- Email: [email protected]
Novel lyophilized hydrogel patches for convenient and effective administration of microneedle-mediated insulin delivery
International Journal of Pharmaceutics 437 (2012) 51–56
Contents lists available at SciVerse ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Pharmaceutical Nanotechnology
Novel lyophilized hydrogel patches for convenient and effective administration of microneedle-mediated insulin delivery Yuqin Qiu a , Guangjiong Qin a,b , Suohui Zhang a , Yan Wu a,b , Bai Xu a , Yunhua Gao a,∗ a b
Key Laboratory of Photochemical Conversion and Optoelectronic Materials, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing 100190, China Graduate University Chinese Academy of Sciences, Beijing 100049, China
a r t i c l e
i n f o
Article history: Received 2 February 2012 Received in revised form 29 June 2012 Accepted 22 July 2012 Available online 27 July 2012 Keywords: Lyophilized hydrogel Transdermal drug delivery Insulin Microneedle Diabetes
a b s t r a c t A lyophilized hydrogel patch system was developed for microneedle-mediated insulin delivery. The matrix of Cross-linked poly(acrylamide-co-acrylic acid) were synthesized by precipitation polymerization. Recombinant human Insulin was loaded into the lyophilized polymer matrix, which can be rehydrated by water. After the hydrated patch was applied to the abdominal skin of diabetic rats after microneedle pretreatment, pharmacodynamics and pharmacokinetics evaluation was performed. The blood samples were collected to monitor blood glucose and serum insulin levels for 12 h. Blood glucose was lowered in proportion to the concentration of insulin loaded in lyophilized hydrogel patches (R2 = 0.99), with a longer duration of action compared to subcutaneous injection. Stability study confirmed more than 90% of insulin activity was retained in lyophilized hydrogel after 6 months of storage at 4 ◦ C. In conclusion, hydrogel patches were demonstrated to be appropriate drug reservoir for sustained release of insulin with microneedle mediated transdermal delivery. © 2012 Elsevier B.V. All rights reserved.
1. Introduction All type 1 and many type 2 diabetes mellitus patients need insulin to achieve glycaemic control, and it often requires multiple insulin administration per day (Weng et al., 2008). Typically, insulin is given by subcutaneous needle injection such as self injection, insulin pen (Buysschaert and Lambert, 1989) and catheters connected to insulin pumps (Lenhard and Reeves, 2001). However, these methods are inconvenient and painful, leading to poor compliance and poor diabetes management (Asche et al., 2010). Inhaled insulin, Exubera® (Pfizer, New York, NY), was introduced as the first FDA-approved needle-free alternative. However, it was subsequently discontinued due to various reasons such as high cost, device design, side effects and unknown health risks (Siekmeier and Scheuch, 2008). Other insulin delivery routes have been explored without compromising with the comfort and future health of patients. Recently, though, attention has been drawn to the possibility of using a microneedle system to deliver insulin (McAllister et al., 2003). The system consists of short, micrometer-scale needles, which can temporarily compromise the skin barrier and allow the drug loaded in a patch to diffuse to the rich capillary bed of the dermis for uptake and subsequent systemic distribution in the blood
∗ Corresponding author. Tel.: +86 10 82543581; fax: +86 10 82543581. E-mail address: [email protected] (Y. Gao). 0378-5173/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ijpharm.2012.07.035
stream (Prausnitz, 2004). Since the depths of needles insertion is within the non-innervated layer of the skin, this technique would allow painless delivery and improve patient acceptance (Kaushik et al., 2001). We have previously developed a drug delivery system with microneedle pretreatment to deliver insulin successfully in liquid formulations (Qin et al., 2012; Wu et al., 2010). However, transdermal delivery of liquid formulations was difficult to handle and resulted in incorrect dosage due to leakage. In clinical application, a hydrogel formulation will be easy-touse and will not require any elaborative devices or membranes to prevent drug leakage as in solution formulations. For a microneedle system, a hydrogel patch would be preferable as it could be developed into a unit dosage form. On the other hand, since hydrogel is a three-dimensional polymer matrix that can retain large amount of water, it provides good compatibility with proteins as well as a good biocompatibility (Slaughter et al., 2009). Furthermore, the drug release rates from a hydrogel can be controlled by changing the characteristics of the hydrogel formulation during synthesis (Hoare and Kohane, 2008). Thus, the aim of this work has been to develop a hydrogel matrix for controlled release, and an insulin hydrogel formulation that could be combined with microneedle delivery system for diabetes mellitus treatment. The hydrogel synthesized by precipitation polymerization is composed of a cross-linked poly(acrylamide-coacrylic acid) (poly(AAm-co-AA)) copolymer. The resulting hydrogel was lyophilized and could be quickly rehydrated to hydrogel prior to use. The advantages of lyophilized formulation included
52
Y. Qiu et al. / International Journal of Pharmaceutics 437 (2012) 51–56
convenience in handling, transportation and storage. Moreover, a very important feature is that proteins are more stable in the solid form than in aqueous solution (Wang, 2000). The morphology, in vitro/in vivo skin permeation and stability of the lyophilized formulation was investigated and reported in this paper.
2.4. Morphology observation
2. Materials and methods
A small part of the lyophilized hydrogel was stuck on a metal stub and coated with gold by a gold sputter (6538-H sputter coater, Shimadzu, Japan) under vacuum. The particle size and morphology were analyzed by scanning electron microscopy (SEM) (S-4300, Hitachi, Japan).
2.1. Materials
2.5. Activity of insulin in the lyophilized hydrogel
Acrylamide (AAm) (HPLC grade, Merck, USA), acrylic acid (AA) (HPLC grade, Fluka, USA), human regular insulin (Novolin® R, ≥27.5 USP IU/mg, Novo Nordisk, Denmark) were used as received. N,N’-methylbisacrylamide, (MBAAm) (HPLC grade, Fluka, USA) and azobisisobutyronitrile (AIBN) (>98%, Merck, USA) were recrystallized from a methanol solution. Fluorescein iso-thiocyanate (FITC) labeled insulin (FITC-insulin) was prepared by Beijing Cheng Wen Immunochemical Laboratory, with the molar ratio of FITC to protein of 0.41. All other chemicals were of analytical grade. Double distilled water purified by a Barnstead Easypure system (Dubuque, IA, USA) with resistivity of 18.2 M cm was used throughout. Male Sprague-Dawley (SD) rats were supplied by the Beijing Vitalriver Laboratory Animal and Technology Co. Ltd. All research protocols have been carried out in accordance with The Code of Ethics of the World Medical Association (Directive 2010/63/EU for animal experiments).
The insulin hydrogel patches before/after lyophilization were first dissolved in 100 ml water, and then filtered with a 0.22 m membrane. Activity of insulin in the patches was determined by measuring the activity of insulin in the filtrate using commercial radioimmunoassay kits as described by the manufacturer (HI14HK, LINCO Research, Inc. St Charles, MO, USA). The recovery percentage = activity after lyophilization/activity before lyophilization × 100%.
2.2. Synthesis of poly(AAm-co-AA) hydrogel The polymers were synthesized using AAm/AA monomers by precipitation polymerization method. The AA monomers were added to enhance skin adhesion of hydrogel (Hoare and Kohane, 2008). However, excessive amounts of AA monomers could slow down the swelling rate. Therefore, the proportion of AAm and AA was optimized to 7:1 (mol/mol) for the comprehensive consideration of skin adhesion and swelling rate. The mixture of AA and AAm was dissolved in acetone. N,N’-methylbisacrylamide (MBAAm) was used as the cross-linking agent, and the amount was investigated in the range of 0.5–2.5 mol% of the total mole weight of monomers. The swelling experiments indicated that hydrogel was formed only when cross-linking agent was less than 1.5 mol%. The final amount of cross-linking agent was optimized to 0.5 mol% for obtaining a transparent hydrogel. AIBN was added as the free-radical initiator in the amount of 0.05 mol % of the total mole weight of monomers. Polymerization was carried out for 24 h at 56–60 ◦ C with continuous purging of nitrogen gas and gentle magnetic stirring. The resulted copolymers were allowed to cool to ambient temperature (25 ◦ C) and repeatedly rinsed with ethanol to remove the residual monomers. 2.3. Preparation of lyophilized hydrogel patches The poly(AAm-co-AA) polymer powders were added (2.5% w/w) into insulin solution (Novolin® R diluted to 25, 50, 100 IU/g by water) while stirring. Then 0.15 g of the insulin hydrogel was applied uniformly on 1 cm × 1 cm non-woven fabrics and placed into freeze-drying vials. The freeze-drying process was performed in a LGJ-12 lyophilizer (SongYuanHuaXing, China) as follows: freezing at −85 ◦ C for 24 h, and then drying at −50 ◦ C for 20 h. The chamber pressure was maintained below 15 Pa during the drying process. After lyophilization, the vials were immediately filled with dry nitrogen gas, sealed with rubber stoppers, and then kept in a desiccator until testing. The final amount of insulin in the lyophilized hydrogel patches was 3.75, 7.5, 15.0 IU/patch, respectively.
2.6. Stability of the lyophilized insulin hydrogel patches Lyophilized insulin gel patches with the original insulin activity of 3.75 IU/patch (L-gel-1), 7.5 IU/patch (L-gel-2), 15.0 IU/patch (Lgel-3) were stored in vacuum foil bags at 4 ◦ C with 65% humidity for 6 months. The activity of insulin in the patches was detected at 0, 3, 6 months, using the method described in 2.5. Here, the recovery percentage = activity at n month/activity at 0 month × 100%. The activity at 0 month is equal the activity tested immediately after lyophilization. 2.7. Microneedle assistant Arrays of solid microneedles were fabricated using the methods described in previous studies (Qiu et al., 2008). Each microneedle array has 484 micro-needles perpendicular to the wafer, over an area of 1 cm × 1 cm. Each microneedle has an octagonal pyramidal shape. The length of needles is about 150 m. With an applicator providing an insertion force approximately 2 N (Qin et al., 2012), these microneedles created microchannels that large enough for the delivery of proteins. In the following in vitro and in vivo transdermal delivery studies, the skin was pretreated by microneedles to facilitate the insulin delivery. 2.8. In vitro percutaneous delivery studies 2.8.1. Preparation of rehydrated FITC-insulin lyophilized hydrogel patches The poly(AAm-co-AA) polymer powders (2.5%, w/w) were added to FITC-insulin solution (pH 7.4, 1 mg/ml) under stirring. Then 0.15 g of the formed hydrogel was applied uniformly on 1 cm × 1 cm non-woven fabrics and lyophilized. The lyophilized formulation was rehydrated to form hydrogel by adding water until the concentration of FITC-insulin was the same as that before lyophilization. 2.8.2. Permeation studies Abdominal skin of male Sprague-Dawley rats was used for permeation studies. The skin was prepared using the method described in previous studies (Qiu et al., 2008). Franz diffusion cell having an effective permeation area of 0.636 cm2 and receiver cell volume of 2.5 ml was used for the study. Before insulin administration, rat skin pretreated by the microneedle delivery system for 20 s in an area of 1 cm2 was mounted in the vertical Franz cells. The receptor cells were constantly stirred at 37 ± 0.2 ◦ C using a magnetic stirrer at 280 rpm. The FITC-insulin hydrogel patch was applied on the
Y. Qiu et al. / International Journal of Pharmaceutics 437 (2012) 51–56
53
skin in donor compartment, and the donor compartment was covered with Parafilm® . For the control protocol, 1 mg/ml FITC-insulin solution was applied on microneedle pretreated skin. To investigate the necessity and possible advantages of microneedles, FITC-insulin hydrogel patch and solution was also applied onto the untreated intact skin. Samples were withdrawn through the sampling port of the diffusion cell at predetermined time intervals. An equal volume of receiver medium was replaced to keep a constant volume. Quantitative analysis of FITC-insulin was achieved by fluorescence spectroscopy using a fluorometer (F-2500, Hitachi, Japan), with the excitation wavelength of 490 nm and emission wavelength of 518 nm. 2.9. In vivo percutaneous delivery of insulin in diabetic rats Male SD rats weighing 250 ± 20 g were provided free access to food and water for at least 5 days to recover after transport. After fasted for 16 h, rats were induced diabetes by the intraperitoneal injection of streptozotocin (65 mg/kg, Sigma) in citrate buffer (pH = 4). To verify the induction of diabetes, the fasting blood glucose concentrations of each rat were checked at scheduled time points by FreestyleTM blood glucose meter (Abbott Diabetes Care, Inc., USA); those with the fasting blood glucose concentrations over 300 mg/dl were considered as streptozotocin induced diabetic rats (Ates et al., 2007; Kang et al., 1999). The diabetic rats were randomly divided into five groups containing eight rats each: (a) untreated negative control; (b) subcutaneous injection positive control; (c) L-gel-1 (3.75 IU/patch); (d) L-gel-2 (7.5 IU/patch); (e) L-gel-3 (15 IU/patch). The experiments were repeated four times. The abdominal hair of diabetic rats was carefully shaved 24 h before the experiments. Diabetic rats were fasted but allowed free access to water throughout the tests. In the microneedle groups, the microneedle delivery system was applied on the skin in an area of 1 cm × 2 cm and then moved. The insulin lyophilized hydrogel (L-gel) patches with the concentration of 3.75 IU/patch (L-gel-1), 7.5 IU/patch (L-gel-2), 15.0 IU/patch (L-gel3) were rehydrated and 2 patches were applied on the treated skin, protected with a medical tape. After 8 h, the patches were removed. In the injection group, regular insulin was injected subcutaneously with a hypodermic needle with a dose of 2.0 IU/kg body weight. Blood glucose levels were assessed before, during and after insulin administration at scheduled time points. The normalized blood glucose levels (NBGLs) were taken as the percentage of the mean values of pre-dose blood glucose levels. Before insulin administration and at each time point after insulin administration, 300 l of the blood sample was collected from the orbital. By centrifuging blood samples at 4000 rpm for 7 min, 100 l serum samples were obtained. All these serum samples were frozen immediately in a deep freezer at −80 ◦ C until analysis. The serum insulin levels were measured using commercial radioimmunoassay (RIA) kits as described by the manufacturer (HI-14HK, LINCO Research, Inc. St Charles, MO, USA). 2.10. Statistical analysis Each data point had all the replicate samples measured, from which the mean and standard deviation (SD) were calculated. Probability values (P-values) were calculated using analysis of variance (ANOVA). A significance level of ˛ = 0.05 was considered statistically significant. 3. Results and discussion 3.1. Characterization of the lyophilized hydrogel The lyophilized poly(AAm-co-AA) hydrogel patch was an organic whole that was not changed to a powder, which facilitated
Fig. 1. Scanning electronic microscophy image of lyophilized hydrogel with insulin loaded.
the storage of the patches. As expected, the lyophilized formulation could be rehydrated into hydrogel within 5 min by dropping water on it. Scanning electron micrograph of the lyophilized polymer was exhibited in Fig. 1, showing a porous network structure. The mesh size as determined from the micrograph is 10–30 m. The concentration of residual acrylamide monomer in the polymer detected by HPLC analysis was 0.02 g/g polymer after rinsing 5 times with ethanol, which confirmed the safety of the polymer (Andersen, 2005).
3.2. Activity of insulin in the lyophilized hydrogel Insulin was extracted from the lyophilized gel and tested for its activity in lyophilized hydrogel, right after the lyophilization process. The recovery percentage of insulin were 85.3 ± 10.2%, 87.6 ± 6.53%, 92.5 ± 5.34% for the concentration of 3.75, 7.5, 15.0 IU/patch, respectively. Activity loss appears slightly higher when insulin concentration is lower. This may be caused by higher relative proportion of insulin activity loss during extraction process.
3.3. Stability of lyophilized hydrogel patches A critical problem in the storage and delivery of pharmaceutical proteins is aggregation in the solid state induced by elevated temperature and moisture (Costantino et al., 1994). In particular, the stability of the insulin molecule in formulations is an important issue in insulin-dependent diabetes therapy, since aggregation of insulin is known to lead to severely reduced biological activity (Brange, 1987). In addition, the aggregation of insulin suspended in a polymer matrix may lead to reduced release rate. Our experimental approach was to expose the lyophilized insulin gel to 65% relative humidity at 4 ◦ C and monitor the recovery percentage of insulin. The recovery percentages of insulin with the original insulin activity of 3.75, 7.5, 15.0 IU/patch after 3 months storage were 96.2 ± 6.2%, 98.3 ± 5.7%, 99.7 ± 3.2%; after 6 months storage were 93.2 ± 3.7%, 94.5 ± 6.8%, 96.7 ± 2.5%, respectively. Activity loss appears slightly faster at lower insulin concentration. Such low concentrations presumably resulted in a lower degree of association of insulin molecules, which led to less stable insulin preparations (Brange, 1987). These preliminary results proved good stability of the lyophilized insulin gel.
54
Y. Qiu et al. / International Journal of Pharmaceutics 437 (2012) 51–56 Table 1 Pharmacodynamic parameters associated with NBGL reduction after insulin administration.
2
Cumulative amount ( µg/cm )
250 200 150 100
Group
Tmax , NBGL (h)a
L-gel-1 L-gel-2 L-gel-3 Subcutaneous injection
4.0 4.0 4.0 2.0
a b
50
c
0
2
4
6
8
Time (h) Fig. 2. Comparison of cumulative amount of FITC-Insulin penetrated through rat skin. () Rehydrated lyophilized hydrogel administration to microneedle pretreated skin, () insulin solution administration to microneedle pretreated skin, () rehydrated lyophilized hydrogel administration to intact skin, () insulin solution administration to intact skin.
3.4. Permeation studies in vitro We have investigated the effect of skin from different species on permeability of a model drug l-carnitine (Zhang et al., 2011). The results indicated that microneedle puncture overcomes the stratum corneum, the major skin barrier. Therefore, differences among the resistance of various species of skin were decreased. Based on this, a good permeability correlation was observed between SD rat skin and human skin. As a result, SD rat skin was proved to be a good model to optimizing the formulation with microneedle pretreatment. FITC-insulin as a model drug was characterized in vitro because its concentration can be easily determined by a fluorescence spectrophotometer. The cumulative amounts of FITC-insulin penetrated were calculated in the manner used in the previous study (Wu et al., 2010). In Fig. 2 the cumulative amount of permeated FITC-insulin from a rehydrated lyophilized hydrogel compared to PBS solution is shown. As expected, the cumulative amount of FITC-insulin permeated through intact skin was less than 1 g/cm2 at 8 h. The flux of FITC-insulin through rat skin was 26.0 ± 3.3 g/cm2 /h from rehydrated gel, and 23.7 ± 5.6 g/cm2 /h from the control PBS solution. There was no significant difference of the cumulative amount or skin flux between the two groups (P > 0.05). This may be due to the following reasons: (1) In the in vitro studies, concentration of FITC-insulin was detected by fluorescence analysis but not immunoassay, while stability studies suggested that the activity of insulin detected by immunoassay was slightly decreased after extraction process; (2) the rate-limiting step of in vitro permeation was the step that insulin permeating through the skin, but not that insulin releasing from the gel; (3) the mesh size of the lyophilized poly(AAm-co-AA) hydrogels was 10–30 m in diameter, as visualized using SEM. It was large enough to allow passage of insulin molecules, and the network structure confirmed the relatively fast release of drug from the hydrogels. These results indicated fast insulin release from rehydrated lyophilized gel. 3.5. Pharmacodynamics of insulin in diabetic rats
0.0 0.8 0.0 0.0
49.3 60.2 74.6 58.4
± ± ± ±
AUCNBGL (h %)c
5.0 9.1 3.4 1.6
309.1 460.0 632.8 233.1
± ± ± ±
11.8 85.6 50.7 87.4
Tmax , NBGL is the time until the maximum reduction in NBGL was achieved. NBGLmax is the maximum reduction in NBGL achieved. AUCNBGL is the integral of the NBGL vs. time curve.
untreated negative control group maintained high blood glucose values throughout the experiment. Table 1 showed that all L-gel microneedle groups reached the minimum NBGL at 4 h on average. As shown in Fig. 3 and Table 1, the insulin concentration in the patches significantly affected the pharmacodynamic properties of insulin. Blood glucose was lowered in proportion to the concentration of insulin loaded in lyophilized hydrogel patches. The linear regression of NBGLmax gave an R2 of 0.99. Simultaneously, the AUCNBGL values in the insulin profile increased with rising insulin concentration. The linear regression of AUCNBGL gave an R2 of 0.98. The concentration dependence of insulin efficacy could be explained by the concentration gradient across the skin (Barry, 2001). The relationship between blood glucose and insulin concentration loaded in hydrogel patches will provide a theoretical basis for determination of insulin concentration used during clinical application. In comparison, when regular insulin was injected, blood glucose level declined sharply and reached the minimum at 2 h on average and then rapidly increased. At 8 h, NBGL had increased to 89.8 ± 5.5%. As a result, a pronounced peak in the NBGL profiles was usually observed for subcutaneous injection of regular insulin. Although the NBGLmax and AUCNBGL of positive control was comparable to L-gel-1, the NBGL value was maintained relatively smooth from 2 to 8 h by L-gel microneedle system, especially with high doses. It implied that the duration of action for regular insulin was prolonged by percutaneous delivery. The maximum reduction in NBGL achieved in this study was more than that achieved in previous studies (Qin et al., 2012). It may be due to that in previous studies diabetic rats were treated with a cotton pad dipped with insulin solution or a drug reservoir filled
Normalized blood glucose level (%)
0
± ± ± ±
NBGLmax (%)b
120 100 80 60 40 20 0
2
4
6
8
10
12
Time (h) Changes in glucose levels among five experimental groups were monitored, as shown in Fig 3 and Table 1. All groups had blood collected periodically to measure glucose and insulin concentrations for 12 h. Consistent with our use of a diabetic animal model, the
Fig. 3. Diabetic rats blood glucose level vs. time profiles after percutaneous administration of lyophilized insulin gel and subcutaneous injection. () Untreated negative control, () subcutaneous insulin injection as positive control, () L-gel-1, () L-gel2, (䊉) L-gel-3.
Y. Qiu et al. / International Journal of Pharmaceutics 437 (2012) 51–56
55
with human regular insulin solution, which were in poor contact with rat abdominal skin, especially when the rats moved around in the cage. The poor contact with skin may impede skin permeation of insulin. In comparison, delivery by hydrogel could solve this problem, since the hydrogel had an improved contact with skin. Furthermore, the moisturizing properties of hydrogel would maintain the water content in the preparation, and subsequently prevent the patches from drying (Peppas et al., 2000).
Table 2 Pharmacokinetic parameters associated with serum insulin concentration after insulin administration.
3.6. Insulin pharmacokinetics
a Tmax , insulin is the time until the maximum serum insulin concentration was achieved. b Cmax,insulin is the maximum serum insulin concentration achieved. c AUCinsulin is the integral of the serum insulin concentration vs. time curve.
Serum insulin concentration (μIU/ml)
As a companion to the NBGL measurements, we also determined the total serum insulin concentration by radioimmunoassay. The linear measurement range was 10–200 IU/ml. The recovery for quality control samples with the insulin concentration of 10, 40 and 160 IU/ml was 85 ± 1%, 93 ± 2% and 101 ± 5%, respectively. The intra-assay and inter-assay coefficients of variation were lower than 9%. Insulin levels in each rat before receiving exogenous insulin were taken to be baseline values and subtracted from insulin levels at all time points after insulin administration respectively. The baseline values for untreated negative control, subcutaneous injection positive control, and L-gel-1, L-gel-2, L-gel-3 groups were 8.5 ± 0.5, 7.1 ± 0.8, 9.2 ± 0.2, 10.6 ± 3.4, 14.6 ± 1.7 IU/ml, respectively. This background correction was necessary because the radioimmunoassay kit used for analysis could not totally distinguish between rat endogenous insulin and delivered regular insulin. Thus, insulin levels at zero-time point were zero for all groups. Consistent with the NBGL measurements, serum insulin concentration increased after insulin administration, extent and kinetics lay in the method of administration, as shown in Fig. 4. The main pharmacokinetic parameters are presented in Table 2. Consistent with the expected dose–response curve, Cmax,insulin and AUCinsulin by transdermal delivery of L-gel depended on the insulin concentration loaded. The linear regression of Cmax,insulin and AUCinsulin gave an R2 of (0.99) and (0.99), respectively. As the administered insulin concentration increased, the extent of changes with Cmax,insulin was more than that with NBGLmax . It may be caused by the lack of self secretion of insulin in diabetic rats. When insulin treatment was started, there was a strong stress response, the blood glucose levels decreased significantly. But when the serum insulin concentration increased to a certain extent, the stress response in rats declined, and the blood glucose levels decreased slightly.
200
150
100
50
0 0
2
4
6
8
10
Group
Tmax , insulin (h)a
L-gel-1 L-gel-2 L-gel-3 Subcutaneous injection
4.0 4.0 6.5 2.0
± ± ± ±
0.0 0.0 0.5 0.0
Cmax,insulin (IU/ml)b 43.0 94.4 182.0 155.9
± ± ± ±
6.6 30.8 17.6 13.2
AUCinsulin (IU h/ml)c 226.5 708.7 1348.5 354.5
± ± ± ±
89.9 182.9 82.7 36.0
As it shown in Table 2, the AUCinsulin of subcutaneous injection was relative to L-gel-1 (P > 0.05). The insulin dose for subcutaneous injection was 2 IU/kg body weight. For the test subjects, with a mean body weight of 250 ± 20 g, the dose for subcutaneous injection is 0.5 ± 0.04 IU per rat. The amount of insulin incorporated in L-gel-1 was 3.75 IU/patch. For each rat, 2 patches were administrated, so the dose was 7.5 IU/rat. The relative bioavailability of microneedle-mediated L-gel patch versus subcutaneous injection was 6.7%. It demonstrated the effectiveness of noninvasive insulin delivery using the novel system. After subcutaneous injection, the serum insulin was up to the maximum concentration at 2 h, but rapidly decreased to 7.4 ± 6.8 IU/ml at 4 h. By microneedle-mediated L-gel delivery, the serum insulin level was maintained relatively smooth from 2 to 8 h, showing a controlled release of insulin. After patch removal at 8 h, serum level of insulin rapidly declined, suggesting insulin percutaneously delivery into systemic circulation. The pharmacokinetic study demonstrated that there was a sustained and steady release of insulin during the whole administration period of 8 h by L-gel microneedle system. Therefore, microneedle-mediated L-gel delivery would provide a convenient and effective administration strategy for sustained and controlled release of insulin, and potentially improve patient compliance with additional benefits such as reduced hypoglycemia. 4. Conclusions In this study, we showed for the first time that lyophilized hydrogel patches could be prepared for insulin delivery. The patches provided a convenient and effective administration strategy for microneedle-mediated insulin delivery. The L-gel patches were easy to storage and could be rehydrated into hydrogel within minutes by dropping water on them. In vitro studies showed similar release behavior between the insulin loaded rehydrated L-gel and solution. Pharmacodynamic studies showed sustained and steady hypoglycemic effect could be obtained by insulin L-gel patch, compared to peak and valley curve associated with subcutaneous injection. The pharmacokinetic studies were consistent with the pharmacodynamic results. Overall, lyophilized hydrogel patches present a promising and beneficial addition to the microneedle-mediated insulin delivery. The optimization based on this formulation provides a promising foundation for mironeedlemediated delivery of other macromolecules such as proteins and peptides.
12
Time (h) Fig. 4. Diabetic rats’ serum insulin concentration vs. time profiles after percutaneous administration of lyophilized insulin gel and subcutaneous injection. () Untreated negative control, () subcutaneous insulin injection as positive control, () L-gel-1, () L-gel-2, (䊉) L-gel-3.
Acknowledgements This work was supported by the Science Foundation of the Chinese Academy of Sciences and Suzhou Natong Bionanotechnology Co. Ltd, Jiangsu, China.
56
Y. Qiu et al. / International Journal of Pharmaceutics 437 (2012) 51–56
References Andersen, F.A., 2005. Amended final report on the safety assessment of polyacrylamide and acrylamide residues in cosmetics. Int. J. Toxicol. 24, 21–50. Asche, C.V., Shane-McWhorter, L., Raparla, S., 2010. Health economics and compliance of vials/syringes versus pen devices: a review of the evidence. Diabetes Technol. Ther. 12, S101–S108. Ates, O., Cayli, S.R., Yucel, N., Altinoz, E., Kocak, A., Durak, M.A., Turkoz, Y., Yologlu, S., 2007. Central nervous system protection by resveratrol in streptozotocininduced diabetic rats. J. Clin. Neurosci. 14, 256–260. Barry, B.W., 2001. Novel mechanisms and devices to enable successful transdermal drug delivery. Eur. J. Pharm. Sci. 14, 101–114. Brange, J., 1987. Galenics of insulin: the physico-chemical and pharmaceutical aspects of insulin and insulin preparations. Springer-Verlag, Berlin. Buysschaert, M., Lambert, A.E., 1989. The insulin pen. Lancet 1, 965. Costantino, H.R., Langer, R., Klibanov, A.M., 1994. Moisture-induced aggregation of lyophilized insulin. Pharm. Res. 11, 21–29. Hoare, T.R., Kohane, D.S., 2008. Hydrogels in drug delivery: progress and challenges. Polymer 49, 1993–2007. Kang, N.L., Alexander, G., Park, J.K., Maasch, C., Buchwalow, I., Luft, F.C., Haller, H., 1999. Differential expression of protein kinase C isoforms in streptozotocininduced diabetic rats. Kidney Int. 56, 1737–1750. Kaushik, S., Hord, A.H., Denson, D.D., McAllister, D.V., Smitra, S., Allen, M.G., Prausnitz, M.R., 2001. Lack of pain associated with microfabricated microneedles. Anesth. Analg. 92, 502–504. Lenhard, M.J., Reeves, G.D., 2001. Continuous subcutaneous insulin infusion: a comprehensive review of insulin pump therapy. Arch. Inter. Med. 161, 2293–2300. McAllister, D.V., Wang, P.M., Davis, S.P., Park, J.H., Canatella, P.J., Allen, M.G., Prausnitz, M.R., 2003. Microfabricated needles for transdermal delivery of
macromolecules and nanoparticles: fabrication methods and transport studies. Proc. Natl. Acad. Sci. U.S.A. 100, 13755–13760. Peppas, N.A., Bures, P., Leobandung, W., Ichikawa, H., 2000. Hydrogels in pharmaceutical formulations. Eur. J. Pharm. Biopharm. 50, 27–46. Prausnitz, M.R., 2004. Microneedles for transdermal drug delivery. Adv. Drug Deliv. Rev. 56, 581–587. Qin, G., Gao, Y., Wu, Y., Zhang, S., Qiu, Y., Li, F., Xu, B., 2012. Simultaneous basalbolus delivery of fast-acting insulin and its significance in diabetes management. Nanomed.: Nanotechnol. Biol. Med. 8, 221–227. Qiu, Y., Gao, Y., Hu, K., Li, F., 2008. Enhancement of skin permeation of docetaxel: a novel approach combining microneedle and elastic liposomes. J. Control. Release 129, 144–150. Siekmeier, R., Scheuch, G., 2008. Inhaled insulin: does it become reality. J. Physiol. Pharmacol. 59, 81–113. Slaughter, B.V., Khurshid, S.S., Fisher, O.Z., Khademhosseini, A., Peppas, N.A., 2009. Hydrogels in regenerative medicine. Adv. Mater. 21, 3307–3329. Wang, W., 2000. Lyophilization and development of solid protein pharmaceuticals. Int. J. Pharm. 203, 1–60. Weng, J., Li, Y., Xu, W., Shi, L., Zhang, Q., Zhu, D., Hu, Y., Zhou, Z., Yan, X., Tian, H., Ran, X., Luo, Z., Xian, J., Yan, L., Li, F., Zeng, L., Chen, Y., Yang, L., Yan, S., Liu, J., Li, M., Fu, Z., Cheng, H., 2008. Effect of intensive insulin therapy on betacell function and glycaemic control in patients with newly diagnosed type 2 diabetes: a multicentre randomised parallel-group trial. Lancet 371, 1753–1760. Wu, Y., Gao, Y., Qin, G., Zhang, S., Qiu, Y., Li, F., Xu, B., 2010. Sustained release of insulin through skin by intradermal microdelivery system. Biomed. Microdevices 12, 665–671. Zhang, S.H., Qin, G.J., Wu, Y., Gao, Y.H., Qiu, Y.Q., Li, F., Xu, B., 2011. Enhanced bioavailability of l-carnitine after painless intradermal delivery vs. oral administration in rats. Pharm. Res. 28, 117–123.
Contents lists available at SciVerse ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Pharmaceutical Nanotechnology
Novel lyophilized hydrogel patches for convenient and effective administration of microneedle-mediated insulin delivery Yuqin Qiu a , Guangjiong Qin a,b , Suohui Zhang a , Yan Wu a,b , Bai Xu a , Yunhua Gao a,∗ a b
Key Laboratory of Photochemical Conversion and Optoelectronic Materials, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing 100190, China Graduate University Chinese Academy of Sciences, Beijing 100049, China
a r t i c l e
i n f o
Article history: Received 2 February 2012 Received in revised form 29 June 2012 Accepted 22 July 2012 Available online 27 July 2012 Keywords: Lyophilized hydrogel Transdermal drug delivery Insulin Microneedle Diabetes
a b s t r a c t A lyophilized hydrogel patch system was developed for microneedle-mediated insulin delivery. The matrix of Cross-linked poly(acrylamide-co-acrylic acid) were synthesized by precipitation polymerization. Recombinant human Insulin was loaded into the lyophilized polymer matrix, which can be rehydrated by water. After the hydrated patch was applied to the abdominal skin of diabetic rats after microneedle pretreatment, pharmacodynamics and pharmacokinetics evaluation was performed. The blood samples were collected to monitor blood glucose and serum insulin levels for 12 h. Blood glucose was lowered in proportion to the concentration of insulin loaded in lyophilized hydrogel patches (R2 = 0.99), with a longer duration of action compared to subcutaneous injection. Stability study confirmed more than 90% of insulin activity was retained in lyophilized hydrogel after 6 months of storage at 4 ◦ C. In conclusion, hydrogel patches were demonstrated to be appropriate drug reservoir for sustained release of insulin with microneedle mediated transdermal delivery. © 2012 Elsevier B.V. All rights reserved.
1. Introduction All type 1 and many type 2 diabetes mellitus patients need insulin to achieve glycaemic control, and it often requires multiple insulin administration per day (Weng et al., 2008). Typically, insulin is given by subcutaneous needle injection such as self injection, insulin pen (Buysschaert and Lambert, 1989) and catheters connected to insulin pumps (Lenhard and Reeves, 2001). However, these methods are inconvenient and painful, leading to poor compliance and poor diabetes management (Asche et al., 2010). Inhaled insulin, Exubera® (Pfizer, New York, NY), was introduced as the first FDA-approved needle-free alternative. However, it was subsequently discontinued due to various reasons such as high cost, device design, side effects and unknown health risks (Siekmeier and Scheuch, 2008). Other insulin delivery routes have been explored without compromising with the comfort and future health of patients. Recently, though, attention has been drawn to the possibility of using a microneedle system to deliver insulin (McAllister et al., 2003). The system consists of short, micrometer-scale needles, which can temporarily compromise the skin barrier and allow the drug loaded in a patch to diffuse to the rich capillary bed of the dermis for uptake and subsequent systemic distribution in the blood
∗ Corresponding author. Tel.: +86 10 82543581; fax: +86 10 82543581. E-mail address: [email protected] (Y. Gao). 0378-5173/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ijpharm.2012.07.035
stream (Prausnitz, 2004). Since the depths of needles insertion is within the non-innervated layer of the skin, this technique would allow painless delivery and improve patient acceptance (Kaushik et al., 2001). We have previously developed a drug delivery system with microneedle pretreatment to deliver insulin successfully in liquid formulations (Qin et al., 2012; Wu et al., 2010). However, transdermal delivery of liquid formulations was difficult to handle and resulted in incorrect dosage due to leakage. In clinical application, a hydrogel formulation will be easy-touse and will not require any elaborative devices or membranes to prevent drug leakage as in solution formulations. For a microneedle system, a hydrogel patch would be preferable as it could be developed into a unit dosage form. On the other hand, since hydrogel is a three-dimensional polymer matrix that can retain large amount of water, it provides good compatibility with proteins as well as a good biocompatibility (Slaughter et al., 2009). Furthermore, the drug release rates from a hydrogel can be controlled by changing the characteristics of the hydrogel formulation during synthesis (Hoare and Kohane, 2008). Thus, the aim of this work has been to develop a hydrogel matrix for controlled release, and an insulin hydrogel formulation that could be combined with microneedle delivery system for diabetes mellitus treatment. The hydrogel synthesized by precipitation polymerization is composed of a cross-linked poly(acrylamide-coacrylic acid) (poly(AAm-co-AA)) copolymer. The resulting hydrogel was lyophilized and could be quickly rehydrated to hydrogel prior to use. The advantages of lyophilized formulation included
52
Y. Qiu et al. / International Journal of Pharmaceutics 437 (2012) 51–56
convenience in handling, transportation and storage. Moreover, a very important feature is that proteins are more stable in the solid form than in aqueous solution (Wang, 2000). The morphology, in vitro/in vivo skin permeation and stability of the lyophilized formulation was investigated and reported in this paper.
2.4. Morphology observation
2. Materials and methods
A small part of the lyophilized hydrogel was stuck on a metal stub and coated with gold by a gold sputter (6538-H sputter coater, Shimadzu, Japan) under vacuum. The particle size and morphology were analyzed by scanning electron microscopy (SEM) (S-4300, Hitachi, Japan).
2.1. Materials
2.5. Activity of insulin in the lyophilized hydrogel
Acrylamide (AAm) (HPLC grade, Merck, USA), acrylic acid (AA) (HPLC grade, Fluka, USA), human regular insulin (Novolin® R, ≥27.5 USP IU/mg, Novo Nordisk, Denmark) were used as received. N,N’-methylbisacrylamide, (MBAAm) (HPLC grade, Fluka, USA) and azobisisobutyronitrile (AIBN) (>98%, Merck, USA) were recrystallized from a methanol solution. Fluorescein iso-thiocyanate (FITC) labeled insulin (FITC-insulin) was prepared by Beijing Cheng Wen Immunochemical Laboratory, with the molar ratio of FITC to protein of 0.41. All other chemicals were of analytical grade. Double distilled water purified by a Barnstead Easypure system (Dubuque, IA, USA) with resistivity of 18.2 M cm was used throughout. Male Sprague-Dawley (SD) rats were supplied by the Beijing Vitalriver Laboratory Animal and Technology Co. Ltd. All research protocols have been carried out in accordance with The Code of Ethics of the World Medical Association (Directive 2010/63/EU for animal experiments).
The insulin hydrogel patches before/after lyophilization were first dissolved in 100 ml water, and then filtered with a 0.22 m membrane. Activity of insulin in the patches was determined by measuring the activity of insulin in the filtrate using commercial radioimmunoassay kits as described by the manufacturer (HI14HK, LINCO Research, Inc. St Charles, MO, USA). The recovery percentage = activity after lyophilization/activity before lyophilization × 100%.
2.2. Synthesis of poly(AAm-co-AA) hydrogel The polymers were synthesized using AAm/AA monomers by precipitation polymerization method. The AA monomers were added to enhance skin adhesion of hydrogel (Hoare and Kohane, 2008). However, excessive amounts of AA monomers could slow down the swelling rate. Therefore, the proportion of AAm and AA was optimized to 7:1 (mol/mol) for the comprehensive consideration of skin adhesion and swelling rate. The mixture of AA and AAm was dissolved in acetone. N,N’-methylbisacrylamide (MBAAm) was used as the cross-linking agent, and the amount was investigated in the range of 0.5–2.5 mol% of the total mole weight of monomers. The swelling experiments indicated that hydrogel was formed only when cross-linking agent was less than 1.5 mol%. The final amount of cross-linking agent was optimized to 0.5 mol% for obtaining a transparent hydrogel. AIBN was added as the free-radical initiator in the amount of 0.05 mol % of the total mole weight of monomers. Polymerization was carried out for 24 h at 56–60 ◦ C with continuous purging of nitrogen gas and gentle magnetic stirring. The resulted copolymers were allowed to cool to ambient temperature (25 ◦ C) and repeatedly rinsed with ethanol to remove the residual monomers. 2.3. Preparation of lyophilized hydrogel patches The poly(AAm-co-AA) polymer powders were added (2.5% w/w) into insulin solution (Novolin® R diluted to 25, 50, 100 IU/g by water) while stirring. Then 0.15 g of the insulin hydrogel was applied uniformly on 1 cm × 1 cm non-woven fabrics and placed into freeze-drying vials. The freeze-drying process was performed in a LGJ-12 lyophilizer (SongYuanHuaXing, China) as follows: freezing at −85 ◦ C for 24 h, and then drying at −50 ◦ C for 20 h. The chamber pressure was maintained below 15 Pa during the drying process. After lyophilization, the vials were immediately filled with dry nitrogen gas, sealed with rubber stoppers, and then kept in a desiccator until testing. The final amount of insulin in the lyophilized hydrogel patches was 3.75, 7.5, 15.0 IU/patch, respectively.
2.6. Stability of the lyophilized insulin hydrogel patches Lyophilized insulin gel patches with the original insulin activity of 3.75 IU/patch (L-gel-1), 7.5 IU/patch (L-gel-2), 15.0 IU/patch (Lgel-3) were stored in vacuum foil bags at 4 ◦ C with 65% humidity for 6 months. The activity of insulin in the patches was detected at 0, 3, 6 months, using the method described in 2.5. Here, the recovery percentage = activity at n month/activity at 0 month × 100%. The activity at 0 month is equal the activity tested immediately after lyophilization. 2.7. Microneedle assistant Arrays of solid microneedles were fabricated using the methods described in previous studies (Qiu et al., 2008). Each microneedle array has 484 micro-needles perpendicular to the wafer, over an area of 1 cm × 1 cm. Each microneedle has an octagonal pyramidal shape. The length of needles is about 150 m. With an applicator providing an insertion force approximately 2 N (Qin et al., 2012), these microneedles created microchannels that large enough for the delivery of proteins. In the following in vitro and in vivo transdermal delivery studies, the skin was pretreated by microneedles to facilitate the insulin delivery. 2.8. In vitro percutaneous delivery studies 2.8.1. Preparation of rehydrated FITC-insulin lyophilized hydrogel patches The poly(AAm-co-AA) polymer powders (2.5%, w/w) were added to FITC-insulin solution (pH 7.4, 1 mg/ml) under stirring. Then 0.15 g of the formed hydrogel was applied uniformly on 1 cm × 1 cm non-woven fabrics and lyophilized. The lyophilized formulation was rehydrated to form hydrogel by adding water until the concentration of FITC-insulin was the same as that before lyophilization. 2.8.2. Permeation studies Abdominal skin of male Sprague-Dawley rats was used for permeation studies. The skin was prepared using the method described in previous studies (Qiu et al., 2008). Franz diffusion cell having an effective permeation area of 0.636 cm2 and receiver cell volume of 2.5 ml was used for the study. Before insulin administration, rat skin pretreated by the microneedle delivery system for 20 s in an area of 1 cm2 was mounted in the vertical Franz cells. The receptor cells were constantly stirred at 37 ± 0.2 ◦ C using a magnetic stirrer at 280 rpm. The FITC-insulin hydrogel patch was applied on the
Y. Qiu et al. / International Journal of Pharmaceutics 437 (2012) 51–56
53
skin in donor compartment, and the donor compartment was covered with Parafilm® . For the control protocol, 1 mg/ml FITC-insulin solution was applied on microneedle pretreated skin. To investigate the necessity and possible advantages of microneedles, FITC-insulin hydrogel patch and solution was also applied onto the untreated intact skin. Samples were withdrawn through the sampling port of the diffusion cell at predetermined time intervals. An equal volume of receiver medium was replaced to keep a constant volume. Quantitative analysis of FITC-insulin was achieved by fluorescence spectroscopy using a fluorometer (F-2500, Hitachi, Japan), with the excitation wavelength of 490 nm and emission wavelength of 518 nm. 2.9. In vivo percutaneous delivery of insulin in diabetic rats Male SD rats weighing 250 ± 20 g were provided free access to food and water for at least 5 days to recover after transport. After fasted for 16 h, rats were induced diabetes by the intraperitoneal injection of streptozotocin (65 mg/kg, Sigma) in citrate buffer (pH = 4). To verify the induction of diabetes, the fasting blood glucose concentrations of each rat were checked at scheduled time points by FreestyleTM blood glucose meter (Abbott Diabetes Care, Inc., USA); those with the fasting blood glucose concentrations over 300 mg/dl were considered as streptozotocin induced diabetic rats (Ates et al., 2007; Kang et al., 1999). The diabetic rats were randomly divided into five groups containing eight rats each: (a) untreated negative control; (b) subcutaneous injection positive control; (c) L-gel-1 (3.75 IU/patch); (d) L-gel-2 (7.5 IU/patch); (e) L-gel-3 (15 IU/patch). The experiments were repeated four times. The abdominal hair of diabetic rats was carefully shaved 24 h before the experiments. Diabetic rats were fasted but allowed free access to water throughout the tests. In the microneedle groups, the microneedle delivery system was applied on the skin in an area of 1 cm × 2 cm and then moved. The insulin lyophilized hydrogel (L-gel) patches with the concentration of 3.75 IU/patch (L-gel-1), 7.5 IU/patch (L-gel-2), 15.0 IU/patch (L-gel3) were rehydrated and 2 patches were applied on the treated skin, protected with a medical tape. After 8 h, the patches were removed. In the injection group, regular insulin was injected subcutaneously with a hypodermic needle with a dose of 2.0 IU/kg body weight. Blood glucose levels were assessed before, during and after insulin administration at scheduled time points. The normalized blood glucose levels (NBGLs) were taken as the percentage of the mean values of pre-dose blood glucose levels. Before insulin administration and at each time point after insulin administration, 300 l of the blood sample was collected from the orbital. By centrifuging blood samples at 4000 rpm for 7 min, 100 l serum samples were obtained. All these serum samples were frozen immediately in a deep freezer at −80 ◦ C until analysis. The serum insulin levels were measured using commercial radioimmunoassay (RIA) kits as described by the manufacturer (HI-14HK, LINCO Research, Inc. St Charles, MO, USA). 2.10. Statistical analysis Each data point had all the replicate samples measured, from which the mean and standard deviation (SD) were calculated. Probability values (P-values) were calculated using analysis of variance (ANOVA). A significance level of ˛ = 0.05 was considered statistically significant. 3. Results and discussion 3.1. Characterization of the lyophilized hydrogel The lyophilized poly(AAm-co-AA) hydrogel patch was an organic whole that was not changed to a powder, which facilitated
Fig. 1. Scanning electronic microscophy image of lyophilized hydrogel with insulin loaded.
the storage of the patches. As expected, the lyophilized formulation could be rehydrated into hydrogel within 5 min by dropping water on it. Scanning electron micrograph of the lyophilized polymer was exhibited in Fig. 1, showing a porous network structure. The mesh size as determined from the micrograph is 10–30 m. The concentration of residual acrylamide monomer in the polymer detected by HPLC analysis was 0.02 g/g polymer after rinsing 5 times with ethanol, which confirmed the safety of the polymer (Andersen, 2005).
3.2. Activity of insulin in the lyophilized hydrogel Insulin was extracted from the lyophilized gel and tested for its activity in lyophilized hydrogel, right after the lyophilization process. The recovery percentage of insulin were 85.3 ± 10.2%, 87.6 ± 6.53%, 92.5 ± 5.34% for the concentration of 3.75, 7.5, 15.0 IU/patch, respectively. Activity loss appears slightly higher when insulin concentration is lower. This may be caused by higher relative proportion of insulin activity loss during extraction process.
3.3. Stability of lyophilized hydrogel patches A critical problem in the storage and delivery of pharmaceutical proteins is aggregation in the solid state induced by elevated temperature and moisture (Costantino et al., 1994). In particular, the stability of the insulin molecule in formulations is an important issue in insulin-dependent diabetes therapy, since aggregation of insulin is known to lead to severely reduced biological activity (Brange, 1987). In addition, the aggregation of insulin suspended in a polymer matrix may lead to reduced release rate. Our experimental approach was to expose the lyophilized insulin gel to 65% relative humidity at 4 ◦ C and monitor the recovery percentage of insulin. The recovery percentages of insulin with the original insulin activity of 3.75, 7.5, 15.0 IU/patch after 3 months storage were 96.2 ± 6.2%, 98.3 ± 5.7%, 99.7 ± 3.2%; after 6 months storage were 93.2 ± 3.7%, 94.5 ± 6.8%, 96.7 ± 2.5%, respectively. Activity loss appears slightly faster at lower insulin concentration. Such low concentrations presumably resulted in a lower degree of association of insulin molecules, which led to less stable insulin preparations (Brange, 1987). These preliminary results proved good stability of the lyophilized insulin gel.
54
Y. Qiu et al. / International Journal of Pharmaceutics 437 (2012) 51–56 Table 1 Pharmacodynamic parameters associated with NBGL reduction after insulin administration.
2
Cumulative amount ( µg/cm )
250 200 150 100
Group
Tmax , NBGL (h)a
L-gel-1 L-gel-2 L-gel-3 Subcutaneous injection
4.0 4.0 4.0 2.0
a b
50
c
0
2
4
6
8
Time (h) Fig. 2. Comparison of cumulative amount of FITC-Insulin penetrated through rat skin. () Rehydrated lyophilized hydrogel administration to microneedle pretreated skin, () insulin solution administration to microneedle pretreated skin, () rehydrated lyophilized hydrogel administration to intact skin, () insulin solution administration to intact skin.
3.4. Permeation studies in vitro We have investigated the effect of skin from different species on permeability of a model drug l-carnitine (Zhang et al., 2011). The results indicated that microneedle puncture overcomes the stratum corneum, the major skin barrier. Therefore, differences among the resistance of various species of skin were decreased. Based on this, a good permeability correlation was observed between SD rat skin and human skin. As a result, SD rat skin was proved to be a good model to optimizing the formulation with microneedle pretreatment. FITC-insulin as a model drug was characterized in vitro because its concentration can be easily determined by a fluorescence spectrophotometer. The cumulative amounts of FITC-insulin penetrated were calculated in the manner used in the previous study (Wu et al., 2010). In Fig. 2 the cumulative amount of permeated FITC-insulin from a rehydrated lyophilized hydrogel compared to PBS solution is shown. As expected, the cumulative amount of FITC-insulin permeated through intact skin was less than 1 g/cm2 at 8 h. The flux of FITC-insulin through rat skin was 26.0 ± 3.3 g/cm2 /h from rehydrated gel, and 23.7 ± 5.6 g/cm2 /h from the control PBS solution. There was no significant difference of the cumulative amount or skin flux between the two groups (P > 0.05). This may be due to the following reasons: (1) In the in vitro studies, concentration of FITC-insulin was detected by fluorescence analysis but not immunoassay, while stability studies suggested that the activity of insulin detected by immunoassay was slightly decreased after extraction process; (2) the rate-limiting step of in vitro permeation was the step that insulin permeating through the skin, but not that insulin releasing from the gel; (3) the mesh size of the lyophilized poly(AAm-co-AA) hydrogels was 10–30 m in diameter, as visualized using SEM. It was large enough to allow passage of insulin molecules, and the network structure confirmed the relatively fast release of drug from the hydrogels. These results indicated fast insulin release from rehydrated lyophilized gel. 3.5. Pharmacodynamics of insulin in diabetic rats
0.0 0.8 0.0 0.0
49.3 60.2 74.6 58.4
± ± ± ±
AUCNBGL (h %)c
5.0 9.1 3.4 1.6
309.1 460.0 632.8 233.1
± ± ± ±
11.8 85.6 50.7 87.4
Tmax , NBGL is the time until the maximum reduction in NBGL was achieved. NBGLmax is the maximum reduction in NBGL achieved. AUCNBGL is the integral of the NBGL vs. time curve.
untreated negative control group maintained high blood glucose values throughout the experiment. Table 1 showed that all L-gel microneedle groups reached the minimum NBGL at 4 h on average. As shown in Fig. 3 and Table 1, the insulin concentration in the patches significantly affected the pharmacodynamic properties of insulin. Blood glucose was lowered in proportion to the concentration of insulin loaded in lyophilized hydrogel patches. The linear regression of NBGLmax gave an R2 of 0.99. Simultaneously, the AUCNBGL values in the insulin profile increased with rising insulin concentration. The linear regression of AUCNBGL gave an R2 of 0.98. The concentration dependence of insulin efficacy could be explained by the concentration gradient across the skin (Barry, 2001). The relationship between blood glucose and insulin concentration loaded in hydrogel patches will provide a theoretical basis for determination of insulin concentration used during clinical application. In comparison, when regular insulin was injected, blood glucose level declined sharply and reached the minimum at 2 h on average and then rapidly increased. At 8 h, NBGL had increased to 89.8 ± 5.5%. As a result, a pronounced peak in the NBGL profiles was usually observed for subcutaneous injection of regular insulin. Although the NBGLmax and AUCNBGL of positive control was comparable to L-gel-1, the NBGL value was maintained relatively smooth from 2 to 8 h by L-gel microneedle system, especially with high doses. It implied that the duration of action for regular insulin was prolonged by percutaneous delivery. The maximum reduction in NBGL achieved in this study was more than that achieved in previous studies (Qin et al., 2012). It may be due to that in previous studies diabetic rats were treated with a cotton pad dipped with insulin solution or a drug reservoir filled
Normalized blood glucose level (%)
0
± ± ± ±
NBGLmax (%)b
120 100 80 60 40 20 0
2
4
6
8
10
12
Time (h) Changes in glucose levels among five experimental groups were monitored, as shown in Fig 3 and Table 1. All groups had blood collected periodically to measure glucose and insulin concentrations for 12 h. Consistent with our use of a diabetic animal model, the
Fig. 3. Diabetic rats blood glucose level vs. time profiles after percutaneous administration of lyophilized insulin gel and subcutaneous injection. () Untreated negative control, () subcutaneous insulin injection as positive control, () L-gel-1, () L-gel2, (䊉) L-gel-3.
Y. Qiu et al. / International Journal of Pharmaceutics 437 (2012) 51–56
55
with human regular insulin solution, which were in poor contact with rat abdominal skin, especially when the rats moved around in the cage. The poor contact with skin may impede skin permeation of insulin. In comparison, delivery by hydrogel could solve this problem, since the hydrogel had an improved contact with skin. Furthermore, the moisturizing properties of hydrogel would maintain the water content in the preparation, and subsequently prevent the patches from drying (Peppas et al., 2000).
Table 2 Pharmacokinetic parameters associated with serum insulin concentration after insulin administration.
3.6. Insulin pharmacokinetics
a Tmax , insulin is the time until the maximum serum insulin concentration was achieved. b Cmax,insulin is the maximum serum insulin concentration achieved. c AUCinsulin is the integral of the serum insulin concentration vs. time curve.
Serum insulin concentration (μIU/ml)
As a companion to the NBGL measurements, we also determined the total serum insulin concentration by radioimmunoassay. The linear measurement range was 10–200 IU/ml. The recovery for quality control samples with the insulin concentration of 10, 40 and 160 IU/ml was 85 ± 1%, 93 ± 2% and 101 ± 5%, respectively. The intra-assay and inter-assay coefficients of variation were lower than 9%. Insulin levels in each rat before receiving exogenous insulin were taken to be baseline values and subtracted from insulin levels at all time points after insulin administration respectively. The baseline values for untreated negative control, subcutaneous injection positive control, and L-gel-1, L-gel-2, L-gel-3 groups were 8.5 ± 0.5, 7.1 ± 0.8, 9.2 ± 0.2, 10.6 ± 3.4, 14.6 ± 1.7 IU/ml, respectively. This background correction was necessary because the radioimmunoassay kit used for analysis could not totally distinguish between rat endogenous insulin and delivered regular insulin. Thus, insulin levels at zero-time point were zero for all groups. Consistent with the NBGL measurements, serum insulin concentration increased after insulin administration, extent and kinetics lay in the method of administration, as shown in Fig. 4. The main pharmacokinetic parameters are presented in Table 2. Consistent with the expected dose–response curve, Cmax,insulin and AUCinsulin by transdermal delivery of L-gel depended on the insulin concentration loaded. The linear regression of Cmax,insulin and AUCinsulin gave an R2 of (0.99) and (0.99), respectively. As the administered insulin concentration increased, the extent of changes with Cmax,insulin was more than that with NBGLmax . It may be caused by the lack of self secretion of insulin in diabetic rats. When insulin treatment was started, there was a strong stress response, the blood glucose levels decreased significantly. But when the serum insulin concentration increased to a certain extent, the stress response in rats declined, and the blood glucose levels decreased slightly.
200
150
100
50
0 0
2
4
6
8
10
Group
Tmax , insulin (h)a
L-gel-1 L-gel-2 L-gel-3 Subcutaneous injection
4.0 4.0 6.5 2.0
± ± ± ±
0.0 0.0 0.5 0.0
Cmax,insulin (IU/ml)b 43.0 94.4 182.0 155.9
± ± ± ±
6.6 30.8 17.6 13.2
AUCinsulin (IU h/ml)c 226.5 708.7 1348.5 354.5
± ± ± ±
89.9 182.9 82.7 36.0
As it shown in Table 2, the AUCinsulin of subcutaneous injection was relative to L-gel-1 (P > 0.05). The insulin dose for subcutaneous injection was 2 IU/kg body weight. For the test subjects, with a mean body weight of 250 ± 20 g, the dose for subcutaneous injection is 0.5 ± 0.04 IU per rat. The amount of insulin incorporated in L-gel-1 was 3.75 IU/patch. For each rat, 2 patches were administrated, so the dose was 7.5 IU/rat. The relative bioavailability of microneedle-mediated L-gel patch versus subcutaneous injection was 6.7%. It demonstrated the effectiveness of noninvasive insulin delivery using the novel system. After subcutaneous injection, the serum insulin was up to the maximum concentration at 2 h, but rapidly decreased to 7.4 ± 6.8 IU/ml at 4 h. By microneedle-mediated L-gel delivery, the serum insulin level was maintained relatively smooth from 2 to 8 h, showing a controlled release of insulin. After patch removal at 8 h, serum level of insulin rapidly declined, suggesting insulin percutaneously delivery into systemic circulation. The pharmacokinetic study demonstrated that there was a sustained and steady release of insulin during the whole administration period of 8 h by L-gel microneedle system. Therefore, microneedle-mediated L-gel delivery would provide a convenient and effective administration strategy for sustained and controlled release of insulin, and potentially improve patient compliance with additional benefits such as reduced hypoglycemia. 4. Conclusions In this study, we showed for the first time that lyophilized hydrogel patches could be prepared for insulin delivery. The patches provided a convenient and effective administration strategy for microneedle-mediated insulin delivery. The L-gel patches were easy to storage and could be rehydrated into hydrogel within minutes by dropping water on them. In vitro studies showed similar release behavior between the insulin loaded rehydrated L-gel and solution. Pharmacodynamic studies showed sustained and steady hypoglycemic effect could be obtained by insulin L-gel patch, compared to peak and valley curve associated with subcutaneous injection. The pharmacokinetic studies were consistent with the pharmacodynamic results. Overall, lyophilized hydrogel patches present a promising and beneficial addition to the microneedle-mediated insulin delivery. The optimization based on this formulation provides a promising foundation for mironeedlemediated delivery of other macromolecules such as proteins and peptides.
12
Time (h) Fig. 4. Diabetic rats’ serum insulin concentration vs. time profiles after percutaneous administration of lyophilized insulin gel and subcutaneous injection. () Untreated negative control, () subcutaneous insulin injection as positive control, () L-gel-1, () L-gel-2, (䊉) L-gel-3.
Acknowledgements This work was supported by the Science Foundation of the Chinese Academy of Sciences and Suzhou Natong Bionanotechnology Co. Ltd, Jiangsu, China.
56
Y. Qiu et al. / International Journal of Pharmaceutics 437 (2012) 51–56
References Andersen, F.A., 2005. Amended final report on the safety assessment of polyacrylamide and acrylamide residues in cosmetics. Int. J. Toxicol. 24, 21–50. Asche, C.V., Shane-McWhorter, L., Raparla, S., 2010. Health economics and compliance of vials/syringes versus pen devices: a review of the evidence. Diabetes Technol. Ther. 12, S101–S108. Ates, O., Cayli, S.R., Yucel, N., Altinoz, E., Kocak, A., Durak, M.A., Turkoz, Y., Yologlu, S., 2007. Central nervous system protection by resveratrol in streptozotocininduced diabetic rats. J. Clin. Neurosci. 14, 256–260. Barry, B.W., 2001. Novel mechanisms and devices to enable successful transdermal drug delivery. Eur. J. Pharm. Sci. 14, 101–114. Brange, J., 1987. Galenics of insulin: the physico-chemical and pharmaceutical aspects of insulin and insulin preparations. Springer-Verlag, Berlin. Buysschaert, M., Lambert, A.E., 1989. The insulin pen. Lancet 1, 965. Costantino, H.R., Langer, R., Klibanov, A.M., 1994. Moisture-induced aggregation of lyophilized insulin. Pharm. Res. 11, 21–29. Hoare, T.R., Kohane, D.S., 2008. Hydrogels in drug delivery: progress and challenges. Polymer 49, 1993–2007. Kang, N.L., Alexander, G., Park, J.K., Maasch, C., Buchwalow, I., Luft, F.C., Haller, H., 1999. Differential expression of protein kinase C isoforms in streptozotocininduced diabetic rats. Kidney Int. 56, 1737–1750. Kaushik, S., Hord, A.H., Denson, D.D., McAllister, D.V., Smitra, S., Allen, M.G., Prausnitz, M.R., 2001. Lack of pain associated with microfabricated microneedles. Anesth. Analg. 92, 502–504. Lenhard, M.J., Reeves, G.D., 2001. Continuous subcutaneous insulin infusion: a comprehensive review of insulin pump therapy. Arch. Inter. Med. 161, 2293–2300. McAllister, D.V., Wang, P.M., Davis, S.P., Park, J.H., Canatella, P.J., Allen, M.G., Prausnitz, M.R., 2003. Microfabricated needles for transdermal delivery of
macromolecules and nanoparticles: fabrication methods and transport studies. Proc. Natl. Acad. Sci. U.S.A. 100, 13755–13760. Peppas, N.A., Bures, P., Leobandung, W., Ichikawa, H., 2000. Hydrogels in pharmaceutical formulations. Eur. J. Pharm. Biopharm. 50, 27–46. Prausnitz, M.R., 2004. Microneedles for transdermal drug delivery. Adv. Drug Deliv. Rev. 56, 581–587. Qin, G., Gao, Y., Wu, Y., Zhang, S., Qiu, Y., Li, F., Xu, B., 2012. Simultaneous basalbolus delivery of fast-acting insulin and its significance in diabetes management. Nanomed.: Nanotechnol. Biol. Med. 8, 221–227. Qiu, Y., Gao, Y., Hu, K., Li, F., 2008. Enhancement of skin permeation of docetaxel: a novel approach combining microneedle and elastic liposomes. J. Control. Release 129, 144–150. Siekmeier, R., Scheuch, G., 2008. Inhaled insulin: does it become reality. J. Physiol. Pharmacol. 59, 81–113. Slaughter, B.V., Khurshid, S.S., Fisher, O.Z., Khademhosseini, A., Peppas, N.A., 2009. Hydrogels in regenerative medicine. Adv. Mater. 21, 3307–3329. Wang, W., 2000. Lyophilization and development of solid protein pharmaceuticals. Int. J. Pharm. 203, 1–60. Weng, J., Li, Y., Xu, W., Shi, L., Zhang, Q., Zhu, D., Hu, Y., Zhou, Z., Yan, X., Tian, H., Ran, X., Luo, Z., Xian, J., Yan, L., Li, F., Zeng, L., Chen, Y., Yang, L., Yan, S., Liu, J., Li, M., Fu, Z., Cheng, H., 2008. Effect of intensive insulin therapy on betacell function and glycaemic control in patients with newly diagnosed type 2 diabetes: a multicentre randomised parallel-group trial. Lancet 371, 1753–1760. Wu, Y., Gao, Y., Qin, G., Zhang, S., Qiu, Y., Li, F., Xu, B., 2010. Sustained release of insulin through skin by intradermal microdelivery system. Biomed. Microdevices 12, 665–671. Zhang, S.H., Qin, G.J., Wu, Y., Gao, Y.H., Qiu, Y.Q., Li, F., Xu, B., 2011. Enhanced bioavailability of l-carnitine after painless intradermal delivery vs. oral administration in rats. Pharm. Res. 28, 117–123.