Molecular Insights into the Effects of MediaâDrug and CarrierâDrug

Subscriber access provided by Kaohsiung Medical University

Communication

Molecular insights into the effects of media-drug and carrier-drug interactions on pH-responsive drug carriers Ratna S. Katiyar, and Prateek K. Jha Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.8b00151 • Publication Date (Web): 08 May 2018 Downloaded from http://pubs.acs.org on May 9, 2018

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 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 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.

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 12 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

1

Molecular insights into the effects of media-drug and carrier-drug interactions on pH-

2

responsive drug carriers

3

Ratna S. Katiyar and Prateek K. Jha*

4

Department of Chemical Engineering, IIT Roorkee, Uttarakhand, India, 247667

5

Email: [email protected]

6

Table of Content Graphic

7 8

Abstract

9

We have performed two sets of all atom Molecular Dynamics (MD) simulations of polyacrylic acid

10

(PAA) oligomers, considered as a model pH-responsive drug carrier. In the first set, multiple

11

oligomers of PAA are simulated in model gastric and intestinal fluids, where the degree of

12

deprotonation of PAA oligomers are varied with the medium pH. Since the gastric fluid has a pH

13

substantially lower than intestinal fluid, PAA is relatively lesser ionized in gastric fluid and forms

14

aggregates. In the second set, we simulated multiple oligomers of PAA with multiple molecules of a

15

cationic anticancer drug, Doxorubicin (DOX), for a range of pH values representative of various

16

physiological conditions. The diffusion coefficient of DOX decreases with an increase in pH due to an

1 ACS Paragon Plus Environment

Molecular Pharmaceutics 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 2 of 12

1

increase in the ionic complexation of PAA with DOX, despite a decrease in PAA aggregation. Our

2

findings are in agreement with recent experimental reports on pH-triggered targeting of tumor cells by

3

PAA-DOX system. Results of these two sets of studies establish that both carrier aggregation and

4

carrier-drug interactions are competing influences that together determine the drug release from pH-

5

responsive polymers.

6

Keywords: pH; Polyelectrolytes; Macromolecular Drug Delivery; Simulations; Molecular Dynamics;

7

Diffusion; Cancer; Polymeric Drug carrier; Pharmacokinetics; Controlled release

8

1. Introduction

9

Weakly charged polyelectrolytes have tremendous potential as stimuli-responsive drug 1–4

10

carriers, as they can benefit from intrinsic pH differences present inside the body

. In the context of

11

oral drug delivery, they are useful for the delivery of poorly soluble drugs that tend to aggregate in the

12

stomach, thus resulting in low intestinal absorption and bioavailability. In an ideal situation, drug

13

molecules should be physically entrapped in the carrier in the acidic gastric environment with limited

14

or no aggregation, and released as free molecules in the basic intestinal environment. Intravenous

15

drug delivery often has an opposite expectation that the drug molecules are released at lower

16

extracellular pH conditions of affected tissues and contained in the relatively higher pH of the

17

bloodstream . Apart from the requirement that the carrier should rapidly respond to the physiological

18

pH gradient, several other factors such as the role played by various components of physiologically

19

relevant media (e.g., gastric and intestinal fluids for oral delivery), drug chemistry, drug-carrier

20

interactions, and carrier biodegradability must also be considered during carrier design. However, with

21

an exception of carrier biodegradability, a thorough molecular understanding of all these factors is

22

seldom sought and a trial-and-error approach is usually employed to find the best carrier for a given

23

drug. Molecular simulations can fill this gap by providing a detailed molecular insight into the

24

carrier/drug response to pH, media-carrier and drug-carrier interactions, and their effects on drug

25

release 6. This approach has been followed in some of the recent studies to understand the effects of

26

pH on polymer aggregation7, effects of bile salts on digestion

27

concentrated polymer solutions and gels 10.

5

8,9

, and drug release through


Page 3 of 12 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


1

In this communication, we present the results of two sets of atomistic molecular dynamics

2

(MD) simulations: (1) polyacrylic acid (PAA) oligomers in simulated biological fluids representative of

3

gastric and intestinal conditions, and (2) PAA oligomers with doxorubicin (DOX) molecules in water.

4

Both these simulations are performed for a range of pH values characteristic of physiological

5

conditions, by using a PAA degree of deprotonation corresponding to the pH considered. Although, it

6

might appear justified to also perform the second set of simulations in gastric/intestinal fluids, we

7

decided to use water for two reasons. First, as we discuss in this paper, the change in PAA20

8

aggregation with pH of set 1 shows a trend similar to that observed in our earlier study of PAA20

9

chains simulated in water.7 Second, simulations of set 2 can be used to infer behavior for both the oral

10

drug delivery and intravenous drug delivery; blood and gastric/intestinal fluids are majorly water but

11

vary in other components.

12

2. Methodology

13

The simulation method employed in this study is similar to the one used in our previous study 7

14

of PAA oligomers in water at different pH

15

performed using the GROMACS 4.6.7 simulation package with CHARMM27 force field. The simple

16

point charge (SPC) water model is used for all simulations. GROMACS compatible molecular

17

topologies of model syndiotactic PAA oligomers containing 20 repeating units (henceforth referred as

18

PAA20), DOX, and components used in simulated biological fluids (Figure 1) are generated using the

19

automatic topology building tool SwissParam

20

5. Table 1 contains the details of the two sets of simulations. It is worth noting that the inclusion of

21

various components of gastric/intestinal fluids at realistic concentration12 in set 1 results in a

22

requirement of large simulation box size to be able to accommodate a statistically significant (> 10)

23

number of molecules of each component. Several other components of gastric/intestinal fluids that

24

would have had < 10 molecules the box size of 10 nm are not included, since statistical averaging

25

over very few molecules of a component in the simulation box is not likely to provide reliable statistics.

26

Simulations of set 2 are performed in water without these additional components and therefore a

27

smaller box size of 6 nm is considered sufficient. For simulations of PAA20 at different pH, number of

28

groups on the oligomer are determined using the Henderson-Hasselbalch equation using

29

= 6.27,

13

11

and is discussed briefly here. MD simulations are

, after the initial structures are created in GaussView

which are then randomly distributed on the PAA20 oligomer. The degree of



Page 4 of 12

1

deprotonation, , is defined as the ratio of the number of groups to the number of repeating

2

units. As an illustration, for = 0.05, one out of 20 repeating units of PAA20 is randomly selected to

3

contain group and other 19 repeating units contain group. The density of the simulation

4

box is approximately 1 / for all simulations.

5 Figure 1: Chemical structures of compounds simulated in this study.

6 7 8 9 10

Table 1: Systems simulated in this study. The composition of FaSSGF, FeSSGF, and FeSSIF are taken from literature 12 and components with less than 10 molecules in the simulation box size (10 nm) are not included.

Set

Medium

Component Name

Box size (nm)

Number of molecules

Concentration (M)

pH

1

Fasted-State Simulated Gastric Fluid (FaSSGF) Fed- State Simulated Gastric Fluid (FeSSGF)

PAA20, = 0 Sodium chloride

10

74 21

0.123 0.035

1.6

PAA20, = 0.05 Sodium chloride Acetic acid Sodium acetate PAA20, = 0.25 Sodium chloride Maleic acid Sodium hydroxide Sodium oleate PAA20, = 0,0.05,0.25,0.6,0.85 DOX

10

74 143 10 18 74 74 27 39 18 16

0.123 0.237 0.017 0.03 0.123 0.123 0.045 0.065 0.03 0.123

5

50

0.384

Fed- State Simulated Intestinal Fluid (FeSSIF) 2

Water

10

6

5.8

1.6, 5, 5.8, 6.5, 7

11 12

MD simulations involved an energy minimization step, followed by MD equilibration performed for

13

50 , and MD production performed for additional 30 with a typical sampling frequency of 80 .


Page 5 of 12 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


1

Both equilibration and production simulations are performed in the NVT ensemble with a reference

2

temperature of 298 K. Number of PAA20-PAA20 (for both simulation sets of Table 1) and PAA20-

3

DOX (for set 2 of Table 1) contacts with time are monitored to track the equilibration time, where a

4

“contact” is counted if any atom of molecules of one species is within a chosen threshold distance

5

(0.6 ) from any atom of molecules of another species. We observed that beyond 50

6

equilibration, the number of contacts converged to an average value, as shown in Figure 2. The

7

number of hydrogen bonds between the two species is defined as the number of donor-acceptor pairs

8

that are within a threshold distance 0.35 and the hydrogen-donor-acceptor angle less than 30

9

degrees. For the second set of simulations, diffusion coefficient of DOX, , is computed using

10

Einstein’s relation, 〈 〉 = 6", as one-sixth of the slope of mean square displacement 〈 〉 against

11

time ". Here, the slope is determined for the linear part of the mean square displacement against time

12

plot for reasons elaborated in an earlier study .

10



Page 6 of 12

1 2 3 4 5 6 7

Figure 2: Profiles of normalized number of a) PAA20-PAA20 contacts against time for set 1, b) PAA20-PAA20 contacts against time for set 2, and c) PAA20-DOX contacts against time for set 2. See Table 1 for details of simulation sets. #$$ %"& is normalized by the total number of atoms of all PAA20 chains. #$' %"& is normalized by the product of the number of PAA20 chains and the number of DOX molecules. After 50 ns of equilibration run, all these profiles converged around average values indicated by the dashed lines in the figure.

8 9 10 6 ACS Paragon Plus Environment

Page 7 of 12 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


1

3. Results and Discussion

2

Figure 3 shows the simulation snapshots after equilibration for both sets of simulations. Results

3

obtained in simulation set 1 (figs. 3a-c) show close similarity with the results of our recently published

4

study of PAA in water , where we also studied the effects of PAA molecular weight, degree of

5

deprotonation (), deprotonation patterns, and tacticity. As elaborated in that study, three competing

6

changes occur with an increase in solution pH (or equivalently, an increase in ): (1) electrostatic

7

repulsion between PAA segments increases, (2) − hydrogen bonding decreases, and (3)

8

− hydrogen bonding first increases and then decreases. The second and third of these

9

changes are also affected by the components of gastric/intestinal fluids included in this study that

10

contains or group, which are maleic acid, acetic acid, acetate ion, and oleate ion.

11

Interactions of PAA20 with other ions present in gastric/intestinal fluid can be interpreted as a balance

12

of Coulombic interactions and entropy. That is, PAA20 repels oppositely charged chloride and

13

hydroxyl ions, and attracts sodium ions. However, a majority of these ions prefers to be dissolved in

14

the solvent than being condensed on PAA20 chain. In general, the overall PAA20 aggregation

15

behavior is dominated by the electrostatic repulsion between PAA20 segments alone, since PAA20

16

aggregation decreases with an increase in pH (increase in ), as also evident from the decrease in

17

the final average value of #$$ %"& with increase in pH in Figure 2a.

7



Page 8 of 12

1 2 3 4 5

Figure 3: Simulation snapshots (after equilibration) for systems described in Table 1. Colour scheme: PAA20 (red), sodium ion (green), chlorine ion (white), hydroxyl ion (yellow), acetic acid (cyan), acetate (black), maleic acid (violet), oleate (magenta), and DOX (blue). Water molecules are not shown.

6 7

The decrease in PAA20 aggregation with an increase in pH is expected to result in higher entrapment

8

(lower release) of drug molecules at lower pH of gastric fluids than compared to intestinal fluids. This

9

is indeed observed in an earlier simulation study of pH-responsive polymer HPMCAS with phenytoin ( 10

10

modelled as a neutral drug) in water

. However, as shown in Table 2, we observe that the DOX

11

diffusion coefficient generally decreases with an increase in pH. Slight increase in diffusion coefficient

12

occurs between = 5 and = 5.8, since a decrease in − hydrogen bonding with pH

13

increase is accompanied by an increase in − hydrogen bonding and therefore the overall


Page 9 of 12 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


1

inter-chain PAA20 hydrogen bonding is higher for the = 5.8 case than compared to = 5. This

2

issue has been elaborated in detail in our earlier simulations of PAA20 in water , where we

3

systematically varied the degree of deprotonation and deprotonation patterns. Excluding this minor

4

aberration, the general decrease in DOX diffusion coefficient with increase in pH can be attributed to

5

an increase in hydrogen bonding between group of PAA20 and #)* group of DOX. Note that

6

since a purely geometric criteria is used to determine ‘hydrogen bonding’ between and #)*

7

groups, the extent of this ‘hydrogen bonding’ may also be considered as the extent of ‘ionic

8

complexation’ between these groups. Also, in first glance, the fact the number of PAA-DOX contacts

9

decreases with an increase in pH (fig. 2c) appears to contradict the fact that the ‘hydrogen

10

bonding’/’ionic complexation’ between PAA20 and DOX increases with an increase in pH (Table 2).

11

However, on closer inspection, we can observe that it occurred because a relatively larger threshold

12

distance (0.6 ) was used for the computation of number of PAA20-DOX contacts, as compared to

13

the threshold distance of 0.35 used for the computation of hydrogen bonding.

14

Finally, the diffusion coefficient of DOX is substantially reduced in the presence of PAA20 ( ∼ 5 − 50

15

times in the pH range studied in Table 2) compared to the case when PAA20 is not present (also

16

compare figs. 3e-i with fig. 3d), demonstrating the controlled release behavior of PAA20-DOX system.

17

Our observation that the DOX diffusion coefficient decreases with an increase in pH is in close

18

agreement with experimental studies of similar systems

19

by Tian et al16,17, who have synthesized pH-responsive block copolymers of PAA with pluronic

20

copolymers and studied the DOX loading and release behavior. They observe a triggered release

21

the acidic condition at = 5 and sustained release at = 7.2, which can be explained by the

22

decrease in DOX diffusion coefficient with increase in pH observed in our simulations. Other

23

studies

24

DOX in the acidic pH condition.

7

14,15,18

14–18

. One of the closest experimental work is

17

in

that did not use PAA but contained carboxyl groups showed similar triggered release of

25 26 27 28



1 2 3

Page 10 of 12

Table 2: Simulation results for set 2 described in Table 1. The values in brackets indicate the standard deviation of the last digit in decimals, e.g., 3.04%4& is equivalent to 3.04 ± 0.04. The last row represents the diffusion coefficient of DOX in the system containing no PAA in set 2. ./

1.6 5 5.8 6.5 7

f

0 0.05 0.25 0.60 0.85

DOX

Diffusion coefficient (10-7 cm2/s)

Number of hydrogen bonds per monomer

DOX

COO and + NH3

1.30 0.42 0.49 0.16 0.14

-

0.036(3) 0.107(6) 0.220(6) 0.219(9)

PAA20 and PAA20 0.089(9) 0.122(9) 0.17(1) 0.117(6) 0.037(5)

7.00

4 5

In summary, we show a successful demonstration of the application of molecular simulation in

6

providing molecular insights into the behavior of pH-responsive controlled release formulations. We

7

studied the polymer behavior in model gastric/intestinal fluids in the first set of simulations, and

8

analyzed the behavior of a polymer-drug combination in different pH environments in the second set

9

of simulations. We have selected a rather simplified example of polyacrylic acid carrier with a common

10

anticancer drug (doxorubicin), but the methods discussed here can be applied to other polymer and

11

drug chemistries, and more importantly, can be used to access the potential of novel polymer

12

chemistries prior to synthesis. However, there are two major issues with the molecular simulation

13

approach that must be kept in mind. First, actual polymers have molecular weights of at least an order

14

of magnitude higher than that can be simulated with existing computational speeds. Therefore, the

15

drug diffusion coefficients obtained in our simulations cannot be directly compared with the diffusion

16

coefficients calculated from the experimental drug release data, and, at best, are expected to possess

17

a positive correlation. Second, the detailed chemistries of physiological environments are often not

18

known or are difficult to model. More work is needed to explore into these issues.

19

Acknowledgments

20

This work is supported by a research grant from the Science and Engineering Research Board

21

(SERB), India (no. YSS/2015/001228).

22

References

23

(1)

Kocak, G.; Tuncer, C.; Bütün, V. pH-Responsive Polymers. Polym. Chem. 2017, 8 (1), 144–


Page 11 of 12 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


176.

1

2

(2)

Rev. 2006, 58 (15), 1655–1670.

3

4

(3)

(4)

(5)

167.

10

(6)

Katiyar, R. S.; Jha, P. K. Molecular Simulations in Drug Delivery: Opportunities and Challenges. WIREs Comput Mol Sci. 2018, e1358.

12

13

Kanamala, M.; Wilson, W. R.; Yang, M.; Palmer, B. D.; Wu, Z. Mechanisms and Biomaterials in pH-Responsive Tumour Targeted Drug Delivery: A Review. Biomaterials 2016, 85, 152–

9

11

Manallack, D. T.; Prankerd, R. J.; Yuriev, E.; Oprea, T. I.; Chalmers, D. K. The Significance of Acid/base Properties in Drug Discovery. Chem. Soc. Rev. 2013, 42 (2), 485–496.

7

8

Mura, S.; Nicolas, J.; Couvreur, P. Stimuli-Responsive Nanocarriers for Drug Delivery. Nat. Mater. 2013, 12 (11), 991–1003.

5

6

Schmaljohann, D. Thermo- and pH-Responsive Polymers in Drug Delivery. Adv. Drug Deliv.

(7)

Katiyar, R. S.; Jha, P. K. Phase Behavior of Aqueous Polyacrylic Acid Solutions Using

14

Atomistic Molecular Dynamics Simulations of Model Oligomers. Polymer (Guildf). 2017, 114,

15

266–276.

16

(8)

Birru, W. A.; Warren, D. B.; Headey, S. J.; Benameur, H.; Porter, C. J. H.; Pouton, C. W.;

17

Chalmers, D. K. Computational Models of the Gastrointestinal Environment. 1. The Effect of

18

Digestion on the Phase Behavior of Intestinal Fluids. Mol. Pharm. 2017, 14 (3), 566–579.

19

(9)

Birru, W. A.; Warren, D. B.; Han, S.; Benameur, H.; Porter, C. J. H.; Pouton, C. W.; Chalmers,

20

D. K. Computational Models of the Gastrointestinal Environment. 2. Phase Behavior and Drug

21

Solubilization Capacity of a Type I Lipid-Based Drug Formulation after Digestion. Mol. Pharm.

22

2017, 14 (3), 580–592.

23

(10)

Molecular Dynamics Simulations. Mol. Pharm. 2014, 11 (5), 1676–1686.

24

25

Jha, P. K.; Larson, R. G. Assessing the Efficiency of Polymeric Excipients by Atomistic

(11)

Zoete, V.; Cuendet, M. A.; Grosdidier, A.; Michielin, O. SwissParam: A Fast Force Field



Generation Tool for Small Organic Molecules. J. Comput. Chem. 2011, 32 (11), 2359–2368.

1

2

(12)

(13)

(14)

(15)

Kim, J. O.; Kabanov, A. V.; Bronich, T. K. Polymer Micelles with Cross-Linked Polyanion Core for Delivery of a Cationic Drug Doxorubicin. J. Control. Release 2009, 138 (3), 197–204.

9

10

Dadsetan, M.; Taylor, K. E.; Yong, C.; Bajzer, Ž.; Lu, L.; Yaszemski, M. J. Controlled Release of Doxorubicin from pH-Responsive Microgels. Acta Biomater. 2013, 9 (3), 5438–5446.

7

8

Spruijt, E.; Cohen Stuart, M. A.; Van Der Gucht, J. Linear Viscoelasticity of Polyelectrolyte Complex Coacervates. Macromolecules 2013, 46 (4), 1633–1641.

5

6

Marques, M. R. C.; Loebenberg, R.; Almukainzi, M. Simulated Biological Fluids with Possible Application in Dissolution Testing. Dissolution Technol. 2011, 18 (3), 15–28.

3

4

Page 12 of 12

(16)

Tian, Y.; Ravi, P.; Bromberg, L.; Hatton, T. A.; Tam, K. C. Synthesis and Aggregation Behavior

11

of Pluronic F87/poly(acrylic Acid) Block Copolymer in the Presence of Doxorubicin. Langmuir

12

2007, 23 (5), 2638–2646.

13

(17)

Tian, Y.; Bromberg, L.; Lin, S. N.; Alan Hatton, T.; Tam, K. C. Complexation and Release of

14

Doxorubicin from Its Complexes with Pluronic P85-B-Poly(acrylic Acid) Block Copolymers. J.

15

Control. Release 2007, 121 (3), 137–145.

16

(18)

Cunningham, A. J.; Robinson, M.; Banquy, X.; Leblond, J.; Zhu, X. X. Bile Acid-Based Drug

17

Delivery Systems for Enhanced Doxorubicin Encapsulation: Comparing Hydrophobic and Ionic

18

Interactions in Drug Loading and Release. Mol. Pharm. 2018, acs.molpharmaceut.7b01091.

19


Molecular Insights into the Effects of MediaâDrug and CarrierâDrug

Recommend Documents

Molecular Insights into the Effects of MediaâDrug and CarrierâDrug