Organic Nanovesicular Cargoes for Sustained Drug Delivery : Synthesis , Vesicle Formation , Controlling ‘ ‘ Pearling ’ ’ States , and Terfenadine Loading / Release Studies

1 Functional Molecular Nano/Micro Solids Laboratory, School of Chemistry, University of Hyderabad, Hyderabad 500046, India 2Organic & Medicinal Chemistry, Dr. Reddy’s Institute of Life Sciences, University of Hyderabad Campus, Hyderabad 500046, India 3 Biology Department, Dr. Reddy’s Institute of Life Sciences, University of Hyderabad Campus, Hyderabad 500046, India 4 Zephase Therapeutics, Dr. Reddy’s Institute of Life Sciences, University of Hyderabad Campus, Hyderabad 500046, India


Introduction
Synthetic nano-/microscale vesicles are of great interest because they emulate complex properties of biological cell membranes [1][2][3][4].Most of the biological membranes are flexible; therefore they show multidimensional shape transformation to form zero-dimensional (0D) microspheres, one dimensional (1D) cylinders, and long flexible tubular structures.Hence mimicking the properties of naturally occurring vesicles by polymer, peptide, or dendrimer based synthetic vesicles is of great interest since they are potential tool for nanomedicine, biosensors, and biophysical and bioelectronic applications [5][6][7][8].For instance, in the budding area of nanomedicine, nontoxic vesicles act as tiny spherical containers for storage and delivery of chemicals/reagents in biochemical reactions.Nanodrug delivery systems protect drug from degradation and deliver drug at specific sites thereby increasing efficacy of treatment [5,9,10].Sometimes, multiple vesicles fuse due to the interaction of surface molecules between many vesicle membranes, a phenomenon known as "pearling, " which leads to the formation of nanotubular vesicles (NTVs) [11].Synthetic NTVs are also proposed for the transportation of chemicals between vesicles.Hence laboratory synthesis of nontoxic bioinspired molecules capable of mimicking the biological membrane is a challenging task, since they can act as life-saving drug delivery and sustained drug release cargoes.Although a plethora of block copolymer based vesicles are explored for drug delivery applications [5], vesicles from small biologically inspired designer molecules are rare.Earlier, we have found that aromatic rigid core 2,6bispyrazolylpyridine (BPP) connected with long alkyl groups forms vesicles [12].Hence for drug loading application the use of rigid molecules decorated with one of the base pairs found in DNA and RNA is a promising starting point to achieve nontoxic drug delivery systems.One of the ingredient problems with soft vesicles is their tendency to form tubes through fusion mechanism.
In this paper we present our strategy to prepare nontoxic and polymer stabilized vesicles from a novel bolaamphiphile 1, that is, two hydrogen bonding N4-acetyl cytosine molecules connected to 4,4  -positions of the BPP using two flexible octyne chains.The vesicles are loaded with terfenadine drug and successfully utilized to transport and release drugs in zebrafish larvae, which is utilized as an emerging in vivo model system (Figure 1).In this paper we present the synthesis of 1 and prevention of fusion driven nanotube formation by protecting the nanovesicles (NV) with biocompatible PEG 2000, terfenadine (TFN) drug loading, and its sustained drug releasing tendency.The nanovesicles were thoroughly characterized using scanning electron microscope (SEM), transmission electron microscope (TEM), and atomic force microscopy (AFM).

Synthesis of Compound 1.
Molecule 1 was synthesized in six steps from BPP (Scheme 1); compound 2 was synthesized as per our reported procedure [13,14].

Preparation of Nanovesicles
Synthesis of Nanovesicles Stabilized by PEG-400.Compound 1 (0.5 mg, 0.00063 mmol) was dissolved in 0.5 mL of DMSO and rapidly injected at vigorous stirring in 50 mL of deionized water, containing predissolved PEG-400 (0.6 g, 0.0003 mmol).After 5 minutes of stirring at RT, solution was drop casted on silica and analyzed by TEM, AFM, and micro-Raman spectroscopy.Remaining solution was left at moderate stirring at RT for 120 h, and sample was collected and analyzed every 24 h.PEG-2000.The experiment was performed as described above; PEG-2000 was used instead of PEG-400.

Synthesis of TFN-Loaded Stabilized Nanovesicles.
Stock solution of terfenadine (TFN) was prepared by dissolving 2 mg of TFN in 1 mL of deionized water. 1 (0.5 mg, 0.00063 mmol) was dissolved in 0.45 mL of DMSO, mixed with 0.05 mL of TFN stock solution, and rapidly injected at vigorous stirring in 50 mL of deionized water, containing predissolved PEG-2000 (0.6 g, 0.0003 mmol).After 12 h of stirring at RT solution was filtered through syringe with hydrophilic syringe filter (Millipore, pores size 0.45 m, hydrophilic, PVDF) to remove an excess of TFN and PEG-2000 and then vesicles were washed with deionized water and final volume was adjusted to 10 mL (solution V1-TFN).This solution was used to identify concentration of TFN in vesicles, stability of drugloaded vesicles and for zebrafish studies.See also schematic representation of the experiments (Figures 2 and 3).

Characterization of Nanovesicles.
Size and morphology of the nanostructures were examined by scanning electron microscope (SEM), atomic force microscopy (AFM), and transmission electron microscope (TEM).SEM studies were performed using a Philips XL30 ESEM scanning electron microscope with a beam voltage of 20 kV.FESEM measurements were performed on a Hitachi S-4500 SE/N instrument operating at 20 kV.TEM studies were done using Tecnai G2 FEI F12 transmission electron microscope (TEM) at an accelerating voltage of 200 kV.Carbon coated TEM grids (200 mesh type-B) were purchased from TED PELLA INC.USA.Atomic force microscopy (AFM) imaging was carried out on NT-MDT model solver Pro M microscope using a class 2R laser of 650 nm wavelength having maximum output of 1 mW.All calculations and image processing were carried out by a software NOVA 1.0.26.1443 provided by the manufacturer.The images were recorded in a semicontact mode using a noncontact super sharp silicon cantilever (NSG 10 DLC) diamond like carbon (DLC) tip purchased from NT-MDT, Moscow.The dimension of the tip is as follows: cantilever length = 100 (±5) m, cantilever width 35 (±5) m, and cantilever thickness = 1.7-2.3m; resonate frequency = 190-325 kHz; force constant = 5.5-22.5 N/m; chip size = 3.6 × 1.6 × 0.4 mm; reflective side = Au; tip height = 10-20 m; DLC tip curvature radius = 1-3 nm.

AFM, LCFM, SEM, FESEM, and TEM-Samples Preparation.
Two drops of the sample solution were drop casted on a clean glass slide by a capillary for AFM studies.Two drops of the sample solution were drop casted on a carbon coated copper grid (200 Mesh) with different time intervals for TEM studies of nanostructures.
Stability of Drug-Loaded Vesicles Study.Solution V1-TFN was left at moderate stirring at RT for 120 h, and sample was collected and analyzed every 24 h to identify time point when nanovesicles disintegration will begin.

Determination of Concentrations of TFN in Surrounding
Media (Figure 3).TFN concentration in solution resulting from leakage from the nanovesicles was investigated at different time points following procedure described below: 1 mL of solution V1-TFN was collected, filtered through hydrophilic   syringe filter, and washed 1 time with 1 mL of deionized water.TFN was extracted from filtrate with diisopropyl ether; solvent was evaporated and remaining TFN was redissolved in 2 mL of acetonitrile.Concentration of TFN in this solution was measured according to modified published procedure [15,16].TFN solution in acetonitrile (2 mL) was mixed with 2 mL of DDQ solution (5 mg/mL).The colored species were generated and absorbance intensity was measured immediately at 457 nm and compared with calibration graphs, which were constructed by plotting the absorbance of the formed CT complexes versus the final concentration of the drug (g/mL).This study was repeated for the samples collected every 24 h up to 120 h.3).Concentrations of TFN inside vesicles at different time points were investigated following procedure described below: 1 mL of solution V1-TFN was collected, filtered through hydrophilic syringe filter, and washed 1 time with 1 mL of deionized water to remove excess of the drug and PEG-2000, and residue was redispersed in 2 mL of water and subjected to sonication for 30 min at RT to break down all vesicles and release TFN.The ability of ultrasound to induce localized and controlled drug release from liposomes, synthetic diblock copolymers, and polymersomes is well known phenomena [17,18].Next, TFN was extracted from this solution with large volume of diisopropyl ether; solvent was evaporated and remaining TFN was redissolved in 2 mL of acetonitrile.Replacement of solvent with ACN was necessary because it showed super priority over other solvents for the formation of CT complex, at the same time facilitating charge-transfer from donor to acceptor.Absorbance intensity of TFN solution in ACN was then measured at selected absorption maximum (457 nm) and obtained value was compared with values of calibration curve obtained using solutions of TFN:DDQ complex of known concentrations (see also Figures 9 and 10).The concentration of TFN calculated applying Beer-Lambert law was 37.5 M, which allowed us to find out drug loading efficiency using the following equation:

Determination of Concentrations of TFN inside Nanovesicles (Figure
where  0 is an initial concentration of TFN in solution, whereas  1 is concentration of TFN in vesicles after loading.This study was repeated for the samples collected every 24 h up to 120 h.

Zebrafish Studies
Zebrafish Teratogenicity Assessment.Zebrafish maintenance, breeding, and embryo collection was carried out by methods described previously [19,20].Briefly, wild-type adult zebrafish (5-6 months old) were maintained at 28 ∘ C under a 14 : 10 h light and dark cycle.Fish were allowed to breed (in a ratio of 2 females : 3 males) in the morning under the stimulation of light.Embryos were collected into petridish containing embryo medium and incubated at 28 ∘ C temperature.The teratogenicity study was performed using 6 embryos that were incubated with vehicle, positive control, and various concentrations of test compound 1 until 5 dpf.The embryos were then anesthetized using 0.008% tricaine solution and various phenotypic parameters were evaluated; namely, body shape, somites, notochord, tail, intestine, fins, brain, upper jaw, heart, lower jaw, liver, and swim bladder were observed for defects in morphology [21].Each embryo was scored based on their level of toxicity from 5 being nontoxic and 0.5 being highly toxic.The scoring was conducted by blinded observers and was done according to the procedures described earlier [22].
Protocol for Zebrafish Treatment with TFN-Loaded Nanovesicles.For TFN-NV assessment, 3 dpf (days post fertilization) larvae were distributed in 24-well plate along with 250 L of 0.1% DMSO solution.Each well contained 6 embryos with various treatments and the larvae were observed at time points of 24 h, 48 h, and 72 h.Embryos were then treated with different concentrations of TFN loaded microvesicles (5 M, 10 M, 20 M, and 40 M).TFN alone at 20 M concentration was used as a positive control and 0.1% DMSO was used as a negative control.The embryos were incubated in 24-well plate and atrial and ventricular rates were observed at different time points after treatment.All embryos were observed for mortality arising out of effect of TFN.Results of zebrafish embryo toxicity study are summarized in Table 1.

Synthesis and Characterization of Compound 1.
Molecule 1 was synthesized in six steps from BPP (Scheme 1).Iodination of the 4,4  -positions of the BPP to synthesize 2 was carried out as per our reported procedure [13,14].Attachment of "linker molecule" decyn-1-ol to 2 was achieved by adopting Sonogashira cross-coupling conditions to get 3 in good

Investigation of Nanostructures Formation.
To study the aggregation driven nanostructure formation, a solution containing 2 mg of compound 1 in DMF (12 mL)/water (6 mL) was prepared and stabilized for an hour.Drop casting this solution on silica followed by SEM analysis showed the formation of clear fibrous networks of several microns long (Figure 4).To unravel the mechanism of formation of nanofibrous network assemblies, the same solution of 1 was drop casted immediately without stabilization on a carbon coated TEM grid.Interestingly, the TEM micrographs showed the presence of tiny vesicles of size in the range of ∼200 nm (Figure 5(a)) and some aggregated pearl-chain linked vesicles of size in the range of ∼150 nm (Figure 5(b)).
Additionally the generation of 1D pearl chain-like assembly by the fusion of vesicles indicates the formation of nanotubular network (Figures 5(c) and 5(d)).To further verify the fusion of spherical vesicles into nanotubes on silica surface AFM studies were undertaken.The AFM topography clearly revealed the presence of nanovesicles (NVs) of sizes ∼200 nm (Figures 6(a) and 6(b)).Interestingly, some of the samples showed spontaneous formation of pearl-chain like vesicular assemblies or nanotube integrated vesicles, pointing out to vesicles fusion.For every 1 m chain length ca. 5 fused vesicles were found, which is matching with the average diameter of the vesicles (Figures 6(c) and 6(d)).Continuous AFM measurement revealed the evolution of smooth nanotubular structures of diameter ∼160 nm from the pearl-chain like vesicular assemblies due to the molecular reorganization (Figures 6(e) and 6(f)).The finally formed tubes diameter (∼160 nm) is less than the initial vesicle diameter of a single isolated vesicle (∼200 nm) due to the stretching of the fused vesicles during the 1D tube formation.The diameter of the tube is determined by the balance between surface tension and tube bending rigidity, preventing membrane collapse.The outer surface of the tubes is very smooth with a roughness smaller than <1 nm (Figures 6(g) and 6(h)).This result demonstrated the flexible nature of the presented synthetic membrane composed of 1.Another interesting finding of the nanotubes is their tendency to form various types of networkjunctions through self-organization (Figure 6(e), insert).

Control over Nanostructure Using Stabilizing Agents.
To demonstrate application of NVs for drug delivery, in the first step, different conditions were screened to stabilize vesicles in solution state and prevent formation of tubular structures.We intended to use polyethylene glycol (PEG) as stabilizing agent since it is biologically neutral and furthermore FDAapproved biocompatible polymer well suitable for pharmaceutical applications [23].PEG-400 and PEG-2000 were employed for stabilization of nanovesicles.In a typical procedure a DMSO solution of compound 1 (0.5 mg in 0.5 mL) was rapidly injected into a deionized water containing PEG (0.6 g in 50 mL) under vigorous stirring for 5 minutes at RT. NVs PEGylated with polymer of lower molecular weight (PEG-400) did not demonstrate high stability; however the use of PEG-2000 stabilized the vesicles up to 108 h.Even at that time point a significant amount of vesicles retained their structure, although their surface was not as smooth as it was at 24 and 48 hours.The size distribution of the NVs was in the range of <60 nm to 130 nm; an insignificant amount of bigger vesicles (∼250 nm) was also observed (Figures 7(a) and 7(b)).

Toxicity Investigation of Stabilized Nanovesicles Using
Zebrafish Teratogenicity Assay.For NVs based drug delivery studies, zebrafish was selected as a model organism since it is easy to maintain, is relatively inexpensive, gives quick results and, is a good predictor of efficacy and safety of chemicals [24,25].In order to explore toxicity of compound 1 zebrafish teratogenicity assay was performed.Treatment of 24 hpf (hours after fertilization) embryos with solution of 1 at different concentrations and study of their growth and development for 5 days show no abnormalities (Figure 8 and Table 1).Figure 5 represents mean lesion score of all parameters different treatment groups.This experiment confirmed the nontoxic nature of 1 at all tested concentrations.

Terfenadine Loading into Nanovesicles.
For drug delivery study, a well-known antihistamine drug terfenadine (TFN) was selected as a model drug.TFN is also known to affect heartbeat regularity in zebrafish, leading to bradycardia, atrioventricular blockage and ultimately death in less than 12 h [26].Selection of this drug facilitates easy monitoring of     its effect on zebrafish by observing the endpoint of mortality at different time points after treatment.As the nontoxic nature of 1 in the zebrafish teratogenicity assay is already tested, any mortality of zebrafish is only attributable to the effect of TFN.Additionally, this procedure is simple, rapid, and noninvasive method of detection of drug action.To prepare drug-encapsulated nanovesicles, TFN and compound 1 were dissolved in DMSO and rapidly injected in solution of PEG-2000 in water with vigorous stirring.The drugencapsulated nanovesicles solution was left for 12 h for ageing, resultant solution was subsecuently subjected to study the TFN-loading efficiency and nanovesicles stability.Determination of TFN concentration in biological samples is a challenging analytical task.Straightforward application of spectroscopic methods (for instance, measuring intensity of UV absorption and fluorescent emission) is not applicable as there is no correlation between obtained values and drug content.The majority of current reported methods (e.g., HPLC or NMR) lack sensitivity to measure the drug level at concentrations below 10 g/mL [15,16].Hence spectrophotometric determination method based on the charge-transfer (CT) reactions of TFN with 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) was opted [15].This technique is one of the most simple, rapid, and accurate methodologies developed till now.In this method, TFN acts as n-electron donor which forms a highly colored charge transfer (CT) complex upon spontaneous reaction with strong -acceptor DDQ.In order to implement this procedure and find out initial concentration of the drug loaded in vesicles, we have developed an in-house protocol (Figures 3, 9, and 10).Using this method, the calculated drug loading efficiency into the vesicles was ∼35 ± 1%.Further, in order to prove that there is no drug leakage from vesicles into surrounding media over time, we performed stability control experiment: drug-loaded vesicles were gently stirred at RT up to 120 h, collecting sample every 24 h and detecting concentration of TFN inside vesicles and in mother liquid (solution collected prior to sonication) at every time-point as described in Figure 3.The experiment revealed that there is no major leakage observed up to 72 h (Figure 11) and concentration of TFN in media remains below 10 M, which is zebrafish sensitivity threshold (fish development and heartbeat are not affected below this concentration of TFN).Subsequently, drug burst release was observed after 72 h, when TFN concentration in mother liquid suddenly increased and crossed 10 M.Simultaneously, TFN concentration inside  NVs decreased dramatically after 72 h.We assumed that slow dissolution of PEG coating resulted in deformation of NVs and sudden drug release (Figure 6, cartoon).These data correlate well with TEM data recorded for vesicles at different time points.Figures 12(a  well-structured defined shape (Figure 12(a)), while at 72 h we observe degradation of NVs and decrease of NVs amount (Figure 12(b)).This explains well drug burst release after 72 h, an effect commonly seen in case of nano-/microsphere or nano-/microvesicles systems [27,28].

Sustained Drug Delivery of TFN-Loaded Nanovesicles in
Zebrafish Larvae.Zebrafish larvae were treated with TFNloaded nanovesicles (TFN-NV) at four different concentrations (5 M, 10 M, 20 M, and 40 M of TFN) and delivery for the first time.We demonstrated that TFN release in vivo at desired time point can be controlled by altering nanovesicles concentration.The mechanism of the drugrelease in vivo is a subject of separate study.Nevertheless, current experiment undoubtfully demonstrated possibility of nanovesicles application as sustained-release drug delivery system.

Figure 1 :
Figure 1: Illustration of the nanovesicles formation from bolaamphiphile 1 and their subsequent self-assembly into pearl chains and finally nanotubes.The nontoxic vesicles fusion (pearling) is prevented by biocompatible polymer coating and used in terfenadine (TFN) drug delivery in a sustained manner (72 h) in zebrafish model system.

Figure 3 :
Figure 3: Determination of TFN concentration inside vesicles and in surrounding media (repeated at different time points).

Figure 4 :
Figure 4: SEM micrographs of nanofibrous network assemblies formed from compound 1 in DMF/water.

Figure 6 :
Figure 6: AFM mechanistic analysis of the formation of nanotubes from nanovesicles: (a) randomly distributed vesicles, (c) fusion of vesicles to form pearl-chain like structures, (e) nanotubes, and (g) roughness of a tube.The corresponding profile analyses are shown in (b), (d), (f), and (h), respectively.

Figure 9 :Figure 10 :
Figure 9: (a) Absorbance of TFN at different concentrations in acetonitrile as complex with DDQ (DDQ concentration was maintained constant: 22 mM for all samples) and (b) calibration graphs derived from (a) according to Beer-Lambert law.
) and 12(b) represent TEM images of the vesicles at 24 and 72 h, respectively.At 24 h NVs have

Figure 11 :
Figure 11: TFN concentration change inside vesicles and in mother liquid at different time points.Cartoon represents slow deformation of vesicles and TFN "burst" drug release after 72 h.

Table 1 :
Results of zebrafish embryo toxicity study with toxicological indices and major organs/systems affected in positive control and at MTC in test compounds (-: no effect, x: slightly toxic, xx: moderately toxic, and xxx: severely toxic).