buserelin acetate with G ₂ - β -CyD and the protease at relatively low CyD concentrations (approximately up to 30 mM) (Matsubara et al., 1997). Both trypsin used in this study and chymotrypsin belong to the chymotrypsin-like clan of the serine endopeptidases, having similarity in structure and hydrolysis of peptide bonds while they are different in target regions of a polypeptide chain. These results suggest that G ₂ - β -CyD acts as a stabilizer of insulin ^• glargine against enzymatic degradation due to interaction with insulin glargine.

Example 6: Subcutaneous administration of insulin glargine/G ₂ - β -CyD solution to rats We previously reported that G ₂ - β -CyD did not change the plasma immunoreactive insulin level and the plasma glucose level when bovine insulin in the phosphate-buffered saline (pH 6.8) was injected subcutaneously to rats (2 lU/kg) (Tokihiro et al., 1997). In this study, we evaluated the effects of G ₂ - β -CyD on pharmacokinetics and pharmacodynamics of insulin glargine after subcutaneous injection to rats. Figure 7 A and Table 2 show the serum insulin glargine level-time profiles and pharmacokinetic parameters, respectively, after subcutaneous administration of insulin glargine (2 lU/kg) with or without G ₂ - β -CyD (100 mM) in the phosphate buffer (pH 9.5) to rats. When insulin glargine was injected, the time (7 ^" _max ) required to reach maximum level (C _max ) of insulin glargine was at 1 .20 h after injection, and then the serum insulin glargine level decreased to the basal level. On the other hand, T _max in the G ₂ - β -CyD system significantly delayed to 5.82 h although C _max was the same as that of insulin glargine alone. The area under the serum insulin glargine level-time curve (AUC) up to 12 h of the G ₂ - β -CyD system (AUC=732.25 ^U/ml_) ' h) was significantly increased, compared to those of insulin glargine alone

(AUC=596.80 (μυ/mL) · h). Brange et al. reported that insulin molecules are transported into the capillaries as dimer or monomer and then absorbed to demonstrate the blood-glucose lowering effect (Brange et al., 1990). On the other hand, G ₂ - β -CyD (100 mM) increased the monomer and dimer of insulin glargine (Fig. 2B), probably due to the interaction with insulin glargine in solution. Taken together, the reason for the delay of T _max of insulin glargine by G ₂ - β -CyD may be contributed to the retention of dimer or monomer of insulin glargine at injection site and gradually released them into the capillaries. On the other hand the increase of the AUC up to 12 h in the serum insulin glargine level was probably due to 1 ) the inhibitory effects of G ₂ - β -CyD on the enzymatic degradation of insulin glargine (Fig. 6) and 2) the enhancement of solubility and the dissolution rate of insulin glargine by G ₂ - β -CyD (Figs. 3-5).

Figure 7B and Table 3 show the serum glucose level-time profiles and pharmacodynamics parameters after subcutaneous administration of insulin glargine (2 lU/kg) with or without G ₂ - β -CyD (200 mM) in the phosphate buffer (pH 9.5) to rats. When insulin glargine alone was injected, the minimal glucose level occurred at about 2 h after injection and then the serum glucose levels recovered within 6 h to basal level. On the other hand, T _nad ir and Cnadir increased significantly in the system of insulin glargine administered with G ₂ - β -CyD while the area under serum glucose level-time curve (AUCG) did not change notably. Therefore, these results suggest that G ₂ - β -CyD enhanced the persistence of blood-glucose lowering effect of insulin glargine as retaining the bioavailability of insulin glargine. Also the serum glucose level was kept at basal level constantly in the presence of G ₂ - β -CyD from 1 hr after injection up to 24 h without a clear decline in the serum glucose level. It was a peakless profile in comparison with insulin glargine alone. The purpose of treatment of diabetes mellitus is to normalize glycemic control. Normalization of the blood glucose concentration requires normalization of the plasma insulin profile. Endogenous insulin secretion needs a low basal level of plasma insulin during fasting and an appropriate elevation during meals (Owens and Bolli, 2008). In this context, the intensive insulin therapy is intended to give a basal level and a meal-related bolus level by means of various insulin formulations (Kramer, 1999). Neutral protamine Hagedorn insulin (NPH) was mainly used as basal insulin after its launch in 1946 (Owens and Bolli, 2008). However its duration of action is not long enough to cover the entire day, typically 12 to 18 hours in clinical practice (Heinemann et al., 2000, Lepore et al., 2000). And it shows a peak occurring 4 to 6 hours after subcutaneous injection (Heinemann et al., 2000) and this is connected to an increase of risk of hypoglycemia, particularly nocturnal hypoglycemia following bedtime injection (Fanelli et al., 2002). Insulin glargine introduced to the market in 2000 provides a longer duration action to last for 24 hours at least and a nearly flat profile (Heinemann et al., 2000, Lepore et al., 2000). As shown in Figure 7B and Table 3, subcutaneous administration of an insulin glargine solution with G ₂ - β -CyD to rats ameliorated the risk of hypoglycemia caused by insulin glargine and provided a sustained-blood glucose lowering effect, possibly due to the conformational change of insulin glargine and the inhibitory effects of G ₂ - β - CyD on the enzymatic degradation of insulin glargine at the injection site. Such a peakless profile of the blood glucose level decreases risks of hypoglycemia and thus provides patients with a better glycemic control and a higher quality of life. To gain insight into the mechanism, further elaborate study on the adsorption of insulin glargine in the presence of G ₂ - β -CyD onto subcutaneous tissue at injection site is under estimation. In conclusion, in the present study, we revealed that G ₂ - β -CyD provided a sustained-blood-glucose lowering effect of insulin glargine after subcutaneous injection to rats. These findings indicate that G ₂ - β -CyD can be a useful excipient for sustained release and a truly peak-less profile of insulin glargine. References

Blickle, J. F., Doucet, J., Krummel, T., Hannedouche, T., 2007.

Diabetic nephropathy in the elderly. Diabetes Metab., 33 Suppl 1 , S40-55.

Brange, J., Andersen, L, Laursen, E. D., Meyn, G., Rasmussen, E., 1997. Toward understanding insulin fibrillation. J. Pharm. Sci., 86, 517-525.

Brange, J., Owens, D. R., Kang, S., Volund, A., 1990. Monomeric insulins and their experimental and clinical implications. Diabetes Care, 13, 923-954.

Brewster, M. E., Hora, M. S., Simpkins, J. W., Bodor, N., 1991 . Use of 2-hydroxypropyl- β -cyclodextrin as a solubilizing and stabilizing excipient for protein drugs. Pharm. Res., 8, 792-795.

Fanelli, C. G., Pampanelli, S., Porcellati, F., Rossetti, P., Brunetti, P., Bolli, G. B., 2002. Administration of neutral protamine Hagedorn insulin at bedtime versus with dinner in type 1 diabetes mellitus to avoid nocturnal hypoglycemia and improve control. A randomized, controlled trial. Ann. Intern. Med., 136, 504-514.

Goldman, J., Carpenter, F. H., 1974. Zinc binding, circular dichroism, and equilibrium sedimentation studies on insulin (bovine) and several of its derivatives. Biochemistry, 13, 4566-4574.

Havelund, S., Plum, A., Ribel, U., Jonassen, I., Volund, A.,

Markussen, J., Kurtzhals, P., 2004. The mechanism of protraction of insulin detemir, a long-acting, acylated analog of human insulin.

Pharm. Res., 21 , 1498-1504.

Heinemann, L, Linkeschova, R., Rave, K., Hompesch, B., Sedlak, M., Heise, T., 2000. Time-action profile of the long-acting insulin analog insulin glargine (HOE901 ) in comparison with those of NPH insulin and placebo. Diabetes Care, 23, 644-649.

Horton, E. S., 2008. Can newer therapies delay the progression of type 2 diabetes mellitus? Endocr. Pract., 14, 625-638.

Ikeda, K., Uekama, K., Otagiri, M., 1975. Inclusion complexes of β- cyclodextrin with antiinflammatory drugs fenamates in aqueous solution. Chem. Pharm. Bull., 23, 201 -208.

Irie, T., Uekama, K., 1997. Pharmaceutical applications of

cyclodextrins. III. Toxicological issues and safety evaluation. J.

Pharm. ScL, 86, 147-162.

Kramer, W., 1999. New approaches to the treatment of diabetes. Exp.

Clin. Diabetes, 107, S52-61 .

Kurtzhals, P., 2004. Engineering predictability and protraction in a basal insulin analogue: the pharmacology of insulin detemir. Int. J. Obes. Relat. Metab. Disord., 28 Suppl 2, S23-28.

Lepore, M., Pampanelli, S., Fanelli, C, Porcellati, F., Bartocci, L., Di

Vincenzo, A., Cordoni, C, Costa, E., Brunetti, P., Bolli, G. B., 2000. Pharmacokinetics and pharmacodynamics of subcutaneous injection of long-acting human insulin analog glargine, NPH insulin, and ultralente human insulin and continuous subcutaneous infusion of insulin lispro. Diabetes, 49, 2142-2148.

Matsubara, K., Ando, Y. , Irie, T., Uekama, K., 1997. Protection afforded by maltosyl-(3-cyclodextrin against alpha-chymotrypsin- catalyzed hydrolysis of a luteinizing hormone-releasing hormone agonist, buserelin acetate. Pharm. Res., 14, 1401 -1405.

Owens, D. R., Bolli, G. B., 2008. Beyond the era of NPH insulin-long- acting insulin analogs: chemistry, comparative pharmacology, and clinical application. Diabetes Technol. Ther., 10, 333-349.

Patterson, B., Fields, A. V., Shannon, R. P., 2009. New insights into myocardial glucose metabolism: surviving under stress. Curr. Opin. Clin. Nutr. Metab. Care.

Rolla, A., 2008. Pharmacokinetic and pharmacodynamic advantages of insulin analogues and premixed insulin analogues over human insulins: impact on efficacy and safety. Am. J. Med., 121 , S9-S19. Schilling, R. J., Mitra, A. K., 1991 . Degradation of insulin by trypsin and alpha-chymotrypsin. Pharm. Res., 8, 721-727.

Simo, R., Carrasco, E., Garcia-Ramirez, M., Hernandez, O, 2006. Angiogenic and antiangiogenic factors in proliferative diabetic retinopathy. Curr. Diabetes Rev., 2, 71 -98. Stella, V. J., Rajewski, R. A., 1997. Cyclodextrins: their future in drug formulation and delivery. Pharm. Res., 14, 556-567.

Szente, L, Szejtli, J., 1999. Highly soluble cyclodextrin derivatives: chemistry, properties, and trends in development. Adv. Drug Deliv. Rev., 36, 17-28.

Tavornvipas, S., Hirayama, F., Takeda, S., Arima, H., Uekama, K., 2006. Effects of cyclodextrins on chemically and thermally induced unfolding and aggregation of lysozyme and basic fibroblast growth factor. J. Pharm. Sci., 95, 2722-2729.

Tokihiro, K., Arima, H., Tajiri, S., Irie, T., Hirayama, F., Uekama, K., 2000. Improvement of subcutaneous bioavailability of insulin by sulphobutyl ether beta-cyclodextrin in rats. J. Pharm. Pharmacol., 52, 91 1 -917.

Tokihiro, K., Irie, T., Hirayama, F., Uekama, K., 1996. Mass spectroscopic evidence on inhibiting effect of maltosyl- β -cyclodextrin on insulin self-association. Pharm. Sci., 2, 519-522.

Tokihiro, K., Irie, T., Uekama, K., 1995. Potential use of maltosyl-β- cyclodextrin for inhibition of insulin self-association in aqueous solution. Pharm. Sci., 1 , 49-53.

Tokihiro, K., Irie, T., Uekama, K., 1997. Varying effects of cyclodextrin derivatives on aggregation and thermal behavior of insulin in aqueous solution. Chem. Pharm. Bull., 45, 525-531 .

Uekama, K., 2004. Design and evaluation of cyclodextrin-based drug formulation. Chem. Pharm. Bull., 52, 900-915.

Uekama, K., Hirayama, F., Irie, T., 1998. Cyclodextrin Drug Carrier Systems. Chem. Rev., 98, 2045-2076.

Uekama, K., Otagiri, M., 1987. Cyclodextrins in drug carrier systems. Crit. Rev. Ther. Drug Carrier Syst., 3, 1 -40.

Wang, F., Carabino, J. M., Vergara, C. M., 2003. Insulin glargine: a systematic review of a long-acting insulin analogue. Clin. Ther., 25, 1541 -1577, discussion 1539-1540. Figure Legends

Figure 1 . Secondary chemical structure of insulin glargine Figure 2. Effect of G ₂ - β -CyD (10 mM (A) and 100 mM (B and C)) on fluorescence spectrum (A) and circular dichroism spectrum (B and C) of insulin glargine (0.1 mM) in phosphate buffer (pH 9.5, /=0.2) at 25°C. The excitation wavelength in measurement of fluorescence spectrum was 277 nm.

Figure 3. Effect of G ₂ - β -CyD (10 mM) on solubility of insulin glargine in phosphate buffer (pH 9.5, /=0.2) at 25°C. The concentration of insulin glargine was determined by HPLC. Each value represents the mean ± S.E.M. of 3 experiments. *p < 0.05, compared to insulin glargine.

Figure 4. Effect of G ₂ - β -CyD (10 mM) on permeation of insulin glargine (0.1 mM) through ultrafiltration membrane having nominal molecular weight limit of 30,000 in phosphate buffer (pH 9.5, /=0.2) at 25°C. The concentration of insulin glargine was determined by HPLC. Each value represents the mean ± S.E.M. of 5-17 experiments. ^* p < 0.05, compared to insulin glargine.

Figure 5. Effect of G ₂ - β -CyD (10 mM) on the dissolution rate from isoelectric precipitation of insulin glargine in phosphate buffer (pH 9.5, /=0.2) at 25°C. The initial concentration of insulin glargine was 0.1 mM, and then precipitated at pH 7.4. The concentration of insulin glargine was determined by HPLC. Each point represents the mean ± S.E.M. of 3 experiments. ^* p < 0.05, compared to insulin glargine.

Figure 6. Effects of G ₂ - -CyD (5 to 20 mM) on tryptic cleavage (2 IU) of insulin glargine (0.1 mM) in phosphate buffer (pH 9.5, /=0.2) at 37°C. The concentration of insulin glargine was determined by HPLC. Each point represents the mean ± S.E.M. of 3 experiments. Figure 7. Effects of G ₂ - β -CyD (100 mM) on serum insulin glargine (A) and glucose (B) levels after subcutaneous administration of insulin glargine (2 lU/kg) to rats. Each point represents the mean ± S.E.M. of 6-1 1 experiments. ^* p < 0.05, compared to insulin glargine.

Table Legends

Table 1 . Particle size of insulin glargine with or without G ₂ - β -CyD (10 mM) in phosphate buffer (pH 9.5). The particle size was measured by Zetasizer Nano. The concentration of insulin glargine and G ₂ - β -CyD were 0.1 mM and 10 mM, respectively.

Table 2. In vivo pharmacokinetics parameters of insulin glargine with or without G ₂ - β -CyD (100 mM). 1 ) Time required to reach the maximum serum insulin glargine level. 2) Maximum serum insulin glargine level. 3) Area under the serum insulin glargine level-time curve up to 9 h post- administration. Each value represents the mean ± S.E.M. of 6-9 experiments. ^* p < 0.05 , compared to insulin glargine. Table 3. In vivo pharmacodynamics parameters of insulin glargine with or without G ₂ - β -CyD (100 mM). 1 ) Time to nadir blood glucose concentration. 2) Nadir blood glucose concentration. 3) The cumulative percentage of change in serum glucose levels up to 9 h post-administration. Each value represents the mean ± S.E.M. of 7-1 1 experiments. ^* p < 0.05, compared to insulin glargine.

Table 1 . Particle Size of Insulin Glargine with or without G ₂ - -CyD (10 mM) in Phosphate Buffer (pH 9.5)

System Diameter (nm)

Insulin glargine 744±82

with G ₂ -p-CyD 796±82

The particle size was measured by Zetasizer Nano. The concentration of insulin glargine and CyD were 0.1 mM and 10 mM, respectively.

Table 2. In vivo Pharmacokinetics Parameter of Insulin Giargine with or without G ₂ -(5-CyD (100 mM)

System T _max D (h) C _max ¾ ^U/mL) AUG ³ ' ((mU/mL)h)

Insulin giargine 1 .20±0.13 124.30± 12.81 596.80±36.55

Insulin giargine *

/ G -p-CyD 5.82±0.18 130.36± 1 1 .68 732.25±33.25

1) Time required to reach the maximum plasma insulin giargine level.

2) Maximum plasma insulin giargine level.

3) Area under the plasma insulin giargine level-time curve up to 12 h post-administration. Each value represents the mean±S.E. of 6 to 9 experiments. *p<0.05 versus Insulin giargine.

Table 3. In vivo Pharmacodynamics Parameter of Insulin Glargine with or without G ₂ -(3-CyD (100 mM)

System T nadir ¹ > (h) (%) AUC _G ³ > (% - h)

Insulin glargine 1 .55± 0.16 48.31 ± :43.75 389.21 : ±22.46

Insulin glargine *

/G ₂ -p-CyD 5.60± : 1 .05 67.60± :3.86 341 .89: ±32.92

1) Time to nadir blood glucose concentration. 2) Nadir blood glucose concentration.

3) The cumulative percentage of change in plasma glucose levels up to 12 h postadmimstration. Each value represents the mean ± S.E. of 11 and 10 experiments for insulin glargine and insulin glargine/G ₂ - -CyD, respectively. *p<0.05 versus Insulin glargine.