- Email: [email protected]
Drug-Loaded Nanoparticles Targeted to the Colon With Polysaccharide Hydrogel Reduce Colitis in a Mouse Model
Imaging and Advanced Technology
Drug-Loaded Nanoparticles Targeted to the Colon With Polysaccharide Hydrogel Reduce Colitis in a Mouse Model HAMED LAROUI, GUILLAUME DALMASSO, HANG THI THU NGUYEN, YUTAO YAN, SHANTHI V. SITARAMAN, and DIDIER MERLIN Division of Digestive Diseases, Department of Medicine, Emory University, Atlanta, Georgia
BACKGROUND & AIMS: One of the challenges to treating inflammatory bowel disease (IBD) is to target the site of inflammation. We engineered nanoparticles (NPs) to deliver an anti-inflammatory tripeptide Lys-Pro-Val (KPV) to the colon and assessed its therapeutic efficacy in a mouse model of colitis. METHODS: NPs were synthesized by double-emulsion/solvent evaporation. KPV was loaded into the NPs during the first emulsion of the synthesis process. To target KPV to the colon, loaded NPs (NP-KPV) were encapsulated into a polysaccharide gel containing 2 polymers: alginate and chitosan. The effect of KPV-loaded NPs on inflammatory parameters was determined in vitro as well as in the dextran sodium sulfate–induced colitis mouse model. RESULTS: NPs (400 nm) did not affect cell viability or barrier functions. A swelling degree study showed that alginate-chitosan hydrogel containing dextran–fluorescein isothiocyanate– labeled NPs collapsed in the colon. Once delivered, NPs quickly released KPV on or within the closed area of colonocytes. The inflammatory responses to lipopolysaccharide were reduced in Caco2–BBE (brush border enterocyte) cells exposed to NP-KPV compared with those exposed to NPs alone, in a dose-dependent fashion. Mice given dextran sodium sulfate (DSS) followed by NP-KPV were protected against inflammatory and histologic parameters, compared with mice given only DSS. CONCLUSIONS: Nanoparticles are a versatile drug delivery system that can overcome physiologic barriers and target anti-inflammatory agents such as the peptide KPV to inflamed areas. By using NPs, KPV can be delivered at a concentration that is 12,000-fold lower than that of KPV in free solution, but with similar therapeutic efficacy. Administration of encapsulated drug-loaded NPs is a novel therapeutic approach for IBD.
icant side effects.1– 4 Newer targeted treatments such as anti–tumor necrosis factor (TNF) agents are effective in a subset of patients but have to be administered systemically and also are associated with significant side effects.1 A major advance in therapeutic strategies in diseases such as IBD would be the ability to target drugs to the site of inflammation in sufficient quantities to maximize local drug concentration and minimize systemic side effects. However, targeting drugs to the site of inflammation has remained a challenge in IBD because of the lack of vehicles that could carry sufficient drugs or that could be released at the site of inflammation. Another problem is created by organs of the gastrointestinal tract, particularly the colon, because it is a challenge to deliver the drug to the colon with minimal digestive enzyme degradation and/or systemic absorption.5 Various carriers have been designed to release the drug at a specific pH value, to be resistant to digestive enzymes, and/or require bacterial cleavage for activation, and several of these carriers currently are being investigated.6 – 8 However, most of these drugs need to be administered in large doses, multiple times a day, resulting in poor patient compliance. Among the various carriers proposed for drug delivery, polymeric nanoparticles (NPs) have been studied for several decades. The most widely used polymers to engineer NPs are aliphatic polyesters such as polylactide (PLA),9 polyglycolide, and co-polymers,10,11 which have been approved by the US Food and Drug Administration and have well-characterized biocompatibility and (bio)degradability properties.12,13 Recent studies have reported using nanotechnology for cell targeting. Thus, micelles have been described as promoting the uptake of 40-nm colloidal gold particles,
Keywords: Colon Targeting; Colitis; KPV-Loaded Nanoparticle; Polysaccharide Hydrogel.
Abbreviations used in this paper: BSA, bovine serum albumin; COX-2, cyclooxygenase-2; DSS, dextran sodium sulfate; ECIS, electrical impedance sensing; FITC, fluorescein isothiocyanate; IL, interleukin; LPS, lipopolysaccharide; MPO, myeloperoxidase; NP, nanoparticle; PCS, photon correlation spectroscopy; PLA, polylactide; PVA, polyvinyl alcohol; TNF, tumor necrosis factor. © 2010 by the AGA Institute 0016-5085/10/$36.00 doi:10.1053/j.gastro.2009.11.003
I
nflammatory bowel disease (IBD), which includes Crohn’s disease and ulcerative colitis, is a relapsing and remitting chronic disease for which treatment options are limited. Most existing treatments are associated with signif-
GASTROENTEROLOGY 2010;138:843– 853
Imaging and Advanced Technology continued
particularly in the rectal area.14 Destruction of extracellular matrix (colonic mucosal) in the apical side of epithelia and alterations in the barrier function organization, mainly caused by destruction of epithelia tight junctions, dramatically increase the potential permeability and interactions of the epithelia to NPs. As described,15 patients with ulcerative colitis showed a 15-fold greater uptake of Evans Blue dye in the plasma of treated individuals vs control individuals. In addition, after administration by enema, colonic absorption of both uncoated and ligand-modified latex NPs has been reported. This uptake seems to be size-dependent; approximately 16% of 100-nm polystyrene particles were absorbed over a 10-day period, whereas only approximately 2% of 1-m microparticles were absorbed under the same conditions.16 The attachment of bioadhesive tomato lectin to 500-mm NPs decreased uptake by rat colonic epithelial cells by 7-fold, indicating an alteration in surface particle characteristics or a lack of appropriate receptors on the colonic epithelia. In the present study, we sought to deliver drug-loaded NPs to the colon. We used the tripeptide KPV as an anti-inflammatory drug because it has been shown that KPV, added to drinking water at 205 g/day (assuming a water uptake of 6 mL17 of a 100-mol/L KPV solution) or administrated intraperitoneally at 10 g/day, decreased colitis in mice.18 –20 Furthermore, we compared the effective KPV dose when delivered in free solution with the effective dose when loaded onto NPs in the inducedcolitis model. With the aim of packaging the KPV-loaded NPs into a biocompatible and biodegradable polymer, we sought to identify a biomaterial made of polysaccharides that degrades in the colon and thus delivers the KPVloaded NPs to the inflamed colonic area. The biological properties of polysaccharides have been used for decades in applications such as wound healing, cell encapsulation, or as an oral delivery vehicle.21–24 We used a polysaccharide gel containing 2 polymers, alginate and chitosan, as a vector. Chitosan is a biocompatible and biodegradable polymer.25,26 Alginate and chitosan form gels that chelate with a calcium or sulfate solution, respectively.27 The combination of alginate and chitosan chelation should maintain the 3-dimensional form of the bead and the chitosan also has therapeutic effects on inflammatory cells in the colon.27,28
Materials and Methods Preparation of Nanoparticles Loaded With KPV and Characterization See Supplementary Materials and Methods section.
Size Determination of the KPV-Loaded NPs See Supplementary Materials and Methods section. 844
Scattering Electron Microscopy of NPs See Supplementary Materials and Methods section.
NP-Loaded KPV KPV and bovine serum albumin (BSA) form a homogenous internal aqueous phase (see the NP synthesis process described in Figure 1). To measure KPV loaded onto NPs, BSA concentration was measured by ultraviolet spectroscopy in the final washing solutions. The encapsulation rate of BSA then was determined based on the initial BSA concentration. After each final washing to remove extra polyvinyl alcohol (PVA) by centrifugation of NP suspension, the supernatant was collected. The accumulated washing volumes were used to determine the concentration of BSA present in the supernatant, thus providing the mass of protein not encapsulated. A mass balance was performed to determine the amount of BSA that was loaded into the NP according to the known initial concentration. The amount of BSA is determined by ultraviolet quantification at 280 nm: A ⫽ log (I0/I) ⫽ .l.c, where A is the solution absorbance, Io is the initial intensity, I is the intensity after sample, is the absorbance coefficient of BSA (L/g.cm), l is the cuvette length (cm), and c is the BSA concentration (g/L). Finally, the encapsulation rate was used to determine the KPV loading, assuming that KPV and BSA are loaded in the same way in a homogenous aqueous phase: encapsulation rate ⫽ 71% ⫾ 4%. This rate means that the loaded KPV concentration (in ng) compared with the dry weight in NPs (in mg) is 84 ⫾ 3.3 ng/mg. To compare KPV-loaded NPs with free KPV, we calculated a KPV concentration in water that was equivalent to a 500-g/mL concentration of KPV-loaded NPs. The calculated free KPV solution was equivalent to an 814-fold dilution of a 100-mol/L initial KPV solution.
Encapsulation of KPV-Loaded NPs Into Biomaterials Chitosan powder was solubilized in acetic acid then neutralized by addition of NaOH (0.1 mol/L) to give a final chitosan concentration of 0.6% (wt/vol). Mediumviscosity sodium alginate was prepared in NaCl (0.15 mol/L) 1.4% (wt/vol). Before mixing with NaCl, or acetic acid, the polymer powders were weighed, placed in glass tubes, and autoclaved. Alginate solution and chitosan solutions were mixed at a 1:1 ratio for a final concentration of 7 and 3 g/L, respectively. The polymer suspension was homogenized for 24 hours. Different NPs were added to obtain a 2-mg/mL concentration of hydrogel solution and stirred to disperse NPs throughout the polymer solution. A chelation solution containing 70 mmol/L of calcium chloride and 30 mmol/L of sodium sulfate was prepared. Figures 2 and 3 show the procedure for inclusion of loaded NPs into
Imaging and Advanced Technology continued
Figure 1. (A) Schematic process of NP synthesis. A hydrophilic drug is encapsulated by double emulsion water in oil in water (W/O/W). (B) Schematic representation of PLA NPs loaded with KPV, a hydrophilic drug, and coated with PVA.
biomaterials and the gavage method, respectively. A detailed protocol is available at http://www.natureprotocols. com/2009/09/03/a_method_to_target_bioactive_c.php.
Drug Release From NPs in Aqueous Solution The drug release kinetic calculation is based on the homogenous phase formed by KPV and BSA during the first aqueous emulsion. We assume that KPV and BSA similarly release from NPs. Thus, NPs covered with PVA and containing BSA– fluorescein isothiocyanate (FITC) and KPV were incubated in a solution simulating intestinal fluid, composed of phosphate buffered saline (PBS) containing 5 g/mL pepsin (Sigma, pepsin from porcine gastric mucosa) and 30 g/mL trypsin (Sigma, trypsin from porcine pancreas) at pH 6. The amount of BSA–FITC was analyzed using fluorescence spectroscopy. The measurement parameters were as follows: excitation wavelength was 490 nm and emission wavelength was 525 nm. The drug release was measured by sampling 100 L of the intestinal fluid at the indicated times. The total amount of BSA on the NPs is known; a concentration of 113 g BSA per mg NPs corresponds to a 70% encapsulation rate. This rate was calculated by measuring BSA in the washing solution of NPs at the end of the process and after centrifugation of the NPs (see NP-Loaded KPV section). For the presented release curve (Figure 4A), we made a 2-mg/mL solution of NPs in intestinal fluid pH 6 at 37°C. At each time point of the
kinetic curve, we sampled 100 L of the solution. Release of BSA is expressed as a percentage of the total initial amount in NPs and the results mirror the kinetic curve of KPV release.
Cell Culture See Supplementary Materials and Methods section.
Cellular Toxicity See Supplementary Materials and Methods section.
Swelling Degree of the Biomaterial Into the Digestive Tract To study polymer behavior in the digestive tract, the swelling degree was calculated. To be reliable, a minimum of 50 mg of polymer beads was required to calculate the swelling degree using the following equation: sd ⫽
mass t ⫺ mass t0 mass t0
Mass t and mass t0 are the mass in milligrams of the bead at time t and time 0 of the study, respectively; sd is the swelling degree. The swelling degree of the biomaterial was studied in: (1) gastric fluid, simulated by acetate buffer solution (0.2 mol/L) containing 50 g/mL pepsin (Sigma, pepsin from porcine gastric mucosa) and 5 g/mL trypsin (Sigma, trypsin from porcine pancreas) 845
Imaging and Advanced Technology continued
Figure 2. Schematic representation of biomaterial encapsulation of KPV-loaded NPs. (A) An alginate and chitosan hydrogel was formed by double linking of Ca2⫹ and SO42- ions. Suspension of NPs in the polysaccharide solution loads the final formed hydrogel with NPs. (B) Optical microscopy image of KPV-loaded NPs encapsulated into a hydrogel bead of alginate-chitosan linked via Ca2⫹ and SO42- ions.
and adjusted to pH 1 or 3; and (2) intestinal fluid, simulated by phosphate buffer solution (0.1 mol/L) containing 5 g/mL pepsin and 30 g/mL trypsin and adjusted to pH 4 or 6.
Animals See Supplementary Materials and Methods section.
Induction of Colitis See Supplementary Materials and Methods section.
Myeloperoxidase Assay See Supplementary Materials and Methods section.
RNA Extraction, Reverse-Transcription Polymerase Chain Reaction, and Real-Time Reverse-Transcription Polymerase Chain Reaction See Supplementary Materials and Methods section.
Results Characterization of the KPV-Loaded NPs As shown in Figure 4A, analysis of the KPV-loaded NPs revealed the average particle size evaluated by photon correlation spectroscopy (PCS) to be about 366 nm, 846
whereas that evaluated by scattering electron microscopy (SEM) was about 300 nm (Figure 4B). There are several possible explanations for this difference. First, the particles shown in Figure 4B were studied immediately after solvent evaporation and before washing, centrifugation, and freeze-drying; by contrast, the other particles were measured after these processes. Some aggregation therefore may have occurred, particularly during the freezedrying process, which could explain the greater size seen in Figure 4A. A second reason likely is related to the SEM technique that requires the study of the dry particles under vacuum, which can cause particle shrinkage. Finally, PCS calculates an intensity average size, whereas SEM evaluates a number average size, which generally is lower than the size calculated by PCS. We also showed that all NPs loaded with KPV are of similar size, as shown by the low polydispersity index of the size distribution (Figure 4A and Supplementary Figure 1).
Drug Release From the KPV-Loaded NPs As shown in Figure 4C, about 25% of the total initial amount of KPV loaded onto the NPs is released into an intestinal fluid solution at pH 6.2 within 30 minutes. This release of KPV from the NPs likely is owing to the release of KPV near the surface of the NPs. After 30
Imaging and Advanced Technology continued
Figure 3. Double-gavage procedure of encapsulated KPV-loaded NPs. (A) The first gavage delivered 100 L of the polymer mix solution (alginate, 7 g/L; chitosan, 3 g/L) containing a homogenous suspension of NPs (2 mg/mL). Red Ponceau has been added for visualization. (B) The second gavage delivered 50 L of a solution containing 70 mmol/L calcium chloride and 30 mmol/L sodium sulfate. (C) Visualization of the mixed hydrogel formed by chelation of the polymers in the stomach. (D) The hydrogel after extraction from a mouse stomach 5 minutes after the double-gavage method.
minutes, the release kinetics of KPV is more linear and about 35% of the total initial amount of KPV is released within 30 hours. This later KPV release phase seems to be the result of KPV diffusion across the NPs. Together, these results indicate that free KPV will be delivered rapidly (burst release) to the intestinal mucosa no more than 30 minutes after the KPV-loaded NPs come into contact with the intestinal fluid at colonic pH (pH 6.2).
KPV-Loaded NPs Do Not Affect Cell Viability For any biomaterial, including a combination of hydrogel and NPs, cytotoxicity assessments need to be performed. First, we performed a simple cytotoxicity test using water-soluble tetrazolium salt (WST-1 assay). As shown in Figure 4D, KPV-loaded NPs at a concentration of 1 mg/mL do not cause cellular death of Caco2–BBE cells over a 72-hour period compared with control materials. Similar results were found by using the normal intestinal enterocyte cell IEC6 cell line (data not shown). However, the WST-1 assay is not suitable for continuous, automated, and real-time analysis, and may be associated with several disadvantages. Of particular interest is the development of electrical impedance sensing (ECIS) as a tool for in vitro toxicity testing.29 This technique uses a small electrode that is deposited on the bottom of tissue culture wells and immersed in culture medium. As shown in Figure 4E, inoculated Caco2–BBE cells attach to the electrode surface to form a confluent layer and reach a resistance of 40,000 Ohms. KPV-loaded NPs at a 1-mg/mL concentration and latex particles at the same concentration (control) were added to a confluent Caco2–BBE monolayer. Even 80 hours of treatment with KPV-loaded NPs did not affect the Caco2–BBE monolayer resistance (before addition of KPV-loaded NPs, the
resistance was 40,000 Ohms compared with 43,000 Ohms 80 hours after addition of KPV-loaded NPs). These data are supported by results with the control biocompatible latex beads (diameter, 200 nm) on the resistance of Caco2–BBE monolayers (40,000 vs 43,000 Ohms before and after administration of beads, respectively). Together, these results show that KPV-loaded NPs coated with PVA do not affect cell growth or intestinal barrier function.
KPV Is Released Extracellularly and Is Taken up by Caco2–BBE Cells We next determined the interaction of NPs with epithelial cells. Caco2–BBE cells were treated with NPs (500 g/mL) for 48 hours, at which time cells were fixed and examined by transmission electron microscopy as described in the Materials and Methods section. As shown in Figure 5A, NPs released drug on the extracellular domain and were detected in intracellular vesicles. These data show that the NP effectively delivered the drug in the local environment of epithelial cells and was able to cross apical barriers into the cells.
Lipopolysaccharide-Induced Inflammatory Responses in Caco2–BBE Monolayers Are Reduced by KPV-Loaded NPs After stimulation with lipopolysaccharide (LPS) 10 mg/mL for 1 hour, as determined in preliminary experiments, Caco2–BBE monolayers show marked inflammatory responses, including increases in levels of the proinflammatory cytokine interleukin (IL)-8 and enzymes such as cyclooxygenase-2 (COX-2) (Figure 5B). Interestingly, we found that pretreatment of Caco2–BBE monolayers with KPV-loaded NPs for 72 hours reduced 847
Imaging and Advanced Technology continued
Figure 4. Physicochemical characterization and biocompatibility of KPV-loaded NPs. (A) Nanoparticles size average (nm) and polydispersity index of KPV-loaded NPs and latex NPs (size calibration standard) by photon correlation spectroscopy method (n ⫽ 5). (B) Scanning electron microscopy picture of KPV-loaded NP suspension (5 g/mL). (C) Cumulative drug release kinetic (percentage of total encapsulated drug). NPs at 5 mg/mL containing BSA–FITC and KPV were incubated in PBS containing 5 g/mL pepsin and 30 g/mL trypsin at pH 6 (n ⫽ 5). (D) WST-1 cell viability test on Caco2–BBE (percentage over control) after 72 hours of exposure to a 1-mg/mL suspension of 3 different NP samples. (Control, culture medium; NanoKPV, KPV-loaded NPs; NanoFITC, dextran–FITC–loaded NPs; NanoBSA, BSA-loaded NPs) (n ⫽ 8). (E) Electrical impedance sensing method on Caco2–BBE cells to determine cell viability (percentage over control) after a long exposure to a 1-mg/mL suspension of 2 different NP samples. (culture medium without NP as control; latex particles [culture medium without latex NP as control] and KPV-NP) (n ⫽ 3).
the increases in LPS-induced IL-8 and COX-2 expression levels (Figure 5B). This reduction was dose-dependent because a 50-g/mL concentration of KPV-loaded NPs reduced IL-8 and COX-2 expression levels by 50% and 50%, respectively, whereas concentrations of 200 g/mL and 500 g/mL reduced IL-8 and COX-2 expression levels by 70% and 80%, and 90% and 100%, respectively. Importantly, we showed that an equivalent concentration of free KPV did not affect the IL-8 and COX-2 expression levels. Together, these results show that KPV-loaded NPs are more effective than free KPV in reducing inflammatory responses induced by LPS in the intestinal epithelia.
Swelling Degree of the Biomaterial Into the Digestive Tract Kinetic of the swelling degree measured the evolution of water trapping in the hydrogel structure. Typically in an ionic hydrogel, the swelling degree evolution is inversely proportional to the power of ionic interactions between ions and polysaccharides. Thus, the higher swelling degree indicates the hydrogel is closer to collapsing. In the present study, we assessed the swelling degree of the biomaterial in solutions simulating intestinal fluid at 848
pH levels from 1 to 3 and simulating gastric fluid at a pH level from 4 to 6. The hydrogel comprises a mixture of 2 biocompatible polysaccharides: alginate and chitosan. This mixture is dropped by needle in a solution of Ca2⫹ and SO42- to form beads (Figure 2B). All compositions of those 2 polysaccharides had a total concentration of 10 g/L of polysaccharides. First, we tested alginate alone. With this composition, the hydrogel beads did not swell in a gastric solution of pH 1 or 2 over a time period of 24 hours (Supplementary Figure 2A). However, the hydrogel beads showed low levels of swelling in intestinal solutions of pH 3, 4, 5, or 6, and the level of swelling increased with time at each pH value (Supplementary Figure 2A). Together, these results show that hydrogel beads composed of alginate alone show minimal swelling at pH 3, 4, 5, or 6, and do not collapse. Next, we modified the hydrogel composition to a mix of alginate and chitosan at a ratio of 7/3 (wt/wt). We observed that the hydrogel beads exposed to a gastric solution of pH 1, 2, or 3 for 24 hours were stable. However, the hydrogel beads exposed to the intestinal solution of pH 4 for 24 hours shrank and those exposed to a solution of pH 5 completely collapsed (Supplementary Figure 2B). In ad-
Imaging and Advanced Technology continued
Figure 5. Cell interactions of KPV-loaded NPs and messenger RNA (mRNA) expression of proinflammatory cytokines (IL-8 and COX-2) on Caco2–BBE monolayers treated or not with KPV-loaded NPs. (A) Transmission electron microcopy picture of KPV–NPs (500 g/mL) on monolayer of Caco2–BBE cells after a LPS stimulation (10 g/mL, 1 h). Bar, 1 m. KPV-loaded NP in the extracellular domain (1). KPV-loaded NP interacting with the cell membrane starting an endocytose process (2). KPV-loaded NP in intracellular domain after endocytosis (3). (B) Caco2–BBE cells were stimulated with LPS (10 g/mL, 1 h), alone or after pretreatment with different concentrations of KPV-loaded NPs (50, 200, or 500 g/mL) or the equivalent concentration of free KPV in solution (free KPV 41 g/L corresponds to 500 g/mL KPV-loaded NPs) (41 g/L) for 72 hours (n ⫽ 3). (**P ⬍ .001, *P ⬍ .01 for IL-8; and ##P ⬍ .001, #P ⬍ .01 for COX-2).
dition, we observed that the collapsing effect of the hydrogel beads in the intestinal solution was time-dependent (Supplementary Figure 2B). Together, these results indicate that this biomaterial could be used to deliver KPV-loaded NPs to intestinal solutions of pH 5 and 6 (the approximate pH of the colonic fluid) but not at acidic pH values of 1, 2, or 3 (the approximate pH of the gastric fluid). In other words, these data indicate that the hydrogel composed of alginate and chitosan at a ratio of 7/3 (wt/wt) could be used to deliver KPV-loaded NPs to the colon but not to the gastric gland.
model for the study of human ulcerative colitis is dextran sodium sulfate (DSS)-induced acute and chronic colitis with ulceration in mice, which is characterized by a marked loss in body weight, rectal bleeding, reduction in
Targeting Drug-Loaded NPs to the Colon Next, we determined the rate of drug release from encapsulated NPs in the colon. Encapsulated dextran– FITC–loaded NPs (5 mg/mL of hydrogel) were administered to mice by daily gavage for 3 days. The mice then were killed and FITC fluorescence in each tissue was quantified along the gastrointestinal tract. The data were expressed by g of tissue protein. As shown in Figure 6, we detected lower levels of dextran–FITC in the stomach, the proximal small intestine, distal intestine, and cecum, with a higher level in the colon, indicating that the FITC was delivered mostly to the colon.
Encapsulated KPV-Loaded NPs Reduce Induced Colitis Many experimental animal models have been used to study human IBD. One well-characterized animal
Figure 6. Localization of encapsulated dextran–FITC NPs (g dextran–FITC/g protein of tissue) on the digestive tract after 3 days of gavage. Localization of dextran–FITC NPs (g dextran–FITC/g protein of tissue) on the digestive tract after 3 days of gavage. Digestive tract was divided into 6 sections as follows: stomach, proximal (1/3) small intestine, medial (2/3) small intestine, and distal (3/3) small intestine, cecum, and colon (n ⫽ 12). 849
Imaging and Advanced Technology continued
Figure 7. Histology of DSS-induced colitis in 7 different biomaterial treatments on mice.During in vivo experiments, all mice received a daily gavage of 150 L of biomaterial. Control group received the hydrogel with empty NPs (alginate, 7 g/L; chitosan, 3 g/L) and regular water as drinking solution. DSS treated group received the hydrogel with empty NPs (alginate, 7 g/L; chitosan, 3 g/L) and 3% DSS solution as drinking solution. KPV–NP– treated group received the hydrogel (alginate, 7 g/L; chitosan, 3 g/L) loaded with encapsulated KPV-loaded NPs and 3% DSS solution as drinking solution. Histologic section of colon after 7 days of a daily gavage in groups 1, 2, and 3.
colon length, and destruction of the intestinal epithelium. We investigated the anti-inflammatory effects of KPV-loaded NPs encapsulated in the hydrogel (alginate and chitosan: 7 g/L and 3 g/L, respectively) in mice with active DSS-induced colitis. C57BL/6 mice received regular water and a gavage of 150 L of hydrogel daily for 7 days (group 1), water with 3% DSS and gavage of 150 L of hydrogel daily for 7 days (group 2), water with 3% DSS and gavage of 150 L of hydrogel encapsulated with KPV-loaded NPs daily for 7 days (group 3), water with 3% DSS and gavage of 150 L of hydrogel encapsulated with KPV-loaded NPs daily for 7 days (group 4), water with 3% DSS and gavage of 150 L of hydrogel containing 41 g/L of free KPV daily for 7 days (group 5), water with 3% DSS and gavage of 150 L of hydrogel encapsulated with dextran–FITC–loaded NPs daily for 7 days (group 6), and water with 3% DSS for 7 days (group 7). This treatment regimen resulted in intestinal colitis and histologic studies revealed DSS-induced cell-wall damage with epithelial exfoliation when compared with mice that received water alone (Figure 7). In the DSS-treated mice, we also observed interstitial edema, dilation of vessels, and a general increase in the number of inflammatory cells (lymphocytes and polynuclear cells) in the lamina propria (Figure 7). To evaluate the role of KPV in the inflammatory process, encapsulated KPV-loaded NPs were administered by daily gavage for 7 days to the mice that also received DSS in the drinking water; after 7 days the mice were killed. Our histologic studies showed reduced intestinal inflammation in mice with DSS-induced colitis treated with encapsulated KPV-loaded NPs, compared with control mice that received only water (Figure 7). We additionally observed weight loss in the DSS-treated mice (Figure 8A), which started at day 4 and reached a ma850
ximum at day 7. The administration of encapsulated KPV-loaded NPs reduced the weight loss of DSS-induced colitis (Figure 8A). However, administration of encapsulated NPs to mice with DSS-induced colitis did not reduce the severity of weight loss compared with mice that received DSS alone (Figure 8A). Intestinal myeloperoxidase (MPO) activity was measured to indicate the extent of neutrophilic infiltration. MPO levels in the colon were increased significantly in mice with DSSinduced colitis compared with control mice that received only water (Figure 8B). Administration of encapsulated KPV-loaded NPs significantly reduced DSS-induced colonic inflammation, whereas administrations of biomaterial with free KPV at an equivalent concentration (41 g/mL) had no effect on the basal MPO levels in the colonic mucosa (Figure 8B). These results were reflected in the proinflammatory cytokine levels (TNF-␣ and IL1), which were measured by real-time reverse-transcription polymerase chain reaction in mucosa obtained from the colitic mice (Figure 8C and D). As controls, biomaterials without or encapsulated with empty NPs did not have anti-inflammatory effects on DSS-induced colitis (Figure 8).
Discussion Most currently accepted treatment regimens for IBD require frequent administration of high-dose antiinflammatory drugs, which can lead to high rates of adverse events. Sustained drug-release devices, such as pellets, capsules, or tablets designed to deliver drugs specifically to the colon over longer periods of time, have been developed, but their efficiencies have been limited by the effects of IBD-related diarrhea, which increases
Imaging and Advanced Technology continued
Figure 8. In vivo parameters of DSS-induced colitis in 7 different biomaterial treatments on mice. During in vivo experiments, mice received (or not for group 7) a daily gavage of 150 L of biomaterial with or without NPs. All groups had DSS as drinkable water excepted group 1 (water). Groups numbers were defined as followed: number ⫽ biomaterial ⫹ daily drink; hydrogel, alginate 7 g/L and chitosan 3 g/L; DSS, 30 g/L; Ø-NP, empty NP; 1, hydrogel ⫹ water; 2, hydrogel ⫹ DSS; 3, hydrogel ⫹ KPV–NP (encapsulated KPV-loaded NPs) ⫹ DSS; 4, hydrogel ⫹ Ø-NP ⫹ DSS; 5, hydrogel with 41 g/L free KPV (free KPV in hydrogel) ⫹ DSS; 6, hydrogel ⫹ dextran–FITC NP (encapsulated FITC-loaded NPs); 7, DSS only. ##P ⬍ .001. (A) Percentage of initial body weight (%) after 7 days of a daily gavage in groups 1–7. (B) Percentage increase over control of MPO activity after 7 days in groups 1, 2, 3 4, 5, and 7. (C) mRNA expression over control of TNF-␣ after 7 days in groups 1, 2, 3, 4, and 5. (D) mRNA expression over control of IL- after 7 days in groups 1, 2, 3, 4, and 5.
pellet elimination and reduces the duration of drug release. Thus, a carrier system capable of delivering the drug specifically and exclusively to the inflamed regions for a prolonged period is needed. Such a system significantly could reduce the side effects of conventional antiinflammatory compounds. Controlled drug delivery involves the association of a drug with a carrier system that modulates its pharmacokinetic properties and biodistribution. Several drug carriers and delivery systems have been investigated; among these, the NP is an attractive carrier that recently has shown promise for promoting drug efficacy.6,7,30,31 NPs are solid colloidal particles of approximately 10 –1000 nm in diameter. They consist of macromolecular materials in which the active component is dissolved, entrapped, or encapsulated, or to which the active component is absorbed or attached.32,33 Here, we have shown that KPV-loaded NPs are biocompatible and rapidly release KPV in intestinal solutions at pH 6.2. In addition, we have shown that this release of KPV reduced the intestinal inflammatory responses to LPS in vitro, and that the effective dose of KPV was 12,000 times lower when administered via NPs compared with free solution, showing that the NP method can
successfully reduce effective drug doses. One possibility for this difference is that in contrast to free KPV in solution, KPV delivered by NPs is in close proximity to the cell membrane. Another explanation could be that epithelial cells endocytose KPV-loaded NPs, providing an efficient route for cellular uptake of KPV. As reported previously, the NP is an attractive carrier that has shown great promise for increasing drug efficacy.6,7,30,31 After administration of KPV-loaded NPs, free KPV likely will be generated by initial drug release during passage through the small intestine and by other processes such as degradation of the NP building matrix polyester by digestive enzymes. To deliver the KPV-loaded NPs to the colonic lumen, we encapsulated them into a biomaterial comprised of alginate and chitosan at a ratio of 7/3 (wt/wt), and showed that this biomaterial collapses in intestinal solution at pH 5 or 6, which is the colonic pH under inflamed and noninflamed states.34 Thus, we showed that NP release from the hydrogel (and so KPV release from NP) occurs mostly in the colonic lumen compared with other parts of the gastrointestinal tract such as the stomach and small intestine. Together, these results show that the KPV-loaded NPs are well 851
Imaging and Advanced Technology continued
protected during transit through the gastric gland and the small intestine. The encapsulated KPV-loaded NPs delivered by gavage at 2 mg/mL/day for 6 days provide a KPV concentration of 25.2 ng/day to the colonic lumen. In contrast, the effective dose of KPV in free solution able to reduce DSS-induced colitis was approximately 200 g/day. In other words, the effective dose of KPV was 12,000 times lower when administrated via encapsulated KPV-loaded NPs when compared with KPV in free solution. We also investigated whether such a low KPV concentration delivered to the colonic lumen was sufficient to reduce intestinal inflammation. By using the DSS colitis model, we showed that KPV delivered by gavage of encapsulated KPV-loaded NPs efficiently reduced the severity of intestinal colitis, as shown by a reduction in DSSinduced MPO activity and histologic examination. Interestingly, we found that, using our delivery method, 1 pmol/L KPV more effectively reduced intestinal inflammation compared with 100 mol/L KPV administered in the drinking water. Based on our in vitro results, we suggest that KPV delivered by NPs is in close proximity to the cell membrane and was taken up by the hPepT1 membrane transporter.18 In addition, epithelial cells might endocytose KPV-loaded NPs and this could be an efficient route for cellular uptake of KPV as well. It is known in intestinal colitis that the intestinal barrier function is disrupted in areas of inflammation. In these inflamed areas, NPs might have been translocated to the submucosa and the KPV release could occur at close proximity to the immune cells. Given that KPV affects inflammatory responses in both epithelial and immune cells,18,35 it is reasonable to suggest that, under the delivery conditions used in the present study, low doses of KPV would affect the inflammatory responses of these 2 cell types. It is well known that high levels of colonic mucus are secreted during intestinal inflammation. The biomaterial used in the present study was composed of chitosan; hence, our encapsulated KPV-loaded NPs could interact with the mucin layer of the inflamed colonic tissue and deliver KPV (released from NPs) to the inflamed colonic area. Therefore, chitosan has been investigated as a superior muco-adhesive cationic polymer owing to its ability to develop molecular attraction forces by electrostatic interactions with the negative charges of mucin; this ability has been shown by the formation of hydrogen bonds or ionic interactions between the positively charged amino groups of chitosan and the negatively charged sialic acid residues of mucin, depending on the environmental pH.36 Further studies should be performed to investigate the therapeutic effects of encapsulated KPV-loaded NPs on the clinical symptoms of experimental colitis and histologic mucosal repair. 852
Our results show that encapsulated drug-loaded NPs are a versatile drug-delivery system, with the ability to overcome physiologic barriers and target low concentrations of the anti-inflammatory drug KPV to inflamed areas to relieve symptoms of IBD.
Supplementary Material Note: To access the supplementary material accompanying this article, visit the online version of Gastroenterology at www.gastrojournal.org, and at doi: 10.1053/j.gastro.2009.11.003. References 1. Ballinger A. Adverse effects of nonsteroidal anti-inflammatory drugs on the colon. Curr Gastroenterol Rep 2008;10:485– 489. 2. Langan RC, Gotsch PB, Krafczyk MA, et al. Ulcerative colitis: diagnosis and treatment. Am Fam Physician 2007;76:1323– 1330. 3. Truelove SC. Systemic and local corticosteroid therapy in ulcerative colitis. Br Med J 1960;1:464 – 467. 4. Reshef R, Varkel J, Shiller M, et al. Systemic effects of rectally administered corticosteroids. Isr J Med Sci 1992;28:98 –100. 5. Hebrard G, Hoffart V, Cardot J, et al. Investigation of coated whey protein/alginate beads as sustained release dosage form in simulated gastrointestinal environment. Drug Dev Ind Pharm 2009;14:1–10. 6. Lamprecht A, Yamamoto H, Takeuchi H, et al. A pH-sensitive microsphere system for the colon delivery of tacrolimus containing nanoparticles. J Control Release 2005;104:337–346. 7. Lamprecht A, Yamamoto H, Takeuchi H, et al. Nanoparticles enhance therapeutic efficiency by selectively increased local drug dose in experimental colitis in rats. J Pharmacol Exp Ther 2005; 315:196 –202. 8. Leserman LD, Machy P, Barbet J. Cell-specific drug transfer from liposomes bearing monoclonal antibodies. Nature 1981;293: 226 –228. 9. Laroui H, Grossin L, Leonard M, et al. Hyaluronate-covered nanoparticles for the therapeutic targeting of cartilage. Biomacromolecules 2007;8:3879 –3885. 10. Makhlof A, Tozuka Y, Takeuchi H. pH-sensitive nanospheres for colon-specific drug delivery in experimentally induced colitis rat model. Eur J Pharm Biopharm 2009;72:1– 8. 11. Meissner Y, Pellequer Y, Lamprecht A. Nanoparticles in inflammatory bowel disease: particle targeting versus pH-sensitive delivery. Int J Pharm 2006;316:138 –143. 12. Feng SS, Mei L, Anitha P, et al. Poly(lactide)-vitamin E derivative/ montmorillonite nanoparticle formulations for the oral delivery of Docetaxel. Biomaterials 2009;30:3297–3306. 13. Yang H, Li K, Liu Y, et al. Poly(D,L-lactide-co-glycolide) nanoparticles encapsulated fluorescent isothiocyanate and paclitaxol: preparation, release kinetics and anticancer effect. J Nanosci Nanotechnol 2009;9:282–287. 14. Fukui H, Murakami M, Yoshikawa H, et al. Studies on the promoting effect of lipid-surfactant mixed micelles (MM) on intestinal absorption of colloidal particles. Dependence on particle size and administration site. J Pharmacobiodyn 1987;10:236 –242. 15. Kitajima S, Takuma S, Morimoto M. Changes in colonic mucosal permeability in mouse colitis induced with dextran sulfate sodium. Exp Anim 1999;48:137–143. 16. Lamprecht A, Schafer U, Lehr CM. Size-dependent bioadhesion of micro- and nanoparticulate carriers to the inflamed colonic mucosa. Pharm Res 2001;18:788 –793.
Imaging and Advanced Technology continued
17. Bachmanov AA, Reed DR, Beauchamp GK, et al. Food intake, water intake, and drinking spout side preference of 28 mouse strains. Behav Genet 2002;32:435– 443. 18. Dalmasso G, Charrier-Hisamuddin L, Thu Nguyen HT, et al. PepT1-mediated tripeptide KPV uptake reduces intestinal inflammation. Gastroenterology 2008;134:166 –178. 19. Brzoska T, Luger TA, Maaser C, et al. Alpha-melanocyte-stimulating hormone and related tripeptides: biochemistry, antiinflammatory and protective effects in vitro and in vivo, and future perspectives for the treatment of immune-mediated inflammatory diseases. Endocr Rev 2008;29:581– 602. 20. Kannengiesser K, Maaser C, Heidemann J, et al. Melanocortinderived tripeptide KPV has anti-inflammatory potential in murine models of inflammatory bowel disease. Inflamm Bowel Dis 2008; 14:324 –331. 21. Burd DA, Greco RM, Regauer S, et al. Hyaluronan and wound healing: a new perspective. Br J Plast Surg 1991;44:579 –584. 22. Yotsuyanagi T, Ohkubo T, Ohhashi T, et al. Calcium-induced gelation of alginic acid and pH-sensitive reswelling of dried gels. Chem Pharm Bull 1987;35:1555–1563. 23. Garcia AM, Ghaly ES. Preliminary spherical agglomerates of water soluble drug using natural polymer and cross-linked technique. J Control Release 1996;40:179 –186. 24. Clayton HA, London NJ, Colloby PS, et al. The effect of capsule composition on the biocompatibility of alginate-poly-l-lysine capsules. J Microencapsul 1991;8:221–233. 25. Azab AK, Orkin B, Doviner V, et al. Crosslinked chitosan implants as potential degradable devices for brachytherapy: in vitro and in vivo analysis. J Control Release 2006;111:281–289. 26. Tozaki H, Odoriba T, Okada N, et al. Chitosan capsules for colon-specific drug delivery: enhanced localization of 5-aminosalicylic acid in the large intestine accelerates healing of TNBSinduced colitis in rats. J Control Release 2002;82:51– 61. 27. Mladenovska K, Raicki RS, Janevik EI, et al. Colon-specific delivery of 5-aminosalicylic acid from chitosan-Ca-alginate microparticles. Int J Pharm 2007;342:124 –136. 28. Shi C, Zhu Y, Ran X, et al. Therapeutic potential of chitosan and its derivatives in regenerative medicine. J Surg Res 2006;133: 185–192. 29. Xiao C, Luong JH. Assessment of cytotoxicity by emerging impedance spectroscopy. Toxicol Appl Pharmacol 2005;206:102–112. 30. Jain KK. The role of nanobiotechnology in drug discovery. Drug Discov Today 2005;10:1435–1442.
31. Lamprecht A. Multiparticulate systems in the treatment of inflammatory bowel disease. Curr Drug Targets Inflamm Allergy 2003; 2:137–144. 32. Bertholon I, Ponchel G, Labarre D, et al. Bioadhesive properties of poly(alkylcyanoacrylate) nanoparticles coated with polysaccharide. J Nanosci Nanotechnol 2006;6:3102–3109. 33. Hillaireau H, Le Doan T, Chacun H, et al. Encapsulation of monoand oligo-nucleotides into aqueous-core nanocapsules in presence of various water-soluble polymers. Int J Pharm 2007;331: 148 –152. 34. Nugent SG, Kumar D, Rampton DS, et al. Intestinal luminal pH in inflammatory bowel disease: possible determinants and implications for therapy with aminosalicylates and other drugs. Gut 2001;48:571–577. 35. Luger TA, Scholzen TE, Brzoska T, et al. New insights into the functions of alpha-MSH and related peptides in the immune system. Ann N Y Acad Sci 2003;994:133–140. 36. Sogias IA, Williams AC, Khutoryanskiy VV. Why is chitosan mucoadhesive? Biomacromolecules 2008;9:1837–1842.
Received June 26, 2009. Accepted November 4, 2009. Reprint requests Address requests for reprints to: Shanthi V. Sitaraman, MD, PhD, Division of Digestive Diseases, Department of Medicine, Emory University, Atlanta, Georgia 30322. e-mail: [email protected]; fax: (404) 727-5767. Acknowledgments The authors want to thank Dr Obertone for her scientific contribution to this work. Conflicts of interest The authors disclose no conflicts. Funding This work was supported by National Institutes of Health of Digestive Diseases Research center grant (R24-DK-064399), R56DK061941 (D.M.), RO1-DK076825 and RO1-DK064711 (S.V.S.); and a research fellowship award from the Crohn’s and Colitis Foundation of America (G.D.).
853
853.e1
Imaging and Advanced Technology
Supplementary Materials and Methods Size Determination of the KPV-Loaded NPs PLA was purchased from Sigma-Aldrich (MW of 75–120 kg/mol). The NPs were produced by the doubleemulsion/solvent evaporation procedure described previously1 for the elaboration of PVA-covered PLA NPs. This method involves the use of an amphiphilic molecule as the surfactant of the secondary emulsion; however, in the present study, PVA was used, which was purchased from Aldrich (Milwaukee, WI) (Mw ⫽ 13,000 –23,000 g/mol, hydroxylated at 87%– 89%). Typically, a primary water in oil emulsion (w/o) was prepared by mixing an organic phase (4 mL dichloromethane) containing PLA (25 g/L) with an internal aqueous phase (400 L) containing BSA (50 g/L), and containing or not a tripeptide (Lys-Pro-Val, KPV; 100 nmol/L). The mixture was stirred for 2 minutes with a Vortex mixer (Maxi Mix II; Thermolyne, Dubuque, IA), then sonicated (2 min, power 6, 50% active cycle, in an ice bath) using a Sonifier 450 (Branson). This primary emulsion was poured into a second aqueous phase (8 mL) (external aqueous phase) containing an amphiphilic molecule (PVA 20 g/L). The water in oil in water emulsion (w/o/w) then was obtained by sonication (same conditions as those used for the primary emulsion). This double-emulsion then was transferred to an aqueous 10⫺2 mol/L NaNO3 dispersing phase (40 mL) and stirred for 5 minutes. The organic solvent was evaporated under stirring and ambient air conditions and the collected solid nanospheres were resuspended in water and then centrifuged again to remove the excess PVA. This purification procedure was repeated twice. The final suspension then was freeze-dried. Nanoparticles covered with PVA and containing fluorescent dextran–FITC (molecular weight, 250,000 g/mol; Sigma-Aldrich, St. Louis, MO) or BSA–FITC (Sigma-Aldrich) were prepared according to the procedure described earlier. Dextran–FITC and BSA– FITC were dissolved, respectively, in the internal aqueous phase at a concentration of 2.5 g/L and 50 g/L. Figures 1 and 2 summarize the preparation of KPV-loaded NPs.
PCS The size distribution of NPs (mean diameter and standard deviation) was determined by PCS using Brookhaven equipment (Model 90Plus, Nanoparticle Size analyzer; Brookhaven Instruments Corporation, Holtsville, NY) in 10-3 mol/L NaCl at 25°C.
SEM of NPs Suspensions of NPs on plots were fixed with glutaraldehyde containing 2.5% cacodylate buffer (0.1 mol/L, pH 7) overnight at 4°C. Afterward, the samples were rinsed and dehydrated by 2-minute rinses with a series of increasing alcohol concentrations (30%, 50%, 70%, 80%, 90%, 100% vol/vol), and then incubated for 5 minutes with hexamethyldisilazane and dried under a
GASTROENTEROLOGY Vol. 138, No. 3
range hood. The plots then were fixed on an aluminum support with carbon-adhesive glue and coated with goldpalladium (Spotter Coater SC7640, Guelph, Ontario, Canada). The samples were observed using a scanning electron microscope (Cambridge Instruments Stereoscan S). For cells, they were cultivated (90,000 cells per 13-mm diameter chambers, giving an approximate density of 70,000 cells/cm2) in culture chambers (Lab-Tek, Nalge Nunc International, Naperville, IL) and left to adhere for 48 hours. The wells were washed in Dulbecco’s modified Eagle medium without red phenol for 15 minutes, then fixed with glutaraldehyde containing 2.5% cacodylate buffer 0.1 mol/L, pH 7.2, overnight at 4°C. The samples on plugs then were rinsed and dehydrated by successive 2-minute baths in a series of increasing alcohol concentrations (30%, 50%, 70%, 80%, 90%, 100%), and finally incubated for 5 minutes with hexamethyldisilazane and dried under a range hood. The plates then were fixed on an aluminum support with carbon-adhesive glue and coated with gold-palladium (Spotter Coater SC7640). The samples were examined using the Cambridge Instruments Stereoscan S 240 SEM (Cambridge, MA).
Cell Culture Caco2–BBE cells were cultured to confluency in 75-cm2 flasks at 37°C in a humidified atmosphere containing 5% CO2. The culture medium comprised Dulbecco’s modified Eagle medium/Ham’s F-12 medium (Invitrogen, Carlsbad, CA) supplemented with L-glutamine (2 mmol/L), penicillin (100 U/mL), streptomycin (100 g/mL), and heat-inactivated fetal calf serum (10%) (Invitrogen).
Cellular Toxicity To assess potential toxicity of the KPV-loaded NPs we used 2 different assays, the WST-1 assay and ECIS. For the WST-1 assay, as previously described,2 cells were seeded in 96-well plates at a density of 5 ⫻ 104 cells per well and exposed to several concentrations of PVA-coated NPs loaded with dextran–FITC, KPV, or BSA for 24, 48, or 72 hours; the maximum concentration of NPs was 1 mg/mL. The WST-1 assay (Roche, Nutley, NJ) measures cleavage of the soluble, red tetrazolium salt WST-1 (4-[3(4-iodophenyl)-2-(4-nitrophenyl)-2H-5-tetrazolio]-1,3benzene disulfonate) by dehydrogenase present in intact mitochondria, which leads to formation of dark red formazan crystals. After assessing several NP concentrations and exposure times, 10 L of WST-1 proliferation reagent was added to Caco2–BBE cells (10 L/well) and incubated for 1–2 hours at 37°C. The wavelength for measuring the absorbance of the formazan product was 440 nm. For ECIS, cell-attachment assays were performed using ECIS technology (Applied BioPhysics, Troy, NY).3 The ECIS model 1600R (Applied BioPhysics) was used for these experiments. The measurement system consists of
March 2010
an 8-well culture dish (ECIS 8W1E plate), the surface of which is seeded with cell cultures. Each well contains a small, active electrode (area, 5.10-4 cm2) and a large counter electrode (area, 0.15 cm2) on the bottom of each well. A lock-in amplifier, with an internal oscillator, relays a signal to switch between the different wells, and a personal computer controls the measurement and stores the data. The ECIS system was obtained from Applied BioPhysics. Attachment and spreading of cells on the electrode surface change the impedance in such a way that morphologic information about the attached cells can be inferred. Caco2–BBE cells were seeded in ECIS 8W1E plates coated with 10 g/mL laminin I (Sigma) in Dulbecco’s modified Eagle medium (Invitrogen) supplemented with 10% (vol/vol) heat-inactivated fetal bovine serum (Invitrogen) and 1.5 g/mL plasmocin (Invitrogen). Once cells reached confluency, NPs were added at a concentration of 1 mg/mL. Control cell cultures were used for each experiment; KPV-loaded NPs were replaced by 400-nm latex microspheres (Interfacial Dynamics, Eugene, OR). Basal resistance measurements were performed using the ideal frequency for Caco2–BBE cells, 500 Hz, and a voltage of 1 V.3
Animals Female C57BL/6 mice (age, 8 wk; weight, 18 –22 g; Jackson Laboratories, Bar Harbor, ME) used for this study were group-housed under controlled temperature (25°C) and photoperiod (12:12-hour light– dark cycle) conditions, and given unrestricted access to standard diet and tap water. Mice were allowed to acclimate to these conditions for at least 7 days before inclusion in experiments
MPO Assay Colonic tissue samples were homogenized in icecold potassium phosphate buffer (50 mmol/L K2HPO4 and 50 mmol/L KH2PO4, pH 6.0) containing 0.5% hexadecyltrimethylammonium bromide (Sigma). The homogenates then were sonicated, freeze-thawed 3 times, and centrifuged at 17,500 relative centrifugal force for 15 minutes. Supernatants (20 L) or MPO standard were added to 1 mg/mL o-dianisidine hydrochloride (Sigma) and 0.0005% H2O2, and the change in absorbance at 450 nm was measured. One unit of MPO activity was defined as the amount that degraded 1 mol peroxidase per minute. The results were expressed as absorbance per microgram of protein. Total RNA was extracted using TRIzol reagent (Invitrogen) and reverse-transcribed using the complemen-
Imaging and Advanced Technology
853.e2
tary DNA (cDNA) Synthesis kit (Fermentas, Glen Burnie, MD). Reverse-transcription polymerase chain reaction was performed using the GeneJET Fast polymerase chain reaction kit (Fermentas) and the following specific primers: human IL-8 sense 5=-GTG CAG TTT TGC CAA GGA GT-3=, IL-8 antisense 5=-AAA TTT GGG GTG GAA AGG TT-3=; -actin sense 5=-GTC ACC CAC ACT GTG CCC ATC-3=, and -actin antisense 5=-ACG GAG TAC TTG CGC TCA GGA-3=. Real-time reverse-transcription polymerase chain reaction was performed using an iCycler (Bio-Rad, Hercules, CA). Briefly, cDNA was amplified by 40 cycles of 95°C for 15 seconds and 60°C for 1 minute, using the iQ SYBR Green Supermix system (Bio-Rad) and the following specific primers: IL-1 sense 5=-TCG CTC AGG GTC ACA AGA AA-3=, IL-1 antisense 5=-CAT CAG AGG CAA GGA GGA AAA C-3=; TNF-␣ sense 5=-AGG CTG CCC CGACTA CGT-3=, TNF-␣ antisense 5=-GAC TTT CTC CTG GTA TGA GAT AGC AAA-3=; 18S sense 5=-CCC CTC GAT GAC TTT AGC TGA GTG T-3=, 18S antisense 5=-CGC CGG TCC AAG AAT TTC ACC TCT-3=; mouse 36B4 sense 5=-TCC AGG CTT TGG GCA TCA-3=, and mouse 36B4 antisense 5=-CTT TAT CAG CTG CAC ATC ACT CAG A-3=. 18S and 36B4 were used as housekeeping genes. Fold-induction was calculated using the Ct method, ⌬⌬CT ⫽ (CtTarget-Cthousekeeping)infected ⫺ (CtTarget-Cthousekeeping)uninfected, and the final data were derived from 2⫺⌬⌬CT.
MPO Assay Colitis was induced by 3% (wt/vol) DSS (molecular weight, 40,000 daltons; ICN Biochemicals, Aurora, OH) added to the drinking water. Colonic inflammation was assessed 7 days after DSS treatment. Twelve mice were included in each group. To evaluate the effects of KPV-loaded NPs on induced colitis, the mice were gavaged with KPV-loaded NPs or with control material daily during the 7-day DSS treatment. The gavage procedure is described in Figure 4. References 1. Laroui H, Grossin L, Leonard M, et al. Hyaluronate-covered nanoparticles for the therapeutic targeting of cartilage. Biomacromolecules 2007;8:3879 –3885. 2. Ngamwongsatit P, Banada PP, Panbangred W, et al. WST-1-based cell cytotoxicity assay as a substitute for MTT-based assay for rapid detection of toxigenic Bacillus species using CHO cell line. J Microbiol Methods 2008;73:211–215. 3. Charrier L, Yan Y, Driss A, et al. ADAM-15 inhibits wound healing in human intestinal epithelial cell monolayers. Am J Physiol Gastrointest Liver Physiol 2005;288:G346 –G353.
Drug-Loaded Nanoparticles Targeted to the Colon With Polysaccharide Hydrogel Reduce Colitis in a Mouse Model HAMED LAROUI, GUILLAUME DALMASSO, HANG THI THU NGUYEN, YUTAO YAN, SHANTHI V. SITARAMAN, and DIDIER MERLIN Division of Digestive Diseases, Department of Medicine, Emory University, Atlanta, Georgia
BACKGROUND & AIMS: One of the challenges to treating inflammatory bowel disease (IBD) is to target the site of inflammation. We engineered nanoparticles (NPs) to deliver an anti-inflammatory tripeptide Lys-Pro-Val (KPV) to the colon and assessed its therapeutic efficacy in a mouse model of colitis. METHODS: NPs were synthesized by double-emulsion/solvent evaporation. KPV was loaded into the NPs during the first emulsion of the synthesis process. To target KPV to the colon, loaded NPs (NP-KPV) were encapsulated into a polysaccharide gel containing 2 polymers: alginate and chitosan. The effect of KPV-loaded NPs on inflammatory parameters was determined in vitro as well as in the dextran sodium sulfate–induced colitis mouse model. RESULTS: NPs (400 nm) did not affect cell viability or barrier functions. A swelling degree study showed that alginate-chitosan hydrogel containing dextran–fluorescein isothiocyanate– labeled NPs collapsed in the colon. Once delivered, NPs quickly released KPV on or within the closed area of colonocytes. The inflammatory responses to lipopolysaccharide were reduced in Caco2–BBE (brush border enterocyte) cells exposed to NP-KPV compared with those exposed to NPs alone, in a dose-dependent fashion. Mice given dextran sodium sulfate (DSS) followed by NP-KPV were protected against inflammatory and histologic parameters, compared with mice given only DSS. CONCLUSIONS: Nanoparticles are a versatile drug delivery system that can overcome physiologic barriers and target anti-inflammatory agents such as the peptide KPV to inflamed areas. By using NPs, KPV can be delivered at a concentration that is 12,000-fold lower than that of KPV in free solution, but with similar therapeutic efficacy. Administration of encapsulated drug-loaded NPs is a novel therapeutic approach for IBD.
icant side effects.1– 4 Newer targeted treatments such as anti–tumor necrosis factor (TNF) agents are effective in a subset of patients but have to be administered systemically and also are associated with significant side effects.1 A major advance in therapeutic strategies in diseases such as IBD would be the ability to target drugs to the site of inflammation in sufficient quantities to maximize local drug concentration and minimize systemic side effects. However, targeting drugs to the site of inflammation has remained a challenge in IBD because of the lack of vehicles that could carry sufficient drugs or that could be released at the site of inflammation. Another problem is created by organs of the gastrointestinal tract, particularly the colon, because it is a challenge to deliver the drug to the colon with minimal digestive enzyme degradation and/or systemic absorption.5 Various carriers have been designed to release the drug at a specific pH value, to be resistant to digestive enzymes, and/or require bacterial cleavage for activation, and several of these carriers currently are being investigated.6 – 8 However, most of these drugs need to be administered in large doses, multiple times a day, resulting in poor patient compliance. Among the various carriers proposed for drug delivery, polymeric nanoparticles (NPs) have been studied for several decades. The most widely used polymers to engineer NPs are aliphatic polyesters such as polylactide (PLA),9 polyglycolide, and co-polymers,10,11 which have been approved by the US Food and Drug Administration and have well-characterized biocompatibility and (bio)degradability properties.12,13 Recent studies have reported using nanotechnology for cell targeting. Thus, micelles have been described as promoting the uptake of 40-nm colloidal gold particles,
Keywords: Colon Targeting; Colitis; KPV-Loaded Nanoparticle; Polysaccharide Hydrogel.
Abbreviations used in this paper: BSA, bovine serum albumin; COX-2, cyclooxygenase-2; DSS, dextran sodium sulfate; ECIS, electrical impedance sensing; FITC, fluorescein isothiocyanate; IL, interleukin; LPS, lipopolysaccharide; MPO, myeloperoxidase; NP, nanoparticle; PCS, photon correlation spectroscopy; PLA, polylactide; PVA, polyvinyl alcohol; TNF, tumor necrosis factor. © 2010 by the AGA Institute 0016-5085/10/$36.00 doi:10.1053/j.gastro.2009.11.003
I
nflammatory bowel disease (IBD), which includes Crohn’s disease and ulcerative colitis, is a relapsing and remitting chronic disease for which treatment options are limited. Most existing treatments are associated with signif-
GASTROENTEROLOGY 2010;138:843– 853
Imaging and Advanced Technology continued
particularly in the rectal area.14 Destruction of extracellular matrix (colonic mucosal) in the apical side of epithelia and alterations in the barrier function organization, mainly caused by destruction of epithelia tight junctions, dramatically increase the potential permeability and interactions of the epithelia to NPs. As described,15 patients with ulcerative colitis showed a 15-fold greater uptake of Evans Blue dye in the plasma of treated individuals vs control individuals. In addition, after administration by enema, colonic absorption of both uncoated and ligand-modified latex NPs has been reported. This uptake seems to be size-dependent; approximately 16% of 100-nm polystyrene particles were absorbed over a 10-day period, whereas only approximately 2% of 1-m microparticles were absorbed under the same conditions.16 The attachment of bioadhesive tomato lectin to 500-mm NPs decreased uptake by rat colonic epithelial cells by 7-fold, indicating an alteration in surface particle characteristics or a lack of appropriate receptors on the colonic epithelia. In the present study, we sought to deliver drug-loaded NPs to the colon. We used the tripeptide KPV as an anti-inflammatory drug because it has been shown that KPV, added to drinking water at 205 g/day (assuming a water uptake of 6 mL17 of a 100-mol/L KPV solution) or administrated intraperitoneally at 10 g/day, decreased colitis in mice.18 –20 Furthermore, we compared the effective KPV dose when delivered in free solution with the effective dose when loaded onto NPs in the inducedcolitis model. With the aim of packaging the KPV-loaded NPs into a biocompatible and biodegradable polymer, we sought to identify a biomaterial made of polysaccharides that degrades in the colon and thus delivers the KPVloaded NPs to the inflamed colonic area. The biological properties of polysaccharides have been used for decades in applications such as wound healing, cell encapsulation, or as an oral delivery vehicle.21–24 We used a polysaccharide gel containing 2 polymers, alginate and chitosan, as a vector. Chitosan is a biocompatible and biodegradable polymer.25,26 Alginate and chitosan form gels that chelate with a calcium or sulfate solution, respectively.27 The combination of alginate and chitosan chelation should maintain the 3-dimensional form of the bead and the chitosan also has therapeutic effects on inflammatory cells in the colon.27,28
Materials and Methods Preparation of Nanoparticles Loaded With KPV and Characterization See Supplementary Materials and Methods section.
Size Determination of the KPV-Loaded NPs See Supplementary Materials and Methods section. 844
Scattering Electron Microscopy of NPs See Supplementary Materials and Methods section.
NP-Loaded KPV KPV and bovine serum albumin (BSA) form a homogenous internal aqueous phase (see the NP synthesis process described in Figure 1). To measure KPV loaded onto NPs, BSA concentration was measured by ultraviolet spectroscopy in the final washing solutions. The encapsulation rate of BSA then was determined based on the initial BSA concentration. After each final washing to remove extra polyvinyl alcohol (PVA) by centrifugation of NP suspension, the supernatant was collected. The accumulated washing volumes were used to determine the concentration of BSA present in the supernatant, thus providing the mass of protein not encapsulated. A mass balance was performed to determine the amount of BSA that was loaded into the NP according to the known initial concentration. The amount of BSA is determined by ultraviolet quantification at 280 nm: A ⫽ log (I0/I) ⫽ .l.c, where A is the solution absorbance, Io is the initial intensity, I is the intensity after sample, is the absorbance coefficient of BSA (L/g.cm), l is the cuvette length (cm), and c is the BSA concentration (g/L). Finally, the encapsulation rate was used to determine the KPV loading, assuming that KPV and BSA are loaded in the same way in a homogenous aqueous phase: encapsulation rate ⫽ 71% ⫾ 4%. This rate means that the loaded KPV concentration (in ng) compared with the dry weight in NPs (in mg) is 84 ⫾ 3.3 ng/mg. To compare KPV-loaded NPs with free KPV, we calculated a KPV concentration in water that was equivalent to a 500-g/mL concentration of KPV-loaded NPs. The calculated free KPV solution was equivalent to an 814-fold dilution of a 100-mol/L initial KPV solution.
Encapsulation of KPV-Loaded NPs Into Biomaterials Chitosan powder was solubilized in acetic acid then neutralized by addition of NaOH (0.1 mol/L) to give a final chitosan concentration of 0.6% (wt/vol). Mediumviscosity sodium alginate was prepared in NaCl (0.15 mol/L) 1.4% (wt/vol). Before mixing with NaCl, or acetic acid, the polymer powders were weighed, placed in glass tubes, and autoclaved. Alginate solution and chitosan solutions were mixed at a 1:1 ratio for a final concentration of 7 and 3 g/L, respectively. The polymer suspension was homogenized for 24 hours. Different NPs were added to obtain a 2-mg/mL concentration of hydrogel solution and stirred to disperse NPs throughout the polymer solution. A chelation solution containing 70 mmol/L of calcium chloride and 30 mmol/L of sodium sulfate was prepared. Figures 2 and 3 show the procedure for inclusion of loaded NPs into
Imaging and Advanced Technology continued
Figure 1. (A) Schematic process of NP synthesis. A hydrophilic drug is encapsulated by double emulsion water in oil in water (W/O/W). (B) Schematic representation of PLA NPs loaded with KPV, a hydrophilic drug, and coated with PVA.
biomaterials and the gavage method, respectively. A detailed protocol is available at http://www.natureprotocols. com/2009/09/03/a_method_to_target_bioactive_c.php.
Drug Release From NPs in Aqueous Solution The drug release kinetic calculation is based on the homogenous phase formed by KPV and BSA during the first aqueous emulsion. We assume that KPV and BSA similarly release from NPs. Thus, NPs covered with PVA and containing BSA– fluorescein isothiocyanate (FITC) and KPV were incubated in a solution simulating intestinal fluid, composed of phosphate buffered saline (PBS) containing 5 g/mL pepsin (Sigma, pepsin from porcine gastric mucosa) and 30 g/mL trypsin (Sigma, trypsin from porcine pancreas) at pH 6. The amount of BSA–FITC was analyzed using fluorescence spectroscopy. The measurement parameters were as follows: excitation wavelength was 490 nm and emission wavelength was 525 nm. The drug release was measured by sampling 100 L of the intestinal fluid at the indicated times. The total amount of BSA on the NPs is known; a concentration of 113 g BSA per mg NPs corresponds to a 70% encapsulation rate. This rate was calculated by measuring BSA in the washing solution of NPs at the end of the process and after centrifugation of the NPs (see NP-Loaded KPV section). For the presented release curve (Figure 4A), we made a 2-mg/mL solution of NPs in intestinal fluid pH 6 at 37°C. At each time point of the
kinetic curve, we sampled 100 L of the solution. Release of BSA is expressed as a percentage of the total initial amount in NPs and the results mirror the kinetic curve of KPV release.
Cell Culture See Supplementary Materials and Methods section.
Cellular Toxicity See Supplementary Materials and Methods section.
Swelling Degree of the Biomaterial Into the Digestive Tract To study polymer behavior in the digestive tract, the swelling degree was calculated. To be reliable, a minimum of 50 mg of polymer beads was required to calculate the swelling degree using the following equation: sd ⫽
mass t ⫺ mass t0 mass t0
Mass t and mass t0 are the mass in milligrams of the bead at time t and time 0 of the study, respectively; sd is the swelling degree. The swelling degree of the biomaterial was studied in: (1) gastric fluid, simulated by acetate buffer solution (0.2 mol/L) containing 50 g/mL pepsin (Sigma, pepsin from porcine gastric mucosa) and 5 g/mL trypsin (Sigma, trypsin from porcine pancreas) 845
Imaging and Advanced Technology continued
Figure 2. Schematic representation of biomaterial encapsulation of KPV-loaded NPs. (A) An alginate and chitosan hydrogel was formed by double linking of Ca2⫹ and SO42- ions. Suspension of NPs in the polysaccharide solution loads the final formed hydrogel with NPs. (B) Optical microscopy image of KPV-loaded NPs encapsulated into a hydrogel bead of alginate-chitosan linked via Ca2⫹ and SO42- ions.
and adjusted to pH 1 or 3; and (2) intestinal fluid, simulated by phosphate buffer solution (0.1 mol/L) containing 5 g/mL pepsin and 30 g/mL trypsin and adjusted to pH 4 or 6.
Animals See Supplementary Materials and Methods section.
Induction of Colitis See Supplementary Materials and Methods section.
Myeloperoxidase Assay See Supplementary Materials and Methods section.
RNA Extraction, Reverse-Transcription Polymerase Chain Reaction, and Real-Time Reverse-Transcription Polymerase Chain Reaction See Supplementary Materials and Methods section.
Results Characterization of the KPV-Loaded NPs As shown in Figure 4A, analysis of the KPV-loaded NPs revealed the average particle size evaluated by photon correlation spectroscopy (PCS) to be about 366 nm, 846
whereas that evaluated by scattering electron microscopy (SEM) was about 300 nm (Figure 4B). There are several possible explanations for this difference. First, the particles shown in Figure 4B were studied immediately after solvent evaporation and before washing, centrifugation, and freeze-drying; by contrast, the other particles were measured after these processes. Some aggregation therefore may have occurred, particularly during the freezedrying process, which could explain the greater size seen in Figure 4A. A second reason likely is related to the SEM technique that requires the study of the dry particles under vacuum, which can cause particle shrinkage. Finally, PCS calculates an intensity average size, whereas SEM evaluates a number average size, which generally is lower than the size calculated by PCS. We also showed that all NPs loaded with KPV are of similar size, as shown by the low polydispersity index of the size distribution (Figure 4A and Supplementary Figure 1).
Drug Release From the KPV-Loaded NPs As shown in Figure 4C, about 25% of the total initial amount of KPV loaded onto the NPs is released into an intestinal fluid solution at pH 6.2 within 30 minutes. This release of KPV from the NPs likely is owing to the release of KPV near the surface of the NPs. After 30
Imaging and Advanced Technology continued
Figure 3. Double-gavage procedure of encapsulated KPV-loaded NPs. (A) The first gavage delivered 100 L of the polymer mix solution (alginate, 7 g/L; chitosan, 3 g/L) containing a homogenous suspension of NPs (2 mg/mL). Red Ponceau has been added for visualization. (B) The second gavage delivered 50 L of a solution containing 70 mmol/L calcium chloride and 30 mmol/L sodium sulfate. (C) Visualization of the mixed hydrogel formed by chelation of the polymers in the stomach. (D) The hydrogel after extraction from a mouse stomach 5 minutes after the double-gavage method.
minutes, the release kinetics of KPV is more linear and about 35% of the total initial amount of KPV is released within 30 hours. This later KPV release phase seems to be the result of KPV diffusion across the NPs. Together, these results indicate that free KPV will be delivered rapidly (burst release) to the intestinal mucosa no more than 30 minutes after the KPV-loaded NPs come into contact with the intestinal fluid at colonic pH (pH 6.2).
KPV-Loaded NPs Do Not Affect Cell Viability For any biomaterial, including a combination of hydrogel and NPs, cytotoxicity assessments need to be performed. First, we performed a simple cytotoxicity test using water-soluble tetrazolium salt (WST-1 assay). As shown in Figure 4D, KPV-loaded NPs at a concentration of 1 mg/mL do not cause cellular death of Caco2–BBE cells over a 72-hour period compared with control materials. Similar results were found by using the normal intestinal enterocyte cell IEC6 cell line (data not shown). However, the WST-1 assay is not suitable for continuous, automated, and real-time analysis, and may be associated with several disadvantages. Of particular interest is the development of electrical impedance sensing (ECIS) as a tool for in vitro toxicity testing.29 This technique uses a small electrode that is deposited on the bottom of tissue culture wells and immersed in culture medium. As shown in Figure 4E, inoculated Caco2–BBE cells attach to the electrode surface to form a confluent layer and reach a resistance of 40,000 Ohms. KPV-loaded NPs at a 1-mg/mL concentration and latex particles at the same concentration (control) were added to a confluent Caco2–BBE monolayer. Even 80 hours of treatment with KPV-loaded NPs did not affect the Caco2–BBE monolayer resistance (before addition of KPV-loaded NPs, the
resistance was 40,000 Ohms compared with 43,000 Ohms 80 hours after addition of KPV-loaded NPs). These data are supported by results with the control biocompatible latex beads (diameter, 200 nm) on the resistance of Caco2–BBE monolayers (40,000 vs 43,000 Ohms before and after administration of beads, respectively). Together, these results show that KPV-loaded NPs coated with PVA do not affect cell growth or intestinal barrier function.
KPV Is Released Extracellularly and Is Taken up by Caco2–BBE Cells We next determined the interaction of NPs with epithelial cells. Caco2–BBE cells were treated with NPs (500 g/mL) for 48 hours, at which time cells were fixed and examined by transmission electron microscopy as described in the Materials and Methods section. As shown in Figure 5A, NPs released drug on the extracellular domain and were detected in intracellular vesicles. These data show that the NP effectively delivered the drug in the local environment of epithelial cells and was able to cross apical barriers into the cells.
Lipopolysaccharide-Induced Inflammatory Responses in Caco2–BBE Monolayers Are Reduced by KPV-Loaded NPs After stimulation with lipopolysaccharide (LPS) 10 mg/mL for 1 hour, as determined in preliminary experiments, Caco2–BBE monolayers show marked inflammatory responses, including increases in levels of the proinflammatory cytokine interleukin (IL)-8 and enzymes such as cyclooxygenase-2 (COX-2) (Figure 5B). Interestingly, we found that pretreatment of Caco2–BBE monolayers with KPV-loaded NPs for 72 hours reduced 847
Imaging and Advanced Technology continued
Figure 4. Physicochemical characterization and biocompatibility of KPV-loaded NPs. (A) Nanoparticles size average (nm) and polydispersity index of KPV-loaded NPs and latex NPs (size calibration standard) by photon correlation spectroscopy method (n ⫽ 5). (B) Scanning electron microscopy picture of KPV-loaded NP suspension (5 g/mL). (C) Cumulative drug release kinetic (percentage of total encapsulated drug). NPs at 5 mg/mL containing BSA–FITC and KPV were incubated in PBS containing 5 g/mL pepsin and 30 g/mL trypsin at pH 6 (n ⫽ 5). (D) WST-1 cell viability test on Caco2–BBE (percentage over control) after 72 hours of exposure to a 1-mg/mL suspension of 3 different NP samples. (Control, culture medium; NanoKPV, KPV-loaded NPs; NanoFITC, dextran–FITC–loaded NPs; NanoBSA, BSA-loaded NPs) (n ⫽ 8). (E) Electrical impedance sensing method on Caco2–BBE cells to determine cell viability (percentage over control) after a long exposure to a 1-mg/mL suspension of 2 different NP samples. (culture medium without NP as control; latex particles [culture medium without latex NP as control] and KPV-NP) (n ⫽ 3).
the increases in LPS-induced IL-8 and COX-2 expression levels (Figure 5B). This reduction was dose-dependent because a 50-g/mL concentration of KPV-loaded NPs reduced IL-8 and COX-2 expression levels by 50% and 50%, respectively, whereas concentrations of 200 g/mL and 500 g/mL reduced IL-8 and COX-2 expression levels by 70% and 80%, and 90% and 100%, respectively. Importantly, we showed that an equivalent concentration of free KPV did not affect the IL-8 and COX-2 expression levels. Together, these results show that KPV-loaded NPs are more effective than free KPV in reducing inflammatory responses induced by LPS in the intestinal epithelia.
Swelling Degree of the Biomaterial Into the Digestive Tract Kinetic of the swelling degree measured the evolution of water trapping in the hydrogel structure. Typically in an ionic hydrogel, the swelling degree evolution is inversely proportional to the power of ionic interactions between ions and polysaccharides. Thus, the higher swelling degree indicates the hydrogel is closer to collapsing. In the present study, we assessed the swelling degree of the biomaterial in solutions simulating intestinal fluid at 848
pH levels from 1 to 3 and simulating gastric fluid at a pH level from 4 to 6. The hydrogel comprises a mixture of 2 biocompatible polysaccharides: alginate and chitosan. This mixture is dropped by needle in a solution of Ca2⫹ and SO42- to form beads (Figure 2B). All compositions of those 2 polysaccharides had a total concentration of 10 g/L of polysaccharides. First, we tested alginate alone. With this composition, the hydrogel beads did not swell in a gastric solution of pH 1 or 2 over a time period of 24 hours (Supplementary Figure 2A). However, the hydrogel beads showed low levels of swelling in intestinal solutions of pH 3, 4, 5, or 6, and the level of swelling increased with time at each pH value (Supplementary Figure 2A). Together, these results show that hydrogel beads composed of alginate alone show minimal swelling at pH 3, 4, 5, or 6, and do not collapse. Next, we modified the hydrogel composition to a mix of alginate and chitosan at a ratio of 7/3 (wt/wt). We observed that the hydrogel beads exposed to a gastric solution of pH 1, 2, or 3 for 24 hours were stable. However, the hydrogel beads exposed to the intestinal solution of pH 4 for 24 hours shrank and those exposed to a solution of pH 5 completely collapsed (Supplementary Figure 2B). In ad-
Imaging and Advanced Technology continued
Figure 5. Cell interactions of KPV-loaded NPs and messenger RNA (mRNA) expression of proinflammatory cytokines (IL-8 and COX-2) on Caco2–BBE monolayers treated or not with KPV-loaded NPs. (A) Transmission electron microcopy picture of KPV–NPs (500 g/mL) on monolayer of Caco2–BBE cells after a LPS stimulation (10 g/mL, 1 h). Bar, 1 m. KPV-loaded NP in the extracellular domain (1). KPV-loaded NP interacting with the cell membrane starting an endocytose process (2). KPV-loaded NP in intracellular domain after endocytosis (3). (B) Caco2–BBE cells were stimulated with LPS (10 g/mL, 1 h), alone or after pretreatment with different concentrations of KPV-loaded NPs (50, 200, or 500 g/mL) or the equivalent concentration of free KPV in solution (free KPV 41 g/L corresponds to 500 g/mL KPV-loaded NPs) (41 g/L) for 72 hours (n ⫽ 3). (**P ⬍ .001, *P ⬍ .01 for IL-8; and ##P ⬍ .001, #P ⬍ .01 for COX-2).
dition, we observed that the collapsing effect of the hydrogel beads in the intestinal solution was time-dependent (Supplementary Figure 2B). Together, these results indicate that this biomaterial could be used to deliver KPV-loaded NPs to intestinal solutions of pH 5 and 6 (the approximate pH of the colonic fluid) but not at acidic pH values of 1, 2, or 3 (the approximate pH of the gastric fluid). In other words, these data indicate that the hydrogel composed of alginate and chitosan at a ratio of 7/3 (wt/wt) could be used to deliver KPV-loaded NPs to the colon but not to the gastric gland.
model for the study of human ulcerative colitis is dextran sodium sulfate (DSS)-induced acute and chronic colitis with ulceration in mice, which is characterized by a marked loss in body weight, rectal bleeding, reduction in
Targeting Drug-Loaded NPs to the Colon Next, we determined the rate of drug release from encapsulated NPs in the colon. Encapsulated dextran– FITC–loaded NPs (5 mg/mL of hydrogel) were administered to mice by daily gavage for 3 days. The mice then were killed and FITC fluorescence in each tissue was quantified along the gastrointestinal tract. The data were expressed by g of tissue protein. As shown in Figure 6, we detected lower levels of dextran–FITC in the stomach, the proximal small intestine, distal intestine, and cecum, with a higher level in the colon, indicating that the FITC was delivered mostly to the colon.
Encapsulated KPV-Loaded NPs Reduce Induced Colitis Many experimental animal models have been used to study human IBD. One well-characterized animal
Figure 6. Localization of encapsulated dextran–FITC NPs (g dextran–FITC/g protein of tissue) on the digestive tract after 3 days of gavage. Localization of dextran–FITC NPs (g dextran–FITC/g protein of tissue) on the digestive tract after 3 days of gavage. Digestive tract was divided into 6 sections as follows: stomach, proximal (1/3) small intestine, medial (2/3) small intestine, and distal (3/3) small intestine, cecum, and colon (n ⫽ 12). 849
Imaging and Advanced Technology continued
Figure 7. Histology of DSS-induced colitis in 7 different biomaterial treatments on mice.During in vivo experiments, all mice received a daily gavage of 150 L of biomaterial. Control group received the hydrogel with empty NPs (alginate, 7 g/L; chitosan, 3 g/L) and regular water as drinking solution. DSS treated group received the hydrogel with empty NPs (alginate, 7 g/L; chitosan, 3 g/L) and 3% DSS solution as drinking solution. KPV–NP– treated group received the hydrogel (alginate, 7 g/L; chitosan, 3 g/L) loaded with encapsulated KPV-loaded NPs and 3% DSS solution as drinking solution. Histologic section of colon after 7 days of a daily gavage in groups 1, 2, and 3.
colon length, and destruction of the intestinal epithelium. We investigated the anti-inflammatory effects of KPV-loaded NPs encapsulated in the hydrogel (alginate and chitosan: 7 g/L and 3 g/L, respectively) in mice with active DSS-induced colitis. C57BL/6 mice received regular water and a gavage of 150 L of hydrogel daily for 7 days (group 1), water with 3% DSS and gavage of 150 L of hydrogel daily for 7 days (group 2), water with 3% DSS and gavage of 150 L of hydrogel encapsulated with KPV-loaded NPs daily for 7 days (group 3), water with 3% DSS and gavage of 150 L of hydrogel encapsulated with KPV-loaded NPs daily for 7 days (group 4), water with 3% DSS and gavage of 150 L of hydrogel containing 41 g/L of free KPV daily for 7 days (group 5), water with 3% DSS and gavage of 150 L of hydrogel encapsulated with dextran–FITC–loaded NPs daily for 7 days (group 6), and water with 3% DSS for 7 days (group 7). This treatment regimen resulted in intestinal colitis and histologic studies revealed DSS-induced cell-wall damage with epithelial exfoliation when compared with mice that received water alone (Figure 7). In the DSS-treated mice, we also observed interstitial edema, dilation of vessels, and a general increase in the number of inflammatory cells (lymphocytes and polynuclear cells) in the lamina propria (Figure 7). To evaluate the role of KPV in the inflammatory process, encapsulated KPV-loaded NPs were administered by daily gavage for 7 days to the mice that also received DSS in the drinking water; after 7 days the mice were killed. Our histologic studies showed reduced intestinal inflammation in mice with DSS-induced colitis treated with encapsulated KPV-loaded NPs, compared with control mice that received only water (Figure 7). We additionally observed weight loss in the DSS-treated mice (Figure 8A), which started at day 4 and reached a ma850
ximum at day 7. The administration of encapsulated KPV-loaded NPs reduced the weight loss of DSS-induced colitis (Figure 8A). However, administration of encapsulated NPs to mice with DSS-induced colitis did not reduce the severity of weight loss compared with mice that received DSS alone (Figure 8A). Intestinal myeloperoxidase (MPO) activity was measured to indicate the extent of neutrophilic infiltration. MPO levels in the colon were increased significantly in mice with DSSinduced colitis compared with control mice that received only water (Figure 8B). Administration of encapsulated KPV-loaded NPs significantly reduced DSS-induced colonic inflammation, whereas administrations of biomaterial with free KPV at an equivalent concentration (41 g/mL) had no effect on the basal MPO levels in the colonic mucosa (Figure 8B). These results were reflected in the proinflammatory cytokine levels (TNF-␣ and IL1), which were measured by real-time reverse-transcription polymerase chain reaction in mucosa obtained from the colitic mice (Figure 8C and D). As controls, biomaterials without or encapsulated with empty NPs did not have anti-inflammatory effects on DSS-induced colitis (Figure 8).
Discussion Most currently accepted treatment regimens for IBD require frequent administration of high-dose antiinflammatory drugs, which can lead to high rates of adverse events. Sustained drug-release devices, such as pellets, capsules, or tablets designed to deliver drugs specifically to the colon over longer periods of time, have been developed, but their efficiencies have been limited by the effects of IBD-related diarrhea, which increases
Imaging and Advanced Technology continued
Figure 8. In vivo parameters of DSS-induced colitis in 7 different biomaterial treatments on mice. During in vivo experiments, mice received (or not for group 7) a daily gavage of 150 L of biomaterial with or without NPs. All groups had DSS as drinkable water excepted group 1 (water). Groups numbers were defined as followed: number ⫽ biomaterial ⫹ daily drink; hydrogel, alginate 7 g/L and chitosan 3 g/L; DSS, 30 g/L; Ø-NP, empty NP; 1, hydrogel ⫹ water; 2, hydrogel ⫹ DSS; 3, hydrogel ⫹ KPV–NP (encapsulated KPV-loaded NPs) ⫹ DSS; 4, hydrogel ⫹ Ø-NP ⫹ DSS; 5, hydrogel with 41 g/L free KPV (free KPV in hydrogel) ⫹ DSS; 6, hydrogel ⫹ dextran–FITC NP (encapsulated FITC-loaded NPs); 7, DSS only. ##P ⬍ .001. (A) Percentage of initial body weight (%) after 7 days of a daily gavage in groups 1–7. (B) Percentage increase over control of MPO activity after 7 days in groups 1, 2, 3 4, 5, and 7. (C) mRNA expression over control of TNF-␣ after 7 days in groups 1, 2, 3, 4, and 5. (D) mRNA expression over control of IL- after 7 days in groups 1, 2, 3, 4, and 5.
pellet elimination and reduces the duration of drug release. Thus, a carrier system capable of delivering the drug specifically and exclusively to the inflamed regions for a prolonged period is needed. Such a system significantly could reduce the side effects of conventional antiinflammatory compounds. Controlled drug delivery involves the association of a drug with a carrier system that modulates its pharmacokinetic properties and biodistribution. Several drug carriers and delivery systems have been investigated; among these, the NP is an attractive carrier that recently has shown promise for promoting drug efficacy.6,7,30,31 NPs are solid colloidal particles of approximately 10 –1000 nm in diameter. They consist of macromolecular materials in which the active component is dissolved, entrapped, or encapsulated, or to which the active component is absorbed or attached.32,33 Here, we have shown that KPV-loaded NPs are biocompatible and rapidly release KPV in intestinal solutions at pH 6.2. In addition, we have shown that this release of KPV reduced the intestinal inflammatory responses to LPS in vitro, and that the effective dose of KPV was 12,000 times lower when administered via NPs compared with free solution, showing that the NP method can
successfully reduce effective drug doses. One possibility for this difference is that in contrast to free KPV in solution, KPV delivered by NPs is in close proximity to the cell membrane. Another explanation could be that epithelial cells endocytose KPV-loaded NPs, providing an efficient route for cellular uptake of KPV. As reported previously, the NP is an attractive carrier that has shown great promise for increasing drug efficacy.6,7,30,31 After administration of KPV-loaded NPs, free KPV likely will be generated by initial drug release during passage through the small intestine and by other processes such as degradation of the NP building matrix polyester by digestive enzymes. To deliver the KPV-loaded NPs to the colonic lumen, we encapsulated them into a biomaterial comprised of alginate and chitosan at a ratio of 7/3 (wt/wt), and showed that this biomaterial collapses in intestinal solution at pH 5 or 6, which is the colonic pH under inflamed and noninflamed states.34 Thus, we showed that NP release from the hydrogel (and so KPV release from NP) occurs mostly in the colonic lumen compared with other parts of the gastrointestinal tract such as the stomach and small intestine. Together, these results show that the KPV-loaded NPs are well 851
Imaging and Advanced Technology continued
protected during transit through the gastric gland and the small intestine. The encapsulated KPV-loaded NPs delivered by gavage at 2 mg/mL/day for 6 days provide a KPV concentration of 25.2 ng/day to the colonic lumen. In contrast, the effective dose of KPV in free solution able to reduce DSS-induced colitis was approximately 200 g/day. In other words, the effective dose of KPV was 12,000 times lower when administrated via encapsulated KPV-loaded NPs when compared with KPV in free solution. We also investigated whether such a low KPV concentration delivered to the colonic lumen was sufficient to reduce intestinal inflammation. By using the DSS colitis model, we showed that KPV delivered by gavage of encapsulated KPV-loaded NPs efficiently reduced the severity of intestinal colitis, as shown by a reduction in DSSinduced MPO activity and histologic examination. Interestingly, we found that, using our delivery method, 1 pmol/L KPV more effectively reduced intestinal inflammation compared with 100 mol/L KPV administered in the drinking water. Based on our in vitro results, we suggest that KPV delivered by NPs is in close proximity to the cell membrane and was taken up by the hPepT1 membrane transporter.18 In addition, epithelial cells might endocytose KPV-loaded NPs and this could be an efficient route for cellular uptake of KPV as well. It is known in intestinal colitis that the intestinal barrier function is disrupted in areas of inflammation. In these inflamed areas, NPs might have been translocated to the submucosa and the KPV release could occur at close proximity to the immune cells. Given that KPV affects inflammatory responses in both epithelial and immune cells,18,35 it is reasonable to suggest that, under the delivery conditions used in the present study, low doses of KPV would affect the inflammatory responses of these 2 cell types. It is well known that high levels of colonic mucus are secreted during intestinal inflammation. The biomaterial used in the present study was composed of chitosan; hence, our encapsulated KPV-loaded NPs could interact with the mucin layer of the inflamed colonic tissue and deliver KPV (released from NPs) to the inflamed colonic area. Therefore, chitosan has been investigated as a superior muco-adhesive cationic polymer owing to its ability to develop molecular attraction forces by electrostatic interactions with the negative charges of mucin; this ability has been shown by the formation of hydrogen bonds or ionic interactions between the positively charged amino groups of chitosan and the negatively charged sialic acid residues of mucin, depending on the environmental pH.36 Further studies should be performed to investigate the therapeutic effects of encapsulated KPV-loaded NPs on the clinical symptoms of experimental colitis and histologic mucosal repair. 852
Our results show that encapsulated drug-loaded NPs are a versatile drug-delivery system, with the ability to overcome physiologic barriers and target low concentrations of the anti-inflammatory drug KPV to inflamed areas to relieve symptoms of IBD.
Supplementary Material Note: To access the supplementary material accompanying this article, visit the online version of Gastroenterology at www.gastrojournal.org, and at doi: 10.1053/j.gastro.2009.11.003. References 1. Ballinger A. Adverse effects of nonsteroidal anti-inflammatory drugs on the colon. Curr Gastroenterol Rep 2008;10:485– 489. 2. Langan RC, Gotsch PB, Krafczyk MA, et al. Ulcerative colitis: diagnosis and treatment. Am Fam Physician 2007;76:1323– 1330. 3. Truelove SC. Systemic and local corticosteroid therapy in ulcerative colitis. Br Med J 1960;1:464 – 467. 4. Reshef R, Varkel J, Shiller M, et al. Systemic effects of rectally administered corticosteroids. Isr J Med Sci 1992;28:98 –100. 5. Hebrard G, Hoffart V, Cardot J, et al. Investigation of coated whey protein/alginate beads as sustained release dosage form in simulated gastrointestinal environment. Drug Dev Ind Pharm 2009;14:1–10. 6. Lamprecht A, Yamamoto H, Takeuchi H, et al. A pH-sensitive microsphere system for the colon delivery of tacrolimus containing nanoparticles. J Control Release 2005;104:337–346. 7. Lamprecht A, Yamamoto H, Takeuchi H, et al. Nanoparticles enhance therapeutic efficiency by selectively increased local drug dose in experimental colitis in rats. J Pharmacol Exp Ther 2005; 315:196 –202. 8. Leserman LD, Machy P, Barbet J. Cell-specific drug transfer from liposomes bearing monoclonal antibodies. Nature 1981;293: 226 –228. 9. Laroui H, Grossin L, Leonard M, et al. Hyaluronate-covered nanoparticles for the therapeutic targeting of cartilage. Biomacromolecules 2007;8:3879 –3885. 10. Makhlof A, Tozuka Y, Takeuchi H. pH-sensitive nanospheres for colon-specific drug delivery in experimentally induced colitis rat model. Eur J Pharm Biopharm 2009;72:1– 8. 11. Meissner Y, Pellequer Y, Lamprecht A. Nanoparticles in inflammatory bowel disease: particle targeting versus pH-sensitive delivery. Int J Pharm 2006;316:138 –143. 12. Feng SS, Mei L, Anitha P, et al. Poly(lactide)-vitamin E derivative/ montmorillonite nanoparticle formulations for the oral delivery of Docetaxel. Biomaterials 2009;30:3297–3306. 13. Yang H, Li K, Liu Y, et al. Poly(D,L-lactide-co-glycolide) nanoparticles encapsulated fluorescent isothiocyanate and paclitaxol: preparation, release kinetics and anticancer effect. J Nanosci Nanotechnol 2009;9:282–287. 14. Fukui H, Murakami M, Yoshikawa H, et al. Studies on the promoting effect of lipid-surfactant mixed micelles (MM) on intestinal absorption of colloidal particles. Dependence on particle size and administration site. J Pharmacobiodyn 1987;10:236 –242. 15. Kitajima S, Takuma S, Morimoto M. Changes in colonic mucosal permeability in mouse colitis induced with dextran sulfate sodium. Exp Anim 1999;48:137–143. 16. Lamprecht A, Schafer U, Lehr CM. Size-dependent bioadhesion of micro- and nanoparticulate carriers to the inflamed colonic mucosa. Pharm Res 2001;18:788 –793.
Imaging and Advanced Technology continued
17. Bachmanov AA, Reed DR, Beauchamp GK, et al. Food intake, water intake, and drinking spout side preference of 28 mouse strains. Behav Genet 2002;32:435– 443. 18. Dalmasso G, Charrier-Hisamuddin L, Thu Nguyen HT, et al. PepT1-mediated tripeptide KPV uptake reduces intestinal inflammation. Gastroenterology 2008;134:166 –178. 19. Brzoska T, Luger TA, Maaser C, et al. Alpha-melanocyte-stimulating hormone and related tripeptides: biochemistry, antiinflammatory and protective effects in vitro and in vivo, and future perspectives for the treatment of immune-mediated inflammatory diseases. Endocr Rev 2008;29:581– 602. 20. Kannengiesser K, Maaser C, Heidemann J, et al. Melanocortinderived tripeptide KPV has anti-inflammatory potential in murine models of inflammatory bowel disease. Inflamm Bowel Dis 2008; 14:324 –331. 21. Burd DA, Greco RM, Regauer S, et al. Hyaluronan and wound healing: a new perspective. Br J Plast Surg 1991;44:579 –584. 22. Yotsuyanagi T, Ohkubo T, Ohhashi T, et al. Calcium-induced gelation of alginic acid and pH-sensitive reswelling of dried gels. Chem Pharm Bull 1987;35:1555–1563. 23. Garcia AM, Ghaly ES. Preliminary spherical agglomerates of water soluble drug using natural polymer and cross-linked technique. J Control Release 1996;40:179 –186. 24. Clayton HA, London NJ, Colloby PS, et al. The effect of capsule composition on the biocompatibility of alginate-poly-l-lysine capsules. J Microencapsul 1991;8:221–233. 25. Azab AK, Orkin B, Doviner V, et al. Crosslinked chitosan implants as potential degradable devices for brachytherapy: in vitro and in vivo analysis. J Control Release 2006;111:281–289. 26. Tozaki H, Odoriba T, Okada N, et al. Chitosan capsules for colon-specific drug delivery: enhanced localization of 5-aminosalicylic acid in the large intestine accelerates healing of TNBSinduced colitis in rats. J Control Release 2002;82:51– 61. 27. Mladenovska K, Raicki RS, Janevik EI, et al. Colon-specific delivery of 5-aminosalicylic acid from chitosan-Ca-alginate microparticles. Int J Pharm 2007;342:124 –136. 28. Shi C, Zhu Y, Ran X, et al. Therapeutic potential of chitosan and its derivatives in regenerative medicine. J Surg Res 2006;133: 185–192. 29. Xiao C, Luong JH. Assessment of cytotoxicity by emerging impedance spectroscopy. Toxicol Appl Pharmacol 2005;206:102–112. 30. Jain KK. The role of nanobiotechnology in drug discovery. Drug Discov Today 2005;10:1435–1442.
31. Lamprecht A. Multiparticulate systems in the treatment of inflammatory bowel disease. Curr Drug Targets Inflamm Allergy 2003; 2:137–144. 32. Bertholon I, Ponchel G, Labarre D, et al. Bioadhesive properties of poly(alkylcyanoacrylate) nanoparticles coated with polysaccharide. J Nanosci Nanotechnol 2006;6:3102–3109. 33. Hillaireau H, Le Doan T, Chacun H, et al. Encapsulation of monoand oligo-nucleotides into aqueous-core nanocapsules in presence of various water-soluble polymers. Int J Pharm 2007;331: 148 –152. 34. Nugent SG, Kumar D, Rampton DS, et al. Intestinal luminal pH in inflammatory bowel disease: possible determinants and implications for therapy with aminosalicylates and other drugs. Gut 2001;48:571–577. 35. Luger TA, Scholzen TE, Brzoska T, et al. New insights into the functions of alpha-MSH and related peptides in the immune system. Ann N Y Acad Sci 2003;994:133–140. 36. Sogias IA, Williams AC, Khutoryanskiy VV. Why is chitosan mucoadhesive? Biomacromolecules 2008;9:1837–1842.
Received June 26, 2009. Accepted November 4, 2009. Reprint requests Address requests for reprints to: Shanthi V. Sitaraman, MD, PhD, Division of Digestive Diseases, Department of Medicine, Emory University, Atlanta, Georgia 30322. e-mail: [email protected]; fax: (404) 727-5767. Acknowledgments The authors want to thank Dr Obertone for her scientific contribution to this work. Conflicts of interest The authors disclose no conflicts. Funding This work was supported by National Institutes of Health of Digestive Diseases Research center grant (R24-DK-064399), R56DK061941 (D.M.), RO1-DK076825 and RO1-DK064711 (S.V.S.); and a research fellowship award from the Crohn’s and Colitis Foundation of America (G.D.).
853
853.e1
Imaging and Advanced Technology
Supplementary Materials and Methods Size Determination of the KPV-Loaded NPs PLA was purchased from Sigma-Aldrich (MW of 75–120 kg/mol). The NPs were produced by the doubleemulsion/solvent evaporation procedure described previously1 for the elaboration of PVA-covered PLA NPs. This method involves the use of an amphiphilic molecule as the surfactant of the secondary emulsion; however, in the present study, PVA was used, which was purchased from Aldrich (Milwaukee, WI) (Mw ⫽ 13,000 –23,000 g/mol, hydroxylated at 87%– 89%). Typically, a primary water in oil emulsion (w/o) was prepared by mixing an organic phase (4 mL dichloromethane) containing PLA (25 g/L) with an internal aqueous phase (400 L) containing BSA (50 g/L), and containing or not a tripeptide (Lys-Pro-Val, KPV; 100 nmol/L). The mixture was stirred for 2 minutes with a Vortex mixer (Maxi Mix II; Thermolyne, Dubuque, IA), then sonicated (2 min, power 6, 50% active cycle, in an ice bath) using a Sonifier 450 (Branson). This primary emulsion was poured into a second aqueous phase (8 mL) (external aqueous phase) containing an amphiphilic molecule (PVA 20 g/L). The water in oil in water emulsion (w/o/w) then was obtained by sonication (same conditions as those used for the primary emulsion). This double-emulsion then was transferred to an aqueous 10⫺2 mol/L NaNO3 dispersing phase (40 mL) and stirred for 5 minutes. The organic solvent was evaporated under stirring and ambient air conditions and the collected solid nanospheres were resuspended in water and then centrifuged again to remove the excess PVA. This purification procedure was repeated twice. The final suspension then was freeze-dried. Nanoparticles covered with PVA and containing fluorescent dextran–FITC (molecular weight, 250,000 g/mol; Sigma-Aldrich, St. Louis, MO) or BSA–FITC (Sigma-Aldrich) were prepared according to the procedure described earlier. Dextran–FITC and BSA– FITC were dissolved, respectively, in the internal aqueous phase at a concentration of 2.5 g/L and 50 g/L. Figures 1 and 2 summarize the preparation of KPV-loaded NPs.
PCS The size distribution of NPs (mean diameter and standard deviation) was determined by PCS using Brookhaven equipment (Model 90Plus, Nanoparticle Size analyzer; Brookhaven Instruments Corporation, Holtsville, NY) in 10-3 mol/L NaCl at 25°C.
SEM of NPs Suspensions of NPs on plots were fixed with glutaraldehyde containing 2.5% cacodylate buffer (0.1 mol/L, pH 7) overnight at 4°C. Afterward, the samples were rinsed and dehydrated by 2-minute rinses with a series of increasing alcohol concentrations (30%, 50%, 70%, 80%, 90%, 100% vol/vol), and then incubated for 5 minutes with hexamethyldisilazane and dried under a
GASTROENTEROLOGY Vol. 138, No. 3
range hood. The plots then were fixed on an aluminum support with carbon-adhesive glue and coated with goldpalladium (Spotter Coater SC7640, Guelph, Ontario, Canada). The samples were observed using a scanning electron microscope (Cambridge Instruments Stereoscan S). For cells, they were cultivated (90,000 cells per 13-mm diameter chambers, giving an approximate density of 70,000 cells/cm2) in culture chambers (Lab-Tek, Nalge Nunc International, Naperville, IL) and left to adhere for 48 hours. The wells were washed in Dulbecco’s modified Eagle medium without red phenol for 15 minutes, then fixed with glutaraldehyde containing 2.5% cacodylate buffer 0.1 mol/L, pH 7.2, overnight at 4°C. The samples on plugs then were rinsed and dehydrated by successive 2-minute baths in a series of increasing alcohol concentrations (30%, 50%, 70%, 80%, 90%, 100%), and finally incubated for 5 minutes with hexamethyldisilazane and dried under a range hood. The plates then were fixed on an aluminum support with carbon-adhesive glue and coated with gold-palladium (Spotter Coater SC7640). The samples were examined using the Cambridge Instruments Stereoscan S 240 SEM (Cambridge, MA).
Cell Culture Caco2–BBE cells were cultured to confluency in 75-cm2 flasks at 37°C in a humidified atmosphere containing 5% CO2. The culture medium comprised Dulbecco’s modified Eagle medium/Ham’s F-12 medium (Invitrogen, Carlsbad, CA) supplemented with L-glutamine (2 mmol/L), penicillin (100 U/mL), streptomycin (100 g/mL), and heat-inactivated fetal calf serum (10%) (Invitrogen).
Cellular Toxicity To assess potential toxicity of the KPV-loaded NPs we used 2 different assays, the WST-1 assay and ECIS. For the WST-1 assay, as previously described,2 cells were seeded in 96-well plates at a density of 5 ⫻ 104 cells per well and exposed to several concentrations of PVA-coated NPs loaded with dextran–FITC, KPV, or BSA for 24, 48, or 72 hours; the maximum concentration of NPs was 1 mg/mL. The WST-1 assay (Roche, Nutley, NJ) measures cleavage of the soluble, red tetrazolium salt WST-1 (4-[3(4-iodophenyl)-2-(4-nitrophenyl)-2H-5-tetrazolio]-1,3benzene disulfonate) by dehydrogenase present in intact mitochondria, which leads to formation of dark red formazan crystals. After assessing several NP concentrations and exposure times, 10 L of WST-1 proliferation reagent was added to Caco2–BBE cells (10 L/well) and incubated for 1–2 hours at 37°C. The wavelength for measuring the absorbance of the formazan product was 440 nm. For ECIS, cell-attachment assays were performed using ECIS technology (Applied BioPhysics, Troy, NY).3 The ECIS model 1600R (Applied BioPhysics) was used for these experiments. The measurement system consists of
March 2010
an 8-well culture dish (ECIS 8W1E plate), the surface of which is seeded with cell cultures. Each well contains a small, active electrode (area, 5.10-4 cm2) and a large counter electrode (area, 0.15 cm2) on the bottom of each well. A lock-in amplifier, with an internal oscillator, relays a signal to switch between the different wells, and a personal computer controls the measurement and stores the data. The ECIS system was obtained from Applied BioPhysics. Attachment and spreading of cells on the electrode surface change the impedance in such a way that morphologic information about the attached cells can be inferred. Caco2–BBE cells were seeded in ECIS 8W1E plates coated with 10 g/mL laminin I (Sigma) in Dulbecco’s modified Eagle medium (Invitrogen) supplemented with 10% (vol/vol) heat-inactivated fetal bovine serum (Invitrogen) and 1.5 g/mL plasmocin (Invitrogen). Once cells reached confluency, NPs were added at a concentration of 1 mg/mL. Control cell cultures were used for each experiment; KPV-loaded NPs were replaced by 400-nm latex microspheres (Interfacial Dynamics, Eugene, OR). Basal resistance measurements were performed using the ideal frequency for Caco2–BBE cells, 500 Hz, and a voltage of 1 V.3
Animals Female C57BL/6 mice (age, 8 wk; weight, 18 –22 g; Jackson Laboratories, Bar Harbor, ME) used for this study were group-housed under controlled temperature (25°C) and photoperiod (12:12-hour light– dark cycle) conditions, and given unrestricted access to standard diet and tap water. Mice were allowed to acclimate to these conditions for at least 7 days before inclusion in experiments
MPO Assay Colonic tissue samples were homogenized in icecold potassium phosphate buffer (50 mmol/L K2HPO4 and 50 mmol/L KH2PO4, pH 6.0) containing 0.5% hexadecyltrimethylammonium bromide (Sigma). The homogenates then were sonicated, freeze-thawed 3 times, and centrifuged at 17,500 relative centrifugal force for 15 minutes. Supernatants (20 L) or MPO standard were added to 1 mg/mL o-dianisidine hydrochloride (Sigma) and 0.0005% H2O2, and the change in absorbance at 450 nm was measured. One unit of MPO activity was defined as the amount that degraded 1 mol peroxidase per minute. The results were expressed as absorbance per microgram of protein. Total RNA was extracted using TRIzol reagent (Invitrogen) and reverse-transcribed using the complemen-
Imaging and Advanced Technology
853.e2
tary DNA (cDNA) Synthesis kit (Fermentas, Glen Burnie, MD). Reverse-transcription polymerase chain reaction was performed using the GeneJET Fast polymerase chain reaction kit (Fermentas) and the following specific primers: human IL-8 sense 5=-GTG CAG TTT TGC CAA GGA GT-3=, IL-8 antisense 5=-AAA TTT GGG GTG GAA AGG TT-3=; -actin sense 5=-GTC ACC CAC ACT GTG CCC ATC-3=, and -actin antisense 5=-ACG GAG TAC TTG CGC TCA GGA-3=. Real-time reverse-transcription polymerase chain reaction was performed using an iCycler (Bio-Rad, Hercules, CA). Briefly, cDNA was amplified by 40 cycles of 95°C for 15 seconds and 60°C for 1 minute, using the iQ SYBR Green Supermix system (Bio-Rad) and the following specific primers: IL-1 sense 5=-TCG CTC AGG GTC ACA AGA AA-3=, IL-1 antisense 5=-CAT CAG AGG CAA GGA GGA AAA C-3=; TNF-␣ sense 5=-AGG CTG CCC CGACTA CGT-3=, TNF-␣ antisense 5=-GAC TTT CTC CTG GTA TGA GAT AGC AAA-3=; 18S sense 5=-CCC CTC GAT GAC TTT AGC TGA GTG T-3=, 18S antisense 5=-CGC CGG TCC AAG AAT TTC ACC TCT-3=; mouse 36B4 sense 5=-TCC AGG CTT TGG GCA TCA-3=, and mouse 36B4 antisense 5=-CTT TAT CAG CTG CAC ATC ACT CAG A-3=. 18S and 36B4 were used as housekeeping genes. Fold-induction was calculated using the Ct method, ⌬⌬CT ⫽ (CtTarget-Cthousekeeping)infected ⫺ (CtTarget-Cthousekeeping)uninfected, and the final data were derived from 2⫺⌬⌬CT.
MPO Assay Colitis was induced by 3% (wt/vol) DSS (molecular weight, 40,000 daltons; ICN Biochemicals, Aurora, OH) added to the drinking water. Colonic inflammation was assessed 7 days after DSS treatment. Twelve mice were included in each group. To evaluate the effects of KPV-loaded NPs on induced colitis, the mice were gavaged with KPV-loaded NPs or with control material daily during the 7-day DSS treatment. The gavage procedure is described in Figure 4. References 1. Laroui H, Grossin L, Leonard M, et al. Hyaluronate-covered nanoparticles for the therapeutic targeting of cartilage. Biomacromolecules 2007;8:3879 –3885. 2. Ngamwongsatit P, Banada PP, Panbangred W, et al. WST-1-based cell cytotoxicity assay as a substitute for MTT-based assay for rapid detection of toxigenic Bacillus species using CHO cell line. J Microbiol Methods 2008;73:211–215. 3. Charrier L, Yan Y, Driss A, et al. ADAM-15 inhibits wound healing in human intestinal epithelial cell monolayers. Am J Physiol Gastrointest Liver Physiol 2005;288:G346 –G353.