Nanomedicines Constructed By Carboxyl-containing Drugs-mediated Self-assembly Of Hydrophilic Polymers

In modern drug devolvement, the low water solubility, lowbioavailability,andtoxicityissueshave prohibited the clinical application of a large number of drugcandidates. In addition, the poorsolubility ofmany clinically used smallmoleculedrugsrestricts their broad applications. To solve these problems, aseries of newdrug deliverysystems such as nanoparticles-based drug delivery platforms have been developed, whichhave many unique advantagesbecauseoftheirsmallparticle size andspecial structure.Numerous studies have indicated that nanocarriers can significantly increasethesolubilityofpoorly soluble drugs, and therefore greatly improve their oral bioavailability. Inaddition,nanovehicles can significantly improve the drug’s efficacyand reduce its sideeffectsor adverse reactions by their passive or active targeting. Self-assembled polymernanocarriers, one ofthe most widelystudiednanosystems, havebeen widelyusedfor deliveryof avarietyoftherapeutics including small molecule drugs, peptides, proteins,and nucleicacids. Polymeric assemblies are generally assembly by hydrophilicoramphiphilicpolymersvia non-covalentinteractions like hydrophobic, electrostatic, andhydrogen bonding forces. Hydrophobic drugsaregenerallywrappedinthehydrophobicdomainofpolymerassemblies based on hydrophobic interaction. However, hydrophobicinteraction aloneoften leads tolowdrugloading, while enhancing drug feeding will result inthe formation of drug crystals.Although the expected efficacy may be achieved by increaseddrug dose, the concomitant increase in the dose of polymer vehicles would raisethecost oftreatmentand produce side effects. Besides, amphiphilic copolymers employed forself-assembly of drug delivery nanoplatforms often possess complex chemical structureandneed to be synthesized by multiple processes, which strongly hinder thebench-to-bedside translation of nanomedicines based on polymer assemblies.Accordingly, it is highly necessary to develop an effective, high drug loading, and facilestrategy to fabricate nanoplatforms by polymer self-assembly. To circumvent this issue, we hypothesize that multiple non-convalent interactions between polymer and drug mayenhance the encapsulation of drug molecules into nanoassemblies.In this study, we firstlyselected indomethacin (IND) as a model drug and homopolymer polyethyleneimine (PEI)as a modelcarrier material to prove this concept. Since PEI contains primary, secondary, andtertiary amines as well as long carbon chain, while IND contains carboxyl group andhydrophobicunit, there should be electrostatic, hydrogen-bonding, and hydrophobicinteractions between PEI and IND. In addition, the hydrophobic interactions of drugmolecules may also facilitate the construction of highly efficient nanomedicines. Carefulcharacterization was performed to elucidate the physicochemical properties and IND/PEInanoassemblies and interactions between IND and PEI.By using a series ofcarboxyl-containing drugs with different structures, we proved the universality of this facileassembly approach.In addition to pharmacokinetic studies,in vivo pharmacodynamicevaluations were conducted based on acute and chronic inflammation models.Furthermore,acute toxicity of1800PEI and25000PEI was evaluated after oral administration to addressthe potential clinical applications of these assembled PEI nanomedicines.Based on above results, β-cyclodextrin (β-CD) conjugated PEI (PEI-CD) was synthesizedto improve the biocompatibility of PEI. On the one hand,the introduction of β-CDmayreduce the toxic effects of the PEI carrier material. On the other hand, it can provideadditional non-covalent force between the carrier material and drug, i.e. the host-guestinteraction between cyclodextrin cavity and hydrophobic group, which is beneficial to theformation of nanoassemblies.In order to achieve oral targeting, we used the yeast wall to entrap IND/PEInanoparticles.Targeting of inflammatory sites may be achieved by recognition of β-1,3glucan on the yeast wall with its receptor on macrophages. Similarly, the pharmacokineticand pharmacodynamic studies were carried out for IND/PEI nanoassemblies loaded yeastcapsules.Methods1. Fabrication of nanoassemblies based on PEI or PEI-CDPEI/drug nanoassemblies were prepared through a dialysis method. First, a certainproportion of drug and PEI was dissolved in a hydrophilic organic solvent, and the resultingsolution was dialyzed against deionized water at room temperature to remove the organic solvent. To fabricate drug/PEI-CD nanomedicines, a modified dialysis process was used.Briefly, an appropriate amount of drug dissolved in DMSO was gradually added into anaqueous solution of PEI-CD (10.0mg/mL) under bath sonication. The obtained mixture wasthen dialyzed against deionized water to form drug-containing nanoassemblies. Theinvolved drugs include IND, naproxen, ibuprofen, flurbiprofen, and diflunisal.2. Preparation of YS and drug loaded YSYeast shells were prepared from yeast cells by treatment using acid, alkali, and organicsolvent to remove yeast cell contents. After YS were dried, they were mixed with PEIaqueous solution to minimally hydrate the particles and incubated for2h to allow YScompletely swelled and adsorb PEI molecules. Then IND in DMSO solution was added andthe particles were resuspended by homogenization or sonication. The suspension wascentrifuged to remove DMSO and uncoated IND and PEI, the precipitate was freeze-driedto give drug-loaded YS.3. Physicochemical characterization of drug delivery systemsThe drug loading and encapsulation efficiency were detected by UV method. Themorphology of nanoparticles was observed by scanning electron microscopy (SEM),transmission electron microscopy (TEM), and atomic force microscopy (AFM). Dynamiclight scattering (DLS) was used for characterize the size distribution and surface potentialof nanoparticles.4. Characterization of the drug form in assembliesRaw IND, IND/PEI or IND/PEI-CD physical mixture, and IND/PEI or IND/PEI-CDassemblies containing various contents of IND were examined by DSC and XRDmeasurement.5. Characterization of multiple interactions between polymer and drugRaw IND, IND/PEI or IND/PEI-CD physical mixture, and IND/PEI or IND/PEI-CDassemblies containing various contents of IND were examined by FT-IR,1H and1H-1HRoesy NMR measurements.6. Synthesis and characterization of β-cyclodextrin-conjugated polyethyleneimine(PEI-CD)PEI-CD was synthesized by a nucleophilic substitution reaction between branched PEIand6-monotosyl β-CD. The structure of PEI-CD was characterized by1H-NMR and FT-IR. The molar ratio of ethyleneimine units in PEI to β-CD groups was calculated via1H-NMRspectrum7. In vitro release testsFor in vitro release study,0.5mL of IND formations was placed into dialysis tubing,which was immerged into40mL of PBS (pH7.4). At predetermined time intervals,4.0mLof release medium was withdrawn, and fresh PBS was added.To simulate release profiles under GI tract conditions, aqueous solution of HCl (pH1.2)was employed within the first two hours, and then the release medium was switched into0.01M PBS (pH7.4). IND concentration in release buffer was quantified by UV at310nm.8. Pharmacokinetic studyIND formations were orally administered via gastric gavage. Blood samples werecollected at specific time points post-dose. Plasma was obtained after10min ofcentrifugation at3000rpm. Then,100mL of acetonitrile and400mL of acetonitrile wasadded into50μL plasma, followed by votexing for1min, and the supernatant waswithdrawn after centrifugation at8000rpm for10min. After being dried under N2atmosphere,100μL of mobile phase was added for sampling. IND concentration in wasquantified by high performance liquid chromatography. The chromatographic conditionsare as follows: mobile phase, acetonitrile-6μM H3PO4(55:45, v:v); eluent rate,1.0mL/min;column temperature,40°C; detection wavelength,245nm; sample volume,20μL; column,C18reverse column (5μm×250mm).9. Study on the retention and gastrointestinal irritationIND/PEI-CD assemblies were orally administered to SD male rats at10mg/kg. Atpredetermined time points, they were euthanized and isolated segments of stomach andsmall intestine tissues were fixed. Histological sections were made and stained withhematoxylin-eosin (HE). Using the fluorescence-labeled PEI-CD, the distribution ofIND-containing assemblies in intestinal tract was detected at various time points.10. Preparation of quantum dots-yeast shell (QDs-YS)The QD-YS were prepared by the spontaneous deposition of the quantum dots (QDs)with positive surface charge. To this end, YS were mixed with a certain volume of quantumdots, and incubated at37°C for a period of time. After centrifugation, they were washedwith deionized water to remove free quantum dots. 11. Ex vivo imagingThe nude mice with acute paw edema were orally administrated with a certain amountof QDs-YS. The major organs and right hind paws were collected after24h for ex vivoimaging. After fluorescence imaging, the fluorescence intensity of various parts werequantified to evaluate the arthritis targeting capability of the yeast microcapsules.12. In vivo pharmacodynamic studyIn order to substantiate the therapeutic advantages of newly fabricated IND/PEInanomedicines, preliminary pharmacodynamic evaluation was carried out based on acarrageenan-induced acute inflammation and Freund’s adjuvant induced arthritis model inrats. The anti-inflammatory efficacy was evaluated by the swelling degree defined as thedifference between the paw volume before and after inflammation.Swelling degree=Vt-V0(mL)Where V0and Vtrepresent the paw volume before and after inflammation,respectively.13. Acute toxicity evaluation.Male Kunming mice (22-27g) were randomly assigned into4groups (n=6),including three experimental groups and one control group. In the experimental groups,mice were administered via oral gavage of PEI aqueous solution (1.0mL) at doses rangingfrom0.625,1.25, to2.5g/kg. Mice in the control group were orally administered with1.0mL of saline solution. Each day post-injection, mice were weighed and their behaviors wereobserved for any signs of illness. After14days, animals were sacrificed by cervicaldislocation after anaesthesia. Blood samples were collected for the quantification ofhematological parameters and biochemical markers relevant to liver/kidney functions.Organs including heart, liver, kidney, lung, and spleen were harvested and weighed tocalculate the organ index. Histopathological sections of organs and gastrointestinal tissueswere made and stained with HE.14. Statistical Analysis.Statistical analysis was performed by SPSS12.0using one-way ANOVA test forexperiments consisting of more than two groups, and with a two-tailed, unpaired t-test inexperiments with two groups. Statistical significance was assessed at p <0.05. Results1. Nanoassemblies based on carboxyl-containing drugs and PEI or PEI-CD werefacilely prepared by dialysis, taking advantage of multiple interactions between carriermaterials and drugs. Morphology observation combined with size determination revealedthus obtained nanoassemblies displayed spherical shape with size controllable viaprocessing parameters. In addition, these drug-containing assemblies are positively chargedaccording to the zeta-potential measurement. For IND assemblies, they exhibited drugloading content higher than71%and encapsulation efficiency>90%. The similarassembling profiles were found in other carboxyl-containing drugs like naproxen, diflunisal,ibuprofen, and flurbiprofen.2. Measurements based on FT-IR,1H and1H-1H Roesy NMR confirmed the presenceof non-covalent multiple interactions between IND and PEI or PEI-CD, includinghydrophobic, host-guest recognition, hydrogen-bonding, and electrostatic forces. DSCand XRD measurements as well as CLSM observation indicated the IND moleculesencapsulated in nanoparticles were essentially amorphous other than drug crystal.3. For IND/PEI nanomedicines, fast release of IND was achieved at pH7.4. In vitrorelease in media simulating GI conditions suggested that the drug release was largelysuppressed in acidic conditions, which can be dramatically accelerated when the releasemedium was switched into release buffer of pH7.4. In addition, the release of IND wasfaster than that of raw IND and commercial tablet. Consistent with in vitro release profiles,in vivo pharmacokinetic study in rats after oral administration indicated that the area underthe plasma concentration-time curve (AUC) of nanomedicines was120%larger than that ofraw IND. This implied that the oral bioavailability of IND was significantly enhanced byformulating into nanomedicines via molecular self-assembly. In vivo therapeutic effect washighly improved compared to the raw IND control, while no significant toxicities werefound.4. No animal death occurred at PEI doses lower than2.5g/kg, and the median lethaldose (LD50) was higher than2.5g/kg. Calculation of the organ index, measurements oftypical hematological parameters, and biomarkers relevant to liver/kidney functionsshowed no significant toxicities at PEI doses lower than2.5g/kg.5. Similarly, high drug-loading property,rapid drug release, high bioavailability, andimproved therapeutic effect were found for IND/PEI-CD nanoparticles. Moreover, PEI-CD was proved to be more biocompatible than PEI, because of the introuction of β-CD units.6. IND YS and QDs-YS were successfully prepared as demonstrated by measurementsof zeta-potential, TEM, and fluorescence microscopy. IND loading capacity was up to30%,with an entrapment efficiency of70%. Pharmacokinetic study showed that drug in IND YSwas rapidly released at pH7.4, while almost no release occurred at pH1.2.7. The bioavailability of IND in the case of the yeast system was highly improvedwhen compared with raw IND. In vivo evaluation based on models of carrageenan inducedacute paw edema and Freund’s adjuvant induced arthritis suggested that theanti-inflammatory efficacy was also highly improved after microencapsulation of IND inYS.8. Pathological study showed nanoassemblies can significantly suppress the GIirritation of IND, while improving its anti-inflammatory efficacy9. En vivo imaging indicated that oral administration of QDs-YS presented strongfluorescence intensity, which is comparable to that by i.v. injection of QDs, while thedistribution in liver was significantly deceased.Conclusions1. In this study, we proposed innovatively that polymer assemblies with high drugloading can be achieved by multiple non-covalent forces. In order to verify this hypothesis,we constructed IND/PEI nanomedicines by one-pot self-assembly mediated via multipleinteractions between drug and PEI. The physicochemical properties and drug loadingcapacity of obtained nanoassemblies were carefully characterized. Also, the self-assemblyprofiles and the versatility of this facile assembly strategy under different conditions wereinvestigated. In vitro release study showed that the new nanosystem displayed a significantpH-sensitivity. Virtually no drug release occurred in the stomach, while rapid release couldbe achieved in the intestine. The oral bioavailability and therapeutic efficacy of the loadeddrug were greatly improved through the assembly of nanosystems, while thegastrointestinal irritation and toxicity were reduced. Acute oral toxicity evaluation of PEIshowed that no side effects or adverse reactions were produced at PEI dose less than2.5g/kg.2. To further improve the biocompatibility of PEI for more safety and effectivepolymer nanoassemblies, we introduced β-cyclodextrin unit into PEI, resulting in a new polymer of β-cyclodextrin conjugated PEI (PEI-CD). The introduction of β-CD cansignificantly improve the biocompatibility of PEI, and provide additional host-guestinteractions for assembly of PEI-CD and various drugs. Similarly, efficient drug loading,rapid cargo release, high bioavailability, and improved therapeutic effect were achieved byPEI-CD/IND nanoparticles.3. After the assembled pH-responsive polymeric nanomedicines were establishedsuccessfully, we further used the yeast wall as microcapsules to entrap IND/PEInanoparticles to achieve inflammation targeting by macrophage surface receptor wich canrecognize β-1,3glucan on the yeast wall. This part of study provided a new way ofdesigning orally targeted drug delivery system through biotechnology and bioengineeringapproaches.4. In summary, we discovered a facile, convenient, cost-effective, and easily scalableone-pot assembly strategy to formulate various lipophilic therapeutics bearing carboxylgroup into nanomedicines. The present study provides new insights into the design of novelnanomedicines, and opens a new direction for the molecular self-assembly of polymerdriven by small molecules. Additionally, the application of biomimetic technology in yeastmicrocapsules can highly improve the biocompatibility of drug payload, and innovativelyachieve the oral targeting of drugs to inflammatory sites.