Gold Nanoparticles as Carriers for Efficient Transmucosal Insulin

Subinoy Rana, Avinash Bajaj, Rubul Mout, Vincent M. Rotello. Monolayer coated gold nanoparticles for delivery applications. Advanced Drug Delivery Rev...
0 downloads 0 Views 236KB Size
Download PDF
300

Langmuir 2006, 22, 300-305

Gold Nanoparticles as Carriers for Efficient Transmucosal Insulin Delivery Hrushikesh M. Joshi,† Devika R. Bhumkar,‡ Kalpana Joshi,*,§ Varsha Pokharkar,*,‡ and Murali Sastry*,†,| Nanoscience Group, Materials Chemistry DiVision, National Chemical Laboratory, Pune - 411 008, India, Bharati Vidyapeeth Deemed UniVersity, Poona College of Pharmacy, Pune - 411 038, India, and School of Health Sciences, UniVersity of Pune, Pune - 411 007, India ReceiVed July 22, 2005. In Final Form: October 24, 2005 Nanomaterials have gained tremendous importance in biology and medicine because they can be used as carriers for delivering small molecules such as drugs, proteins, and genes. We report herein the binding of the hormone insulin to gold nanoparticles and its application in transmucosal delivery for the therapeutic treatment of diabetes mellitus. Insulin was loaded onto bare gold nanoparticles and aspartic acid-capped gold nanoparticles and delivered in diabetic Wistar rats by both oral and intranasal (transmucosal) routes. Our principle observations are that there is a significant reduction of blood glucose levels (postprandial hyperglycemia) when insulin is delivered using gold nanoparticles as carriers by the transmucosal route in diabetic rats. Furthermore, control of postprandial hyperglycemia by the intranasal delivery protocol is comparable to that achieved using the standard subcutaneous administration used for type I diabetes mellitus, thus showing considerable promise for further development.

Introduction Nanomaterials have received considerable attention because of their potential for application in a wide spectrum of areas that include biology and medicine.1-3 Much of this interest stems from the fact that nanomaterials possess interesting optoelectronic properties and is also due to their size compatibility with a variety of biologically active molecules. Examples of applications include DNA detection by gold nanoparticles,4 cell imaging using fluorescent semiconductor nanocrystals,5 rational assembly of quantum dots using antigen-antibody interactions,6 and magnetic nanoparticles as contrast enhancement agents in magnetic resonance imaging (MRI).7 Nanoparticles could also serve as excellent delivery vehicles for a variety of biomolecules such as proteins, DNA, and drugs. The tunable shape- and size-dependent optical properties of gold nanoparticles have been exploited in various surface coatings8 and biomedical applications. They are biocompatible, nontoxic,9 bind readily to a large range of biomolecules such as amino acids,10 proteins/enzymes,11 and DNA,12 and expose large surface areas for the immobilization of such biomolecules. The ability * Corresponding authors. E-mail: [email protected]. [email protected]@hotmail.com. † National Chemical Laboratory. ‡ Bharati Vidyapeeth Deemed University. § University of Pune. | Current address: Tata Chemicals Innovation Centre, Mumbai - 400 059, India. (1) (a) Albers, W. M.; Vikholm, I.; Viitala, T.; Peltonen, J. Interfacial and Materials Aspects of the Immobilization of Biomolecules onto Solid Surfaces. Handbook of Surfaces and Interfaces of Materials; Academic Press: San Diego, CA, 2001; Vol. 5, Chapter 1. (2) Rembaum, A.; Dreyer, W. J. Science 1980, 208, 364. (3) Park, S. A.; Taton, T. A.; Mirkin, C. A. Science 2002, 295, 1503. (4) Rosi, N. L.; Mirkin, C. A. Chem. ReV. 2005, 105, 1547. (5) Bruchez, M., Jr.; Moronne, M.; Gin, P.; Weiss, S.; Alivisatos, A. P. Science 1998, 281, 2013. (6) Goldman, E. R.; Balighian, E. D.; Mattoussi, H.; Kuno, K.; Mauro, J. M.; Tran, P. T.; Anderson, G. P. J. Am. Chem. Soc. 2002, 124, 6378. (7) Perez, J. M.; Simeone, F. J.; Saeki, Y.; Josephson, L.; Weissleder, R. J. Am. Chem. Soc. 2003, 125, 10192. (8) Shankar, S. S.; Rai, A.; Ahmad, A. Sastry, M. Chem. Mater. 2005, 17, 566. (9) Shukla, R.; Bansal, V.; Chaudhary, M.; Basu, A.; Bhonde, R. R.; Sastry, M. Langmuir 2005, 21, 10644.

to modulate the surface chemistry of gold nanoparticles by binding suitable ligands has important applications in many areas such as novel organic reactions,13 sensors (both inorganic and biological entities), drug/DNA delivery,14 and imaging.15 Branched polyethylenimine covalently attached to gold nanoparticles has been investigated for the delivery of plasmid.16 The antitumor drug cisplatin was adsorbed on Au-Au2S nanoparticles via 11mercaptoundecanoic acid (MUA) layers.17 Gu et al. have shown that gold nanoparticles in toluene react with bis(vancomycine) cystamide in water under vigorous stirring conditions to form vancomycin-capped gold nanoparticles; the antibiotic-capped gold nanoparticles showed enhanced antibacterial activity against E. coli strains.18 Recently, SiO2-core-Au-shell nanoparticles designed to absorb in the near-infrared have been used in cancer hyperthermia.19 (10) (a) Selvakannan, PR.; Mandal, S.; Phadtare, S.; Gole, A.; Pasricha, R.; Adyanthaya, S.; Sastry, M. J. Colloid Interface Sci. 2004, 269, 97. (b) Selvakannan, PR.; Mandal, S.; Phadtare, S.; Pasricha, R.; Sastry, M. Langmuir 2003, 19, 3545. (c) Mandal, S.; Selvakannan, PR.; Phadtare, S.; Pasricha, R.; Sastry, M. Proc. Indian Acad. Sci., Chem. Sci. 2002, 114, 513. (11) (a) Patolsky, F.; Gabriel, T.; Willner, I. J. Electroanal. Chem. 1999, 479, 69. (b) Niemeyer, C. M.; Ceyhan, B. Angew. Chem., Int. Ed. 2001, 40, 3685. (c) Gole, A.; Dash, C.; Ramakrishnan, V.; Sainkar, S. R.; Mandle, A. B.; Rao, M.; Sastry, M. Langmuir 2001, 17, 1674. (d) Gole, A.; Dash, C.; Soman, C.; Sainkar, S. R.; Rao, M.; Sastry, M. Bioconjugate Chem. 2001, 12, 684. (e) Gole, A.; Vyas, S.; Phadtare, S.; Lachke, A.; Sastry, M. Colloids Surf., B 2002, 25, 129. (f) Zhao, J.; O’Daly, J. P.; Henkens, R. W.; Stonehuerner, J.; Crumblis, A. L. Biosens. Bioelectron. 1996, 11, 493. (g) Keating, C. D.; Kovaleski, K. M.; Natan, M. J. J. Phys. Chem. B 1998, 102, 9404. (12) (a) Alivisatos, A. P.; Peng, X.; Wilson, T. E.; Loweth, C. L.; Bruchez, M. P., Jr.; Schultz, P. G. Nature 1996, 382, 609. (b) Kumar, A.; Pattarkine, M.; Bhadbhade, M.; Mandale, A. B.; Ganesh, K. N.; Datar, S. S.; Dharmadhikari, C. V.; Sastry, M. AdV. Mater. 2001, 13, 341. (c) Kanaras, A. G.; Wang, Z.; Bates, A. D.; Cosstick, R.; Brust, M. Angew. Chem., Int. Ed. 2003, 42, 191. (13) Ingram, R. S.; Hostetler, M. J.; Murray, R. W. J. Am. Chem. Soc. 1997, 119, 9175. (14) Niemeyer, C. M. Angew. Chem., Int. Ed. 2001, 40, 4128. (15) Bielinska, A.; Eichman, J. D.; Lee, I.; Baker, J. R., Jr.; Balogh, L. J. Nanopart. Res. 2002, 4, 395. (16) Thomas, M.; Klibanov, A. M. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 9138. (17) Ren, L.; Chow, G. M. Mater. Sci. Eng., C 2003, 23, 113. (18) Gu, H.; Ho, P. L.; Tong, E.; Wang, L.; Xu, B. Nano Lett. 2003, 3, 1261. (19) Hirsch, L. R.; Stafford, R. J.; Bankson, J. A.; Sershen, S. R.; Rivera, B.; Price, R. E.; Hazle, J. D.; Halas, N. J.; West, J. L. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 13549.

10.1021/la051982u CCC: $33.50 © 2006 American Chemical Society Published on Web 12/03/2005

Au Nanoparticles as Carriers for Insulin DeliVery

Insulin is a very important biologically active molecule that is used to control glucose levels in the blood. Insulin-dependent diabetes mellitus (type I diabetes, IDDM) is caused by the destruction of insulin-secreting beta cells of the pancreatic Islets of Langerhans and afflicts 0.3% of the world’s population every year. In normal human beings, the pancreatic human insulin response is described as an early burst of insulin release, followed by a gradual increasing phase of insulin secretion lasting for several hours. In postprandial hyperglycemia (PPHG), there is a loss of this first phase of insulin secretion, which may contribute to a reduced suppression of hepatic glucose production leading to higher glucose appearance in the blood.20 Hence it is important that the delivery systems provide this first phase of insulin to treat PPHG. Current dosage regimens of insulin that maintain low serum glucose levels comprise up to four subcutaneous injections per day. A subcutaneous therapeutic regimen fails to deliver physiological patterns of insulin because of adverse insulin pharmacokinetics, and hence normoglycaemia is not achieved. Insulin absorption via the subcutaneous route is sustained and thus does not mimic the normal pattern of endogenous insulin secretion. However, because of the sustained absorption, insulin concentrations between meals may be inappropriately high, which commonly results in episodes of hypoglycemia. Subcutaneous or intramuscular injections of insulin will result in peripheral hyperinsulinemia, leading to exacerbation of the macrovascular complications of diabetes. The difficulties in achieving a normal physiological profile of insulin by injectable therapy have led to the investigation of alternative nonparenteral routes for the delivery of insulin in an attempt to provide peak post-prandial insulin concentrations and improve glycemic control. Oral administration of insulin remains a significant challenge because insulin is susceptible to hydrolysis and digestion by acids and enzymes in the gastrointestinal tract. Furthermore, the bioavailability of insulin is extremely low because of poor membrane permeability. Insulin administration could be made much less traumatic and more efficient if other delivery routes could be identified. Despite the large number of studies that have demonstrated that gold nanoparticles are nontoxic to living cells, are biocompatible, and can be conjugated with a spectrum of biological ligands, reports on their application in drug delivery have been rather limited. In this article, we address this lacuna and report on the application of gold nanoparticles in insulin delivery. We have developed a methodology wherein insulin is bound to the surface of amino acid-capped gold nanoparticles; this formulation may be readily administered transmucosally. Trials carried out on diabetic rats show that there is significant transmucosal insulin uptake, which is mirrored by a significant fall in their blood glucose levels. Particularly gratifying is our observation that the transmucosal delivery methodology using gold nanoparticles leads to levels of blood glucose control that are comparable to those of parenteral insulin administration (via subcutaneous injections). Presented below are details of the investigation. Experimental Details Chemicals and Materials. Bovine insulin (C254H377N65O75S6, molecular weight 5800 Da) and dialysis tubing (12 kDa cutoff) were obtained from Sigma. Chloroauric acid, aspartic acid (molecular weight 133, C4H7NO4), and sodium borohydride were obtained from Aldrich and used as received. Alloxan monohydrate was obtained from Spectrochem, Mumbai. The glucose oxidase peroxidase (GODPOD) kit was purchased from Accurex Biomedicals Pvt. Ltd., Thane, (20) Gin, H.; Rigalleau, V. Diabetes Metab. 2000, 26, 265.

Langmuir, Vol. 22, No. 1, 2006 301 India. ELISA was performed on an AxSYM/Imx fully automated immunoassay system, Abbott; the ELISA kit was obtained from Abbott. Synthesis of Gold Nanoparticles. In a typical experiment, 100 mL of 1.25 × 10-4 M concentrated aqueous solution of chloroauric acid (HAuCl4) was reduced by 0.01 g of sodium borohydride (NaBH4) at room temperature to yield a ruby-red solution containing 35 ( 7 Å diameter gold nanoparticles.21 The ruby-red solution yielded an absorbance maximum at 518 nm. The gold nanoparticle (nanogold) solution was thoroughly dialyzed for 12 h to remove byproducts of the reaction. After dialysis, the pH of the nanogold solution was measured to be ∼7. Synthesis of Aspartic Acid-Capped Gold Nanoparticles. Aspartic acid was added to the gold nanoparticle solution to yield an overall amino acid concentration of 1 × 10-4 M in solution. This solution was incubated for 12 h and then thoroughly dialyzed for 24 h to remove uncoordinated aspartic acid. Loading of Insulin onto Gold Nanoparticles. A calculated amount of insulin was added to solutions of uncapped and aspartic acid-capped gold nanoparticles to yield an insulin concentration of 0.063 mg/mL in solution. These solutions were then incubated for 16 h at 2-8 °C and then centrifuged at 30 000 rpm for 0.5 h. The pellets thus obtained were separated from the supernatant solution and redispersed in milli Q water prior to further characterization. The free insulin present in the supernatant was determined by the standard method of ELISA. The free insulin in supernatant binds to antibodies coated on the plate, which is a specific reaction. To this complex was added anti-antibodies conjugated with peroxidase. The substrate then added gives a color, the intensity of which is directly proportional to the antigen present in the supernatant. The percentage loading of insulin on the nanoparticles was determined by the following formula: % loading ) total amount of insulin added - amount of insulin in supernatant × total amount of insulin added 100

UV-Visible Spectroscopy Measurements. The binding of insulin to bare gold nanoparticles and to aspartic acid-capped gold nanoparticles was monitored by UV-visible spectroscopy measurements carried out on a model V-570 Jasco dual-beam spectrophotometer. Transmission Electron Microscopy (TEM) Measurements. TEM measurements of the gold nanoparticles at various stages of surface treatment were performed on a JEOL model 1200EX instrument operated at an accelerating voltage of 120 kV. Samples for TEM analysis were prepared by placing drops of the gold nanoparticle solutions on carbon-coated TEM copper grids. The mixtures were allowed to dry for 1 min, after which the extra solution was removed using blotting paper. Fourier Transform Infrared (FTIR) Spectroscopy Measurements. FTIR spectroscopy was used to study the secondary structure of insulin after binding with bare gold and aspartic acid-capped gold nanoparticles. FTIR spectra of insulin bound to the gold nanoparticles and aspartic acid-capped gold nanoparticles were recorded on a Perkin-Elmer Spectrum One spectrometer operated in the horizontal attenuated total reflection (HATR) mode at a resolution of 4 cm-1 in the range of 650-4000 cm-1. For comparison, an FTIR spectrum of pure insulin was also recorded. In-Vivo Insulin Delivery Studies. The insulin-gold nanoparticle formulations were tested for oral and nasal administration on diabetic rats. Twelve- to thirteen-week-old Wistar male and female rats weighing 180-200 g each were provided by the National Toxicology Centre, Pune, India. The animals were housed under standard conditions of temperature (25 °C) in 12h/12h light and dark cycles and were fed with a standard pellet diet and water ad libitum. Animal handling was performed according to good laboratory practices (GLP). The Wistar male and female rats were rendered diabetic (21) Patil, V.; Malvankar, R. B.; Sastry, M. Langmuir 1999, 15, 8197.

302 Langmuir, Vol. 22, No. 1, 2006 prior to the study by intravenous injection of 70 mg/kg alloxan in distilled water.22 The animals were divided into groups containing six animals each. They were considered to be diabetic when the baseline glucose levels were above 200 mg/dL. The diabetic rats were fasted for 12 h before the experiments. The dose was given in IU/kg, where 1 IU ) 45.5 µg. The formulations administered to the rats were as follows: (1) insulin-loaded gold nanoparticles (Au-Ins, formulation A) administered perorally (p.o., 50 IU/kg) and intranasally (i.n., 20 IU/kg); (2) bare (uncapped) gold nanoparticles administered p.o. and i.n; (3) insulin-loaded aspartic acid-capped gold nanoparticles (Au-AspIns, formulation B) administered p.o. (50 IU/kg) and i.n. (20 IU/kg); (4) aspartic acid-capped gold nanoparticles administered p.o. and i.n, and (6) insulin solution administered p.o. (50 IU/kg), i.n. (20 IU/kg), and subcutaneously (5 IU/kg). Blood samples were collected from the retro-orbital plexus of the rats prior to administration of the formulation to establish baseline glucose levels and for the determination of insulin and gold concentration levels in the blood serum. Similarly, after dosing the animals with the various insulin formulations, blood samples were collected at different time intervals over a period of 6 h. Glycemia was determined in the serum samples by the standard glucose oxidase peroxidase method. Glucose oxidase (GOD) converts glucose to gluconic acid and hydrogen peroxide. Hydrogen peroxide formed in this reaction, in the presence of peroxidase (POD), oxidatively couples with 4-aminoantipyrine/phenol to produce red quinoeimine dye. This dye has an absorbance maximum at 505 nm. The intensity of the colored complex is directly proportional to glucose in the serum sample.23,24 Serum insulin levels were determined by ELISA tests. The results shown are the mean values of serum glucose levels and insulin levels of animals in each group (six each). The research proposals for these studies were prepared according to form B of the guidelines of the Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA). The Institutional Animal Ethics Committee (IAEC) of the Poona College of Pharmacy approved the proposal. Inductively Coupled Plasma (ICP) Measurements. The inductively coupled plasma technique was used to determine the gold concentration in the blood serum after administration of the different gold nanoparticle formulations to diabetic rats. Samples were prepared by adding 0.3 mL of 35% 36 N HCl and 0.1 mL of 55% 12 N HNO3 to a weighed amount of blood plasma and diluting it up to 2 mL. The gold concentration in the above samples were determined on an ICP OES Perkin-Elmer Optima 3000 instrument.

Results and Discussion

Joshi et al. Scheme 1. Diagram Illustrating Insulin Loading on Bare Gold Nanoparticles and Aspartic Acid-Capped Gold Nanoparticles

capped gold nanoparticles via weaker interactions, presumably hydrogen bonding (Scheme 1). In this case, the release of insulin would be expected to be more facile. The loading of insulin on bare and aspartic acid-capped gold nanoparticles was determined to be 86 and 95%, respectively. Curves 1 and 2 in Figure 1A correspond to UV-vis absorption spectra recorded from the dialyzed gold nanoparticle solution before and after loading of insulin directly onto the gold surface. In the case of the as-prepared solution, a prominent absorption at 520 nm is observed that arises because of the excitation of surface plasmon (SP) vibrations in the gold nanoparticles. Consequent to exposure to insulin, the SP band shifts to 534 nm indicating binding of insulin to the gold nanoparticles. After surface modification of the gold nanoparticles with aspartic acid, the SP band shifts to 522 nm (curve 3, Figure 1B). Further complexation with insulin results in an additional shift to 535 nm indicating surface complexation with insulin (curve 4, Figure 1B). In both cases (insulin loaded onto bare gold and aspartic acid-capped gold nanoparticles), we do not observe a broadening of the SP band suggesting that aggregation of the particles is not occurring. The presence of insulin on the surface of both bare

The nature of binding of insulin to gold nanoparticles is an important parameter that would impact the stability of the formulation as well as determine the facility with which insulin is released from the gold particles. We have studied this aspect by preparing formulations wherein insulin is directly bound to bare gold nanoparticles (presumably via a covalent linkage) and with insulin bound via hydrogen bonds with amino acid-modified gold nanoparticles (illustrated in Scheme 1). The molecular structure of insulin is also shown in Scheme 1; direct binding of insulin to the surface of bare gold nanoparticles is believed to occur via complexation of the amine or thiol groups with the gold surface.11c-e,25 It is known that both thiol and amine functionalities form covalent linkages with nanogold.25,26 In the second formulation, insulin was immobilized on aspartic acid(22) Rameshkumar, K.; Shah, S. N.; Goswami, D. B.; Mohan, V.; Bodhankar, S. L. Toxicol. Int. 2004, 11, 75. (23) Takenaga, M.; Yamaguchi, Y.; Kitagawa, A.; Ogawa, Y.; Mizushima, Y.; Igarashi, R. J. Controlled Release 2002, 79, 81. (24) Hosny, E. A.; Al-Shora, H. I.; Elmazar, M. M. A. Int. J. Pharm. 2002, 237, 71. (25) Joshi, H.; Shirude, P.; Bansal, V.; Ganesh, K. N.; Sastry, M. J. Phys. Chem. B 2004, 108, 11535. (26) Brust, M.; Walker, M.; Bethell, D.; Schiffrin, D. J.; Whyman, R. J. J. Chem. Soc., Chem. Commun. 1994, 801.

Figure 1. (A) UV-vis absorption spectra recorded from the gold nanoparticle solution before (curve 1) and after loading with insulin (curve 2). (B) UV-vis absorption spectra recorded from the aspartic acid-capped gold nanoparticle solution before (curve 3) and after loading with insulin (curve 4). (C) FTIR spectra of the (a) insulin solution, (b) Au-insulin nanoparticles, and (c) Au-Asp-insulin nanoparticles.

Au Nanoparticles as Carriers for Insulin DeliVery

Langmuir, Vol. 22, No. 1, 2006 303

Figure 2. Representative TEM images recorded from (A) as-prepared Au nanoparticles, (B) Au-Asp nanoparticles, (C) Au-insulin nanoparticles, and (D) Au-Asp-insulin nanoparticles. The scale bar for each image is 50 nm.

(curve b, Figure 1C) and aspartic acid-capped gold nanoparticles (curve c, Figure 1C) was determined by FTIR spectroscopy. A comparison of these two spectra with that recorded from the insulin solution (curve a, Figure 1C) reveals the presence of two prominent resonances at 1644 and 1546 cm-1. These two absorption bands are identified as the amide I and II vibrational modes, respectively, and can arise only from the insulin bound to the surface of the nanoparticles. The position of the amide I and II bands also indicates that the secondary structure of insulin is not perturbed after complexation with the gold nanoparticles and is important to the biological activity of this protein. These bands were absent in FTIR spectra recorded from the bare gold and aspartic acid-capped gold nanoparticle samples (data not shown). The TEM images recorded from the as-prepared dialyzed, borohydride-reduced gold nanoparticles (Figure 2A), aspartic acid-capped gold nanoparticles (Figure 1B), insulin-loaded gold nanoparticles (formulation A, Figure 2C), and insulin-loaded aspartic acid-capped gold nanoparticles (formulation B, Figure 2D) indicate that the overall structure of the nanoparticles and their assemblies is not significantly different in each case. The preparation procedure had some effect on nanoparticle assembly, most probably during the drying process, but this suggests that modification of the gold nanoparticles either with aspartic acid or insulin does not lead to effect such as nanoparticle aggregation. Consequently, the shifts in the SP band of the gold nanoparticles after loading of insulin (Figure 1A and B) may be attributed to surface complexation of insulin and not secondary effects such as aggregation of the nanoparticles. The insulin-gold nanoparticle solutions were extremely stable over time and did not show a detectable variation in either the UV-vis spectra or particle size distribution after storage for 1 month. The mean serum glucose levels determined in samples from diabetic rats collected before administration of the insulin-gold nanoparticle formulations were taken as the baseline values. The percentage blood glucose levels relative to this baseline (taken as 100%) at each time interval after dosing the diabetic rats with the different insulin formulations was measured. The mean percentage blood glucose level versus time profile obtained after subcutaneous administration of insulin is plotted in Figure 3A

Figure 3. (A) Plot of the time variation of the blood glucose level (expressed in %) in diabetic rats after subcutaneous administration of insulin; dose 5 IU/kg. (B) Plot of the time variation of the blood glucose level (expressed in %) in diabetic rats after peroral administration of insulin. Control insulin, dose 50 IU/kg (curve 1); bare Au nanoparticles, 1 mL (curve 2); Au-Asp nanoparticles, 1 mL (curve 3); Au-insulin nanoparticles, dose 50 IU/kg (curve 4); and Au-Asp-insulin nanoparticles, dose 50 IU/kg (curve 5). (C) Plot of time variation of the blood glucose level (expressed in %) in diabetic rates after intranasal administration of insulin. Control insulin, dose 20 IU/kg, (curve 1); bare Au nanoparticles, 1 mL (curve 2); Au-Asp nanoparticles, 1 mL (curve 3); Au-insulin nanoparticles, dose 20 IU/kg (curve 4); and Au-Asp-insulin nanoparticles, dose 20 IU/kg (curve 5).

and serves as a reference for subsequent studies using the nanoparticle-based insulin formulations. In the case of subcutaneous administration of insulin solution (dose 5 IU/kg), one observes a rapid decrease in the blood glucose level in diabetic rats; this reduction amounted to 38 and 53% within 1 and 2 h

304 Langmuir, Vol. 22, No. 1, 2006

after administration, respectively (Figure 3A). The depression in the blood glucose levels is sustained over a period of ca. 4 h. The time-dependent variation in blood glucose levels in diabetic rats that were administered the different insulin formulations perorally (p.o) and intranasally (i.n.) is plotted in Figure 3B and C, respectively. In some of the experiments, a small increase in blood glucose levels above the baseline is observed (curve 1, Figure 3B; curves 2 and 4, Figure 3C). We believe that this increase may be attributed to metabolic changes and endogenous secretion of glucagon in the diabetic rats due to stress in the animals during blood sampling.27 In both the p.o. (Figure 3B) and i.n. administration protocols (Figure 3C), we observe that bare gold nanoparticles (curve 2) and aspartic acid-capped gold nanoparticles (curve 3) do not result in a lowering of blood glucose levels in the diabetic rats. Oral administration of insulin solution (curve 1, Figure 3B) did not result in a detectable reduction in blood glucose levels even after 5 h of dosage, which is mainly attributed to poor uptake and enzymatic degradation of insulin in the gastrointestinal tract. However, oral administration of both formulation A, Au-Ins (curve 4, Figure 3B) and formulation B, Au-Asp-Ins (curve 5, Figure 3B) resulted in a detectable reduction in blood glucose levels of up to 19 and 31%, respectively. This result indicates that in both cases the gold-nanoparticle-surface-bound insulin is bioactive and that the absorption of insulin is enhanced in the formulation where insulin is bound to the amino acid-modified nanogold surface. In both of these experiments, the maximal decrease in the blood glucose level occurred 180 min after administration of the respective formulations. In contrast with oral administration of insulin solution where a significant drop in the blood glucose level was not observed, nasal administration of insulin solution did yield a small drop in blood glucose levels in diabetic rats (curve 1, Figure 3C). The average blood glucose level fell by ca. 18% in this experiment. As briefly mentioned above, nasal administration of both bare gold nanoparticles and aspartic acid-capped gold nanoparticles did not yield positive results. The experiments with Au-Ins (curve 4, Figure 3C) and Au-Asp-Ins (curve 5, Figure 3C) are much more interesting. The decrease in blood glucose levels in both of these experiments is quite significant and reached a maximum reduction of 50 and 55% for Au-Ins and Au-AspIns, respectively. The rate of release of insulin and consequent uptake in both cases is quite different; maximum blood glucose reduction occurs 180 min after administration of the Au-Ins formulation whereas the corresponding time period is ca. 120 min in the case of Au-Asp-Ins. This result clearly shows that the uptake of insulin by the intranasal gold nanoparticle delivery mechanism is much more rapid in the case of insulin loaded onto aspartic acid-capped gold nanoparticles relative to that for the formulation in which insulin was directly bound to nanogold. Furthermore, a comparison of the maximum blood glucose reduction values in the two formulations administered p.o. and i.n. indicates that the insulin uptake is much higher in the i.n. delivery protocol. Clearly, membrane permeability of the nanogold-insulin formulations across nasal mucosal cells is much better than across g.i. mucosa. This result is particularly gratifying considering that the insulin-nanogold dosage administered to the diabetic rats in the i.n. experiment (20 IU/kg) is smaller than that in the p.o. administration (50 IU/kg). The maximum blood glucose level reduction in the case of the Au-Asp-Ins i.n. administration (55%) is also comparable to that observed in the (27) Dorkoosh, F. A.; Verhoef, J. C.; Borchard, G.; Rafiee-Tehrani, M.; Verheijden, J. H. M.; Junginger, H. E. Int. J. Pharm. 2002, 247, 47.

Joshi et al. Table 1. Insulin Levels in Serum Expressed in mIU/mL time Au-insulin Au-Asp-insulin (min) oral nasal oral nasal 120 240 360

1.7 1.2 0.9

1.2 2.0

0.9 0.9 0.8

1.9 1.6 1.9

control suboral nasal cutaneous not detectable 0.90 0.60 0.20 0.50 0.60

2.50 1.60 1.60

Table 2. Levels of Au in Serum Determined by ICP Analysis (in ppm) time (min) 120 240 360

Au-insulin oral nasal 0.031 0.031 0.029

0.024 0.021 0.032

Au-Asp-insulin oral nasal 0.030 0.025 0.029

0.020 0.026 0.022

subcutaneous administration (53%, Figure 3A). This is an important result and indicates that the transmucosal delivery of insulin by gold nanoparticles could be an exciting alternative to painful and traumatic subcutaneous delivery. The nature of interaction between insulin and gold nanoparticle carriers would be expected to play a major role in the release of insulin. The formulation Au-Ins consists of insulin covalently linked to gold nanoparticles. The release of insulin from this formulation Au-Ins is observed to be slower and less intense when compared with the release of insulin from the Au-AspIns formulation. In the latter case, insulin is believed to be linked to the aspartic acid-gold nanoparticle surface coating via much weaker hydrogen bonding and electrostatic interactions. This weak interaction may be responsible for the faster and larger release of insulin. Insulin and gold concentrations in blood serum of diabetic rats were also monitored at different times after dosing, as in the above study with various insulin-nanogold formulations by ELISA and ICP measurements, respectively; the data obtained is shown in Tables 1 (serum insulin levels) and 2 (serum gold levels). When the nanogold-insulin formulations are used either in the oral or nasal protocol, it is observed that the blood insulin levels are consistently higher than in experiments wherein insulin solution was administered perorally and intranasally (Table 1). This clearly indicates that insulin uptake is enhanced when administered as Au-Ins and Au-Asp-Ins nanoparticles, resulting in the lowering of blood glucose level as observed and discussed earlier (Figure 3). A comparison of blood insulin levels at a particular time for a given formulation in the peroral and intranasal protocols shows that the levels are in general higher for intranasal administration (Table 1). This result is also in agreement with our observations of the higher depression of blood glucose levels when insulin bound to gold nanoparticles is administered intranasally (Figure 3B and C). The blood insulin levels in a subcutaneous administration experiment vis-a`-vis intranasal administration of Au-Ins and Au-Asp-Ins formulations are not significantly different (Table 1) and result in similar depressions of blood glucose levels. The gold levels in the blood at different times after administration of the Au-Ins and AuAsp-Ins formulations in the oral and nasal forms are almost constant over the time interval studied (Table 2). This suggests that whereas the overall uptake of the nanoparticles in both modes of administration is of the same order of magnitude the quantum of insulin absorbed across nasal mucosa and g.i. mucosa is significantly different; insulin absorption is much higher for the nasal mucosa. As briefly mentioned earlier, the relatively poor blood glucose level depression in the peroral administration protocol now may be ascribed with better confidence limits for either poor membrane permeability of insulin in the gastrointestinal tract or degradation of insulin by gastric juices; clearly,

Au Nanoparticles as Carriers for Insulin DeliVery

poor uptake of the carrier nanoparticles is not the cause of the difference between peroral and intranasal administrations. In conclusion, we have demonstrated that gold nanoparticles could serve as excellent carriers for insulin in the treatment of diabetes mellitus. The purpose of the study was to demonstrate the effectiveness of the delivery of insulin-loaded gold nanoparticles by oral and nasal routes. Different delivery systems such as nasal sprays and pumps are under investigation. With different delivery routes, the dose changes. Our studies indicate that even at a lower dose, insulin absorption via the nasal route is more effective than that via the oral route and leads to a reduction of blood glucose levels (postprandial hyperglycemia) that is comparable to that of the subcutaneous administration of insulin. The presence of an amino acid layer between insulin and the gold nanoparticles is observed to promote the uptake of insulin. Thus this approach has demonstrated considerable promise for

Langmuir, Vol. 22, No. 1, 2006 305

the development of a nonparenteral delivery system for insulin and possesses the potential to alleviate the pain and trauma associated with the subcutaneous administration of insulin. The excretion of the gold nanoparticle carriers from the body, identification of potential sites of accumulation, and toxicity of the nanoparticles during sustained and regular use of the gold nanoparticle formulations (as would be the case for type I diabetes mellitus) are critical issues that require study. Acknowledgment. H.M.J. thanks the Department of Science and Technology (DST) of the Government of India for financial assistant. D.R.B. is thankful to the Fair & Lovely Foundation, Project Saraswati for financial assistance. The TEM assistance of Ms. Renu Pasricha is gratefully acknowledged. LA051982U