Local and Sustained Activity of Doxycycline ... - ACS Publications

Mar 11, 2016 - Bone and Joint Research Unit, Queen Mary University of London, London, United Kingdom. •S Supporting Information. ABSTRACT: Achieving...
1 downloads 0 Views 3MB Size
Subscriber access provided by LAURENTIAN UNIV

Article

Local and sustained activity of doxycycline delivered with layer-by-layer microcapsules Dong Luo, David James Gould, and Gleb B. Sukhorukov Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.6b00070 • Publication Date (Web): 11 Mar 2016 Downloaded from http://pubs.acs.org on March 15, 2016

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.

Biomacromolecules 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 37

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

Biomacromolecules

1

Local and sustained activity of doxycycline

2

delivered with layer-by-layer microcapsules

3

Dong Luo1, David J Gould2*‡, Gleb B. Sukhorukov1*‡

4

1

5

4NS, UK

6

2

7

Keywords: antibiotic, doxycycline, layer-by-layer, microcapsules, drug

School of Engineering & Materials Science, Queen Mary University of London, London, E1

Bone & Joint Research Unit, Queen Mary University of London, London, UK

8

9

ABSTRACT

10

Achieving localized delivery of small molecule drugs has the potential to increase efficacy and

11

reduce off target and side effects associated with systemic distribution. Herein, we explore the

12

potential use of layer-by-layer (LbL) assembled microcapsules for the delivery of doxycycline.

13

Absorbance of doxycycline onto core dextran sulphate of pre-assembled microcapsules provides

14

an efficient method to load both synthetic and biodegradable microcapsules with the drug.

15

Application of an outer layer lipid coat enhances the sustained in vitro release of doxycycline

16

from both microcapsule types. To monitor doxycycline delivery in a biological system C2C12

17

mouse myoblasts are engineered to express EGFP under the control of the optimized components

1 Environment ACS Paragon Plus

Biomacromolecules

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

of the tetracycline regulated gene expression system. Microcapsules are not toxic to these cells

2

and upon delivery to the cells, EGFP is more efficiently induced in those cells that contain

3

engulfed microcapsules and monitored EGFP expression clearly demonstrates that synthetic

4

microcapsules with a DPPC coat are the most efficient for sustain intracellular delivery.

5

Doxycycline released from microcapsules also displayed sustained activity in an antimicrobial

6

growth inhibition assay compared with doxycycline solution. This study reveals the potential for

7

LbL microcapsules in small molecule drug delivery and their feasible use for achieving

8

prolonged doxycycline activity.

9

Introduction

10

Polyelectrolyte microcapsules generated on a dissolvable matrix core using the layer-by-layer

11

(LbL) assembly method were first described in 1998.1 Their construction utilizes electrostatic

12

interactions between oppositely charged polyelectrolytes to add layers to the core. Assembly is

13

performed under native conditions which permit the inclusion of biologically active molecules

14

within the structure, the positioning of these, influences activity in intact capsules as well as

15

stability and release kinetics.2, 3 The inclusion of metal nanoparticles as a charged species in

16

microcapsule layers enables control over movement (magnetism)4 and can provide a remote

17

mechanism to release microcapsule contents.5-7 Microcapsules have been extensively studied in

18

isolation and have been successfully employed for the delivery of macromolecules to cells, when

19

they are multifunctionalised they can be utilized in more intricate applications such as ATP

20

synthesis8 and biosensors.9 LbL microcapsules are usually assembled on a calcium carbonate

21

core which yields microcapsules that are 3 to 4 µm in diameter. Their size has implications for

22

their use in cellular systems, firstly, they will be engulfed by phagocytes which facilitates their

2 Environment ACS Paragon Plus

Page 2 of 37

Page 3 of 37

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

Biomacromolecules

1

use in antigen delivery,10 secondly, if they are delivered to extravascular compartments of the

2

body they will not enter the blood and will be retained within the delivered tissue unless removed

3

mobile cells. Their fate within cells depends upon their structure that those composed of

4

biodegradable polymers will be broken down and release their contents whilst those assembled

5

with synthetic polymers can potentially remain intact. LbL microcapsules can be used to deliver

6

macromolecules to cells but the incorporation and delivery of small molecule drugs is more

7

problematic because they can diffuse out of the capsule structure. Strategies that have been

8

employed to encapsulate small molecule drugs include the use of precipitated11 or crystallized

9

material in the capsule core12 and adsorbing molecules into the microcapsule structure after

10

assembly.13 In addition other attributes such as heat shrinkage, crosslinking and outer lipid

11

layers14 can be applied to promote retention of molecules.

12

Small molecule drugs are usually delivered orally and distribute systemically when local

13

pharmacological effects are actually required. Systemic distribution can cause ‘off target’,

14

adverse and side effects. If local delivery can be achieved such drugs could have greater efficacy

15

and improved safety. Doxycycline is a widely used antibiotic in the treatment of infections in

16

humans. At low concentrations below that required for antimicrobial effects doxycycline also

17

exerts anti-inflammatory effects due to inhibition of matrix metalloproteinases15 so has potential

18

use in treatment of inflammatory conditions.16 Doxycycline is typically administered orally as

19

tablets or suspensions and also topically in cream. There have been no previous reports on the

20

incorporation of doxycycline into LbL microcapsules but doxycycline has been incorporated into

21

both nano and microstructures however sustained biological activity of the antibiotic has not

22

been explored in these studies.17, 18

3 Environment ACS Paragon Plus

Biomacromolecules

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 37

1

In this work, doxycycline was encapsulated into LbL microcapsules by forming a complex

2

with dextran sulphate sodium salt in the microcapsule core or structure and sustained effect was

3

monitored by both an intracellular assay and antimicrobial assay. The microcapsule design was

4

optimized by coating microcapsules with lipid. The lipid coating altered the microcapsule

5

appearance;,the release kinetics of doxycycline, and biological activity of doxycycline.

6

Experimental section

7

Materials

8

Poly(allylamine hydrochloride) (PAH, 56 kDa), Poly(sodium 4-styrenesulfonate) (PSS, 70

9

kDa), Dextran sulphate sodium salt (DS, 100 kDa), Poly-L-arginine hydrochloride (Parg, 15–70

10

kDa), 1,2-Dipalmitoyl-sn-glycero-3-phosphocholine (DPPC),L-α-Phosphatidylcholine (lecithin,

11

from egg yolk), Doxycycline Hydrochloride, Rhodamine B isothiocyanate,

12

Tetrazolium Bromide (MTT), Cell Lysis Reagent (CelLytic™ M ), Protease Inhibitor Cocktail,

13

Ethylenediaminetetraacetic Acid disodium salt (EDTA), Calcium Chloride, Sodium Carbonate

14

and all other salts were purchased from Sigma-Aldrich UK.1, 2-dipalmitoyl-sn-glycero-3-

15

phosphoethanolamine-N-(lissaminerhodamine B sulfonyl) (ammonium salt) was purchased from

16

Avanti Polar-lipids. Agar (Fisher BioReagents) and LB Broth, Miller (Fisher BioReagents) were

17

used for antimicrobial studies. Other biological reagents, including DMEM, PBS, foetal calf

18

serum, penicillin and trypsin were purchased from Lonza Biologics Inc (Newington, NH, United

19

States).

20

Microcapsule synthesis and doxycycline encapsulation

Thiazolyl Blue

21

The layer-by-layer self-assembly technique was used for microcapsule construction on a

22

CaCO3 sacrificial template for oppositely charged polyelectrolytes. For subsequent loading of

23

doxycycline into capsules, DS was co-precipitated within CaCO3 according to the method for

4 Environment ACS Paragon Plus

Page 5 of 37

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

Biomacromolecules

1

encapsulation of large molecules3. DS (20 mg/ml) was added to 0.33 M CaCl2 solution and was

2

then immediately mixed with 0.33 M Na2CO3 by vigorous stirring for 20 seconds and then

3

incubated for a further 10 seconds. The hybrid CaCO3 particles were washed three times with

4

deionized H2O, and were immediately used for microcapsule assembly. PAH and PSS were used

5

for synthetic shells, and Parg and DS as biodegradable ones. Due to the DS trapped in the

6

templates, the cores were initially negatively charged so PAH or Parg was assembled as the first

7

layer. All the polymer solutions were prepared at 2 mg/ml with NaCl (0.2 M), and after

8

absorbing each layer, microcapsules were triple washed with deionized H2O. For fluorescent

9

visualization, the second positive layer was replaced with TRITC-PAH or TRITC-Parg. Six

10

layers were deposited for all the capsules before dissolving the templates with 0.2 M EDTA

11

solution.

12

DS trapped in the hollow cavity of microcapsules was used as attraction for doxycycline

13

molecules. This method has been used for encapsulating soluble low molecular weight drugs,19

14

which are loaded via electrostatic interaction or as a shell component. Herein, doxycycline

15

molecules penetrate the shells and react with DS and form a complex within the microcapsule.

16

Microcapsules were counted with a haemocytometer and the concentration adjusted to 6×107

17

capsules per ml before mixing with 1 ml of doxycycline solution (10 mg/ml) and shaking for 8

18

hours. At the end of this incubation maximum encapsulation was achieved by placing the

19

mixture at 60℃ for 10 min to shrink the capsules according to a previously reported procedure.20

20

Then the capsules were washed three times with deionized water and supernatants were collected

21

to determine amount of loaded doxycycline. In order to determine the amount of encapsulated

22

doxycycline per gram of microcapsules, equal quantity of microcapsules (both empty capsules

23

and doxycycline loaded ones) were freeze dried (ScanVac Cool Safe Freeze Drying, Denmark) at

5 Environment ACS Paragon Plus

Biomacromolecules

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

-107 ℃, 0.009 mBar for 2 days. Then each sample was weighed (CP2-P microbalance, Sartorius

2

AG Gottingen, Germany) and encapsulated doxycycline weight was calculated.

3

Lipid layer assembly

4

Phospholipids, DPPC or lecithin, were coated as outmost layers on microcapsules to further

5

reduce the permeability and thereby reduce the doxycycline leakage. DPPC (for synthetic

6

capsules) and lecithin (for biodegradable capsules) were dissolved in chloroform at 0.5 mg/ml

7

and were then evaporated using a rota vap at 42℃, after which a thin layer of phospholipids was

8

formed.21, 22 Liposomes were then constructed by adding water (final lipid concentration of

9

1mg/ml) to the lipid film and sonicating for 1 hour at 60℃. The formed lipid vesicles were added

10

to the doxycycline loaded capsules and shaken for 20 min. Then the mixture was washed 3 times

11

with deionized water to remove excessive lipid, and supernatants were also collected for

12

doxycycline concentration measurement. To visualize the lipid coating, Rhodamine B labelled

13

lipid (0.2 mg/ml) was mixed with the DPPC or lecithin. Permeability of microcapsules was

14

tested with FITC under the confocal microscope.

15

Release profile of doxycycline from microcapsules

16

Release of doxycycline from microcapsules was determined by a fluorescence spectrum assay.

17

The doxycycline concentration in solution was calculated according to the emission fluorescence

18

intensity at 536 nm. Typically, a series of standard doxycycline solutions (40, 30, 20, 15, 10, 5,

19

4, 3, 2, 1, 0.5, 0.25, 0.1, 0.05 ug/ml) were prepared and emission intensity at 536 nm was

20

measured (excited at 350nm). The calibration curve was used to determine amount of

21

doxycycline in supernatants collected during the assembly process and therefore enable

22

calculation of the amount of doxycycline retained in each capsule (Supplementary Figure S1).

6 Environment ACS Paragon Plus

Page 6 of 37

Page 7 of 37

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

Biomacromolecules

1

In the release experiments, 12×107of synthetic or biodegradable capsules were prepared in a 1

2

ml suspension in either deionized water, PBS or NaCl (0.15M). The suspensions were shaken on

3

a vortex for specified intervals, after which it was centrifuged and the supernatant was collected

4

for evaluation of the doxycycline released. The removed supernatant was replaced with 1 ml of

5

fresh solution and the incubation continued. To demonstrate 100% release, doxycycline loaded

6

microcapsules were mixed with 2 M NaOH.23 At this pH, microcapsules disassemble and the

7

doxycycline concentration in the solution was measured.

8

Engineering C2C12 myoblasts as doxycycline reporter cells

9

The lentiviral vectors used to engineer mouse C2C12 myoblast cells (ATCC® CRL-1772™)

10

were pLOX-TwGFP which encodes EGFP from the Ptet and pUCL-MIK which encodes the

11

rtTA-2SM2 transactivator24 and the tetRKRAB repressor25 from a bicistrontic cassette26 under

12

the control of the constitutive SFFV promoter.27 Lentivirus was prepared as previously

13

described28 then C2C12 cells were multiply infected with pUCL-MIK before multiple infection

14

with pLOX-TwGFP. The resulting cell population was then stimulated with doxycycline

15

hydrochloride (1 µg/ml) and after 24 hours the cells with the highest expression of EGFP were

16

sorted using a BD FACS Aria II. These cells were plated on a 96 well plate at 1 cell per well and

17

clones expanded. The selected clone was not fluorescent in the unstimulated state but

18

ubiquitously expressed high levels of EGFP upon doxycycline stimulation.

19

MTT assay

20

The myoblast cells were cultured in the Dulbecco's minimum essential media, containing 10%

21

fetal bovine serum (FBS) and 1% penicillin–streptomyosin, at 37 °C and 5% CO2. The standard

22

MTT assay was used to assess the cell viability after treatment with doxycycline solution, empty

23

capsules and doxycycline-loaded capsules for 48 hours. Cells were seeded in 96-well plates with

7 Environment ACS Paragon Plus

Biomacromolecules

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

1×104 cells per well and capsules were added at 10:1 ratio. After 48 hours, medium was

2

removed, 20 µl of MTT solution (in PBS) was added to each well and cells were cultured for 3.5

3

hours. Then 150 µl MTT solvent (4 mM HCl, 0.1% Nondet P-40 in isopropanol) was added to

4

each well. The plate was read in a BMG Fluostar Galaxy plate reader (filter 570 nm).

5

Cell uptake of doxycycline capsules and induced EGFP expression

6

Cellular uptake of capsules and induced EGFP expression was visualized using a Leica

7

confocal microscope with 40× oil emersion objective. Cells were seeded on 8-well glass chamber

8

slides (well size 10.7 mm×9.4 mm, ibidi USA Inc) at 1×104 cells per well and 10 times the

9

number of doxycycline loaded capsules were added immediately. For the control, the equivalent

10

amount of free doxycycline was mixed with the medium. After 2 days, free capsules were

11

removed by washing with PBS, and fluorescence and phase contrast microscopy images were

12

taken.

13

To observe the induced EGFP expression, cells co-cultured with doxycycline-loaded capsules

14

for one day were trypsinized and centrifuged to remove any free or cell surface attached

15

capsules, and cells were re-seeded in the chamber. Eight hours later, images in green, red and

16

transmission channels were taken at the same time from one side of the chamber to the other to

17

form a large panoramic image. Cells were counted in the large image, and classified into four

18

groups, namely green and red (cells with both EGFP and capsules), green (cells with EGFP

19

only), red (cells with capsules only) and none (cell without EGFP or capsules).

20

In order to evaluate the EGFP expression kinetics, cells with both free doxycycline and

21

doxycycline-loaded capsules (10:1) were cultured in the chamber for 34 days, and confocal

22

images were taken every two days. To reduce the effect of residual doxycycline, after 2 days

23

incubation with microcapsules medium was changed every day. These studies on EGFP

8 Environment ACS Paragon Plus

Page 8 of 37

Page 9 of 37

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

Biomacromolecules

1

expression enable comparison of expression kinetics for doxycycline in solution, doxycycline-

2

loaded synthetic capsules and biodegradable capsules.

3

EGFP fluorescence intensity assay

4

Quantitative analysis of EGFP expression kinetics was performed with a Fluostar Galaxy plate

5

reader. Engineered C2C12 cells (1×104 cells per well) in a 96-well plate were treated with free

6

doxycycline, doxycycline-loaded synthetic or biodegradable capsules (7 wells for each group).

7

After specific intervals (2, 4, 6, 8, 10, 12 days), culture medium were removed, and cells were

8

washed with PBS and lysed for 30 min by addition of 20 µl lysis reagent (CelLytic™ M, with

9

0.05% Protease Inhibitor Cocktail added) to each well. Then 80 µl PBS was added, and EGFP

10

fluorescence intensity was measured in the reader with excitation and emission wavelengths

11

fixed at 480 nm and 520 nm accordingly. Control lysates were prepared using untreated cells and

12

the assay blank consisted of lysis reagent and PBS up to 100 µl volume. EGFP intensity was

13

calculated according to the following equation, EGFP fluorescence intensity = (intensity of test

14

well – intensity of control well) / intensity of control well ×100%.

15

Bacterial Inhibition assay

16

The antibacterial activity of doxycycline loaded microcapsules was tested by growth inhibition

17

of Escherichia coli (E. coli, DH5a). Bacterial suspensions in LB Broth were cultured at

18

37°C/200rpm shaking until they were in the linear phase of growth indicated by an optical

19

density between 0.4-0.6 measured using a spectrophotometer (Cecil CE2021) at 400nm.

20

Bacterial suspension (0.4 ml) was spread on the surface of dried LB agar plates. The sensitivity

21

of the E. coli to inhibition with doxycycline was firstly demonstrated using filters (diameter 7

22

mm) treated with doxycycline concentrations from 5 mg/ml to 8 µg/ml and placed on an agar

23

plate pre-spread with E. coli. Filters remained on the plate which was incubated at 37°C

9 Environment ACS Paragon Plus

Biomacromolecules

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

overnight and then diameter of zones of inhibition were measured in mm. Sustained release of

2

doxycycline from microcapsules was examined in a transfer experiment in which filters were

3

initially loaded with 40 µl of doxycycline solution (0.1 or 1 mg/ml) or 40 µl of microcapsule

4

suspension equivalent 1 mg/ml of doxycycline. After incubation for 90 minutes on an agar plate

5

pre-spread with E. Coli, filters were transferred to a second pre-spread plate with addition of 10

6

µl sterile water to the filter, the process was repeated every 90 minutes. After 12 hours, all the

7

filters were disposed and all the plates were incubated overnight at 37°C. Images were taken and

8

the inhibitory zones against E. coli growth were measured using Nano Measurer(1.2).

9

Statistical Methods

Page 10 of 37

10

Descriptive statistics and significant differences between groups were calculated using the

11

Student's t tests for two-sample data of unequal variance (Microsoft® Excel 07 software);

12

significance refers to p values less than 0.05 and 0.01.

13

Results and discussion

14

Doxycycline-loaded microcapsule preparation

15

Small molecule drugs can be incorporated into microcapsules through electrostatic interactions

16

following a post loading approach.29, 30 Doxycycline is a small molecule drug (structure shown in

17

Figure 1a), and was encapsulated by precipitating with DS (Supplementary Figure S2).

18

Typically, two types of capsule were assembled, synthetic capsules consisting of PAH/PSS

19

layers and biodegradable ones composed of Parg/DS, both of which had DS in the interior for

20

doxycycline loading (Figure 2a and2d). Upon incubation, doxycycline penetrated the

21

microcapsule layers forming a less soluble complex with DS in the core, as well as being trapped

22

within the layers. The synthetic capsules became smaller in size and had a more protuberant

23

structure after doxycycline loading (Figure 2b), indicating that doxycycline-DS complex was

10 Environment ACS Paragon Plus

Page 11 of 37

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

Biomacromolecules

1

formed in the core. Whilst, the biodegradable capsules did not undergo such notable

2

morphological change, but according to the increased capsule shell thickness (Figure 2e), we

3

could assume that more doxycycline was absorbed in the microcapsule walls. The principle of

4

the complex formation is based on the ion paring ability of dextran sulphate, as indicated by the

5

FTIR spectrum (Figure 1b), which was also used to promote minocycline loading efficiency in

6

PLGA nanoparticles.31 Similar phenomenon were observed when ciprofloxacin was deposited

7

into capsules, with a black shadow seen in the capsule interior under TEM and an increase in

8

average capsule height by 500 nm.32 According to measurement of supernatants from

9

encapsulation, doxycycline loaded in synthetic and biodegradable capsules was determined to be

10

5.4 pg and 4.8 pg per capsule, respectively. Doxycycline amount per gram of capsules was

11

determined to be around 97.4 mg according to the net weight increase compared to the empty

12

microcapsules. The results were confirmed by fluorescence imaging, synthetic capsules with

13

doxycycline had more blue fluorescence in the core, while blue fluorescence in biodegradable

14

capsules was within the rings of microcapsule shells (Supplementary Figure S3) in both

15

instances doxycycline is complexed with DS. a

b Doxy DS DS-Doxy

-

-SO3

-OH

-NH

3500

3000

-C=O

4000

2500

2000

1500

1000

500

16

Figure 1. Doxycycline and FTIR spectrum. Molecular structure of doxycycline (a), FTIR

17

spectrum of doxycycline (blue), DS sodium salt (black) and doxycycline-DS complex (red) (b).

11 Environment ACS Paragon Plus

Biomacromolecules

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 12 of 37

1

Since the microcapsule shells are permeable, extra layers are often assembled to enable longer

2

drug retention. Such layers are either more hydrophobic or have a denser meshwork. The

3

phospholipids DPPC and lecithin were used as an outer layer for synthetic and biodegradable

4

capsules respectively. As could be seen by changes in zeta potential (Supplementary Figure S4),

5

the outmost DPPC coating altered the negative charge of PSS to slightly positive, which

6

confirmed addition of the lipid layer on the surface. In addition, confocal images of capsules

7

show red circles when the lipid was labelled with Rhodamine B. In the permeability test, the

8

outer lipid layer also effectively hindered penetration of FITC through capsules. Whilst an

9

incomplete lipid layer on the microcapsule surface only rendered them partially permeable to

10

FITC (Supplementary Figure S5). These results provide evidence that microcapsule permeability

11

was reduced by the lipid coating. SEM images of lipid coated capsules (Figure 2c and f) shows

12

that DPPC, or lecithin coating smoothed the capsule surface (inset image), which was evidenced

13

as a more compact outer shell was formed. Similar morphological changes were previously

14

observed when biotinylated lipid was assembled on the surface of PAH/PSS capsules resulting in

15

an uncollapsed structure.33 According to measurement of supernatants, doxycycline loaded in

16

DPPC-coated synthetic and lecithin-coated biodegradable capsules was 5.3 pg and 5.4 pg per

17

capsule, respectively.

18

12 Environment ACS Paragon Plus

Page 13 of 37

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

Biomacromolecules

1

Figure 2. Microcapsule images. SEM images of synthetic capsules (a-c) and biodegradable

2

capsules (d-f), before being loaded with doxycycline (a, d), after loading with doxycycline (b, e),

3

and coated with a lipid layer on the surface (c, f) (scale bar of 20 µm); the inset images are

4

individual capsules at high magnification (×40000) (scale bar of 3 µm).

5 6

Release kinetics of doxycycline in different mediums

7

Release kinetics of a drug from LbL assembled microcapsules is primarily dependent on

8

physio-chemical properties of the polymeric shells34 and the binding characteristics between the

9

drug molecules and capsules.30, 35 In our study, doxycycline was trapped by forming a complex

10

with DS, and the release of payload was affected by both factors. Due to the porous structure of

11

polymeric shells, doxycycline was able to diffuse through, but the capsules shell thickness,

12

polyelectrolyte molecular weight and ionic strength of the medium all affect the diffusion rate.34

13

Changes in the ionic strength of solution alter the strength of electrostatic interactions within the

13 Environment ACS Paragon Plus

Biomacromolecules

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 37

1

microcapsule shell structure and in the binding between doxycycline and DS. By shaking in

2

water, 0.15M NaCl or PBS, doxycycline release from all microcapsules was monitored, and the

3

rate of doxycycline release from all microcapsules was PBS>0.15M NaCl>water (Figure 3). In

4

low ionic strength solution, a tightly packed microcapsule structure is observed, but when the

5

polymeric structure is transferred to high ionic strength solution, the charge of polymers is

6

shielded and the rearranged microcapsule structure is more permeable to small molecules.36 With

7

regards to doxycycline and DS binding, in water, there would be a stronger binding between the

8

two molecules, due to charge interaction of negatively charged DS and amino groups of

9

doxycycline, but PBS and NaCl might accelerate the dissociation of the DS-doxycycline

10

complex via charge decoupling in salt solution which facilitates release of free doxycycline from

11

the capsule interior. In general the release of doxycycline from microcapsules was bi-phasic; a

12

primary burst release within the first 40 minutes and a slower secondary release after 40 minutes

13

(Table 1). The first phase is relevant to detaching of more loosely bound doxycycline and the

14

second phase is salt induced dissolution of the complex between DS and doxycycline. In the case

15

of pure water these two phases were less distinct. The sustained release measured in these

16

solutions contrasts to the complete dissociation of microcapsules with 100% doxycycline release

17

observed in 2M NaOH (data not shown).

18

The release kinetics were however slower when microcapsules were coated with a lipid layer

19

as shown in Figure 3c and 3d. The retention of doxycycline with the lipid coated capsules can be

20

attributed to lipophilic nature of doxycycline37 and to the hydrophobic barrier formed at the

21

outmost capsule surface by phospholipids which would be less permeable aqueous solutions. A

22

phospholipid outer layer seal was previously employed in the encapsulation of the small

23

molecule drug daunorubincin.22 The difference in the release kinetics for lipid coated synthetic

14 Environment ACS Paragon Plus

Page 15 of 37

1

and biodegradable capsules may be due to the fact that lecithin, a natural phospholipid, is a

2

heterogeneous molecule with a range of molecular weights which may prohibit the formation of

3

a more compact coating layer observed with synthetic DPPC.

4 a

b in H2O in 0.15 M NaCl in PBS

120

80

doxycycline released (%)

doxycycline released (%)

in H2O in 0.15 M NaCl in PBS

100

100

80

60

40

60

40

20

20

0

0 0

c

50

100

150

200

250

300

time (min)

0

d

in H2O in 0.15 M NaCl in PBS

100

50

100

150

200

250

300

time (min)

in H2O in 0.15 M NaCl in PBS

100

80

doxycycline released (%)

80

doxycycline released (%)

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

Biomacromolecules

60

40

20

60

40

20

0

0 0

50

100

150

200

250

300

0

50

100

time (min)

150

200

250

300

time (min)

5

Figure 3. Kinetics of doxycycline release from microcapsules. Release profile of doxycycline

6

from synthetic capsules (a), biodegradable capsules (b), synthetic capsules with DPPC coating

7

(c) and biodegradable capsules with lecithin coating (d) in water, 0.15 M NaCl, and PBS. Values

8

are the mean of 3 groups and vertical bars represent the SD.

9 10 11 12

15 Environment ACS Paragon Plus

Biomacromolecules

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 16 of 37

1

Table 1 Rate of doxycycline release rates from microcapsules in different solutions. Phase 1

2

release is the kineic in the first 40 minutes whilst phase 2 release relates to the kinetic iduring the

3

subsequent 235 minutes. Doxycycline release rate

In H2O

(ug/min)

Phase1 Phase2

Phase1 Phase2

Phase1 Phase2

Syn capsules

0.229

0.067

0.700

0.019

1.595

0.032

DPPC coated syn capsules

0.263

0.016

0.428

0.029

1.045

0.043

Bio capsules

0.305

0.043

0.769

0.021

1.794

0.039

Lecithin coated bio capsules

0.313

0.019

0.599

0.029

1.476

0.059

In NaCl

In PBS

4 5 6

Cytotoxicity of microcapsules

7

Viability of cells treated with doxycycline solution, synthetic and biodegradable capsules (w/o

8

doxycycline encapsulated) is shown in Figure 4. Compared to the control group, cells treated

9

with doxycycline solution (2.5 µg/ml) had a viability of 89% after 48 hours incubation. With

10

both synthetic and biodegradable capsules, over 80% cell viability was observed at the same time

11

point. In addition, no significant difference was observed between empty and doxycycline loaded

12

capsules which was consistent with previous studies of capsule cytotoxicity in vitro and in vivo.4,

13

38, 39

16 Environment ACS Paragon Plus

Page 17 of 37

w/o doxy with doxy 100

80

viability, %

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

Biomacromolecules

60

40

20

0

control

syn capsules

bio capsules

1

Figure 4. Microcapsules cytotoxicity. Cell viability of engineered C2C12 cells incubated with

2

doxycycline solution, synthetic and biodegradable capsules with and without (w/o) doxycycline.

3

Values are the mean of six values and vertical bars represent the SD.

4

Induced expression of EGFP in C2C12 cells

5

When reporter cells were treated with doxycycline solution or encapsulated doxycycline the

6

different cell delivery routes impacts on the kinetics of EGFP expression from the tetracycline

7

responsive promoter (Ptet). The promoter is activated by rtTA-2SM2 in the presence of

8

doxycycline and is actively switched off upon removal of doxycycline through binding of the

9

repressor tetRKRAB. This tet-on system has shown high gene regulation efficiency in our

10

previous study,40 and this was confirmed here. Confocal images were taken to assess whether

11

doxycycline loaded capsules were internalized by myoblast cells and subsequently induce EGFP

12

expression in C2C12 cells via the tet-on system. Synthetic capsules were incubated with cells for

13

2 days at a ratio of 10 capsules per cell, and the same amount of free doxycycline was used as

14

control. As illustrated in Figure 5, EGFP expression was successfully switched on by

15

doxycycline solution with homogenous fluorescence in every cell (Figure 5A). However, when

16

delivered as cargo by microcapsules, doxycycline was released after cell uptake of capsules, so

17

cells without internalized capsules had no fluorescence as indicated by red circles in Figure 5B.

17 Environment ACS Paragon Plus

Biomacromolecules

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 37

1

The results highlighted that doxycycline delivery from microcapsules can be targeted to specific

2

cells via capsule internalization which then function in situ.

3

To evaluate the efficiency of doxycycline delivered by capsules, cells were classified into four

4

categories (Green/red, green, red and none – see methods for details) and the number of cells in

5

each category was counted within the panoramic confocal image (Supplementary Figure S6), and

6

was presented as a percentage in Figure 6. The capsule uptake ratio by cells was about 70%, and

7

doxycycline capsules induced EGFP expression was as high as 61.8% (shown as green and red in

8

the histogram), but there were a small proportion of these cells with red microcapsules that were

9

devoid of EGFP expression (8.9%). When no capsules were engulfed 18% of cells were not

10

green as expected, but 11% were fluorescent. There are potential explanations for this

11

observation, firstly, capsules were not visible in these cells (hidden under the nucleus); secondly,

12

after cell division a daughter cell could contain EGFP but not a microcapsule from the parent

13

cell; alternatively doxycycline could leak from extracellular microcapsules located nearby or in

14

other cells. A similar phenomenon was observed when EGFP was delivered as cargo in capsules

15

to A549 cells, when cells with no engulfed capsules had EGFP in the cytosol.41 As for the non-

16

fluorescent cells that had internalized microcapsules, the absence of fluorescence may be due to

17

capsules being devoid of doxycycline or containing an amount of doxycycline below the

18

threshold to induce EGFP expression, though the baseline for doxycycline to switch on gene

19

expression is known to be very low (1 pg/ml).40 Previous studies have tried to chemically modify

20

doxycycline to achieve spatial and temporal gene expression regulation,42,

21

doxycycline was inactivated in a ‘cage’ so that transcription was only promoted after photo-

22

activation. But, delivery by versatile microcapsule carriers is a preferable approach as minimum

18 Environment ACS Paragon Plus

43

in which

Page 19 of 37

1

modification to molecular structure is introduced and side products of chemical reactions are not

2

generated.

3 4

Figure 5. Images of doxycycline induced EGFP expression. Cells treated with a solution of

5

doxycycline (A) and doxycycline-loaded synthetic capsules (B) (capsules to cells ratio 10:1,

6

incubate time 48 hours, red circles indicate cells that have no internalized capsules or EGFP

7

expression) (scale bar of 50 µm).

70

60

50

percentage/ %

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

Biomacromolecules

40

30

20

10

0 red/green

green

red

none

color of cells

19 Environment ACS Paragon Plus

Biomacromolecules

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 20 of 37

1

Figure 6. Correlation of microcapsule delivery and EGFP expression in C2C12 cells. After

2

incubation with doxycycline-loaded synthetic microcapsules that were labelled red due to use of

3

TRITC labelled PAH cells exposed to doxycycline express EGFP. The distribution of cells in the

4

four categories (Red/green: cells with microcapsules and expressing EGFP; Green: cells

5

expressing EGFP but have no capsules; Red: cells with microcapsules have no EGFP expression;

6

None: neither red or green) was determined by microscopic examination with results expressed

7

as a percentage of all cells (total number 459).

8

Kinetics of EGFP expression

9

To demonstrate the sustained effect of doxycycline delivered by microcapsules, the kinetics of

10

EGFP expression was monitored by fluorescence intensity change in long-term experiments.

11

Four types of capsules, synthetic (PAH/PSS) and biodegradable (DS/Parg) were compared with

12

their lipid coated counterparts. EGFP fluorescence intensity at specific time intervals was

13

measured for each type of capsule. Cells treated with free doxycycline had the maximum

14

fluorescence intensity at day 2 and declined to baseline values after 10 days (Figure 7a). For all

15

the capsule-treated cells, the fluorescence intensity increased and peaked on the fourth day, then

16

fluorescence decreased gradually, and there were significant differences compared with the free

17

doxycycline treated cells after the sixth day (p< 0.05). Doxycycline molecules are lipophilic

18

enabling them to rapidly enter cells and induce gene expression, while microcapsules enter cells

19

following membrane adherence, phagocytosis and macropinocytosis mechanisms, and are finally

20

targeted to phagolysosomes38, 44 doxycycline released from microcapsules in these compartments

21

can still penetrate lipid membranes but clearly the kinetics of nuclear delivery are slower than the

22

free doxycycline. For the synthetic capsules, DPPC coating resulted in elevated and prolonged

23

EGFP expression compared with the uncoated microcapsules (p