Intermolecular Interactions between Coencapsulated Drugs Inhibit

Subscriber access provided by PEPPERDINE UNIV

Article

Intermolecular interactions between co-encapsulated drugs inhibit drug crystallization and enhance colloidal stability of polymeric micelles Ling Zhang, Hanzi Sun, Zhen Chen, Zhengsheng Liu, Niu Huang, and Feng Qian Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.7b00591 • Publication Date (Web): 22 Aug 2017 Downloaded from http://pubs.acs.org on August 23, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Molecular Pharmaceutics is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Molecular Pharmaceutics

Intermolecular interactions between co-encapsulated drugs inhibit drug crystallization and enhance colloidal stability of polymeric micelles

Ling Zhang 1 †, Hanzi Sun 2 †, Chen Zhen 1, Zhengsheng Liu 1, Niu Huang 2 *, and Feng Qian 1 *

1

School of Pharmaceutical Sciences and Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Tsinghua University, Beijing 100084, P.R.China 2

National Institute of Biological Sciences, Beijing 100084, P.R.China

Manuscript for Molecular Pharmaceutics

* To whom correspondence should be address: Niu Huang, ([email protected]) and Feng Qian ([email protected]) †These authors contributed equally to this work.

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT: Novel “pairs” of drugs possessing pharmacological synergies could be encapsulated into polymeric micelles and exert superb therapeutic effects in vivo upon intravenous administration, with the prerequisite that the micelles remain stable. NADP(H) quinone oxidoreductase 1 (NQO1) inhibitors, such as β-lapachone (LPC) and tanshinone IIA (THA) are structurally and pharmacologically similar molecules that are both poorly water soluble, crystallize extremely fast, and demonstrate synergistic anticancer effect when used together with paclitaxel (PTX). However, when co-encapsulated with PTX in poly (ethylene glycol)-b-poly (D, L-lactic acid) (PEG-PLA) micelles, only PTX/LPC but not PTX/THA pair yield satisfactory colloidal stability. To reveal the molecular mechanism contributing to the colloidal stability of the co-encapsulated micelles, we investigated the molecular interactions of PTX/LPC and PTX/THA, through both experimental methods (crystallization kinetics,

13

C-NMR) and molecular dynamic

simulation. We observed that PTX was capable of inhibiting LPC but not THA crystallization both in aqueous environment and in solid state, which could be attributed to the strong hetero-intermolecular interactions (π-π, H-bonding) between LPC and PTX, which disrupted the homo-intermolecular interactions between LPC molecules and thus formed a favorable miscible binary system. In comparison, the lack of a strong PTX/THA interaction left the strong THA/THA stacking interaction undisturbed and the fast THA crystallization tendency unrestrained. We conclude that the intermolecular interactions, i.e., the “pharmaceutical synergy”, between the co-encapsulated drugs critically control the colloidal stability of polymeric micelles, therefore shall be evaluated when design co-encapsulated drug delivery systems for optimal therapeutic benefits.

KEYWORDS: β-Lapachone, tanshinone IIA, paclitaxel, polymer micelles, crystallization, molecular dynamic simulation, synergistic anticancer therapy


Page 2 of 23

Page 3 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Abstract Figure



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 23

1. INTRODUCTION Polymeric micelle is one of the most widely studied nano-carriers for solubility enhancement of poorly water soluble drugs, through drug encapsulation within their hydrophobic cores

1-2

.

Cancer-targeted drug delivery using such self-assembled nano-carriers is envisioned to be able to optimize the systemic pharmacokinetics, enhance drug accumulation at the site of action (e.g., tumor), manipulate intracellular trafficking, etc.1-5, with an indispensable prerequisite that the drug encapsulated polymeric micelles remain intact in aqueous formulation before administration, and in blood circulation after administration 6-7. Not all drugs are inherently suitable to be encapsulated in the commonly used biocompatible and biodegradable polymeric micelles with satisfactory colloidal stability. Loading drugs with high crystallization propensity within commonly used polymeric micelles, such as those based on different grades of poly (ethylene glycol)-b-poly (D,L-lactic acid) (PEG-PLA) 1, 8-9, is particularly challenging possibly due to the lack of sufficient miscibility and intermolecular interactions between the drug molecules and the hydrophobic PLA segments 10. Chemical modifications on the hydrophobic polymer segments to introduce interfacial drug interactive motifs 10-12, or on the drug molecule to allow complexation with the polymer 5, or to covalently link the drug and polymer

13

are certainly common strategies to improve the drug encapsulation efficiency. However, concerns such as insufficient drug loading due to certain stoichiometric requirement between drug and polymer, potential toxicity of new chemical entities and/or impurities, impaired biodegradability and biocompatibility, etc., could all hinder further clinical evaluation and application 14-15. Non-covalent approaches to improve drug encapsulation efficiency without facing the aforementioned complications, could serve as an alternative strategy to enhance colloidal stability of polymeric micelles for drug delivery

14, 16-17

. Co-encapsulation of more than one drug into the

hydrophobic core of polymeric micelles was previously observed to be able to greatly improve drug loading and micelle stability, likely attributed to molecular interactions between co-loaded drugs (π-π interaction or hydrogen bonding, etc.)

16

. Meanwhile, combination anticancer drug

therapies are intensively investigated and routinely used in the clinic, to achieve improved therapeutic outcomes with better disease control, lower side effects and less drug resistance

18-22

.

With the research advances in precision medicine and drug repurposing, drugs with novel


Page 5 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


biological mechanisms have been identified, which often demonstrate pharmacological synergies with other existing therapeutic agents

19, 23

. Therefore, we are interested in identifying and

evaluating novel “pairs” of drugs that possess both pharmacological and pharmaceutical synergies, which could form stable polymeric micelles when co-encapsulated within micelles and subsequently exert superb therapeutic effects in vivo upon intravenous administration. β-Lapachone (LPC, Figure 1A) has drawn continuous research interests for more than 30 years, as a substrate of NADP(H): quinone oxidoreductase 1 (NQO-1), a potential anticancer target that overexpresses on a variety of cancers 24-27. However, the poor water solubility and the extremely fast crystallization propensity of LPC largely hindered it from being formulated with minimal excipient-related toxicity for clinical evaluation. Previously, we have demonstrated that through co-encapsulation with paclitaxel (PTX, Figure 1C), a first line chemotherapy drug used in clinic for NSCLC, breast cancer, PDAC, etc., the drug encapsulation efficiency of LPC in PEG-PLA micelles increased ~10 fold

16

. At the same time, combinations of PTX/NQO-1

substrates have demonstrated synergistic anticancer effects against cancers including ovarian, breast, prostate, melanoma, lung, colon, pancreatic cancer and retinoblastoma 28-29. In this study, we aim to investigate the molecular interactions between PTX and two structurally related NQO-1 substrates, LPC and tanshinone IIA 30-31 (THA, Figure 1B), to reveal the mechanisms dictating the colloidal stability of polymeric micelles co-encapsulating with multiple drugs. Both LPC and THA are poorly water soluble quinine compounds with extremely fast crystallization tendency, representing a class of drug molecules that are particularly challenging to be loaded inside the hydrophobic cores of PEG-PLA micelles. The intermolecular interactions between the two pairs of PTX/LPC and PTX/THA were compared side-by-side using both experimental and computational approaches.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. Chemical structures of (A) β-lapachone (LPC), (B) tanshinoneⅡA (THA), (C) paclitaxel (PTX) and (D) poly (ethylene glycol)-b-poly (D, L-lactic acid) (PEG-PLA). THA and LPC have common quinone structures (highlighted in red in A and B).

2. EXPERIMENTAL SECTION

2.1 Materials β-Lapachone (MW: 242.27 Da, Tm: 154-162ºC, LogP: 2.5, aqueous solubility: ~38µg/mL) was synthesized as previously described 16, 32. Tanshinone IIA (MW: 294.33 Da, Tm: 209-210ºC, LogP: 5.4, aqueous solubility: ~0.018µg/mL) and paclitaxel (MW: 853.91 Da, Tm: 213-216ºC, LogP: 3.0, aqueous solubility: ~0.7µg/mL) were obtained from Ouhe Chemical Co., Ltd (Beijing, China). PEG5k-PLA5k block copolymer (MW: 10, 000 Da) was purchased from Daigang Biotechnology Co., Ltd (Jinan, China). All organic solvents were of analytical grade.

2.2 Preparation and characterization of drug encapsulated PEG-PLA micelles PEG-PLA polymeric micelles loaded with LPC (20 mg), THA (20 mg), LPC (20 mg)/PTX (60 mg), or THA (20 mg)/PTX (60 mg) were prepared by a film hydration method as previously described

16, 33

. Certain amount of drug(s) and 300 mg PEG-PLA were dissolved in an

eggplant-shaped bottle with 5 mL acetonitrile (theoretical drug loading listed in Table1). The


Page 6 of 23

Page 7 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


solvent was removed by a rotary evaporator at 60ºC to form a thin film. Then, 10 mL normal saline at 60ºC was added to hydrate the film following sonication for 5 min. Particle sizes of the resulted micelle solutions were characterized by dynamic light scattering (Malvern Instruments Inc., U.K.) analysis. There was no filtering treatment on the above prepared micelle solutions.

Table 1. PEG-PLA micelles with LPC, THA, PTX/LPC, or PTX/THA encapsulated Micelle systems (abbreviation) 100L 25L75P 100T 25T75P

LPC(THA)/PTX weight ratio

Theoretical LPC or THA Theoretical PTX weight Theoretical total drugs weight percentage (wt%) percentage (wt%) weight percentage (wt%)

N/A 1/3 N/A 1/3

5.26 5.26 5.26 5.26

N/A 15.79 N/A 15.79

5.26 21.05 5.26 21.05

2.3 Crystallization of LPC and THA with/without the coexistence of PTX observed by polarized optical microscope (POM) To study the crystallization inhibition effect of PTX on either LPC or THA in solid state, thin film samples were prepared using a spin-coating method. Briefly, 5 mg LPC (or THA) with or without the coexistence of 15 mg PTX was dissolved in 200 µL dichloromethane (CH2Cl2), and the solution was sonicated for 5 min. The obtained solution was then spin-coated on the glass substrates at 3000 rpm for 60 s. The crystallization rates and morphologies of the coated samples were monitored under polarized optical microscopy (Zeiss Axio Imager A2m microscope, Germany) at room temperature immediately after preparation and after three months storage (room temperature, stored in desiccator with P2O5).

2.4 Molecular interactions between THA and PTX investigated by one dimensional 13C-nuclear magnetic resonance (13C-NMR) NMR was used to study the molecular interactions between THA and PTX in solution, similarly as our previous report on LPC and PTX molecular interactions 16. NMR samples were prepared by dissolving THA, PTX, or their mixtures at weight ratio of 25:75 (25T75P) in deuterated chloroform (CDCl3), respectively. THA concentration in all solutions was 40 mg/mL. Samples were then analyzed by a 100MHz NMR spectrometer (13C-NMR, Bruker Biospin GmbH, Rheinstetten, Germany) at 25 ºC.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 23

2.5 Effect of PTX on the supersaturation kinetics of LPC and THA in aqueous solution We compared the supersaturation kinetics of LPC and THA in aqueous medium with or without the coexistence of PTX, using previously reported method

16

. First, LPC, THA, PTX/LPC or

PTX/THA mixture at weight ratio of 50:50 were dissolved in methanol, with 5 mg/mL LPC or THA concentration. Then, 2 mL stock solution was dispersed into 18 mL PBS (PH: 7.4) to generate LPC or THA supersaturation on a shaker at 37ºC. After 10, 15, 20, 25, and 30 min, 0.5 mL of solution was withdraw and filtered through 0.45 µm nylon filters to remove precipitates of drug aggregates in PBS solution. The drug concentration of filtered solutions was determined by HPLC/UV-vis. All experiments were performed in triplicate. The degree of supersaturation was evaluated using the supersaturation ratio defined in the following equation:

ܵ=

‫ܥ‬′ ‫ܥ‬

Where ܵ is the supersaturation ratio, ‫ܥ‬′ and ‫ ܥ‬is the LPC or THA concentration with and without the coexistence of PTX in PBS solution at each time point, respectively.

2.6 Molecular dynamics (MD) simulation All molecular simulations were performed using the Gromacs program package version 4.5.6 with the General Amber Force Field (GAFF)

35

34

parameters assigned for solute molecules and

TIP3P water model for solvent. The initial 3D structures of LPC, THA, and PTX were obtained from the Cambridge Structural Database (CSD)

36

. And each simulation system was built as a

40×40×40 Å box with periodic boundary conditions (PBC). The solute molecules were randomly placed into the simulation box and solvated by the solvent molecules. All calculations applied an atom-based truncation scheme, updated heuristically with a list cutoff of 14 Å, a nonbond cutoff of 12 Å, and the Lennard-Jones (LJ) smoothing function initiated at 10 Å. Long-range electrostatic interactions were computed by the Particle Mesh Ewald (PME)

37

and LJ interactions were

calculated using force shift algorithm. To equilibrate solvent molecules around the solute, the system was minimized using 5000 steps of a steepest descent algorithm, followed by 500 ps of NVT MD simulation and a subsequent 500 ps of NPT simulation. The 5 ns NPT MD production simulation was performed at a temperature of 300 K via the v-rescale temperature coupling


Page 9 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


scheme and Parrinello-Rahman pressure coupling scheme 38. The LINCS algorithm 39 was applied to constrain all bonds involving hydrogen atoms with a time step of 2 fs. The structural and energetics analysis was performed for the last 4 ns of the production run. For each mixture system (Table 3), 10 independent simulation systems were built up using different random seeds, and the simulation snapshots of these 10 systems were combined for further analysis.

2.7 Geometry of molecule stacking interaction To qualify the formation of the stacking interactions between LPC/LPC or THA/THA molecule pairs, we defined three structural parameters to describe the stacking interaction: the angle α between two molecular planes; the distance r1 from the center of one molecule to the plane of the other molecule; and the distance r2 between the centers of molecules (Figure 2). Two molecules form intermolecular stacking interaction by satisfying the criteria of cos(α) > 0.95, 2.5 Å < r1 < 4.5 Å and 3 Å < r2 < 6 Å. Similarly, the stacking interaction between PTX/LPC and PTX/THA was defined as: the angle α between each phenyl ring plane and LPC or THA plane; the distance r1 between the center of each phenyl ring to the plane of the LPC or THA; and the distance r2 between the center of each phenyl ring to the center of the LPC or THA. The stacking interaction forms between one PTX phenyl ring with LPC or THA by satisfying the criteria of cos(α) > 0.95, 3 Å < r1 1µm) while blue columns represents the nano-sized particles within 100nm. (B) Visual appearance of micelle solutions at 0h. All red-line circled samples were cloudy with significant drug precipitates thus the font “Micelle” written underneath the petri-dish was invisible; while blue-line circled samples remained as clear solution so that the font “Micelle” was clearly visible. (C) The particle size distribution of 25L75P micelles at 48h. All micelle solutions were measured by dynamic light scattering (DLS) without filtration.

We prepared PEG-PLA micelles co-encapsulated with two drug pairs, i.e., PTX/LPC and


Page 10 of 23

Page 11 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


PTX/THA, with the same theoretical drug loading density of 21.05% (Table 1). The theoretical apparent solubility of LPC (or THA), PTX and PEG-PLA was 2 mg/mL, 6 mg/mL and 30 mg/mL, respectively, in all micelle solutions. We observed that stable PEG-PLA micelles with high drug loading could not be obtained with either LPC (100L) or THA (100T) alone, because of their fast crystallization rates. The 100L and 100T micelle solutions appeared to be cloudy with red, crystalline drug precipitates (Figure 3A, B). However, when combined with PTX at weight ratio of 25:75 (25L75P), the same amount of LPC was able to be completely encapsulated within the micelles, and form a clear and orange color solution, which is consistent with our previous report

16

. Without any filtration, the 25L75P

micelles exhibited a narrow and monodispersive size distribution with a mean particle size of 60-70nm, which is similar with PTX (100P) and blank micelles (Figure 3A). Note that solubilities of LPC (~38µg/mL), THA (~0.018µg/mL), or PTX (~0.7µg/mL) in water without micelle encapsulation are all much lower than the ~mg/mL apparent drug solubility obtained in some systems (e.g., 25L/75P), we conclude that in those cases almost all drugs were encapsulated within the hydrophobic core of the micelles. Furthermore, no significant change on particle size and size distribution was observed over 48h (Figure 3C), indicating the superior colloidal stability of 25L75P micelles. In comparison, PTX/THA co-encapsulation (i.e., the 25T75P micelle) did not increase the encapsulation efficiency and concentration of THA similarly as the 25L75P micelle. Formation of large size (>1µm) aggregates and drug precipitates were observed immediately after the preparation of 25T75P micelle similarly as 100T (Figure 3A, B), indicating that PTX had little effect on inhibiting the crystallization of THA or enhancing the colloidal stability of THA loaded micelles through co-encapsulation. It’s worth noting that, the hydrophobic PLA segments co-exist with either drug pairs within the hydrophobic core. However, we conclude that any molecular interactions between the PLA segments with the drugs are unlikely the critical mechanism behind the different stabilities between PTX/LPC and PTX/THA col-loaded micelles, since PEG segments are the common components in all micelles. Therefore, in the following investigation, we focus on the interactions between the co-loaded drugs, which also simplified the computational comparison research.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

3.2 Crystallization of LPC or THA in the presence of PTX Despite the structural similarities between LPC and THA, PTX appeared to be only capable of inhibiting the crystallization of LPC, but not THA, thus enhancing the micelle stability through co-encapsulation. We further confirmed the different crystallization inhibition capability of PTX against LPC and THA, both in aqueous environment and in dry state. In aqueous environment, the supersaturation ratios (S) of LPC and THA (see method section) in the presence of PTX were measured to compare the effect of supersaturation enhancement of both drugs by PTX. In this study, S=1 indicates no enhancement of supersaturation, while S>1 indicates improved supersaturation with the co-existence of another molecule (i.e, PTX). As shown in Figure 4, PTX can significantly improve the supersaturation of LPC in PBS buffer (SLPC>1) over 30 mins, in agreement with our previous observation

16

, while there was no detectable

supersaturation of THA with the coexistence of PTX (STHA≈1). This result clearly demonstrated that PTX was unable to prevent THA crystallization from supersaturation in aqueous solution.

Figure 4. Supersaturation kinetics of LPC and THA in the presence of PTX, measured by superstation ratio, S (n=3).

We also compared the crystallization inhibition effect of PTX on LPC and THA in the solid state, by visual observation of spin-coated films of LPC, THA, PTX/LPC, and PTX/THA under POM. Both LPC and THA crystallizes extremely fast when spin-coated as pure amorphous drug films, and complete crystallization of either drug was evident immediately after preparation (Figure 5A-B). With the presence of 75% PTX, the crystallization of initially amorphous 25L75P and


Page 12 of 23

Page 13 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


25T75P films was slowed down significantly, presumably due to both dilution and decreased molecular mobility with the introducing of large amount of high Tg (152 ºC) PTX 40. After three months (RT/desiccator), the 25L75P film still remained crystal free, while crystal formation was observed in the 25T75P film (Figure 5C-D). These findings again validated that PTX was able to inhibit the crystallization of LPC much more efficiently than THA in solid state.

Figure 5. Crystallization of spin-coated films observed by polarized optical microscope (POM). (A) Pure LPC (100L) film at 0h. (B) Pure THA (100T) film at 0h. (C) LPC-PTX (25L75P) film and (D) THA-PTX (25T75P) film after three months storage in RT/ desiccator. (Scar bar: 100µm)

3.3 Comparison of molecular interactions between PTX/LPC and PTX/THA by

13

C-NMR

analysis We have previously reported that, the coexistence of LPC significantly influenced the chemical shifts on the benzene rings (ring1, ring2, ring3, highlighted in Figure 1C), -NH, and –OH groups of PTX (Table 2) by 13C-NMR analysis 16. These results suggested the formation of considerable hydrogen bonding and/or π-π interaction between LPC and PTX, which might have contributed to crystallization inhibition effects of PTX towards LPC. In comparison with the PTX/LPC system, the coexistence of THA caused the PTX chemical shifts of C1’, C10 Aco CO, N3’ Ph C=O (ring 1), C3’NPh-i (ring 1), C3’NPh-p (ring 1), C3’Ph-m (ring 2), C3’NPh-m (ring 1), C3’Ph-p (ring 2), C2Ph-i (ring 3), C4, C7, C13 to shift to a less extent towards upfield; and that of C12, C3’Ph-i



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 23

(ring 2), C2’,C3’ to a less extent towards downfield. Meanwhile, the presence of PTX did not change any chemical shift of THA (data did not show), indicating the absence of any specific molecular interaction between THA and PTX.

Table 2. The change of 13C NMR chemical shifts (∆δ 13C ) of PTX in coexistence of LPC or THA in CDCl3 solution comparing with the pure drug (PTX) solution at the same concentration (data of PTX chemical shifts in PTX/LPC mixture cited from reference 16). ∆δ13C of PTX Carbon number PTX/LPC 16

PTX/THA

C1'

-0.13

-0.04

C10 AcO CO

-0.14

-0.03

N3'Ph C=O

-0.21

-0.05

C12

0.15

0.08

C3'Ph-i

0.15

0.07

C3'NPh-i

-0.33

-0.09

C3'NPh-p

-0.28

-0.07

C3'Ph-m C3'NPh-m

-0.28

-0.08

C3'Ph-p

-0.26

-0.07

C2Ph-i

-0.33

-0.09

C4

-0.24

-0.06

C2'

0.12

0.05

C7

-0.29

-0.07

C13

-0.37

-0.02

C3'

0.18

0.06

*The chemical shift of 77.16 ppm from CDCl3 as reference signal.

The NMR results indicated that LPC but not THA is likely to be more miscible with PTX, which


Page 15 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


may be due to the existence of favorable intermolecular interactions between LPC and PTX. Without any substantial molecular interactions between THA and PTX, the crystallization of THA would likely remain un-altered with the co-encapsulation of PTX.

3.4 Computational simulations of the microscopic interactions occurring in LPC, THA, PTX and their mixture systems We further carried out MD simulations on the solvated systems of PTX, LPC, THA and their mixtures to investigate the microscopic molecular interactions, which led to total of 9 different simulation systems (Table 3). For each system, we extracted total of 10,000 frames from 10 random simulations for structural and energetics analysis.

Table 3. PTX, LPC, THA and their mixture systems for MD simulation studies. PTX (number)

LPC (number)

THA (number)

PTX concentration (mol/L)

LPC concentration (mol/L)

THA concentration (mol/L)

0

3

0

0

0.075

0

0

10

0

0

0.250

0

0

0

3

0

0

0.075

0

0

10

0

0

0.250

1

3

0

0.025

0.075

0

1

10

0

0.025

0.250

0

1

0

3

0.025

0

0.075

1

0

10

0.025

0

0.250

1

0

0

0.025

0

0

All the MD simulation snapshots of mixture systems were aligned based on the central 8-member aliphatic ring of the PTX molecule, and we computed the distribution density of LPC or THA molecules around PTX molecule using a grid-based algorithm (Figure 6). Clearly, LPC is more



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

averagely distributed around PTX (Figure 6A and 6C) while THA shows self-aggregation with lamellar type of distribution (Figure 6B and 6D). These results suggest that LPC molecules favor the hetero-intermolecular interactions with PTX while THA molecules prefer to form self-stacking interactions in the mixture systems, which is consistent with the earlier experimental findings.

Figure 6. Calculated density distributions of LPC or THA around PTX in the mixture simulation systems. In the 3 LPC/1 PTX simulation system (A) and in the 3 THA/1 PTX simulation system (B), density cutoff was set to be 0.002/ Å3, individually. In the 10 LPC/1 PTX simulation system (C) and in the 10 THA/1 PTX simulation system (D), density cutoff was set to be 0.004/ Å3.

The interaction energies calculated between PTX, LPC and THA molecules in our simulations are summarized in Table 4. Clearly, the LPC molecules interact with PTX more favorably than THA molecules in the mixture systems. Consistently, the intermolecular interaction between LPC molecules is reduced significantly with the presence of PTX molecule, by 4 kcal/mol in 1 PTX/3


Page 16 of 23

Page 17 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


LPC system and 9 kcal/mol in 1 PTX/10 LPC system, individually. However, the presence of PTX only slightly reduces the intermolecular interaction energy between THA molecules. Therefore, PTX molecules intend to disrupt the self-stacking interactions between LPC molecules by forming more favorable PTX/LPC hetero-intermolecular interactions, while PTX/THA interactions are not adequate to break the self-stacking interactions between THA molecules.

Table 4. Calculated molecular interaction energy between LPC molecules, THA molecules, PTX/LPC and PTX/THA (X represents LPC or THA) Simulation system

Averaged molecular interaction energy (kcal/mol) PTX/X 0 0 0 0 -19.80±3.40 -41.51±2.98 -17.72±3.95 -38.80±5.76

Interaction pair 3 LPC 10 LPC 3 THA 10 THA 1 PTX/3 LPC 1PTX/10 LPC 1 PTX/3 THA 1PTX/10 THA

X/X -42.15±0.43 -175.97±0.78 -42.10±0.44 -188.51±0.84 -38.16±2.70 -167.21±3.58 -40.40±1.89 -185.52±2.21

3.5 Structural analysis of stacking interaction

Table 5. Calculated ring stacking ratio of LPC and THA molecules.

Simulation system 3 LPC 10 LPC 3 THA 10 THA 1 PTX/3 LPC 1 PTX/10 LPC 1 PTX/3 THA 1 PTX/10 THA

Stacking ratio 1.096 1.261 1.198 1.534 0.896 1.144 1.144 1.480

To further analyze the intermolecular stacking interactions, we calculated the stacking ratios between LPC molecules, THA molecules and the PTX/LPC, PTX/THA in the mixture simulation system. If the simulation system only contains 3 LPC or 3 THA molecules, the averaged stacking ratio is 1.096 and 1.198. The stacking ratios were 0.896 and 1.144 in the 1 PTX/3 LPC and



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 23

1PTX/3 THA systems, reduced by 0.2 and 0.054 respectively (Figure 2 & Table 5). When the system contains 10 LPC or THA molecules, the data generally shows the same trend. Clearly, the presence of PTX molecule introduces significant disruption to the stacking interaction between LPC molecules, but much less disruption to THA molecules.

We also decomposed the stacking interaction in PTX/LPC and PTX/THA systems by considering the individual phenyl rings of PTX (Table 6). Interestingly, the phenyl ring 1 and the ring 3 form much better stacking interaction with LPC than with THA, which is consist with what we observed in the 13C NMR chemical shifts experiment, where the coexistence of LPC was shown to perturb the chemical shifts on the ring1 and ring3 significantly. These results disclose that the direct hetero-intermolecular stacking interactions contribute to the compensation of the loss of the LPC self-stacking interactions.

Table 6. Calculated stacking contacts of each phenyl ring in the mixture simulation result, note that ring 1, ring 2 and ring 3 were defined as shown in Figure 1C.

Simulation system LPC THA

Ring 1

Ring 2

Ring 3

3

2096

4

958

10

3297

4

1369

3

1090

5

1017

10

2713

0

499

In addition, we calculated the direct atomic contacts between PTX molecule with LPC or THA in the PTX/LPC or PTX/THA mixture simulation systems. For each heavy atom of PTX molecule, we counted the number of heavy atoms of LPC or THA molecules inside the radius of 5Ᾰ and averaged them by the number of frames. These averaged contacts generally indicate the favorable interactions formed between PTX molecules with LPC or THA. We further projected the averaged contacts to the surface of the crystal conformation of PTX molecule (Figure 7). The results demonstrate that the interactions between PTX and LPC are in the polar distribution pattern on the surface of PTX molecule. Combined with the stacking interaction analysis between PTX benzyl group and LPC molecules, these results suggest that LPC molecules form favorable stacking


Page 19 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


interactions with the benzyl groups of PTX in the PTX/LPC mixture simulation system, while the hydrophilic part of the PTX molecule is still exposed in the solvent to facilitate the aqueous solvation. This stabilizes the LPC molecules in the solution phase and inhibits the crystallization and precipitation of LPC molecules. While for the PTX/THA mixture simulation system, because of the stronger stacking interaction between THA molecules themselves, PTX molecules cannot form such stable interaction with THA molecules and interrupt their original stacking interaction. So PTX molecule cannot inhibit the crystallization and precipitation of THA molecule from the solution phase.

Figure 7. Averaged atomic contact of each PTX atom with LPC or THA molecule is projected on the surface of PTX molecule. The calculated atomic contact is represented with red (high degree of contact) or blue (low degree of contact) color. (A) Atomic contacts with LPC molecule in 1 PTX/3 LPC simulation system. (B) Atomic contacts with THA molecule in 1 PTX/3 THA simulation system. (C) Atomic contacts with LPC molecule in 1 PTX/10 LPC simulation system. (D) Atomic contacts with THA molecule in 1 PTX/10 THA simulation system.

4. CONCLUSION We investigated the mechanisms dictating the drastically different colloidal stability of polymeric micelles co-encapsulated with PTX/LPC or PTX/THA. Based on experimental investigation and



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

molecular modeling, we observed that PTX inhibits the crystallization of LPC but not THA effectively, due to the formation of strong PTX-LPC but not PTX-THA molecular interactions (π-π stacking and H-bonding), despite the structural similarity between LPC and THA. The self-stacking interactions between LPC molecules but not THA molecules could be disrupted by PTX, because the PTX/LPC hetero-intermolecular interaction is stronger than LPC/LPC self-stacking while the PTX/THA interaction is weaker than THA/THA stacking interactions. We conclude that, within the drug encapsulated polymeric micelles, the existence of strong hetero-intermolecular interactions between the co-encapsulated drugs is the key factor to ensure miscibility, inhibit crystallization, and maintain the colloidal stability of the micelles. Therefore, to encapsulate drug pairs with pharmacological synergy within micellar systems to achieve desired therapeutic effects, their pharmaceutical compatibility, represented by the intermolecular interactions between them, shall be carefully assessed.

ACKNOWLEDGMENTS This research is supported by a research grant provided by The Beijing Municipal Science and Technology Commission, and by China National Nature Science Foundation (project number 81573355). FQ also thank the start-up funds provided by the Center for Life Sciences at Tsinghua and Peking Universities (Beijing, China), and by the China Recruitment Program of Global Experts.

REFERENCES 1.

Lu, Y.; Park, K., Polymeric micelles and alternative nanonized delivery vehicles for poorly

soluble drugs. Int J Pharm 2013, 453 (1), 198-214. 2.

Kim, J. Y.; Kim, S.; Pinal, R.; Park, K., Hydrotropic polymer micelles as versatile vehicles for

delivery of poorly water-soluble drugs. Journal of Controlled Release 2011, 152 (1), 13-20. 3.

Cabral, H.; Kataoka, K., Progress of drug-loaded polymeric micelles into clinical studies. J

Control Release 2014, 190, 465-76. 4.

Sakai-Kato, K.; Nishiyama, N.; Kozaki, M.; Nakanishi, T.; Matsuda, Y.; Hirano, M.; Hanada, H.;

Hisada, S.; Onodera, H.; Harashima, H.; Matsumura, Y.; Kataoka, K.; Goda, Y.; Okuda, H.; Kawanishi, T., General considerations regarding the in vitro and in vivo properties of block copolymer micelle products and their evaluation. J Control Release 2015, 210, 76-83. 5.

Murakami, M.; Cabral, H.; Matsumoto, Y.; Wu, S.; Kano, M. R.; Yamori, T.; Nishiyama, N.;


Page 20 of 23

Page 21 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Kataoka, K., Improving drug potency and efficacy by nanocarrier-mediated subcellular targeting. Sci Transl Med 2011, 3 (64), 64ra2. 6.

Wilhelm, S.; Tavares, A. J.; Dai, Q.; Ohta, S.; Audet, J.; Dvorak, H. F.; Chan, W. C. W., Analysis

of nanoparticle delivery to tumours. Nat Rev Mater 2016, 1 (5). 7.

Owen, S. C.; Chan, D. P. Y.; Shoichet, M. S., Polymeric micelle stability. Nano Today 2012, 7 (1),

53-65. 8.

Wang, A.; Cummings, N. D.; Sethi, M.; Wang, E.; Sukumar, R.; Moore, D. T.; Werner, M. E.,

Preclinical evaluation of genexol-PM, a nanoparticle formulation of paclitaxel, as a novel radiosensitizer for the treatment of non-small cell lung cancer. J Clin Oncol 2013, 31 (15). 9.

Zheng, N.; Dai, W.; Du, W.; Zhang, H.; Lei, L.; Zhang, H.; Wang, X.; Wang, J.; Zhang, X.; Gao, J.;

Zhang, Q., A novel lanreotide-encoded micelle system targets paclitaxel to the tumors with overexpression of somatostatin receptors. Mol Pharm 2012, 9 (5), 1175-88. 10. Gao, X.; Huang, Y.; Makhov, A. M.; Epperly, M.; Lu, J.; Grab, S.; Zhang, P.; Rohan, L.; Xie, X. Q.; Wipf, P.; Greenberger, J.; Li, S., Nanoassembly of surfactants with interfacial drug-interactive motifs as tailor-designed drug carriers. Mol Pharm 2013, 10 (1), 187-98. 11. Yamamoto, T.; Yokoyama, M.; Opanasopit, P.; Hayama, A.; Kawano, K.; Maitani, Y., What are determining factors for micelle carriers? Consideration stable drug incorporation into polymeric on physical and chemical characters of the micelle inner core. Journal of Controlled Release 2007, 123 (1), 11-18. 12. Lee, S. C.; Huh, K. M.; Lee, J.; Cho, Y. W.; Galinsky, R. E.; Park, K., Hydrotropic polymeric micelles for enhanced paclitaxel solubility: In vitro and in vivo characterization. Biomacromolecules 2007, 8 (1), 202-208. 13. Bae, Y.; Diezi, T. A.; Zhao, A.; Kwon, G. S., Mixed polymeric micelles for combination cancer chemotherapy through the concurrent delivery of multiple chemotherapeutic agents. J Control Release 2007, 122 (3), 324-30. 14. Shin, H. C.; Alani, A. W.; Cho, H.; Bae, Y.; Kolesar, J. M.; Kwon, G. S., A 3-in-1 polymeric micelle nanocontainer for poorly water-soluble drugs. Mol Pharm 2011, 8 (4), 1257-65. 15. Shin, H. C.; Alani, A. W.; Rao, D. A.; Rockich, N. C.; Kwon, G. S., Multi-drug loaded polymeric micelles for simultaneous delivery of poorly soluble anticancer drugs. J Control Release 2009, 140 (3), 294-300. 16. Zhang, L.; Chen, Z.; Yang, K.; Liu, C.; Gao, J.; Qian, F., beta-Lapachone and Paclitaxel Combination Micelles with Improved Drug Encapsulation and Therapeutic Synergy as Novel Nanotherapeutics for NQO1-Targeted Cancer Therapy. Mol Pharm 2015, 12 (11), 3999-4010. 17. Shin, H. C.; Alani, A. W. G.; Rao, D. A.; Rockich, N. C.; Kwon, G. S., Multi-drug loaded polymeric micelles for simultaneous delivery of poorly soluble anticancer drugs. Journal of Controlled Release 2009, 140 (3), 294-300. 18. Greco, F.; Vicent, M. J., Combination therapy: Opportunities and challenges for polymer-drug conjugates as anticancer nanomedicines. Adv Drug Deliver Rev 2009, 61 (13), 1203-1213. 19. Kim, J.; Aftab, B. T.; Tang, J. Y.; Kim, D.; Lee, A. H.; Rezaee, M.; Kim, J.; Chen, B. Z.; King, E. M.; Borodovsky, A.; Riggins, G. J.; Epstein, E. H.; Beachy, P. A.; Rudin, C. M., Itraconazole and Arsenic Trioxide Inhibit Hedgehog Pathway Activation and Tumor Growth Associated with Acquired Resistance to Smoothened Antagonists. Cancer Cell 2013, 23 (1), 23-34. 20. Eldar-Boock, A.; Polyak, D.; Scomparin, A.; Satchi-Fainaro, R., Nano-sized polymers and liposomes designed to deliver combination therapy for cancer. Curr Opin Biotech 2013, 24 (4),



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

682-689. 21. Park, K., True combination therapy using synergistic drug combination. Journal of Controlled Release 2014, 187, 198-198. 22. Sandler, A.; Gray, R.; Perry, M. C.; Brahmer, J.; Schiller, J. H.; Dowlati, A.; Lilenbaum, R.; Johnson, D. H., Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer. New England Journal of Medicine 2006, 355 (24), 2542-2550. 23. Tsubamoto, H.; Sonoda, T.; Yamasaki, M.; Inoue, K., Impact of combination chemotherapy with itraconazole on survival for patients with recurrent or persistent ovarian clear cell carcinoma. Anticancer Res 2014, 34 (4), 2007-14. 24. Bentle, M. S.; Bey, E. A.; Dong, Y.; Reinicke, K. E.; Boothman, D. A., New tricks for old drugs: the anticarcinogenic potential of DNA repair inhibitors. J Mol Histol 2006, 37 (5-7), 203-18. 25. Belinsky, M.; Jaiswal, A. K., NAD(P)H:quinone oxidoreductase1 (DT-diaphorase) expression in normal and tumor tissues. Cancer Metastasis Rev 1993, 12 (2), 103-17. 26. Ough, M.; Lewis, A.; Bey, E. A.; Gao, J.; Ritchie, J. M.; Bornmann, W.; Boothman, D. A.; Oberley, L. W.; Cullen, J. J., Efficacy of beta-lapachone in pancreatic cancer treatment: exploiting the novel, therapeutic target NQO1. Cancer Biol Ther 2005, 4 (1), 95-102. 27. Gomez Castellanos, J. R.; Prieto, J. M.; Heinrich, M., Red Lapacho (Tabebuia impetiginosa)--a global ethnopharmacological commodity? J Ethnopharmacol 2009, 121 (1), 1-13. 28. D'Anneo, A.; Augello, G.; Santulli, A.; Giuliano, M.; di Fiore, R.; Messina, C.; Tesoriere, G.; Vento, R., Paclitaxel and beta-lapachone synergistically induce apoptosis in human retinoblastoma Y79 cells by downregulating the levels of phospho-Akt. J Cell Physiol 2010, 222 (2), 433-43. 29. Li, C. J.; Li, Y. Z.; Pinto, A. V.; Pardee, A. B., Potent inhibition of tumor survival in vivo by beta-lapachone plus taxol: combining drugs imposes different artificial checkpoints. Proc Natl Acad Sci U S A 1999, 96 (23), 13369-74. 30. Zhou, L.; Chan, W. K.; Xu, N.; Xiao, K.; Luo, H.; Luo, K. Q.; Chang, D. C., Tanshinone IIA, an isolated compound from Salvia miltiorrhiza Bunge, induces apoptosis in HeLa cells through mitotic arrest. Life Sci 2008, 83 (11-12), 394-403. 31. Liu, F.; Yu, G.; Wang, G.; Liu, H.; Wu, X.; Wang, Q.; Liu, M.; Liao, K.; Wu, M.; Cheng, X.; Hao, H., An NQO1-initiated and p53-independent apoptotic pathway determines the anti-tumor effect of tanshinone IIA against non-small cell lung cancer. PLoS One 2012, 7 (7), e42138. 32. Planchon, S. M.; Wuerzberger, S.; Frydman, B.; Witiak, D. T.; Hutson, P.; Church, D. R.; Wilding, G.; Boothman, D. A., Beta-lapachone-mediated apoptosis in human promyelocytic leukemia (HL-60) and human prostate cancer cells: a p53-independent response. Cancer Res 1995, 55 (17), 3706-11. 33. Blanco, E.; Bey, E. A.; Dong, Y.; Weinberg, B. D.; Sutton, D. M.; Boothman, D. A.; Gao, J., Beta-lapachone-containing PEG-PLA polymer micelles as novel nanotherapeutics against NQO1-overexpressing tumor cells. J Control Release 2007, 122 (3), 365-74. 34. Van Der Spoel, D.; Lindahl, E.; Hess, B.; Groenhof, G.; Mark, A. E.; Berendsen, H. J., GROMACS: fast, flexible, and free. J Comput Chem 2005, 26 (16), 1701-1718. 35. Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D. A., Development and testing of a general amber force field. J Comput Chem 2004, 25 (9), 1157-74. 36. Watson, D. G., The Cambridge Structural Database (CSD): Current Activities and Future Plans. J Res Natl Inst Stand Technol 1996, 101 (3), 227-229. 37. Sciortino, F.; Essmann, U.; Stanley, H. E.; Hemmati, M.; Shao, J.; Wolf, G. H.; Angell, C. A., Crystal stability limits at positive and negative pressures, and crystal-to-glass transitions. Phys Rev E


Page 22 of 23

Page 23 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Stat Phys Plasmas Fluids Relat Interdiscip Topics 1995, 52 (6), 6484-6491. 38. Laio, A.; Parrinello, M., Escaping free-energy minima. Proc Natl Acad Sci U S A 2002, 99 (20), 12562-6. 39. Hess, B.; Bekker, H.; Berendsen, H. J. C.; Fraaije, J. G. E. M., LINCS: A linear constraint solver for molecular simulations. Journal of Computational Chemistry 1997, 18 (12), 1463-1472. 40. Liggins, R. T.; Hunter, W. L.; Burt, H. M., Solid-state characterization of paclitaxel. J Pharm Sci 1997, 86 (12), 1458-1463.


Intermolecular Interactions between Coencapsulated Drugs Inhibit

Recommend Documents