- Email: [email protected]
A multifunctional mesoporous silica nanocomposite for targeted delivery, controlled release of doxorubicin and bioimaging
Accepted Manuscript Title: A multifunctional mesoporous silica nanocomposite for targeted delivery, controlled release of doxorubicin and bioimaging Author: Meng Xie Hui Shi Zhen Li Haijun Shen Kun Ma Bo Li Song Shen Yi Jin PII: DOI: Reference:
S0927-7765(13)00256-7 http://dx.doi.org/doi:10.1016/j.colsurfb.2013.04.009 COLSUB 5737
To appear in:
Colloids and Surfaces B: Biointerfaces
Received date: Revised date: Accepted date:
25-1-2013 25-3-2013 12-4-2013
Please cite this article as: M. Xie, H. Shi, Z. Li, H. Shen, K. Ma, B. Li, S. Shen, Y. Jin, A multifunctional mesoporous silica nanocomposite for targeted delivery, controlled release of doxorubicin and bioimaging, Colloids and Surfaces B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.04.009 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 2
A multifunctional mesoporous silica nanocomposite for targeted
4
delivery, controlled release of doxorubicin and bioimaging
ip t
3
5
Meng Xiea, Hui Shib, Zhen Lia, Haijun Shena, Kun Maa, Bo Lia, Song Shena and Yi
7
Jina*
8
a
9
Hangzhou, 310058, China b
us
Institute of Pharmaceutics, College of Pharmaceutical Sciences, Zhejiang University,
an
10
cr
6
Department of Pharmacy, Zhejiang Medical College, Hangzhou, 310053, China
M
11 12
* Corresponding author: Yi Jin
14
Address: 866 Yuhangtang Road, College of Pharmaceutical sciences, Zhejiang
15
University, Hangzhou, 310058, China.
16
Tel/fax: +86-571-88208435
18 19
te
Ac ce p
17
d
13
E-mail: [email protected]
1
Page 1 of 34
ABSTRACT
2
In this study, a targeting drug delivery system based on mesoporous silica
3
nanoparticle (MSN) was successfully developed for anti-cancer drug delivery and
4
bioimaging. Carboxyl functionalized MSN (MSN/COOH) was firstly prepared and
5
then modified with folate as the cancer targeting moiety and a near infrared
6
fluorescent dye as labelling segment. Folate was conjugated to MSN/COOH via
7
functional polyethyleneglycol (PEG), constructing the vector MSN/COOH–PEG–FA.
8
The functionalization with carboxyl caused the pore surface of the nanocarrier more
9
negative than native MSN,
us
cr
ip t
1
an
which could provide attractive forces between the
nanoparticles and positively charged doxorubicin hydrochloride (DOX). Meanwhile,
11
the folate modification significantly enhanced the cellular uptake of the delivery
12
system compared to unmodified counterparts. Furthermore, the introduction of PEG
13
increased the water dispersibility. Besides, the modification with the near infrared
14
fluorescent dye Cy5 made the system effective for live cell and in vivo imaging.
15
Therefore, the Cy5–MSN/COOH–PEG–FA system
16
nanocarrier for simultaneous diagnosis and treatment of diseases.
18 19 20
d
te
could be a promising
Ac ce p
17
M
10
KEYWORDS: Doxorubicin, mesoporous silica nanoparticles, folate, drug delivery, bioimaging
2
Page 2 of 34
1
1.
Introduction Cancer remains a major cause of death in most countries in the world, and the
3
incidence of cancer increases with age [1]. Nowadays, nanotechnology-based drug
4
delivery systems (NDDS) are increasingly applied in cancer therapy and diagnosis.
5
Compared to conventional chemo-therapy, nanotherapeutic systems have several
6
potential advantages for cancer treatment. These include increased stability of
7
anti-cancer drugs in blood, decreased non-specific toxicity, easy modification of
8
particle surface for targeting systems, reduced resistance of P-glycoprotein (P-gp)
9
expressing cells and responses to environmental stimuli such as temperature, pH, salt,
10
and ultra-sound [2-5]. The mostly studied drug delivery systems nowadays are
11
polymeric micelles systems [6-9], lipid-based drug delivery systems [10-13],
12
polymeric nanoparticles [14-19],
M
an
us
cr
ip t
2
etc.
However, during the fabrication of these systems, the use of certain organic
14
solvents and additives are unavoidable, which may bring the bioactive substances
15
with undesirable side effects. Meanwhile, these organic materials often suffer from
16
physical instabilities and non-ideal drug encapsulation and delivery performance [12].
21
Ac ce p
te
d
13
22
specific surface area, large pore volume, tunable pore structures and excellent
23
physicochemical stability [20-28]. Previously, it was observed that mesoporous silica
24
nanoparticle (MSN) could be effectively endocytosed and was able to escape the
25
endolysosomal entrapment [29-32]. These new developments rendered the possibility
17 18 19 20
In contrast, porous inorganic materials, such as mesoporous silica materials, can easily be synthesized without using any of the above mentioned organic solvents or additives under very mild conditions. Mesoporous silica nanoparticles (MSN) possess not only some common properties such as facile multifunctionalization, excellent biocompatibility and biodegradability, but also several unique features such as high
3
Page 3 of 34
1
of designing a new generation of drug delivery system for intracellular controlled
2
release and imaging applications. Despite the great progresses in using MSNs as drug delivery vehicles, a precise
4
control over the release kinetics of the loaded drug by decorating the pore surface
5
with functional groups is challenging [33], and often an initial burst effect is observed
6
during release. Besides, there is still plenty of work to be done to study the
7
distribution behavior of MSN in vivo. In this work, we have constructed a
8
multifunctional mesoporous silica nanocomposite (Cy5–MSN/COOH–PEG–FA) for
9
targeted delivery, controlled drug release and bioimaging.
us
cr
ip t
3
an
The weak base type drug
doxorubicin hydrochloride (DOX) was used as the model drug, as many anti-cancer
11
drugs in clinic are positively charged at physiological conditions. The nanocomposite
12
consisted of three parts. Firstly, carboxyl functionalized MSN (MSN/COOH) was
13
prepared as the drug carrier. The functionalization with carboxyl caused the pore
14
surface of the nanocarrier more negative than native MSN, which made it appropriate
15
to load weak base type drug DOX. So, MSN/COOH has a high drug loading capacity
16
with noticeable pH-sensitive release behavior for DOX molecules. Secondly, folate
21
Ac ce p
te
d
M
10
22
dye Cy5 was connected to the nanocomposite as the labeling segment. The
23
modification made the system effective for live cell and in vivo imaging. To our
24
knowledge, this is the first attempt to prepare negative charge functionalized
25
mesoporous silica nanoparticles with both active targeting and near-infrared
17 18 19 20
was employed as the cancer targeting moiety. MSN/COOH and folate were conjugated by a hydrophilic spacer PEG 2000. The introduction of PEG increased the water
dispersibility and that folate modification significantly enhanced the cellular
uptake of the drug delivery system in human nasopharyngeal carcinoma cells (KB cells) compared to unmodified counterparts. What’s more, a near infrared fluorescent
4
Page 4 of 34
1
fluorescent dye labeling. This Cy5–MSN/COOH–PEG–FA system
could be a
2
promising nanocarrier for simultaneous diagnosis and treatment of diseases.
3 4
2.
5
2.1 Materials
6
Cetyltrimethylammonium bromide (CTAB, 98%), tetraethyl orthosilicate (TEOS,
7
99%), 2-cyanopropyltriethoxysilane (CPTES, 99%), N,N’-Dicyclohexylcarbodimide
8
(DCC), N-(3-Dimethylaminopropyl)-N’-ethylcarbodiimide hydrochloride (EDC),
9
2-N-morpholino-ethanesulfonic acid (MES), N-Hydroxysulfosuccinimide sodium salt
10
(sulfo-NHS), Folate were purchased from Sigma-Aldrich. Cy5-hydrazide was
11
purchased from Lumiprobe. Doxorubicin hydrochloride (DOX) was purchased from
12
Dalian Meilun Biology Technology Co. Ltd. (Dalian, China). NH2–PEG–NH2 and
13
NH2–PEG–OH (weight average MW: 2000) were purchased from Yarebio (Shanghai,
14
China). Lyso-tracker Red and DAPI were purchased from Beyotime (Shanghai,
15
China). Water was purified by distillation and deionization (MilliQ Plus). All other
16
chemicals were of analytical grade and used without further purification.
21
Ac ce p
te
d
M
an
us
cr
ip t
Materials and methods Materials
22
vigorous stirring for about 30 min at 60 °C, 2.0 mL TEOS and 0.4 mL
23
2-cyanopropyltriethoxysilane were rapidly added to the mixture. The resulting
24
mixture was stirred for another 2 h at 60 °C and was then kept statically at the same
25
temperature for 24 h. Samples were collected by centrifugation at 20000 rpm for 20
17 18 19 20
2.2 Synthesis of carboxyl functionalized MSN (MSN/COOH) Mesoporous silica nanoparticles functionalized with carboxyl was synthesized by co-condensation. In a typical synthesis, 1.2 g CTAB was dissolved in a solution containing water (180 mL) and ammonia aqueous solution (5.5 mL, 25%). After
5
Page 5 of 34
min, washed, and re-dispersed with deionized water and ethanol several times. Then,
2
the dried product was treated with 9 mol·L-1 H2SO4 solution at 100 °C to produce
3
carboxyl functionalized MSN. At last, the surfactant templates CTAB were removed
4
by extraction in acidic ethanol (about 9 mL HCl in 100 mL ethanol at 65 °C for 24 h),
5
and separated by centrifugation at 20000 rpm to obtain the product. The synthesized
6
nanoparticles were freeze-dried for further use.
cr
ip t
1
7
2.3 Synthesis of NH2–PEG–FA
9
The carboxylic group of folate was firstly activated by DCC and NHS. Briefly, 200
10
mg of folate dissolved in 10 ml of dimethylsulfoxide (DMSO) was reacted with 180
11
mg of NHS and 100 mg of DCC at room temperature for 12 h (folate/NHS/DCC
12
molar ratio=1:2:2). The activated folate (90 mg) was then reacted with 400 mg of
13
NH2–PEG–NH2 (weight average MW: 2000, PEG/folate molar ratio=1/1.1) dissolved
14
in 5 ml of DMSO. The reaction was performed at room temperature for 24 h. The
15
resultant solution was diluted with 20 ml of deionized water and centrifuged for three
16
times. Then the supernatant was purified by dialysis with a 10000 MW cutoff
21
Ac ce p
te
d
M
an
us
8
22
10 mL isotonic 0.1 M MES saline buffer pH 5.5, and then reacting them with EDAC
23
(10 equiv.) and NHS (10 equiv.) for 1 h. MES buffer was used to slow down
24
hydrolysis of the NHS esters on MSN/COOH, as it lacks amino and carboxyl groups,
25
which could compete in the reaction. The NPs were then centrifuged to remove excess
17 18 19 20
membrane and the dialyzed solution was finally freeze-dried. All the processes were carried out in the dark.
2.4 Surface coating of MSN/COOH with NH2–PEG–FA Carboxyl on the surface of the NPs (60 mg) were activated by redispersing the NPs in
6
Page 6 of 34
EDAC/NHS and the water-soluble isourea byproduct. Activated NPs were
2
re-dispersed in 10 mL PBS buffer and reacted with 40 mg of NH2–PEG–FA or
3
NH2–PEG–OH (weight average MW: 2000) for 24 h. The coated NPs were
4
centrifuged and washed with PBS buffer for three times to remove any unbound PEG.
5
The dialyzed solution was finally freeze-dried. All the processes were carried out in
6
the dark.
cr
ip t
1
7
2.5 Synthesis of fluorescent nanoparticles
9
Cy5-hydrazide was reacted with the carboxyl on the surface of the nanoparticles via
10
EDC/NHS coupling method to fabricate fluorescent nanoparticles for imaging. A total
11
of
12
MSN/COOH–PEG–FA) were re-dispersed in 5 mL MES saline buffer (0.1 mol·L-1,
13
pH 5.5) and activated using EDC (19.1 mg) and sulfo-NHS (21.7 mg) for 1 h,
14
respectively. The NPs were then centrifuged to remove excess EDAC/NHS and the
15
water-soluble isourea byproduct. Activated NPs were re-dispersed in 5 mL MES
16
saline buffer and reacted with Cy5-hydrazide (0.3 mg) for 24 h. The NPs were then
of
nanoparticles
an
mg
(MSN/COOH,
MSN/COOH–PEG
or
21
Ac ce p
te
d
M
30
us
8
22
(AMF500, Bruker). Thermogravimetric analysis (TGA, Perkin-Elmer Pyris Diamond
23
TG) of MSN/COOH–PEG–FA was performed at a heating rate of 10 °C min-1 in N2.
24
The infrared (IR) spectra were measured using a JASCO FT/IR-4100 spectrometer
25
(Tokyo, Japan). The UV–Vis absorption spectra were performed using a Persee
17 18 19 20
centrifuged and washed with deionized water for three times to remove any unbound Cy5-hydrazide and freeze-dried for further use.
2.6 Characterization of the nanoparticles The structure of NH2–PEG–FA was measured by 1H NMR in DMSO-d6 at 500 MHz
7
Page 7 of 34
TU-1800PC spectrophotometer (Beijing, China). The pore characteristics of the
2
samples were studied by determining the nitrogen adsorption using a Quantachrome
3
Autosorb-1-C surface area and pore size analyzer (Florida, USA) at -196 °C. The
4
morphology and structure of MSN/COOH and MSN/COOH–PEG–FA samples were
5
characterized via transmission electron microscopy (TEM; JEOL JEM-1200EX
6
microscope, Japan). The size distributions and zeta potentials of MSN, MSN/COOH
7
and MSN/COOH–PEG–FA were induced by different pH values of aqueous solutions
8
were measured using a Malvern Zetasizer Nano-S90 (England) particle size and zeta
9
potential analyzer, respectively.
an
us
cr
ip t
1
10
2.7 Preparation of drug-loaded nanoparticles
12
Typically, 10 mg of MSN/COOH–PEG–FA were immersed in PBS solutions (5mL,
13
pH 6.5) containing 5 mg of DOX. After stirring for 24 h under light-sealed conditions,
14
the mixture was centrifuged and the supernatant was removed. The drug-loaded
15
MSN/COOH–PEG–FA samples were washed twice with PBS solution to remove
16
doxorubicin that was adsorbed on the surface but not inside the pores. The DOX
18 19
d
te
Ac ce p
17
M
11
concentration in original solution and supernatant solution were determined via UV–Vis spectroscopy. DOX loading amount and loading efficiency in the nanoparticles were calculated by the following equations: DOXin feed - free DOX ! 100 nanopartic les in feed DOXin feed - free DOX (%, w/w) ! ! 100 DOXin feed
Loading amount(%,w/w) !
20 21
Loading efficiency
22 23
2.8 Cell culture
24
Human nasopharyngeal carcinoma cell line (KB cell line), known as folate receptor
8
Page 8 of 34
positive tumor cell, was obtained from the institute of biochemistry and cell biology
2
of Chinese academy of sciences (IBCB, Shanghai, China). Cells were grown at 37 °C
3
and 5% CO2 in RPMI 1640 medium supplemented with 10% (v/v) FBS. For all
4
experiments, cells were harvested by the use of D-Hank’s-trypsin solution and
5
redispersed in fresh medium before platinge.
ip t
1
cr
6
2.9 Intracellular distribution study
8
The cellular uptake and intracellular distribution experiments in KB cells were
9
performed using CLSM (CarlZeiss LSM 710 NLO, Dresden, Germany). The cells
10
were seeded into six-well plates at a density of 2×105 cells per well medium
11
respectively. After a 24 h incubation period, the cells were treated with Cy5 labeled
12
nanoparticles or DOX (free DOX in solution or DOX-loaded nanoparticle with a
13
DOX concentration of 10 µg mL-1). After incubation for 1 h in folate-free medium,
14
the culture medium was removed. The cells were then carefully washed three times
15
with PBS to remove any free DOX or nanoparticles and fixed with 4% (w/v)
16
paraformaldehyde for 0.5 h. The lysosomes were stained with Lyso-tracker Red and
21
Ac ce p
te
d
M
an
us
7
22
cells were cultured for another 1 h, washed with sterilized PBS, stained with
23
Lysotracker Red and DAPI, and then analyzed by CLSM.
17 18 19 20
the nuclei were stained with DAPI according to the operation manuals. Then, the cells were washed three times with PBS. The intracellular localization of was observed with CLSM.
To further evaluate the role of FA in the cellular uptake, the cells were pretreated with free FA (0.5 mg mL-1) for 2 h, Cy5 labeled nanoparticles were then added and the
24 25
2. 10
Cell viability assay
9
Page 9 of 34
In vitro cytotoxcity of blank and DOX loaded nanoparticles against KB cells were
2
assessed by the standard MTT assay. Cells were seeded in 96-well plates at a density
3
of 1×104 cells per well and allowed to attach overnight. After a 24 h incubation period
4
at 37 °C in 100 µL RPMI medium containing 10% FBS, culture medium was
5
discarded and then cells were treated with pH 7.4 PBS solution of free DOX,
6
drug-loaded or blank nanoparticles at various concentrations. After the incubation in
7
folate-free medium for 24 h or 48 h, culture medium was removed, 31.5 μL of
8
5mg·mL-1 MTT solution was added, and cells were incubated for another 4 h. Then,
9
media was removed, the MTT crystals were dissolved in DMSO (0.1mL·well-1), and
10
the optical density of each sample was read at 570 nm on an Automated Microplate
11
Reader 550 (Bio-Rad, USA). A culture medium without particles was used as the
12
blank control. The cytotoxicity was expressed as the percentage of the cell viability as
13
compared with the blank control. The time and dose dependences of the cytotoxicity
14
were investigated at different particle concentrations.
15
2.11
Intravital near-infrared (NIR) optical imaging of tumor-bearing mice
21
Ac ce p
16
te
d
M
an
us
cr
ip t
1
22
tumor-bearing mice were treated with Cy5 labeled MSN/COOH–PEG–FA via the
23
intratumor injection. The mice were anesthetized and placed on an animal plate. The
24
fluorescent scans were performed at various time points using the Maestro in-vivo
25
imaging system (Cambridge Research & Instrumentation, CRi, USA).
17 18 19 20
In order to observe the real-time distribution and tumor accumulation ability of near-infrared dye Cy5 labeled nanoparticles in mice bearing KB xenografts, non-invasive optical imaging systems were utilized. Briefly, male Balb/c nude mice of 4 weeks were injected with 0.2 mL of a cell suspension containing 5×106 KB cells subcutaneously. After 15 days, when the tumor reached 0.8×0.8 cm in diameter, the
10
Page 10 of 34
1 2
2.12
3
All experiments were run in triplicate and the acquired data are expressed as mean ±
4
SD. Statistical significance was determined using Student’s t-test.
ip t
Statistical analysis
5 6
3.
7
3.1 Synthesis of FA modified system
8
The procedure for the synthesis of mesoporous silica nanoparticles functionalized
9
with carboxyl (MSN/COOH), folate and near-infrared dye Cy5 was illustrated in
10
scheme 1. The functionalization was performed by a four-step surface modification
11
process. 1. The pore surface of the MSN was functionalized by co-condensation with
12
CPTES to form MSN/CN; 2. MSN/CN was hydrolyzed with 9 mol·L-1 H2SO4 solution
13
at 100 °C to yield the carboxyl functionalized MSN (MSN/COOH); 3. Cy5-hydrazide
14
was reacted with the carboxyl on the surface of MSN/COOH to fabricate fluorescent
15
nanoparticles; 4. MSN/COOH–Cy5 was reacted with NH2–PEG–FA to form folate
16
modified nanoparticles (MSN/COOH–PEG–FA).
us
an
M
d
Ac ce p
te
21 22
biocompatible.
17 18 19 20
cr
Results and discussion
MSN/COOH–PEG–FA was synthesized according to the method outlined in fig.
1. Carboxyl (–COOH) of folate and carboxyl of MSN/COOH were connected through the end-amino hydrophilic spacer, NH2–PEG–NH2. PEG was chosen as a linker in order (1) to get optimized length of linker to reach accessible receptor sites, (2) to increase the water dispersibility, and (3) to make the folate-conjugate system more
23 24
3.2 Characterization of MSN/COOH–PEG–FA
25
The synthesis of NH2–PEG–FA was carried out as shown in fig. 1. Folate was firstly
11
Page 11 of 34
conjugated to NH2–PEG–NH2 following the usual NHS/DCC reaction to get
2
NH2–PEG–FA. The formation was confirmed by the 1H NMR spectrum, as shown in
3
fig. S1 (ESI). In the 1H NMR spectrum, the peak at 8.67 ppm belongs to the proton on
4
the C7 of folate. The peaks at 6.65 ppm (3’, 5’-H, 2H) and 7.66 ppm (2’,6’-H, 2H)
5
belong to the protons on the benzene ring. The peaks at 1.9, 2.1–2.4 and 4.3 ppm
6
originate from the α, β and γ –CH2– protons of glutamic acid part in the folate
7
structure. The peak at 3.6 ppm belongs to the –CH2–protons of PEG.
cr
ip t
1
MSN/COOH–PEG–FA was prepared from the reaction between MSN/COOH and
9
NH2–PEG–FA using EDC between amino group of NH2–PEG–FA and acid group of
10
MSN/COOH. IR spectra of pure MSN/CN, MSN/COOH and MSN/COOH–PEG–FA
11
were shown in fig. S2 (ESI). All three samples showed the same peak at 1630 cm-1,
12
which attributed to H2O adsorbed in the mesopores. The IR spectrum of MSN/COOH
13
samples revealed a peak at 1718 cm-1, the stretching vibration frequency of C=O of
14
carboxyl group. The appearance of the new intensive peak at 2927 cm-1 was assigned
15
to the C–H stretching frequencies of –CH2–CH2– groups attributed to PEG, which
16
indicated the successful attachment of NH2–PEG–FA onto the surface of MSN/COOH.
21
Ac ce p
te
d
M
an
us
8
22
performed under a constant N2 flow, as shown in fig. S3 (ESI). Upon heating, the
23
materials respectively underwent a total weight loss of 16.0 % (H1) and 28.7 % (H2)
24
up to 800 °C. The graft ratio (W) in MSN/COOH–PEG–FA was 15.05 %. The
25
calculating process was shown in supporting information, following literature
17 18 19 20
Both MSN/COOH and MSN/COOH–PEG–FA samples had no peak at 2250 cm-1, a peak appeared in MSN/CN IR spectrum attributed to the stretching vibration frequency of C≡N of cyano group, indicating the completion of hydrolysis of CN catalyzed by H2SO4 acid.
TGA of CTAB-extracted MSN/COOH and MSN/COOH–PEG–FA were
12
Page 12 of 34
1
procedures [34] with slight change. Cy5-hydrazide was reacted with the carboxyl group on the surface of
3
MSN/COOH–PEG–FA via EDC/NHS coupling method to fabricate fluorescent
4
nanoparticles for imaging. Compared with the co-condensation method [35], this
5
coupling method does not compromise the framework structure of the parent
6
mesoporous materials. The Conjugation of Cy5-hydrazide was confirmed by UV–Vis
7
absorption spectroscopy. As shown in fig. S4 (ESI), Cy5-hydrazide in water exhibited
8
the main absorption bands at 639 nm. The band was present in the spectrum of
9
Cy5–MSN/COOH–PEG–FA, indicating conjugation of Cy5. Also, upon conjugation
10
of Cy5 with MSN/COOH–PEG–FA, the peak was slightly red shifted to 645 nm,
11
indicative of conjugation between Cy5 and MSN/COOH–PEG–FA.
M
an
us
cr
ip t
2
Figure 2A shows the nitrogen ad/desorption isotherms of MSN, MSN/COOH and
13
MSN/COOH–PEG–FA. All three samples showed type IV isotherms according to the
14
IUPAC classification with well-defined steps at relative pressures (P/P0) of 0.2–0.4,
15
which was the typical characteristic of mesoporous silica. This suggested that the
16
functionalization with carboxyl and folate did not change the structure of mesoporous
21
Ac ce p
te
d
12
22
surface-shielding effect of organic materials.
17 18 19 20
channels. The pore-size distribution was determined by the Barret–Joyner–Halenda (BJH) method (fig. 2B). With the introduction of carboxyl and folate, the capillary condensation step (P/P0) shifted to lower relative pressure and consistently the pore diameter decreased from 2.8 to 2.3 nm. Meanwhile, the surface area and pore volume are smaller after functionalization (table S1, ESI). These may be due to the
23 24 25
3.3 Properties of the MSN/COOH-PEG-FA The size and morphology of the MSN/COOH and MSN/COOH–PEG–FA were
13
Page 13 of 34
characterized by TEM (fig. 3). The particle size distribution of the nanoparticles was
2
found to be relatively narrow and quite uniform. The mean particle size of the
3
nanoparticles determined by dynamic light scattering (DLS) measurement was 63.5 ±
4
3.4 nm (fig. 3(c)), which was consistent with its diameter observed by TEM. The
5
physical properties of the NPs before and after the functionalization were given in
6
table S2 (ESI). After the introduction of NH2–PEG–FA, the average size increased 5
7
nm. Compared with MSN/COOH, the size distribution of MSN/COOH–PEG–FA was
8
narrower. There were no obvious differences in the mean particle size and size
9
distribution between MSN/COOH–PEG and MSN/COOH–PEG–FA.
an
us
cr
ip t
1
As illustrated in scheme 1, carboxyl incorporated on mesopore surfaces can make
11
the pore surface of the NPs more negative. Figure S5 (ESI) shows the zeta potential of
12
native and functionalized MSN at various pH levels. As expected, the zeta potential of
13
MSN/COOH was always lower than that of native MSN. At pH 7.4, the zeta potential
14
of native MSN was -9.0 mV, while at pH 6.5, the zeta potential was around zero.
15
However, due to the carboxyl, functionalized MSN showed highly negative zeta
16
potentials in the pH range from 7.4 (-24.5mV) to 5.0 (-13.6 mV) and there was also
21
Ac ce p
te
d
M
10
22
temperature in physiological condition (PBS, pH 7.4) for 24 h (table S2, ESI). It was
23
observed that native MSN showed obvious precipitation. However, the nanoparticle
24
size distributions of MSN/COOH and MSN/COOH–PEG–FA did not change
25
significantly, which was consistent with the results of zeta potential study.
17 18 19 20
no significant difference in the zeta potentials before and after the surface coating with NH2–PEG–OH or NH2–PEG–FA. Surface zeta potential gave an indication of the potential stability of the colloidal
system. The stability of the native and functionalized MSN in aqueous media were evaluated by comparing the visual appearance and average size after stored at room
14
Page 14 of 34
1 2
3.4 Doxorubicin loading and release The utility of the MSN/COOH–PEG–FA as an intracellular carrier for anticancer
4
drug delivery system was evaluated using doxorubicin hydrochloride (DOX) as a
5
model anticancer drug. The loading of DOX into the nanoparticle was performed by
6
diffusion and electrostatic attraction. The pKa of DOX is 8.2 while the nanoparticles
7
carry negative charges when the solution pH is higher than 3.0 (Fig. S5, ESI). The
8
electrostatic attraction between DOX and the nanoparticles happens when the solution
9
pH changes from weak acidic to neutral or weak alkaline condition, which facilitates
10
the DOX loading. Consequently, it was found that the DOX loading content and
11
encapsulation efficiency in MSN/COOH–PEG–FA reached 24.0 ± 0.4 % and 62.4 ±
12
1.0 %, respectively, indicating the MSN/COOH–PEG–FA was indeed an effective
13
DOX loading vehicle.
d
M
an
us
cr
ip t
3
To prove that the DOX molecules could penetrate into the nanopores of the
15
functionalized nanoparticles, BET analysis was also carried out. The change of
16
specific surface area, pore size and pore volume distribution of the nanoparticles
21
Ac ce p
te
14
22
investigated under a simulated physiological condition (PBS, pH 7.4) and in an acidic
23
environment (acetate buffer, pH 5.0 or 6.5) at 37 °C to assess the feasibility of using
24
MSN/COOH as an anticancer drug delivery carrier (fig. 4). Free DOX solution was
25
used as a control. This study clearly showed that the pH of the medium had a strong
17 18 19 20
before and after DOX loading are summarized in table S1 (ESI). As the data clearly indicated,
the
BET
surface
area,
pore
volume
and
pore
diameter
of
MSN/COOH–PEG–FA were all decreased significantly after the loading of DOX, indicating penetration of the DOX molecules into the nanopores. The release patterns of DOX from MSN/COOH–PEG–FA nanoparticles were
15
Page 15 of 34
1
effect on the
2
nanoparticles. The drug release at pH 7.4 was considerably slow and only 40 % of the
3
DOX
4
MSN/COOH–PEG–FA nanoparticles maintained electrostatic interactions between
5
the DOX molecules and the carboxyl under physiological conditions. However, it is
6
interesting to note that the amount of released DOX at pH 5.0 reached to
7
approximately 90%. This phenomenon can be explained by the fact that the
8
electrostatic interactions were decreased as well as the electrostatic repulsion
9
enhanced between the nanoparticles and DOX in an acidic environment, thus the
released
after
8
h.
This
result
suggested
that
us
cr
ip t
were
an
10
molecules
release quantity of DOX from the MSN/COOH–PEG–FA
release amount of DOX increased.
The pH dependent release property of the nanoparticles may benefit the DOX
12
release in tumor cells whose pH is lower than that of the normal cells. It is the most
13
desirable that most DOX molecules encapsulated in the nanoparticles would remain in
14
the mesoporous for a considerable long period of time when the nanoparticles stayed
15
in the plasma at normal physiological conditions (pH 7.4), thus elongating the
16
circulation period of the drug and reducing the side effects. When the nanoparticles
18 19 20 21
d
te
Ac ce p
17
M
11
reached the tumor sites and were internalized inside the tumor cells by endocytosis, faster release might occur inside the endosome/lysosome of tumor cells because of the low pH values, thus enhanced the delivery efficiency.
3.5 In vitro toxicity studies
22
The cytotoxicity of KB cells exposed to
CTAB-extracted nanoparticles was
23
investigated, and the effects of the particle concentration on the cell proliferation
24
activity were also studied. As shown in fig. S6 (ESI), the cell proliferation was not
25
hindered after the cells were exposed to the nanoparticles for 48 h at particle
16
Page 16 of 34
1
concentrations of not higher than 250 µg mL-1. However, the hindrance was indeed
2
observed at the particle concentration 500 µg mL-1. The very low cytotoxicity
3
be attributed to the relatively low concentrations of the CTAB molecules in the
4
nanoparticles. Therefore, MSN/COOH–PEG–FA with the concentration less than 250
5
µg mL-1 was safe for cells within 48 h.
ip t
may
cr
6
3.6 Intracellular uptake of nanoparticles
8
Confocal laser-scanning microscopy (CLSM) was used to compare endocytosis of
9
MSN/COOH–PEG–FA with that of MSN/COOH–PEG using a folate-receptor
10
positive cancer cell line, KB cell line, in the FA-free incubation medium. A
11
near-infrared dye Cy5-hydrazide was reacted with the carboxyl of the nanoparticles to
12
fabricate fluorescent nanoparticles for imaging. The red fluorescence indicated the
13
distribution of Cy5–MSN/COOH–PEG or Cy5–MSN/COOH–PEG–FA, the green
14
fluorescence indicated the distribution of lysosomes which were stained by
15
Lyso-tracker red, and the blue indicated the nuclei stained by DAPI. As shown in Fig.
16
5A, cells exposed to MSN/COOH–PEG showed a weak red fluorescence, and it was
18 19 20 21
an
M
d
te
Ac ce p
17
us
7
observed mainly on the surface of the cells. The failure of MSN/COOH–PEG delivery was partly attributed to the negative charge of the supported MSN/COOH–PEG, which is repelled by the negatively charged cell surface. Besides, after PEGylation, the hydrophilic property of the nanoparticles enhanced, which was against the cellular uptake.
22
As reported, the folate receptor is overexpressed in several types of human cancer,
23
such as brain, kidney, lung, and breast. The FR ligand, folate, is a vitamin that is used
24
for the biosynthesis of nucleotides and is utilized in high levels to meet the needs of
25
proliferating cancer cells. In order to further improve drug delivery efficiency,
17
Page 17 of 34
MSN/COOH–PEG–FA was designed. After incubation for 1 h, the folate conjugated
2
nanoparticles were mostly localized in lysosomes leaving a clear zone of nucleus, as
3
evidenced by colocalization of green fluorescence and red fluorescence, indicating
4
cellular uptake instead of adhesion to the surface and the nanoparticles.
ip t
1
To further evaluate the role of folate in the cellular uptake of nanoparticles, the
6
folate receptor was blocked on the surface of KB cells by pre-treating KB cells with
7
free folate for 2 h. As shown in Fig. 5B, there was little difference in the cellular
8
uptake of MSN/COOH–PEG regardless of the addition of folate in the medium.
9
However, when KB cells were incubated with MSN/COOH–PEG–FA, much weaker
10
red fluorescence was observed in the folate medium than in the folate free medium.
11
This result demonstrated that the uptake of MSN/COOH–PEG–FA and folate is
12
competitive,
13
receptor-mediated endocytosis. This observation clearly inferred that folate
14
conjugated nanoparticle was very much effective as a delivery system for targeted
15
anticancer drugs.
17 18 19 20
us
an
M
MSN/COOH–PEG–FA were
internalized
via
the
folate
te
d
thus
Ac ce p
16
cr
5
3.7 Intracellular distribution of DOX To determine whether encapsulation of DOX in MSN/COOH–PEG–FA affected its trafficking
inside
cells,
the
intracellular
distribution
of
free
DOX,
DOX–MSN/COOH–PEG and DOX–MSN/COOH–PEG–FA in KB cells were further
21
estimated. Since DOX has red fluorescence, its distribution in the KB cells can be
22
easily observed. As shown in fig. 6, after 1 h of incubation with DOX solution in KB
23
cells, strong fluorescence was emitted from the cells, particularly the nuclear regions,
24
indicating a high level of DOX molecules in the nucleus. The intense DOX
25
accumulation in the nucleus for free DOX occurred because intracellular DOX 18
Page 18 of 34
1
molecules in the cytosol could transport rapidly to the nucleus and avidly bound to
2
the chromosomal DNA [36]. In the case of the DOX–MSN/COOH–PEG and DOX–MSN/COOH–PEG–FA,
4
the fluorescence intensity were weaker, indicating that the DOX loaded nanoparticles
5
were initially located within the endosomal intracellular compartments, releasing
6
loaded DOX in the cytosol region in a sustained manner, and then diffused through
7
the endocytic compartment membrane to enter the nucleus. However, the
8
internalization of DOX–MSN/COOH–PEG–FA wasmore obvious compared with the
9
DOX–MSN/COOH–PEG in the absence of free folate in the cell culture medium.
10
This result directly confirms that the cellular uptake of the nanoparticles can be
11
enhanced by attaching folate on their surface, which was in accord with the
12
intracellular uptake of blank nanoparticles.
13
3.8 Cytotoxicity of DOX-loaded nanoparticles
14
DOX is a well-known cytotoxic anticancer drug even at a very low concentration.
15
Since DOX–MSN/COOH–PEG–FA has a sustained DOX release behavior and a
17 18 19 20
cr
us
an
M
d
te
Ac ce p
16
ip t
3
remarkable uptake by KB cells, the drug delivery system was expected to have high cytotoxicity against cancer cells. To verify the pharmacological activity of released DOX from DOX–MSN/COOH–PEG–FA, the cytotoxicity of free DOX, as well as DOX loaded nanoparticles containing the equal molar DOX were investigated using cell viability assay by comparing with the control group, in which neither DOX
21
loaded nanoparticles nor free DOX were added. As shown in fig. S7 (ESI), KB cells
22
demonstrated dose-dependent cytotoxicity for both free DOX and DOX loaded
23
nanoparticles and the cell viability of DOX–MSN/COOH–PEG–FA was equivalent to
24
that of free DOX. On the other hand, cell viability of DOX–MSN/COOH–PEG–FA
25
was much higher than that of DOX–MSN/COOH–PEG. These results clearly showed 19
Page 19 of 34
that the increased cytotoxicty of DOX–MSN/COOH–PEG–FA against KB cells is
2
most likely due to the uptake of DOX–MSN/COOH–PEG–FA via folate-receptor
3
medited endocytosis. Furthermore, at low DOX concentration (<0.1 µg mL-1),
4
DOX–MSN/COOH–PEG–FA showed stronger effect on killing tumor cells than free
5
DOX, while the cytotoxicity of free DOX became more obvious with the increase of
6
DOX concentration, which probably because of the slow release of DOX from
7
DOX–MSN/COOH–PEG–FA.
cr
ip t
1
3.9 In vivo imaging
an
9
us
8
To investigate the accumulation of the nanocarrier in tumor tissues, the non-invasive
11
near-infrared fluorescence (NIRF) imaging was acquired and the near-infrared dye
12
Cy5 was conjugated to MSN/COOH–PEG–FA. The in vivo imaging of
13
Cy5–MSN/COOH–PEG–FA was evaluated in KB tumor xenograft mouse model. As
14
shown in fig. 7, after the injection, fluorescent signals were observed in the tumors of
15
the mice in both Cy5–MSN/COOH–PEG–FA and free Cy5 groups, suggesting that the
17 18 19 20
d
te
Ac ce p
16
M
10
fluorescence of Cy5 was less affected by the body autofluorescence.Interestingly, in the Cy5–MSN/COOH–PEG–FA group, the intense of the fluorescence did not seem to decrease significantly as time increased. 4 days after the injection, the fluorescence intensity was still strong and mostly located on the tumor position. However, in the free Cy5 group, the fluorescence intensity became weaker as time increased. The
21
observation suggested that Cy5 labeled MSN/COOH–PEG–FA could accumulate in
22
tumor tissue for a long period of time for tumor imaging.
23
4. Conclusion
24
In this work, a novel negative-charge functionalized and folate receptor targeted 20
Page 20 of 34
mesoporous silica nanocarrier was synthesized. The experimental results clearly
2
demonstrated that DOX can easily be loaded into the nanoparticles with high loading
3
capacity. This drug delivery system was proved to possess many unique
4
characteristics: 1. enhanced stability at physiological conditions; 2. desirable release
5
acceleration at low pH conditions for cytoplasmic drug delivery; 3. beneficial to be
6
taken up by cells caused by a folate-receptor mediated endocytosis process. More
7
importantly, the drug delivery system was able to accumulate in tumor tissue for a
8
long period of time, thus could simultaneously be used for in vivo imaging. Based on
9
the features, there is the great potential of the multifunctionalized fluorescent
10
mesoporous silica nanoparticles for simultaneous imaging in the near infrared region
11
and targeted drug release.
M
an
us
cr
ip t
1
12
Acknowledgements
14
This work was supported by National Natural Science Foundation of China (No.
15
30973647) and National Basic Research Program of China (973 Program,
16
2009CB930300).
18 19 20
References [1]
J. Fang, H. Nakamura and H. Maeda, Adv. Drug Deliver. Rev., 3 (2011) 136.
[2]
L.S. Jabr-Milane, V.L Van, S. Yadav and M.M. Amiji, Cancer Treat. Rev., 34 (2008) 592.
21 22
te
Ac ce p
17
d
13
[3]
T. Bansal, N. Akhtar, M. Jaggi, R.K. Khar and S. Talegaonkar, Drug Discov. Today, 14 (2009) 1067.
23 24
[4]
J.R. Kanwar, N. Singh and R.K. Kanwar, Nanomedicine-UK, 6 (2011) 701.
25
[5]
K.T. Oh, H.J. Baik, A.H. Lee, Y.T. Oh, Y.S. Youn and E.S. Lee, Int. J. Mol. Sci., 21
Page 21 of 34
10 (2009) 3776.
1
[6]
Int. J. Nanomed., 7 (2012) 187. [7]
and Y. Jin, J Gene Med., 12 (2010) 669. [8]
604.
11 12 13
Colloids Surfaces B., 94 (2012) 266.
an
10
R. Khonkarn, S. Mankhetkorn, M. Talelli, W.E. Hennink and S. Okonogi,
[10] T. Kobayashi, T. Ishida, Y. Okada, S. Ise, H. Harashima and H. Kiwada, Int. J. Pharm., 329 (2007) 94.
[11] K. Ma, H.J. Shen, S. Shen, M.Xie, C.B. Mao, L.Y. Qiu and Y. Jin, J. Gene. Med., 13 (2011) 290.
d
9
[9]
us
7 8
Y.H. Tao, J.F. Han, X.W. Wang and H.Y. Dou, Colloids Surfaces B., 102 (2013)
cr
5 6
K. Ma, M.X. Hu, M. Xie, H.J. Shen, L.Y. Qiu, W.M. Fan, H.Y. Sun , S.Q. Chen
ip t
3 4
B. Li, H. Xu, Z. Li, M.F. Yao, M Xie, H.J. Shen, S. Shen, X.S. Wang and Y. Jin,
M
2
[12] S. Simovic, T.J. Barnes, A. Tan and C.A. Prestidge, Nanoscale, 4 (2012) 1220.
15
[13] Y.C. Kuo and C.T. Liang, Colloids Surfaces B., 85 (2011) 131.
16
[14] Y. Patil, T. Sadhukha, L.N. Ma and J. Panyam, J. Control. Release, 136 (2009)
21
Ac ce p
te
14
22
B., 88 (2011) 741.
17 18 19 20
23 24 25
21.
[15] V. Saxena, Y. Naguib and M.D. Hussain, Colloids Surfaces B., 94 (2012) 274. [16] Y.I. Jeong, D.H. Kim, C.W. Chung, J.J. Yoo, K.H. Choi, C.H. Kim, S.H. Ha and D.H. Kang, Colloids Surfaces B., 90 (2012) 28.
[17] Q.H. Gao, Q. Liang, F. Yu, J. Xu, Q.H. Zhao and B.W. Sun, Colloids Surfaces
[18] W. Xiong, W. Wang, Y. Wang, Y.B. Zhao, H.B. Chen, H.B. Xu and X.L. Yang, Colloids Surfaces B., 84 (2011) 447. [19] L. Jia, J.P. Xu, H. Wang and J. Ji, Colloids Surfaces B., 84 (2011) 49.
22
Page 22 of 34
2 3 4
[20] I.I. Slowing, J.L. Vivero-Escoto, C.W. Wu and V.S. Lin, Adv. Drug Deliver. Rev., 60 (2008) 1278. [21] J.L.Vivero-Escoto, I.I. Slowing, B.G. Trewyn and V.S.Y. Lin, small, 6 (2010) 1952.
ip t
1
[22] Q.J. He and J.L. Shi, J. Mater. Chem., 21 (2011) 5845.
6
[23] J. Rosenholm, C. Sahlgren and M. Lindén, J. Mater. Chem., 20 (2010) 2707.
7
[24] O. Tabasi, C. Falamaki and Z. Khalaj, Colloids Surfaces B., 98 (2012) 18.
8
[25] N.C. Zhang, E.Y. Ding, X. Feng, Y. Xu and H.H. Cai, Colloids Surfaces B., 89
12 13
us
an
11
[26] Z.D. Chen, X. Li, H.Y. He, Z.H. Ren, Y. Liu, J. Wang, Z. Li, G. Shen and G.R. Han, Colloids Surfaces B., 95 (2012) 274.
M
10
(2012) 133.
[27] Z. Guo, X.M. Liu, L. Ma, J. Li, H. Zhang, Y.P. Gao and Y. Yuan, Colloids Surfaces B., 101 (2013) 228.
d
9
cr
5
[28] H. Gustafsson, C. Thörn and K. Holmberg, Colloids Surfaces B., 87 (2011) 464.
15
[29] C.P. Tsai, C.Y. Chen, Y. Hung, F.H. Chang and C.Y. Mou, J. Mater. Chem., 19 (2009) 5737.
21
Ac ce p
16
te
14
22
[34] M.S. Moorthy, S.S. Park, D. Fuping, S.H. Hong, M. Selvaraj and C.S. Ha, J.
17 18 19 20
23
[30] F. Lu, S.H. Wu, Y. Hung and C. Y. Mou, Small, 5 (2009) 1408. [31] Q. He, Z. Zhang, Y. Gao, J. Shi and Y. Li, Small, 5 (2009) 2722. [32] J. Lu, Z.X. Li, J.I. Zink and F. Tamanoi, nanomed-nanotechnol., 8 (2012) 212. [33] M. Manzano, V. Aina, C.O. Arean, F. Balas, V. Cauda, M. Colilla, M.R. Delgado and M. Vallet-Regí, Chem. Eng. J., 137 (2008) 30.
Mater. Chem., 22 (2012) 9100.
24
[35] K. Ma, H. Sai and U. Wiesner, J. Am. Chem. Soc., 134 (2012) 13180.
25
[36] R. Gillies and J.M.J. Frchet, Bioconjugate Chem., 16 (2005) 361.
26 23
Page 23 of 34
1
Figures Captions
2
Scheme 1
3
MSN/COOH–PEG–FA, DOX loading, and pH dependent drug release.
4
Fig. 1
Synthesis scheme of MSN/COOH–PEG–FA.
5
Fig. 2
(A) Nitrogen adsorption-desorption isotherms and (B) pore size distribution
6
of MSN, MSN/COOH, MSN/COOH–PEG–FA and DOX–MSN/COOH–PEG–FA.
7
The inset shows the enlarged figure of DOX–MSN/COOH–PEG–FA.
8
Fig. 3
9
MSN/COOH–PEG–FA (Scale bars: 50 nm). (C) The size distribution of
of
the
synthesis
and
chemical
modification
of
cr
ip t
Illustration
an
us
(A) TEM micrographs of MSN/COOH. (B) TEM micrographs of
10
MSN/COOH–PEG–FA determined by DLS.
11
Fig. 4
12
different pH values. Results are expressed as means ± the standard error from three
13
independent experiments.
14
Fig. 5
15
Cy5–MSN/COOH–PEG or Cy5–MSN/COOH–PEG–FA (10 µg/ml) for 1 h. (B)
16
Folate receptor was blocked on the surface of KB cells by pre-treating KB cells with
d
M
In vitro DOX release profiles from free DOX or MSN/COOH–PEG–FA at
21
Ac ce p
te
(A) Intracellular fluorescence distribution in KB cells after incubated with
22
Fig. 7
23
MSN/COOH–Cy5 and (B) free Cy5 for 1 h, 4 h, 12 h, 24 h, 48 h and 96 h.
17 18 19 20
free folate for 2 h before the incubation. Scale bars correspond to 20 µm. Fig. 6
Fluorescence visualization of DOX intracellular accumulation. KB cells were
treated with DOX–MSN/COOH–PEG or DOX–MSN/COOH–PEG–FA for 1 h (DOX concentration was 10 µg·mL-1). Images were obtained under confocal laser scanning microscopy. Scale bars correspond to 20 µm. In vivo NIRF imaging in KB tumor-bearing mouse after the injection of (A)
24
24
Page 24 of 34
Graphical Abstract
1 2
In Vivo Imaging Drug delivery
cr
ip t
Targeting
A novel carboxyl functionalized, folate modified and near infrared fluorescent dye
6
conjugated multifunctional mesoporous silica nanocarrier was prepared for targeting,
7
drug delivery and bioimaging.
M
an
us
3 4 5
8
Ac ce p
te
d
9
25
Page 25 of 34
1
Highlights Multifunctionalized MSN was synthesized for drug delivery and bioimaging.
3
Carboxyl functionalization made the drug delivery system pH-dependent.
4
Folate modification significantly enhanced the cellular uptake.
5
The introduction of PEG increased the water dispersibility.
6
The conjunction of biolabel Cy5 made the system effective for bioimaging.
cr
ip t
2
Ac ce p
te
d
M
an
us
7 8
26
Page 26 of 34
Ac ce p
te
d
M
an
us
cr
ip t
Scheme 1
Page 27 of 34
Ac ce p
te
d
M
an
us
cr
ip t
Figure 1
Page 28 of 34
Ac ce p
te
d
M
an
us
cr
ip t
Figure 2
Page 29 of 34
Ac ce p
te
d
M
an
us
cr
ip t
Figure 3
Page 30 of 34
Ac
ce
pt
ed
M
an
us
cr
i
Figure 4
Page 31 of 34
Ac
ce
pt
ed
M
an
us
cr
i
Figure 5
Page 32 of 34
Ac ce p
te
d
M
an
us
cr
ip t
Figure 6
Page 33 of 34
Ac
ce
pt
ed
M
an
us
cr
i
Figure 7
Page 34 of 34
S0927-7765(13)00256-7 http://dx.doi.org/doi:10.1016/j.colsurfb.2013.04.009 COLSUB 5737
To appear in:
Colloids and Surfaces B: Biointerfaces
Received date: Revised date: Accepted date:
25-1-2013 25-3-2013 12-4-2013
Please cite this article as: M. Xie, H. Shi, Z. Li, H. Shen, K. Ma, B. Li, S. Shen, Y. Jin, A multifunctional mesoporous silica nanocomposite for targeted delivery, controlled release of doxorubicin and bioimaging, Colloids and Surfaces B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.04.009 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 2
A multifunctional mesoporous silica nanocomposite for targeted
4
delivery, controlled release of doxorubicin and bioimaging
ip t
3
5
Meng Xiea, Hui Shib, Zhen Lia, Haijun Shena, Kun Maa, Bo Lia, Song Shena and Yi
7
Jina*
8
a
9
Hangzhou, 310058, China b
us
Institute of Pharmaceutics, College of Pharmaceutical Sciences, Zhejiang University,
an
10
cr
6
Department of Pharmacy, Zhejiang Medical College, Hangzhou, 310053, China
M
11 12
* Corresponding author: Yi Jin
14
Address: 866 Yuhangtang Road, College of Pharmaceutical sciences, Zhejiang
15
University, Hangzhou, 310058, China.
16
Tel/fax: +86-571-88208435
18 19
te
Ac ce p
17
d
13
E-mail: [email protected]
1
Page 1 of 34
ABSTRACT
2
In this study, a targeting drug delivery system based on mesoporous silica
3
nanoparticle (MSN) was successfully developed for anti-cancer drug delivery and
4
bioimaging. Carboxyl functionalized MSN (MSN/COOH) was firstly prepared and
5
then modified with folate as the cancer targeting moiety and a near infrared
6
fluorescent dye as labelling segment. Folate was conjugated to MSN/COOH via
7
functional polyethyleneglycol (PEG), constructing the vector MSN/COOH–PEG–FA.
8
The functionalization with carboxyl caused the pore surface of the nanocarrier more
9
negative than native MSN,
us
cr
ip t
1
an
which could provide attractive forces between the
nanoparticles and positively charged doxorubicin hydrochloride (DOX). Meanwhile,
11
the folate modification significantly enhanced the cellular uptake of the delivery
12
system compared to unmodified counterparts. Furthermore, the introduction of PEG
13
increased the water dispersibility. Besides, the modification with the near infrared
14
fluorescent dye Cy5 made the system effective for live cell and in vivo imaging.
15
Therefore, the Cy5–MSN/COOH–PEG–FA system
16
nanocarrier for simultaneous diagnosis and treatment of diseases.
18 19 20
d
te
could be a promising
Ac ce p
17
M
10
KEYWORDS: Doxorubicin, mesoporous silica nanoparticles, folate, drug delivery, bioimaging
2
Page 2 of 34
1
1.
Introduction Cancer remains a major cause of death in most countries in the world, and the
3
incidence of cancer increases with age [1]. Nowadays, nanotechnology-based drug
4
delivery systems (NDDS) are increasingly applied in cancer therapy and diagnosis.
5
Compared to conventional chemo-therapy, nanotherapeutic systems have several
6
potential advantages for cancer treatment. These include increased stability of
7
anti-cancer drugs in blood, decreased non-specific toxicity, easy modification of
8
particle surface for targeting systems, reduced resistance of P-glycoprotein (P-gp)
9
expressing cells and responses to environmental stimuli such as temperature, pH, salt,
10
and ultra-sound [2-5]. The mostly studied drug delivery systems nowadays are
11
polymeric micelles systems [6-9], lipid-based drug delivery systems [10-13],
12
polymeric nanoparticles [14-19],
M
an
us
cr
ip t
2
etc.
However, during the fabrication of these systems, the use of certain organic
14
solvents and additives are unavoidable, which may bring the bioactive substances
15
with undesirable side effects. Meanwhile, these organic materials often suffer from
16
physical instabilities and non-ideal drug encapsulation and delivery performance [12].
21
Ac ce p
te
d
13
22
specific surface area, large pore volume, tunable pore structures and excellent
23
physicochemical stability [20-28]. Previously, it was observed that mesoporous silica
24
nanoparticle (MSN) could be effectively endocytosed and was able to escape the
25
endolysosomal entrapment [29-32]. These new developments rendered the possibility
17 18 19 20
In contrast, porous inorganic materials, such as mesoporous silica materials, can easily be synthesized without using any of the above mentioned organic solvents or additives under very mild conditions. Mesoporous silica nanoparticles (MSN) possess not only some common properties such as facile multifunctionalization, excellent biocompatibility and biodegradability, but also several unique features such as high
3
Page 3 of 34
1
of designing a new generation of drug delivery system for intracellular controlled
2
release and imaging applications. Despite the great progresses in using MSNs as drug delivery vehicles, a precise
4
control over the release kinetics of the loaded drug by decorating the pore surface
5
with functional groups is challenging [33], and often an initial burst effect is observed
6
during release. Besides, there is still plenty of work to be done to study the
7
distribution behavior of MSN in vivo. In this work, we have constructed a
8
multifunctional mesoporous silica nanocomposite (Cy5–MSN/COOH–PEG–FA) for
9
targeted delivery, controlled drug release and bioimaging.
us
cr
ip t
3
an
The weak base type drug
doxorubicin hydrochloride (DOX) was used as the model drug, as many anti-cancer
11
drugs in clinic are positively charged at physiological conditions. The nanocomposite
12
consisted of three parts. Firstly, carboxyl functionalized MSN (MSN/COOH) was
13
prepared as the drug carrier. The functionalization with carboxyl caused the pore
14
surface of the nanocarrier more negative than native MSN, which made it appropriate
15
to load weak base type drug DOX. So, MSN/COOH has a high drug loading capacity
16
with noticeable pH-sensitive release behavior for DOX molecules. Secondly, folate
21
Ac ce p
te
d
M
10
22
dye Cy5 was connected to the nanocomposite as the labeling segment. The
23
modification made the system effective for live cell and in vivo imaging. To our
24
knowledge, this is the first attempt to prepare negative charge functionalized
25
mesoporous silica nanoparticles with both active targeting and near-infrared
17 18 19 20
was employed as the cancer targeting moiety. MSN/COOH and folate were conjugated by a hydrophilic spacer PEG 2000. The introduction of PEG increased the water
dispersibility and that folate modification significantly enhanced the cellular
uptake of the drug delivery system in human nasopharyngeal carcinoma cells (KB cells) compared to unmodified counterparts. What’s more, a near infrared fluorescent
4
Page 4 of 34
1
fluorescent dye labeling. This Cy5–MSN/COOH–PEG–FA system
could be a
2
promising nanocarrier for simultaneous diagnosis and treatment of diseases.
3 4
2.
5
2.1 Materials
6
Cetyltrimethylammonium bromide (CTAB, 98%), tetraethyl orthosilicate (TEOS,
7
99%), 2-cyanopropyltriethoxysilane (CPTES, 99%), N,N’-Dicyclohexylcarbodimide
8
(DCC), N-(3-Dimethylaminopropyl)-N’-ethylcarbodiimide hydrochloride (EDC),
9
2-N-morpholino-ethanesulfonic acid (MES), N-Hydroxysulfosuccinimide sodium salt
10
(sulfo-NHS), Folate were purchased from Sigma-Aldrich. Cy5-hydrazide was
11
purchased from Lumiprobe. Doxorubicin hydrochloride (DOX) was purchased from
12
Dalian Meilun Biology Technology Co. Ltd. (Dalian, China). NH2–PEG–NH2 and
13
NH2–PEG–OH (weight average MW: 2000) were purchased from Yarebio (Shanghai,
14
China). Lyso-tracker Red and DAPI were purchased from Beyotime (Shanghai,
15
China). Water was purified by distillation and deionization (MilliQ Plus). All other
16
chemicals were of analytical grade and used without further purification.
21
Ac ce p
te
d
M
an
us
cr
ip t
Materials and methods Materials
22
vigorous stirring for about 30 min at 60 °C, 2.0 mL TEOS and 0.4 mL
23
2-cyanopropyltriethoxysilane were rapidly added to the mixture. The resulting
24
mixture was stirred for another 2 h at 60 °C and was then kept statically at the same
25
temperature for 24 h. Samples were collected by centrifugation at 20000 rpm for 20
17 18 19 20
2.2 Synthesis of carboxyl functionalized MSN (MSN/COOH) Mesoporous silica nanoparticles functionalized with carboxyl was synthesized by co-condensation. In a typical synthesis, 1.2 g CTAB was dissolved in a solution containing water (180 mL) and ammonia aqueous solution (5.5 mL, 25%). After
5
Page 5 of 34
min, washed, and re-dispersed with deionized water and ethanol several times. Then,
2
the dried product was treated with 9 mol·L-1 H2SO4 solution at 100 °C to produce
3
carboxyl functionalized MSN. At last, the surfactant templates CTAB were removed
4
by extraction in acidic ethanol (about 9 mL HCl in 100 mL ethanol at 65 °C for 24 h),
5
and separated by centrifugation at 20000 rpm to obtain the product. The synthesized
6
nanoparticles were freeze-dried for further use.
cr
ip t
1
7
2.3 Synthesis of NH2–PEG–FA
9
The carboxylic group of folate was firstly activated by DCC and NHS. Briefly, 200
10
mg of folate dissolved in 10 ml of dimethylsulfoxide (DMSO) was reacted with 180
11
mg of NHS and 100 mg of DCC at room temperature for 12 h (folate/NHS/DCC
12
molar ratio=1:2:2). The activated folate (90 mg) was then reacted with 400 mg of
13
NH2–PEG–NH2 (weight average MW: 2000, PEG/folate molar ratio=1/1.1) dissolved
14
in 5 ml of DMSO. The reaction was performed at room temperature for 24 h. The
15
resultant solution was diluted with 20 ml of deionized water and centrifuged for three
16
times. Then the supernatant was purified by dialysis with a 10000 MW cutoff
21
Ac ce p
te
d
M
an
us
8
22
10 mL isotonic 0.1 M MES saline buffer pH 5.5, and then reacting them with EDAC
23
(10 equiv.) and NHS (10 equiv.) for 1 h. MES buffer was used to slow down
24
hydrolysis of the NHS esters on MSN/COOH, as it lacks amino and carboxyl groups,
25
which could compete in the reaction. The NPs were then centrifuged to remove excess
17 18 19 20
membrane and the dialyzed solution was finally freeze-dried. All the processes were carried out in the dark.
2.4 Surface coating of MSN/COOH with NH2–PEG–FA Carboxyl on the surface of the NPs (60 mg) were activated by redispersing the NPs in
6
Page 6 of 34
EDAC/NHS and the water-soluble isourea byproduct. Activated NPs were
2
re-dispersed in 10 mL PBS buffer and reacted with 40 mg of NH2–PEG–FA or
3
NH2–PEG–OH (weight average MW: 2000) for 24 h. The coated NPs were
4
centrifuged and washed with PBS buffer for three times to remove any unbound PEG.
5
The dialyzed solution was finally freeze-dried. All the processes were carried out in
6
the dark.
cr
ip t
1
7
2.5 Synthesis of fluorescent nanoparticles
9
Cy5-hydrazide was reacted with the carboxyl on the surface of the nanoparticles via
10
EDC/NHS coupling method to fabricate fluorescent nanoparticles for imaging. A total
11
of
12
MSN/COOH–PEG–FA) were re-dispersed in 5 mL MES saline buffer (0.1 mol·L-1,
13
pH 5.5) and activated using EDC (19.1 mg) and sulfo-NHS (21.7 mg) for 1 h,
14
respectively. The NPs were then centrifuged to remove excess EDAC/NHS and the
15
water-soluble isourea byproduct. Activated NPs were re-dispersed in 5 mL MES
16
saline buffer and reacted with Cy5-hydrazide (0.3 mg) for 24 h. The NPs were then
of
nanoparticles
an
mg
(MSN/COOH,
MSN/COOH–PEG
or
21
Ac ce p
te
d
M
30
us
8
22
(AMF500, Bruker). Thermogravimetric analysis (TGA, Perkin-Elmer Pyris Diamond
23
TG) of MSN/COOH–PEG–FA was performed at a heating rate of 10 °C min-1 in N2.
24
The infrared (IR) spectra were measured using a JASCO FT/IR-4100 spectrometer
25
(Tokyo, Japan). The UV–Vis absorption spectra were performed using a Persee
17 18 19 20
centrifuged and washed with deionized water for three times to remove any unbound Cy5-hydrazide and freeze-dried for further use.
2.6 Characterization of the nanoparticles The structure of NH2–PEG–FA was measured by 1H NMR in DMSO-d6 at 500 MHz
7
Page 7 of 34
TU-1800PC spectrophotometer (Beijing, China). The pore characteristics of the
2
samples were studied by determining the nitrogen adsorption using a Quantachrome
3
Autosorb-1-C surface area and pore size analyzer (Florida, USA) at -196 °C. The
4
morphology and structure of MSN/COOH and MSN/COOH–PEG–FA samples were
5
characterized via transmission electron microscopy (TEM; JEOL JEM-1200EX
6
microscope, Japan). The size distributions and zeta potentials of MSN, MSN/COOH
7
and MSN/COOH–PEG–FA were induced by different pH values of aqueous solutions
8
were measured using a Malvern Zetasizer Nano-S90 (England) particle size and zeta
9
potential analyzer, respectively.
an
us
cr
ip t
1
10
2.7 Preparation of drug-loaded nanoparticles
12
Typically, 10 mg of MSN/COOH–PEG–FA were immersed in PBS solutions (5mL,
13
pH 6.5) containing 5 mg of DOX. After stirring for 24 h under light-sealed conditions,
14
the mixture was centrifuged and the supernatant was removed. The drug-loaded
15
MSN/COOH–PEG–FA samples were washed twice with PBS solution to remove
16
doxorubicin that was adsorbed on the surface but not inside the pores. The DOX
18 19
d
te
Ac ce p
17
M
11
concentration in original solution and supernatant solution were determined via UV–Vis spectroscopy. DOX loading amount and loading efficiency in the nanoparticles were calculated by the following equations: DOXin feed - free DOX ! 100 nanopartic les in feed DOXin feed - free DOX (%, w/w) ! ! 100 DOXin feed
Loading amount(%,w/w) !
20 21
Loading efficiency
22 23
2.8 Cell culture
24
Human nasopharyngeal carcinoma cell line (KB cell line), known as folate receptor
8
Page 8 of 34
positive tumor cell, was obtained from the institute of biochemistry and cell biology
2
of Chinese academy of sciences (IBCB, Shanghai, China). Cells were grown at 37 °C
3
and 5% CO2 in RPMI 1640 medium supplemented with 10% (v/v) FBS. For all
4
experiments, cells were harvested by the use of D-Hank’s-trypsin solution and
5
redispersed in fresh medium before platinge.
ip t
1
cr
6
2.9 Intracellular distribution study
8
The cellular uptake and intracellular distribution experiments in KB cells were
9
performed using CLSM (CarlZeiss LSM 710 NLO, Dresden, Germany). The cells
10
were seeded into six-well plates at a density of 2×105 cells per well medium
11
respectively. After a 24 h incubation period, the cells were treated with Cy5 labeled
12
nanoparticles or DOX (free DOX in solution or DOX-loaded nanoparticle with a
13
DOX concentration of 10 µg mL-1). After incubation for 1 h in folate-free medium,
14
the culture medium was removed. The cells were then carefully washed three times
15
with PBS to remove any free DOX or nanoparticles and fixed with 4% (w/v)
16
paraformaldehyde for 0.5 h. The lysosomes were stained with Lyso-tracker Red and
21
Ac ce p
te
d
M
an
us
7
22
cells were cultured for another 1 h, washed with sterilized PBS, stained with
23
Lysotracker Red and DAPI, and then analyzed by CLSM.
17 18 19 20
the nuclei were stained with DAPI according to the operation manuals. Then, the cells were washed three times with PBS. The intracellular localization of was observed with CLSM.
To further evaluate the role of FA in the cellular uptake, the cells were pretreated with free FA (0.5 mg mL-1) for 2 h, Cy5 labeled nanoparticles were then added and the
24 25
2. 10
Cell viability assay
9
Page 9 of 34
In vitro cytotoxcity of blank and DOX loaded nanoparticles against KB cells were
2
assessed by the standard MTT assay. Cells were seeded in 96-well plates at a density
3
of 1×104 cells per well and allowed to attach overnight. After a 24 h incubation period
4
at 37 °C in 100 µL RPMI medium containing 10% FBS, culture medium was
5
discarded and then cells were treated with pH 7.4 PBS solution of free DOX,
6
drug-loaded or blank nanoparticles at various concentrations. After the incubation in
7
folate-free medium for 24 h or 48 h, culture medium was removed, 31.5 μL of
8
5mg·mL-1 MTT solution was added, and cells were incubated for another 4 h. Then,
9
media was removed, the MTT crystals were dissolved in DMSO (0.1mL·well-1), and
10
the optical density of each sample was read at 570 nm on an Automated Microplate
11
Reader 550 (Bio-Rad, USA). A culture medium without particles was used as the
12
blank control. The cytotoxicity was expressed as the percentage of the cell viability as
13
compared with the blank control. The time and dose dependences of the cytotoxicity
14
were investigated at different particle concentrations.
15
2.11
Intravital near-infrared (NIR) optical imaging of tumor-bearing mice
21
Ac ce p
16
te
d
M
an
us
cr
ip t
1
22
tumor-bearing mice were treated with Cy5 labeled MSN/COOH–PEG–FA via the
23
intratumor injection. The mice were anesthetized and placed on an animal plate. The
24
fluorescent scans were performed at various time points using the Maestro in-vivo
25
imaging system (Cambridge Research & Instrumentation, CRi, USA).
17 18 19 20
In order to observe the real-time distribution and tumor accumulation ability of near-infrared dye Cy5 labeled nanoparticles in mice bearing KB xenografts, non-invasive optical imaging systems were utilized. Briefly, male Balb/c nude mice of 4 weeks were injected with 0.2 mL of a cell suspension containing 5×106 KB cells subcutaneously. After 15 days, when the tumor reached 0.8×0.8 cm in diameter, the
10
Page 10 of 34
1 2
2.12
3
All experiments were run in triplicate and the acquired data are expressed as mean ±
4
SD. Statistical significance was determined using Student’s t-test.
ip t
Statistical analysis
5 6
3.
7
3.1 Synthesis of FA modified system
8
The procedure for the synthesis of mesoporous silica nanoparticles functionalized
9
with carboxyl (MSN/COOH), folate and near-infrared dye Cy5 was illustrated in
10
scheme 1. The functionalization was performed by a four-step surface modification
11
process. 1. The pore surface of the MSN was functionalized by co-condensation with
12
CPTES to form MSN/CN; 2. MSN/CN was hydrolyzed with 9 mol·L-1 H2SO4 solution
13
at 100 °C to yield the carboxyl functionalized MSN (MSN/COOH); 3. Cy5-hydrazide
14
was reacted with the carboxyl on the surface of MSN/COOH to fabricate fluorescent
15
nanoparticles; 4. MSN/COOH–Cy5 was reacted with NH2–PEG–FA to form folate
16
modified nanoparticles (MSN/COOH–PEG–FA).
us
an
M
d
Ac ce p
te
21 22
biocompatible.
17 18 19 20
cr
Results and discussion
MSN/COOH–PEG–FA was synthesized according to the method outlined in fig.
1. Carboxyl (–COOH) of folate and carboxyl of MSN/COOH were connected through the end-amino hydrophilic spacer, NH2–PEG–NH2. PEG was chosen as a linker in order (1) to get optimized length of linker to reach accessible receptor sites, (2) to increase the water dispersibility, and (3) to make the folate-conjugate system more
23 24
3.2 Characterization of MSN/COOH–PEG–FA
25
The synthesis of NH2–PEG–FA was carried out as shown in fig. 1. Folate was firstly
11
Page 11 of 34
conjugated to NH2–PEG–NH2 following the usual NHS/DCC reaction to get
2
NH2–PEG–FA. The formation was confirmed by the 1H NMR spectrum, as shown in
3
fig. S1 (ESI). In the 1H NMR spectrum, the peak at 8.67 ppm belongs to the proton on
4
the C7 of folate. The peaks at 6.65 ppm (3’, 5’-H, 2H) and 7.66 ppm (2’,6’-H, 2H)
5
belong to the protons on the benzene ring. The peaks at 1.9, 2.1–2.4 and 4.3 ppm
6
originate from the α, β and γ –CH2– protons of glutamic acid part in the folate
7
structure. The peak at 3.6 ppm belongs to the –CH2–protons of PEG.
cr
ip t
1
MSN/COOH–PEG–FA was prepared from the reaction between MSN/COOH and
9
NH2–PEG–FA using EDC between amino group of NH2–PEG–FA and acid group of
10
MSN/COOH. IR spectra of pure MSN/CN, MSN/COOH and MSN/COOH–PEG–FA
11
were shown in fig. S2 (ESI). All three samples showed the same peak at 1630 cm-1,
12
which attributed to H2O adsorbed in the mesopores. The IR spectrum of MSN/COOH
13
samples revealed a peak at 1718 cm-1, the stretching vibration frequency of C=O of
14
carboxyl group. The appearance of the new intensive peak at 2927 cm-1 was assigned
15
to the C–H stretching frequencies of –CH2–CH2– groups attributed to PEG, which
16
indicated the successful attachment of NH2–PEG–FA onto the surface of MSN/COOH.
21
Ac ce p
te
d
M
an
us
8
22
performed under a constant N2 flow, as shown in fig. S3 (ESI). Upon heating, the
23
materials respectively underwent a total weight loss of 16.0 % (H1) and 28.7 % (H2)
24
up to 800 °C. The graft ratio (W) in MSN/COOH–PEG–FA was 15.05 %. The
25
calculating process was shown in supporting information, following literature
17 18 19 20
Both MSN/COOH and MSN/COOH–PEG–FA samples had no peak at 2250 cm-1, a peak appeared in MSN/CN IR spectrum attributed to the stretching vibration frequency of C≡N of cyano group, indicating the completion of hydrolysis of CN catalyzed by H2SO4 acid.
TGA of CTAB-extracted MSN/COOH and MSN/COOH–PEG–FA were
12
Page 12 of 34
1
procedures [34] with slight change. Cy5-hydrazide was reacted with the carboxyl group on the surface of
3
MSN/COOH–PEG–FA via EDC/NHS coupling method to fabricate fluorescent
4
nanoparticles for imaging. Compared with the co-condensation method [35], this
5
coupling method does not compromise the framework structure of the parent
6
mesoporous materials. The Conjugation of Cy5-hydrazide was confirmed by UV–Vis
7
absorption spectroscopy. As shown in fig. S4 (ESI), Cy5-hydrazide in water exhibited
8
the main absorption bands at 639 nm. The band was present in the spectrum of
9
Cy5–MSN/COOH–PEG–FA, indicating conjugation of Cy5. Also, upon conjugation
10
of Cy5 with MSN/COOH–PEG–FA, the peak was slightly red shifted to 645 nm,
11
indicative of conjugation between Cy5 and MSN/COOH–PEG–FA.
M
an
us
cr
ip t
2
Figure 2A shows the nitrogen ad/desorption isotherms of MSN, MSN/COOH and
13
MSN/COOH–PEG–FA. All three samples showed type IV isotherms according to the
14
IUPAC classification with well-defined steps at relative pressures (P/P0) of 0.2–0.4,
15
which was the typical characteristic of mesoporous silica. This suggested that the
16
functionalization with carboxyl and folate did not change the structure of mesoporous
21
Ac ce p
te
d
12
22
surface-shielding effect of organic materials.
17 18 19 20
channels. The pore-size distribution was determined by the Barret–Joyner–Halenda (BJH) method (fig. 2B). With the introduction of carboxyl and folate, the capillary condensation step (P/P0) shifted to lower relative pressure and consistently the pore diameter decreased from 2.8 to 2.3 nm. Meanwhile, the surface area and pore volume are smaller after functionalization (table S1, ESI). These may be due to the
23 24 25
3.3 Properties of the MSN/COOH-PEG-FA The size and morphology of the MSN/COOH and MSN/COOH–PEG–FA were
13
Page 13 of 34
characterized by TEM (fig. 3). The particle size distribution of the nanoparticles was
2
found to be relatively narrow and quite uniform. The mean particle size of the
3
nanoparticles determined by dynamic light scattering (DLS) measurement was 63.5 ±
4
3.4 nm (fig. 3(c)), which was consistent with its diameter observed by TEM. The
5
physical properties of the NPs before and after the functionalization were given in
6
table S2 (ESI). After the introduction of NH2–PEG–FA, the average size increased 5
7
nm. Compared with MSN/COOH, the size distribution of MSN/COOH–PEG–FA was
8
narrower. There were no obvious differences in the mean particle size and size
9
distribution between MSN/COOH–PEG and MSN/COOH–PEG–FA.
an
us
cr
ip t
1
As illustrated in scheme 1, carboxyl incorporated on mesopore surfaces can make
11
the pore surface of the NPs more negative. Figure S5 (ESI) shows the zeta potential of
12
native and functionalized MSN at various pH levels. As expected, the zeta potential of
13
MSN/COOH was always lower than that of native MSN. At pH 7.4, the zeta potential
14
of native MSN was -9.0 mV, while at pH 6.5, the zeta potential was around zero.
15
However, due to the carboxyl, functionalized MSN showed highly negative zeta
16
potentials in the pH range from 7.4 (-24.5mV) to 5.0 (-13.6 mV) and there was also
21
Ac ce p
te
d
M
10
22
temperature in physiological condition (PBS, pH 7.4) for 24 h (table S2, ESI). It was
23
observed that native MSN showed obvious precipitation. However, the nanoparticle
24
size distributions of MSN/COOH and MSN/COOH–PEG–FA did not change
25
significantly, which was consistent with the results of zeta potential study.
17 18 19 20
no significant difference in the zeta potentials before and after the surface coating with NH2–PEG–OH or NH2–PEG–FA. Surface zeta potential gave an indication of the potential stability of the colloidal
system. The stability of the native and functionalized MSN in aqueous media were evaluated by comparing the visual appearance and average size after stored at room
14
Page 14 of 34
1 2
3.4 Doxorubicin loading and release The utility of the MSN/COOH–PEG–FA as an intracellular carrier for anticancer
4
drug delivery system was evaluated using doxorubicin hydrochloride (DOX) as a
5
model anticancer drug. The loading of DOX into the nanoparticle was performed by
6
diffusion and electrostatic attraction. The pKa of DOX is 8.2 while the nanoparticles
7
carry negative charges when the solution pH is higher than 3.0 (Fig. S5, ESI). The
8
electrostatic attraction between DOX and the nanoparticles happens when the solution
9
pH changes from weak acidic to neutral or weak alkaline condition, which facilitates
10
the DOX loading. Consequently, it was found that the DOX loading content and
11
encapsulation efficiency in MSN/COOH–PEG–FA reached 24.0 ± 0.4 % and 62.4 ±
12
1.0 %, respectively, indicating the MSN/COOH–PEG–FA was indeed an effective
13
DOX loading vehicle.
d
M
an
us
cr
ip t
3
To prove that the DOX molecules could penetrate into the nanopores of the
15
functionalized nanoparticles, BET analysis was also carried out. The change of
16
specific surface area, pore size and pore volume distribution of the nanoparticles
21
Ac ce p
te
14
22
investigated under a simulated physiological condition (PBS, pH 7.4) and in an acidic
23
environment (acetate buffer, pH 5.0 or 6.5) at 37 °C to assess the feasibility of using
24
MSN/COOH as an anticancer drug delivery carrier (fig. 4). Free DOX solution was
25
used as a control. This study clearly showed that the pH of the medium had a strong
17 18 19 20
before and after DOX loading are summarized in table S1 (ESI). As the data clearly indicated,
the
BET
surface
area,
pore
volume
and
pore
diameter
of
MSN/COOH–PEG–FA were all decreased significantly after the loading of DOX, indicating penetration of the DOX molecules into the nanopores. The release patterns of DOX from MSN/COOH–PEG–FA nanoparticles were
15
Page 15 of 34
1
effect on the
2
nanoparticles. The drug release at pH 7.4 was considerably slow and only 40 % of the
3
DOX
4
MSN/COOH–PEG–FA nanoparticles maintained electrostatic interactions between
5
the DOX molecules and the carboxyl under physiological conditions. However, it is
6
interesting to note that the amount of released DOX at pH 5.0 reached to
7
approximately 90%. This phenomenon can be explained by the fact that the
8
electrostatic interactions were decreased as well as the electrostatic repulsion
9
enhanced between the nanoparticles and DOX in an acidic environment, thus the
released
after
8
h.
This
result
suggested
that
us
cr
ip t
were
an
10
molecules
release quantity of DOX from the MSN/COOH–PEG–FA
release amount of DOX increased.
The pH dependent release property of the nanoparticles may benefit the DOX
12
release in tumor cells whose pH is lower than that of the normal cells. It is the most
13
desirable that most DOX molecules encapsulated in the nanoparticles would remain in
14
the mesoporous for a considerable long period of time when the nanoparticles stayed
15
in the plasma at normal physiological conditions (pH 7.4), thus elongating the
16
circulation period of the drug and reducing the side effects. When the nanoparticles
18 19 20 21
d
te
Ac ce p
17
M
11
reached the tumor sites and were internalized inside the tumor cells by endocytosis, faster release might occur inside the endosome/lysosome of tumor cells because of the low pH values, thus enhanced the delivery efficiency.
3.5 In vitro toxicity studies
22
The cytotoxicity of KB cells exposed to
CTAB-extracted nanoparticles was
23
investigated, and the effects of the particle concentration on the cell proliferation
24
activity were also studied. As shown in fig. S6 (ESI), the cell proliferation was not
25
hindered after the cells were exposed to the nanoparticles for 48 h at particle
16
Page 16 of 34
1
concentrations of not higher than 250 µg mL-1. However, the hindrance was indeed
2
observed at the particle concentration 500 µg mL-1. The very low cytotoxicity
3
be attributed to the relatively low concentrations of the CTAB molecules in the
4
nanoparticles. Therefore, MSN/COOH–PEG–FA with the concentration less than 250
5
µg mL-1 was safe for cells within 48 h.
ip t
may
cr
6
3.6 Intracellular uptake of nanoparticles
8
Confocal laser-scanning microscopy (CLSM) was used to compare endocytosis of
9
MSN/COOH–PEG–FA with that of MSN/COOH–PEG using a folate-receptor
10
positive cancer cell line, KB cell line, in the FA-free incubation medium. A
11
near-infrared dye Cy5-hydrazide was reacted with the carboxyl of the nanoparticles to
12
fabricate fluorescent nanoparticles for imaging. The red fluorescence indicated the
13
distribution of Cy5–MSN/COOH–PEG or Cy5–MSN/COOH–PEG–FA, the green
14
fluorescence indicated the distribution of lysosomes which were stained by
15
Lyso-tracker red, and the blue indicated the nuclei stained by DAPI. As shown in Fig.
16
5A, cells exposed to MSN/COOH–PEG showed a weak red fluorescence, and it was
18 19 20 21
an
M
d
te
Ac ce p
17
us
7
observed mainly on the surface of the cells. The failure of MSN/COOH–PEG delivery was partly attributed to the negative charge of the supported MSN/COOH–PEG, which is repelled by the negatively charged cell surface. Besides, after PEGylation, the hydrophilic property of the nanoparticles enhanced, which was against the cellular uptake.
22
As reported, the folate receptor is overexpressed in several types of human cancer,
23
such as brain, kidney, lung, and breast. The FR ligand, folate, is a vitamin that is used
24
for the biosynthesis of nucleotides and is utilized in high levels to meet the needs of
25
proliferating cancer cells. In order to further improve drug delivery efficiency,
17
Page 17 of 34
MSN/COOH–PEG–FA was designed. After incubation for 1 h, the folate conjugated
2
nanoparticles were mostly localized in lysosomes leaving a clear zone of nucleus, as
3
evidenced by colocalization of green fluorescence and red fluorescence, indicating
4
cellular uptake instead of adhesion to the surface and the nanoparticles.
ip t
1
To further evaluate the role of folate in the cellular uptake of nanoparticles, the
6
folate receptor was blocked on the surface of KB cells by pre-treating KB cells with
7
free folate for 2 h. As shown in Fig. 5B, there was little difference in the cellular
8
uptake of MSN/COOH–PEG regardless of the addition of folate in the medium.
9
However, when KB cells were incubated with MSN/COOH–PEG–FA, much weaker
10
red fluorescence was observed in the folate medium than in the folate free medium.
11
This result demonstrated that the uptake of MSN/COOH–PEG–FA and folate is
12
competitive,
13
receptor-mediated endocytosis. This observation clearly inferred that folate
14
conjugated nanoparticle was very much effective as a delivery system for targeted
15
anticancer drugs.
17 18 19 20
us
an
M
MSN/COOH–PEG–FA were
internalized
via
the
folate
te
d
thus
Ac ce p
16
cr
5
3.7 Intracellular distribution of DOX To determine whether encapsulation of DOX in MSN/COOH–PEG–FA affected its trafficking
inside
cells,
the
intracellular
distribution
of
free
DOX,
DOX–MSN/COOH–PEG and DOX–MSN/COOH–PEG–FA in KB cells were further
21
estimated. Since DOX has red fluorescence, its distribution in the KB cells can be
22
easily observed. As shown in fig. 6, after 1 h of incubation with DOX solution in KB
23
cells, strong fluorescence was emitted from the cells, particularly the nuclear regions,
24
indicating a high level of DOX molecules in the nucleus. The intense DOX
25
accumulation in the nucleus for free DOX occurred because intracellular DOX 18
Page 18 of 34
1
molecules in the cytosol could transport rapidly to the nucleus and avidly bound to
2
the chromosomal DNA [36]. In the case of the DOX–MSN/COOH–PEG and DOX–MSN/COOH–PEG–FA,
4
the fluorescence intensity were weaker, indicating that the DOX loaded nanoparticles
5
were initially located within the endosomal intracellular compartments, releasing
6
loaded DOX in the cytosol region in a sustained manner, and then diffused through
7
the endocytic compartment membrane to enter the nucleus. However, the
8
internalization of DOX–MSN/COOH–PEG–FA wasmore obvious compared with the
9
DOX–MSN/COOH–PEG in the absence of free folate in the cell culture medium.
10
This result directly confirms that the cellular uptake of the nanoparticles can be
11
enhanced by attaching folate on their surface, which was in accord with the
12
intracellular uptake of blank nanoparticles.
13
3.8 Cytotoxicity of DOX-loaded nanoparticles
14
DOX is a well-known cytotoxic anticancer drug even at a very low concentration.
15
Since DOX–MSN/COOH–PEG–FA has a sustained DOX release behavior and a
17 18 19 20
cr
us
an
M
d
te
Ac ce p
16
ip t
3
remarkable uptake by KB cells, the drug delivery system was expected to have high cytotoxicity against cancer cells. To verify the pharmacological activity of released DOX from DOX–MSN/COOH–PEG–FA, the cytotoxicity of free DOX, as well as DOX loaded nanoparticles containing the equal molar DOX were investigated using cell viability assay by comparing with the control group, in which neither DOX
21
loaded nanoparticles nor free DOX were added. As shown in fig. S7 (ESI), KB cells
22
demonstrated dose-dependent cytotoxicity for both free DOX and DOX loaded
23
nanoparticles and the cell viability of DOX–MSN/COOH–PEG–FA was equivalent to
24
that of free DOX. On the other hand, cell viability of DOX–MSN/COOH–PEG–FA
25
was much higher than that of DOX–MSN/COOH–PEG. These results clearly showed 19
Page 19 of 34
that the increased cytotoxicty of DOX–MSN/COOH–PEG–FA against KB cells is
2
most likely due to the uptake of DOX–MSN/COOH–PEG–FA via folate-receptor
3
medited endocytosis. Furthermore, at low DOX concentration (<0.1 µg mL-1),
4
DOX–MSN/COOH–PEG–FA showed stronger effect on killing tumor cells than free
5
DOX, while the cytotoxicity of free DOX became more obvious with the increase of
6
DOX concentration, which probably because of the slow release of DOX from
7
DOX–MSN/COOH–PEG–FA.
cr
ip t
1
3.9 In vivo imaging
an
9
us
8
To investigate the accumulation of the nanocarrier in tumor tissues, the non-invasive
11
near-infrared fluorescence (NIRF) imaging was acquired and the near-infrared dye
12
Cy5 was conjugated to MSN/COOH–PEG–FA. The in vivo imaging of
13
Cy5–MSN/COOH–PEG–FA was evaluated in KB tumor xenograft mouse model. As
14
shown in fig. 7, after the injection, fluorescent signals were observed in the tumors of
15
the mice in both Cy5–MSN/COOH–PEG–FA and free Cy5 groups, suggesting that the
17 18 19 20
d
te
Ac ce p
16
M
10
fluorescence of Cy5 was less affected by the body autofluorescence.Interestingly, in the Cy5–MSN/COOH–PEG–FA group, the intense of the fluorescence did not seem to decrease significantly as time increased. 4 days after the injection, the fluorescence intensity was still strong and mostly located on the tumor position. However, in the free Cy5 group, the fluorescence intensity became weaker as time increased. The
21
observation suggested that Cy5 labeled MSN/COOH–PEG–FA could accumulate in
22
tumor tissue for a long period of time for tumor imaging.
23
4. Conclusion
24
In this work, a novel negative-charge functionalized and folate receptor targeted 20
Page 20 of 34
mesoporous silica nanocarrier was synthesized. The experimental results clearly
2
demonstrated that DOX can easily be loaded into the nanoparticles with high loading
3
capacity. This drug delivery system was proved to possess many unique
4
characteristics: 1. enhanced stability at physiological conditions; 2. desirable release
5
acceleration at low pH conditions for cytoplasmic drug delivery; 3. beneficial to be
6
taken up by cells caused by a folate-receptor mediated endocytosis process. More
7
importantly, the drug delivery system was able to accumulate in tumor tissue for a
8
long period of time, thus could simultaneously be used for in vivo imaging. Based on
9
the features, there is the great potential of the multifunctionalized fluorescent
10
mesoporous silica nanoparticles for simultaneous imaging in the near infrared region
11
and targeted drug release.
M
an
us
cr
ip t
1
12
Acknowledgements
14
This work was supported by National Natural Science Foundation of China (No.
15
30973647) and National Basic Research Program of China (973 Program,
16
2009CB930300).
18 19 20
References [1]
J. Fang, H. Nakamura and H. Maeda, Adv. Drug Deliver. Rev., 3 (2011) 136.
[2]
L.S. Jabr-Milane, V.L Van, S. Yadav and M.M. Amiji, Cancer Treat. Rev., 34 (2008) 592.
21 22
te
Ac ce p
17
d
13
[3]
T. Bansal, N. Akhtar, M. Jaggi, R.K. Khar and S. Talegaonkar, Drug Discov. Today, 14 (2009) 1067.
23 24
[4]
J.R. Kanwar, N. Singh and R.K. Kanwar, Nanomedicine-UK, 6 (2011) 701.
25
[5]
K.T. Oh, H.J. Baik, A.H. Lee, Y.T. Oh, Y.S. Youn and E.S. Lee, Int. J. Mol. Sci., 21
Page 21 of 34
10 (2009) 3776.
1
[6]
Int. J. Nanomed., 7 (2012) 187. [7]
and Y. Jin, J Gene Med., 12 (2010) 669. [8]
604.
11 12 13
Colloids Surfaces B., 94 (2012) 266.
an
10
R. Khonkarn, S. Mankhetkorn, M. Talelli, W.E. Hennink and S. Okonogi,
[10] T. Kobayashi, T. Ishida, Y. Okada, S. Ise, H. Harashima and H. Kiwada, Int. J. Pharm., 329 (2007) 94.
[11] K. Ma, H.J. Shen, S. Shen, M.Xie, C.B. Mao, L.Y. Qiu and Y. Jin, J. Gene. Med., 13 (2011) 290.
d
9
[9]
us
7 8
Y.H. Tao, J.F. Han, X.W. Wang and H.Y. Dou, Colloids Surfaces B., 102 (2013)
cr
5 6
K. Ma, M.X. Hu, M. Xie, H.J. Shen, L.Y. Qiu, W.M. Fan, H.Y. Sun , S.Q. Chen
ip t
3 4
B. Li, H. Xu, Z. Li, M.F. Yao, M Xie, H.J. Shen, S. Shen, X.S. Wang and Y. Jin,
M
2
[12] S. Simovic, T.J. Barnes, A. Tan and C.A. Prestidge, Nanoscale, 4 (2012) 1220.
15
[13] Y.C. Kuo and C.T. Liang, Colloids Surfaces B., 85 (2011) 131.
16
[14] Y. Patil, T. Sadhukha, L.N. Ma and J. Panyam, J. Control. Release, 136 (2009)
21
Ac ce p
te
14
22
B., 88 (2011) 741.
17 18 19 20
23 24 25
21.
[15] V. Saxena, Y. Naguib and M.D. Hussain, Colloids Surfaces B., 94 (2012) 274. [16] Y.I. Jeong, D.H. Kim, C.W. Chung, J.J. Yoo, K.H. Choi, C.H. Kim, S.H. Ha and D.H. Kang, Colloids Surfaces B., 90 (2012) 28.
[17] Q.H. Gao, Q. Liang, F. Yu, J. Xu, Q.H. Zhao and B.W. Sun, Colloids Surfaces
[18] W. Xiong, W. Wang, Y. Wang, Y.B. Zhao, H.B. Chen, H.B. Xu and X.L. Yang, Colloids Surfaces B., 84 (2011) 447. [19] L. Jia, J.P. Xu, H. Wang and J. Ji, Colloids Surfaces B., 84 (2011) 49.
22
Page 22 of 34
2 3 4
[20] I.I. Slowing, J.L. Vivero-Escoto, C.W. Wu and V.S. Lin, Adv. Drug Deliver. Rev., 60 (2008) 1278. [21] J.L.Vivero-Escoto, I.I. Slowing, B.G. Trewyn and V.S.Y. Lin, small, 6 (2010) 1952.
ip t
1
[22] Q.J. He and J.L. Shi, J. Mater. Chem., 21 (2011) 5845.
6
[23] J. Rosenholm, C. Sahlgren and M. Lindén, J. Mater. Chem., 20 (2010) 2707.
7
[24] O. Tabasi, C. Falamaki and Z. Khalaj, Colloids Surfaces B., 98 (2012) 18.
8
[25] N.C. Zhang, E.Y. Ding, X. Feng, Y. Xu and H.H. Cai, Colloids Surfaces B., 89
12 13
us
an
11
[26] Z.D. Chen, X. Li, H.Y. He, Z.H. Ren, Y. Liu, J. Wang, Z. Li, G. Shen and G.R. Han, Colloids Surfaces B., 95 (2012) 274.
M
10
(2012) 133.
[27] Z. Guo, X.M. Liu, L. Ma, J. Li, H. Zhang, Y.P. Gao and Y. Yuan, Colloids Surfaces B., 101 (2013) 228.
d
9
cr
5
[28] H. Gustafsson, C. Thörn and K. Holmberg, Colloids Surfaces B., 87 (2011) 464.
15
[29] C.P. Tsai, C.Y. Chen, Y. Hung, F.H. Chang and C.Y. Mou, J. Mater. Chem., 19 (2009) 5737.
21
Ac ce p
16
te
14
22
[34] M.S. Moorthy, S.S. Park, D. Fuping, S.H. Hong, M. Selvaraj and C.S. Ha, J.
17 18 19 20
23
[30] F. Lu, S.H. Wu, Y. Hung and C. Y. Mou, Small, 5 (2009) 1408. [31] Q. He, Z. Zhang, Y. Gao, J. Shi and Y. Li, Small, 5 (2009) 2722. [32] J. Lu, Z.X. Li, J.I. Zink and F. Tamanoi, nanomed-nanotechnol., 8 (2012) 212. [33] M. Manzano, V. Aina, C.O. Arean, F. Balas, V. Cauda, M. Colilla, M.R. Delgado and M. Vallet-Regí, Chem. Eng. J., 137 (2008) 30.
Mater. Chem., 22 (2012) 9100.
24
[35] K. Ma, H. Sai and U. Wiesner, J. Am. Chem. Soc., 134 (2012) 13180.
25
[36] R. Gillies and J.M.J. Frchet, Bioconjugate Chem., 16 (2005) 361.
26 23
Page 23 of 34
1
Figures Captions
2
Scheme 1
3
MSN/COOH–PEG–FA, DOX loading, and pH dependent drug release.
4
Fig. 1
Synthesis scheme of MSN/COOH–PEG–FA.
5
Fig. 2
(A) Nitrogen adsorption-desorption isotherms and (B) pore size distribution
6
of MSN, MSN/COOH, MSN/COOH–PEG–FA and DOX–MSN/COOH–PEG–FA.
7
The inset shows the enlarged figure of DOX–MSN/COOH–PEG–FA.
8
Fig. 3
9
MSN/COOH–PEG–FA (Scale bars: 50 nm). (C) The size distribution of
of
the
synthesis
and
chemical
modification
of
cr
ip t
Illustration
an
us
(A) TEM micrographs of MSN/COOH. (B) TEM micrographs of
10
MSN/COOH–PEG–FA determined by DLS.
11
Fig. 4
12
different pH values. Results are expressed as means ± the standard error from three
13
independent experiments.
14
Fig. 5
15
Cy5–MSN/COOH–PEG or Cy5–MSN/COOH–PEG–FA (10 µg/ml) for 1 h. (B)
16
Folate receptor was blocked on the surface of KB cells by pre-treating KB cells with
d
M
In vitro DOX release profiles from free DOX or MSN/COOH–PEG–FA at
21
Ac ce p
te
(A) Intracellular fluorescence distribution in KB cells after incubated with
22
Fig. 7
23
MSN/COOH–Cy5 and (B) free Cy5 for 1 h, 4 h, 12 h, 24 h, 48 h and 96 h.
17 18 19 20
free folate for 2 h before the incubation. Scale bars correspond to 20 µm. Fig. 6
Fluorescence visualization of DOX intracellular accumulation. KB cells were
treated with DOX–MSN/COOH–PEG or DOX–MSN/COOH–PEG–FA for 1 h (DOX concentration was 10 µg·mL-1). Images were obtained under confocal laser scanning microscopy. Scale bars correspond to 20 µm. In vivo NIRF imaging in KB tumor-bearing mouse after the injection of (A)
24
24
Page 24 of 34
Graphical Abstract
1 2
In Vivo Imaging Drug delivery
cr
ip t
Targeting
A novel carboxyl functionalized, folate modified and near infrared fluorescent dye
6
conjugated multifunctional mesoporous silica nanocarrier was prepared for targeting,
7
drug delivery and bioimaging.
M
an
us
3 4 5
8
Ac ce p
te
d
9
25
Page 25 of 34
1
Highlights Multifunctionalized MSN was synthesized for drug delivery and bioimaging.
3
Carboxyl functionalization made the drug delivery system pH-dependent.
4
Folate modification significantly enhanced the cellular uptake.
5
The introduction of PEG increased the water dispersibility.
6
The conjunction of biolabel Cy5 made the system effective for bioimaging.
cr
ip t
2
Ac ce p
te
d
M
an
us
7 8
26
Page 26 of 34
Ac ce p
te
d
M
an
us
cr
ip t
Scheme 1
Page 27 of 34
Ac ce p
te
d
M
an
us
cr
ip t
Figure 1
Page 28 of 34
Ac ce p
te
d
M
an
us
cr
ip t
Figure 2
Page 29 of 34
Ac ce p
te
d
M
an
us
cr
ip t
Figure 3
Page 30 of 34
Ac
ce
pt
ed
M
an
us
cr
i
Figure 4
Page 31 of 34
Ac
ce
pt
ed
M
an
us
cr
i
Figure 5
Page 32 of 34
Ac ce p
te
d
M
an
us
cr
ip t
Figure 6
Page 33 of 34
Ac
ce
pt
ed
M
an
us
cr
i
Figure 7
Page 34 of 34