Scribd Logo
Upload

Welcome to Scribd!

  • 20

    Uploaded by

    ヨトミヒト
    6 views10 pages

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    6 views10 pages

    20

    Uploaded by

    ヨトミヒト

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    of 10
    —_—_——— internation® journal of pharmaceutics ESEVIER, 1 s v cagaar is BS Absorption of insulin from Pluronic F-127 gels following subcutaneous administration in rats Jose Mario Barichello. Mariko Morishita * Kozo Takayama, Tsuneji Nagai Iaptment of imei Paes AP nia Cones, Frazee Ste work senunt Noster 1H FOE a 9 Vebewaty 197 Abaract she ane of Proc PETE paslacti-soslente 268 shery of popes and DEST ‘raving sort Ratios PE I2T Formulations 8 wvaluated at Shycos ana stn ne pesto The man as of hs 8 te cnatnate rea raapartes and tt comma sent ting inulin ay a nevi rus, Us 57°C using S sae fpamranetes an yates THM ane Cw men ae oT ' ans ge sls SemeN™ sat hat the ght foesreauen of PETZT Im NS 1 gli From independent af the vehicte wed THE 20 otan ain insulin nissan 9.3 Sar afer amine tears ya aoe pee Theater aiassteaton, By HOAURE eS 2h afer admins pve PTE 1 isin was aa ile ha AE TN TNT i formutaions. FP ound a wives tus. we conan eig7 get formulations containing ether GME OF Firug- nanoparticles ould rowel forthe pertate” ed et atid snes, He oats and poets Sport Hal 18 ca ones A TSS jons containing ithe erent « ahorpton: Po btos™#BTE si anos: HSPN effets ew + si: thn: Pros FT BE Sa ‘Sustamed release " then to target cells OF °8 vrabie tention: im me crab in Biotech oloey M 1. Introduction. fisientdevey of bioactive ides or proteins to the system's SSE , venous apuing stor Te 81ST mom of “Gorrspanding pose OM Moron) vege Tas fem of GINS curasiragy ce om mar © 199 vr STE pay A ens coors AL $0378.517 3199)00819°> tl 0 JM, Barcheto eta nernatoal Journal of Pharmeaceaics 184 (19) 189-198 ‘been poorly accepted by patients. except those suffering from life-threatening diseases. This is the case with insulin, a polypeptide used to treat patients with insulin-dependent (type 1) diabetes mellitus, which uses parenteral administration mainly via the se. route. Such insulin therapy dependent (type 2) diabetes mellitus. The provision of an exogenous supply of insulin that imitates the physiological patter found in non-diabetic populations with currently available therapeutic regimens has been a real challenge for many years, Recently, phar- macokinetic consideration of new insulin formu- Tations and routes of administration has been reviewed (Hoffman and Ziv, 1997). While much developmental activity has been undertaken on fast-acting insulin analogues, less progress has been made with basal (ultralente) insulin (Hoff man and Ziv, 1997). Some approaches have been used to improve the existing parenteral dosage forms to increase patient compliance. One of these approaches is the utilization of Pluronic F-127 (PFI27) gels PFI27 is a block copolymer comprising. poty(- owyethylene) and polyoxypropylene) segments a molecular weight of approximately 12.500 (Schmotka, 1972). The property of reversal ther- ‘mal gelation exhibited by PFI27 aqueous solu- tions in the 20-35% concentration range has been used as u drug delivery system for oph- thalmic (Miller and Donovan, 1982; Bochot et al, 1998; Desai and Blanchard, 1998; Edsman et al, 1998), rectal (Miyazaki et al. 1980). par- enteral (Morikawa et al., 1987: Guzmin et al, 1992; Johnston et al. 1992: Pec et al. 1992: Wang and Johnston, 1995; Paavola et al.. 1995, 1998; Katakam et al., 1997a.b) and percutaneous use (Tobiyama et al,. 1994; Miyazaki et al., 1995: Lee et al, 1997; Sub et al.. 1997), Another impor- tant characteristic of PFI27 gels is the enhance- ment of the stability of proteins loaded into the gel matrix, with completely recovery of their full activity when the gel is dissolved in excess buffer (Stratton et al., 1997), Additionally, PF127 ean easily be administered as a solution, which forms a rigid semisolid gel network upon an increase in temperature, thus avoiding surgical techniques. This kind of “depotlike’ sustained release. gel thas been applied to many drugs, but no report of its use for the delivery of insulin has been pub- lished. ‘Another approach used for improving the delivery of peptides and proteins has been the use of biodegradable micro- and nanospheres. ‘The most widely used and studied class of bio- degradable polymers is the polyesters, including polylactic acid and polylactic-co-glycolic acid (PLGA). Micro. and nanoparticulate systems formulated with these polymers have shown wide applicability. to oral (Ammoury et al, 1993: Chacén et al., 1996) and s.c. delivery (Soriano et al., 1996; Uchida ct al, 1997) of some drugs. In a previous report, we demonstrated that insulin-PLGA nanoparticles (INP) were effective in reducing the blood glucose levels of normal rats alter intra-ileum loop administration (in situ absorption study) (Barichello ct ah. 1999), Since the investigation of the parenteral mode of delivery is also important for_pep- tide drugs, we also studied the possibility of using this formulation for parenteral delivery of insulin. ‘The main objective of this work was to evalu ate the use of PF127 gels, PLGA. nanoparticles, ‘and also the combination of the gel and nanopar- ticles, for parenteral delivery of peptides and proteins having short half-lives using insulin as @ model drug. 2. Materials and methods 2.1. Materials Crystalline porcine insulin (Zn-insulin, 27.0 U; Shizuoka, Japan). Pluronic F-127, pluronic F-68 and bovine insulin (Zn-insulin, 28.6 U/mg) were purchased from Sigma Chemical (St. Louis, MO, USA). PLGA with a molecular weight of 10 KD and a lactide:glycolide ratio of 75:25 was purchased from Wako Pure Chemicals (Osaka, Japan). All references to water imply the use of MilliQ-purified water previously filtered through JM. Barchetta M- Barta eta. Itrnatonal Journal of Phamnaceutcs 184 (1990) 189-198 124m cellulose nitrate membrane. All other denials were at least reagent grade and were ssf without further purification. 22. Preparation of PFI27 solutions The PF127 solutions were prepared by the cold nxthod (Schmolka. 1972). A weighed amount of F127 was slowly added to a cold (S~10°C) solu- ‘oo with gentle mixing until complete dissolution of the polymer. The solutions used included an squeous solution (PF127-4.0), which exhibited a JH of 40 after complete dissolution of the poly- or, phosphate buffered solution of the polymer A pH 5.5 (PF127-5.5) and a phosphate buffered sslution of the polymer at pH 7.0 (PF127-7.0). An ‘Rpropriate amount of insulin was dissolved in UN hydrochloride acid adjusted with 0.1 N ‘odium hydroxide solution to approximately pH 40, and then added with gentle stisring after an ‘nemight refrigeration of the PF127 solution. The final concentration of the PF127 was 20 or 30%. 23, Preparation of INP PLGA nanopartictes were prepared as reported Prsiously (Barichello et al., 1999). Briefly. 75 mg. of PLGA was dissolved in 5 ml of acetone. Insul vas dissolved in 150 yl of 0.1 N hydrochloride ‘id (HCI) adjusted with 0.1 N sodium hydroxide solution to pH 4.0, and added to the polymer Phase. This organic phase was poured into 15 ml ofpH 5.5 phosphate buffer solution containing 75 ng of Pluronic F-68 with moderate stirring at om temperature, Acetone was removed from the colloidal suspension by using a rotary evapo- ‘ator. The nanoparticles were purified from the fee insulin by gel filtration through Sephadex (Sephadex G-50; Pharmacia Biotech, Sweden) and concentrated to a final volume of 2 ml (approxi- ‘mately 39 mg of INP/ml of suspension). An ap- Propriated volume of the final nanoparticle Suspension was centrifuged at 20000 rpm for 30 mnin. The sediment (nanoparticles) was dissolved in dichloromethane previously saturated with wa- ‘er and immediately assayed by using a reverse- Phase high-performance liquid chromatography (HPLC) system previously reported (Barichello et br al., 1999), The insulin content in nanoparticles was 59.8+3.2% The mean particle size of the nanoparticles was around 128429 nm with a relatively narrow particte size distribution. 24. Preparation of INP-loaded PF127 solutions ‘The nanoparticles and the PFI27 solutions were prepared as already described using a pH 5.5 phosphate buffer solution. An appropriated vol ume of the INP suspension was added to the PF127 solution with gentle stirring until complete mixing of both phases gave a final PF127 concen- tration of 20 of 30%. Preparation of the insulin control solution An appropriate amount of porcine insulin was dissolved in 0.1 N hydrochloride acid adjusted with 0.1 IN sodium hydroxide solution to approxi- mately pH 4.0. This solution was added to the necessary volume of phosphate buffer solution (pH 7.4) to give the insulin control solutions. 2.6. In vitro insulin and INP release from PF127 gel formulations A membraneless dissolution model was used for the in vitro studies. The cold PFI27 solutions (3, 8) containing insulin (Img/g of gel) or INP (26 mg/g of gel) were transferred into test tubes and placed in a 37°C water bath. The PF127 solutions equilibration for 20 min at the 37°C. ter of the release medium (a pH 4.0 ‘aqueous solution), pre-equilibrated at the experi- mental temperature, was layered over the surface of the PF127 gel containing insulin or INP. At sampling times. the supernatant was completely removed and replaced with fresh solution, in or- der to maintain sink conditions. The amount of insulin in the released medium was determined as already described. For INP-loaded PF127 gel, the released medium was separated from the nanopa ticles by centrifugation at 20000 rpm for 30 The amount of insulin present in the release medium and the nanoparticles was determined as described earlier. Bovine insulin was used for the in vitro experiments m2 JM. Barco etal. Internal Sours of Phemaccotcs 184 4199" 89.19 27. In cleo study in rats Male Wistar rats weighing 220 g that had fasted for 24h was divided into groups of five animals cach, A volume of 0.2 ml of a formulation was subcuraneously administered to each rat of the respsctive group. For the in vivo experiments porcine insulin in a dose 2 U ke body weight was used. Approximately 5 min before administration, a 0.2-ml sample of blood was taken from the jugular vein. Subsequent blood samples (0.2 ml) ‘were taken at 0.5, 1.2.4. 6 and 12 hafier dosing. Serum was separated by centrifugation at 13000 rpm for I min and kept frozen until analysis, The serum insulin levels were measured by using an insulin enzyme immunoassay kit (Dainabot. Tokyo. Japan). The serum glucose level was deter- mined by the glucose oxidase method using the Glucose B-test kit (Wako Pure Chemical Indus- tries, Osaka, Japan). . Statistical analysis Each value was expressed as the mean + SD. For group comparisons, the one-way layout anal- ysis of variance (ANOVA) with duplication was applied. Significant differences of the mean values \were evaluated by student’s unpaired (est. A P value of <0.05 was considered significant. 3. Results and discussion 3. In vitro insulin release from PF127 get formulations The in vitro release of insulin from the PE127 gel formulations at 37°C is shown in Fig. 1. The slope of the regression fine of insulin released fiom the PF127 gols versus time is shown in Table 1. The results of the in vitro study indicated that 20% gels released insulin approximately twice as fast as 30% gels. The cumulative percentage of insulin eleased at 12 h from 20% PF127-7.0, 20% PF127-40, 20% PFI27-5.5. 30% PF127-7.0. 30% PFI27-5.5 and 30% PFI27-4.0 gel formulations were 142405, I7£15, 99413, 86421, 5806 and 5.6 40.7%, respectively. In general, Time () Fig 1 tn vitro sekase pris of insu from 20 andl WP EAI? gel formoltions prepared in eilerent pH solutions st DPC. 20 PELTAOG § 2- P 70 (=p, Wr PELSTA0 189, 30 PRIZT-TO (WM), Valucs repre the mean of thie exper mens. To improve lait. the eeror hire for eich data point rere mot plotted unless they were larger than the symbols sed the PF127 gel formation oveurs due to the pro- gressive dehydration of the polymer micelles as temperature increases. leading to increased chitin entanglement, This entanglement is more marked her concentrations of PFI27, yielding an inerease of gel strength and consequently. a de= crease of the release rate (Guzman et al, 1992). In any case, release of insulin from the PF127 gels ‘was observed to follow zero-order release Kinetics, Although we have no direct evidence about the mechanism of insulin release from the PF127 for- ulations, the zero-order profiles suggest gel er0- sion and possibly some diffusion as the predominant mechanisms. as previously suggested by others (Guzman et al.. 1992: Johnston et al., 1992; Wang and Johnston. 1995). Table 1 Slope of the regression line of insulin released versus tine from Slope ws.h) 20% PRI Pr 40 28209 16 £068" ss 2st sot magne 10. 3512082 modo INP. 324067 - * Significant (P-<001) difference in the mean values come pared with 20% PEI27 gels using the S by u ares IM Bowral tte ual One important fai ows af stastetitie of PEST gels iy gelation ability un the presence of ganic salty (Panwht ait Kesha, 16) and some addins (Gl 19ST) This is very Spurtant, aan ate commonly sal to control pil in such gels and some other iltivey may be required to stabilize the form fiom, Ay shown in Hae Pan Table 1, PELE ‘O praluved the lastest release of ansulin among ¢ formulanons with the two caneenteations of PEIY? studied. ‘The release profiles of insulin fiom the PEIZ7-S.S and PEIZ-40 rat the concentration of NP. but stightly ifort at the concentration af 20. Pandit and Risks (19961 have demonstrated that salty with Aulivalent anions have at very dramatic effect on the sol-gel transition temperature of PPIIT gel fading 10 a toss of PEIZ7 gelation ability. at stain concentra the alteration ef ranstion temperatures may alter the diffusion of drugs in the gels, hence influencing release fais ay oberved for the vehicles used here The physicoctiemicsl properticy of model di are abo known te affect the drug ech lerstcs from fixemulations (Paavols et tn the PRIZ st ct al sitce me xalts cls were sim- Moreose 1998), $$ formulations, the pH of the Iehicle is closer to the isoelectric point of insulin than itis in the PE27-4.0 and PF127-7.0 formu- Lions, which may” atfcct the physicochemical Froperties of the insulin, at least at the lower PFIZ7 concentration (Paavolt et al., 1998). At tte higher concentration of PFI27, the intermicel far distance and the degree of micellar sellin Asesery for polymers wo contact one anetker sill be reduced. decreasing the gelation tempera {are and, therefore, altering the viscosity and the comsisteney characteristics of the formulation. This i probably the most important factor influ- frcing the release characteristics of the drug (Schmolka, 1972; Chen-Chow and Frank, 1981: Wanka et al. 1990; Lu and Jun, 1998), The PEII7.5.5 formulation was chosen for the in vivo ady based on the results obtained from the in ‘ito study The in viteo ‘of insulin from INP-loaded PFIZ7 gel at 37°C is shown in Fig. 2. A slower ‘ease of insulin was observed for INP- than for ‘nsulin-oaded 20% PFI27 gels. As observed itt st Peat 899 89108 im wo 2 Ba Ese Time In ito insulin release proile from a INPLloaded 20% sa ormlation: INP loud 20° PF12 (Cin in the PETST gol phase (04 and insulin in the BNP phase (=), Values represent the man of ree experiments “SD. To. sroprone clans, the ero hate for each data point were not lovtod unless they were larger tha the symbols Used, just 63% of the total insulin were directly released from the INP-loaded PFI27 gel system. Almost 35% of the total insulin remained in the INP after its release from the PFI27 gel system. It is expected the association of INP and PFI27 Is could provide different insulin absorption profiles to those obtained of « simple PF127 get formulation, 2 Changes im serum msultn and glucose levels falowing subcutancous adninistration of at insulin solution Management of diabetes by the patient in- volves a daily inconvenience, since insulin re quires parenteral adm via the Se. route, The inconvenience of multipte adminis- tration due to the relatively short time of action of the insulin seems to be one of the biggest problems to overcome. As shown in Fig. 3, the 4. administeation of an insulin solution is chat- acterized by an acute and relatively short hypo- glycemic effect. The s.c. region is well supplied by capillary and [ymphatic vessels. Although it has mot been determined whether the absorption route is via capillaries, lymphatic. or both, many drugs in solution injected into s.c. sites behave as if their absorption were taking place by passive diffusion (Banerjee et al, 1991). The major bar- Hier to absorption from the s.. site is believed to 94 JM, Bartel et al / International Journal of Pharmaceutics 184 (1999) 189-198 be the capillary endothel wall, which is Formed by a single layer of flat calls of approximately 0,0025-mm width (Baner- jee et fl, 1991), A prime factor in controlling the ab- sorption of insulin is presumed to be its molecu- lar weight, which is affected by its aggregation behavior in solution (monomer, dimer, hexamer, etc.) {I membrane or cell 3.3. Changes in serum insulin and glucose levels Following subcutaneous administration of an INP “formulation ‘The insulin monomers self-association in aqueous solutions and in the presence of certain. salts, solvents and hydrophobic surfaces has been studied for many years (Chawla et al, 1985; Sluzky ct al, 199; Tokihiro et al., 1997). These known insulin characteristics were considered in preparing INP, in which just 20% of the insulin was encapsulated into the nanoparticles, Surpris- ingly, an in situ absorption study of this INP formulation demonstrated a considerable hypo- lycemic effect of insulin absorbed from the ileum loop of rats (Barichello et al, 1999). In a previously reported study. insulin adsorbed to hydrolyzable nanoparticles was effective in. re- ducing the blood glucose levels of diabetic rats after sc. administration, but no change of the . 4nd 3 ” ws : Z off i " ai ‘ : oo ee @ nee) Fig. 3, Serum glucose and insulin levels following subouta- ‘pews administration of an insulin solution in normal rats elcose (7) and insulin (I). Values represent the means of five rats SD. § Zum ww z 0 & we zo 1805 be mod :™ of Eo . o o3 6 8 Time Fig. 4. Serum glucose and insulin levels following subcut ‘neous administration of an INP suspension In Aocmal ras slucose (O) and insulin (@), Values represent the means of fe rats +SD. glucose level was observed after oral administra- tion of the same hydrolyzable nanoparticles (Couvreur et al., 1980). Since the investigation of the parenteral mode of delivery is also important for peptide drugs, the hypoglycemic effect of the insulin from an INP formulation following sc. administration in normal rats was evaluated (Fig. 4), ‘The time-course of the hypoglycemic effect of insulin in the INP formulation is differ- ent from the time-course of the insulin solution in at least one aspect. The acute hypoglycemic peak characteristics of an insulin solution be- tween 0.5 and 2.0 h after administration (peak at 1 h) are replaced by an almost constant hypo- glycemic eflect with a slower recovery of the serum glucose levels during the 2h after admin- istration of the INP formulation. In a diffusion- controlled process, large molecules would be expected to have lower penetration rates than smaller ones. It appears that molecules of low ‘molecular weight (< 20.000) are absorbed. pri- marily via the capillaries, while molecules having high molecular weights are absorbed primarily ia Iymph vessel (Banerjee et al.. 1991). In addi- tion. the absorption from the sc. site also de- pends on the pH of the vehicle, volume of Injection, concentration of solute, osmolarity of the solution, polymorphic form and particle size «Banerjee et al., 1991). which makes it difficult to ‘compare the hypoglycemic profiles of the insulin solution and INP. REE JM Burchell eta. ncermatinal Journal of Changes in serum insulin and glucose levels flowing subcutaneous administration of insulin lnded PF127 gels Fig. 5 shows the changes in serum glucose and iaalin levels following s.c. administration of isulinloaded PF127 gels in normal rats. The ‘ical profile obtained with an insulin solution vas changed to a delayed and more protonged Iypoelyoemic profile when insulin was loaded ‘ao PFI27 gels. As observed in Fig. 5, the effect of the concentration of PFI27 on the release ot insulin showed an inverse dependence, i. tte higher the concentration of PF127, the ower and more protonged was the release of isin. The decreased absorption rate can beexplained by the decreased diffusion of insulin due to obstruction by the micellar entanglement, ‘epending on the PFI27 concentration (Chen- Chow and Frank, 1981). Viscosity may also detay sr impede diffusion of the drug into body fluids (Lu and Jun, 1998), These results are in good agreement with the in vitro results described ear- let. The use of a PFI27 gel matrix might also seve to stabilize insulin against its self-agerega- tion responsible for the loss of its biological oteney observed in a variety of delivery environ ments (Chawla et al., 1985) and_ processing streses (Wang and Johnston, 1995; Katakam eal, 1997a; Stratton et al., 1997), It might sho protect insulin against local proteolytic en- Serum ghucne (ofl Time) Pig 5, Serum glucose and insula fevels following subcus- ous administration of PP127 gels containing insulin in nor al rats: 20% PFI2T gel (0.6), 3% PFI27 got (S.A) cose (2,0) and insutn (4). Values represent the mars of fixe ats £ SD. Pharmaceutics 184 (199%) 189. 198 ws g i 5 Fig. 6 Serum glucose and insulin levels following subeuta- neous administration of INP-loaded PFI27 ele in normal rate: INP-loaded 20% PEI2T sel (0,4); INPAloaded 30% PFI27 gel (3A): glucose (4.6) and insulin (4,6). Values represent tho moans of fve rats SD. zymes while the peptide is entrapped in the gel mati 3.5. Changes in serum insulin and glucose levels following subcutaneous administration of INP-loaded PF127 gels ‘The effect of INP-loaded PF127 gels on serum glucose and insulin levels fotlowing s.c. adminis- tration in normal rats is seen in Fig. 6. PF127 gel formulations containing insulin-PLGA nanopar- ticles had the most long-lasting hypoglycemic ef- fects of all formulations, The administration of such a formulation resulted in hypoglycemic ef- fect profiles of an almost trapezoidal shape. The local distribution of a solution injected subeuta- neously is of interest because the permeation rate of the drug depends in part on the geometry and the resulting area of the depot exposed to the tissue. As observed, INP-loaded PFI27 gels pro- duced a burst of serum insulin more pronounced at the lower concentration of PF127, not ob- served in the case of insulin-loaded PF127 gels. Usually, the nanoparticles (INP) occupy a larger volume in solution and this might be effective enough at disrupting the stability of the PF127 micellar aggregates, directly affecting the critical ‘eel formation temperature. Such effect might be more pronounced at lower PFI27 concentration than at higher PFI27 concentrations. Further in & 96 JM Barthel eta Interaational Journal of Pharmaceutics 184 (1999) 189-198 rvenore near sar 20rr ne 8 6 150 300 450 600 AUC (hore eau) Fig, 7. AUC sales of the serum glucose levels following subeutancous administration of various insula formotations is normal rats: 1S, insulin solution: INP, insulin nanoparticles, DOPE, 20°% PFT gel, 3OPF, 30% PFI27 gel; INP/20PF, INPoaded 20% PEI27 ge; and INP/SOF, INP-loaded 307% DPFI2T gel. Values represent the mean of five rats SD. Siz- nificant (* P-<0.01) diference in the AUC mean values computed ith the contro! group using the Students es. Signicam ¢ P-

    You might also like