Modular Integration of Upconverting Nanocrystal-Dendrimer Composites for Folate Receptor-Specific NIR Imaging and Light-Triggered Drug Release

full papers

www.MaterialsViews.com

Imaging

Modular Integration of Upconverting Nanocrystal–

Dendrimer Composites for Folate Receptor-Speciﬁc

NIR Imaging and Light-Triggered Drug Release

Pamela T. Wong, Dexin Chen, Shengzhuang Tang, Sean Yanik, Michael Payne,

Jhindan Mukherjee, Alexa Coulter, Kenny Tang, Ke Tao,* Kang Sun,*

JamesR. Bake Jr r. ,a nd Seo Kk i Choi *

U

pconversion nanocrystals (UCNs) display near-infrared (NIR)-responsive

photoluminescent properties for NIR imaging and drug delivery. The development of

effective strategies for UCN integration with other complementary nanostructures for

targeting and drug conjugation is highly desirable. This study reports on a core/shell-

based theranostic system designed by UCN integration with a folate (FA)-conjugated

dendrimer for tumor targeting and with photocaged doxorubicin as a cytotoxic

agent. Two types of UCNs (NaYF :Yb/Er (or Yb/Tm); diameter = ≈50 to 54 nm) are

4

described, each displaying distinct emission properties upon NIR (980 nm) excitation.

The UCNs are surface modiﬁed through covalent attachment of photocaged

doxorubicin (ONB-Dox) and a multivalent FA-conjugated polyamidoamine

(

PAMAM) dendrimer G5(FA) to prepare UCN@(ONB-Dox)(G5FA). Surface

6

plasmon resonance experiments performed with G5(FA) dendrimer alone show

6

−

9

nanomolar binding avidity (K = 5.9 × 10 M) to the folate binding protein. This

dendrimer binding corresponds with selective binding and uptake of UCN@

D

(ONB-Dox)(G5FA) by FAR-positive KB carcinoma cells in vitro. Furthermore,

UCN@(ONB-Dox)(G5FA) treatment of FAR(+) KB cells inhibits cell growth in

a light dependent manner. These results validate the utility of modularly integrated

UCN-dendrimer nanocomposites for cell type speciﬁc NIR imaging and light-

controlled drug release, thus serving as a new theranostic system.

Dr. P. T. Wong, S. Tang, S. Yanik, M. Payne,

Dr. J. Mukherjee, A. Coulter, K. Tang,

Prof. J. R. Baker Jr., Prof. S. K. Choi

Michigan Nanotechnology Institute

for Medicine and Biological Sciences

Department of Internal Medicine

University of Michigan Medical School

Ann Arbor, MI 48109, USA

1. Introduction

Rapid developments in biological nanotechnology have

led to the identiﬁcation of numerous types of photolumi-

[

1]

nescent nanomaterials

with promising applications in

[1c–e,2]

[1c]

[1a]

imaging,

catalysis, and therapeutics. These include

[1d]

[3]

gold (Au) nanoparticles (NPs),

quantum dots, titanium

[

1c,1e]

and upconversion nanocrystals^[⁴^](UCNs). The

dioxides,

E-mail: skchoi@umich.edu

tailored applications of these nanomaterials are achieved typ-

ically through their surface modiﬁcation or modular integra-

tion with other functionally designed NPs that bring together

complementary capabilities needed for speciﬁc cell targeting

and delivery of payloads. In particular, modular integration

of NPs with near-infrared (NIR)-excited UCNs has shown

promising potential for deep-tissue imaging and controlled

D. Chen, Prof. K. Tao, Prof. K. Sun

State Key Laboratory of Metal Matrix Composites

School of Materials Science and Engineering

Shanghai Jiao Tong University

Shanghai 200240, P. R. China

E-mail: ktao@sjtu.edu.cn; ksun@sjtu.edu.cn

[

5]

DOI: 10.1002/smll.201501575

6078 www.small-journal.com

small 2015, 11, No. 45, 6078–6090

www.MaterialsViews.com

Figure 1. Modular assembly of a near-infrared (NIR)-excited upconversion nanocrystal (UCN) covalently linked with multivalent folate (FA)-

conjugated dendrimer for folate receptor (FAR)-targeted cell imaging and with photocaged doxorubicin for light triggered release of doxorubicin.

payload delivery.^[²^,⁵^b^,⁶^]However, despite the proven beneﬁts as for the release of entrapped or attached guest molecules

imparted by such modularly integrated UCNs, the design (cisplatin,^[Dox,

5d]

[12b,13]

bovine serum albumin ) and acti-

[5g]

[

14]

principles of each integrated UCN nanostructure have not vation of photocaged molecules (platinum agent,

been addressed through a systematic assessment of the requi- payload,^[⁶^d^,¹⁵^]FA ligand ).

siRNA

[6a]

site design parameters. In this study, we report on the design

As a shell module, a ﬁfth generation (G5) PAMAM den-

and characterization of a novel core/shell nanostructure drimer is coated onto the UCN surface and is responsible for

[

7]

consisting of three functionally orthogonal modules: UCN

the speciﬁc targeting of the composite to a tumor cell. The

covalently coated with poly(amidoamine) (PAMAM) den- G5 dendrimer is the prototype of dendritic polymers. It is

[

8]

[9]

[8,16]

drimer and photocaged doxorubicin (ONB-Dox). This characterized by a globular shape (diameter = 5.4 nm)

study addresses the key design aspects important for the consisting of a large number of amine-terminated branches

efﬁcacy of these nanocomposites in tumor receptor-targeted, (n_t_h_e_o_r= 124), each chemically amenable to ligand or drug

temporally controlled drug delivery by assessing the effec- conjugation in a predeﬁned orientation with conformational

tiveness of each of the modular components in their func- ﬂexibility. Surface modiﬁcation of the dendrimer with mul-

tional roles, including cell speciﬁc targeting, imaging, and tiple targeting ligands offers the added advantage of multi-

[

17]

drug payload release.

valent binding,

allowing for extremely tight and selective

As a core module, UCNs are an emerging class of photo- adsorption to target receptors on the cell surface. Together,

luminescent nanostructures made of host materials such as these properties make the G5 dendrimer an ideal platform

[

7,10]

[18]

NaYF doped with lanthanide ions (Yb, Er, Tm, and Nd).

for cell targeting. The use of G5 dendrimers for targeting

4

[

10e]

[18a,b,d]

These UCNs are excited by NIR light (980, 800 nm

cancer biomarkers including folate receptor-α (FAR ),

α

[

5f,19]

[20]

and have the unique photophysical ability to convert the riboﬂavin (RF) receptor,

α β integrin,

and prostate-

has been studied

extensively. Thus, the G5 dendrimer has been established as

v

3

[

21]

absorbed NIR to shorter visible (vis) and ultraviolet (UV) speciﬁc membrane antigen (PSMA

[

5b,7b,10d,e]

light with excellent photostability and brightness.

Thus, the incorporation of UCNs offers two primary advan- one of the most effective NPs

[

18b,18d,21b,22]

for tumor-targeted

and therapeutic payloads.^[¹⁸^a^,^d^,²¹^a^,²⁴^]

The present study incorporates these two nanoparticle

[

19c,20b,23]

tages to modular applications. First, the NIR-excited UCN delivery of genes

core allows for sustained tracking of tumor cells due to the

lack of photobleaching which commonly limits the utility of modules in combination with photocaged doxorubicin as

organic ﬂuorophores. Furthermore, NIR radiation can pene- an anticancer therapeutic agent to generate a novel core/

trate signiﬁcantly deeper into tissue than UV radiation or vis- shell nanostructure, UCN@(ONB-Dox)(G5FA), which

[

6b,c]

ible light which is used for photodynamic therapy (PDT).

refers to a UCN core coated with a FA-conjugated G5 den-

[

9]

Together, these properties make UCNs highly valuable for drimer G5(FA) and linked to a photocaged Dox molecule

[

2,5b,6e,11]

use in in vivo imaging studies.

Second, UV emis- (Figure 1). We designed two types of NIR-excited UCNs,

sion from the UCN core upon excitation by NIR light can one with the ability to emit vis light (540, 650 nm) suitable

be harnessed to trigger the release or activation of a conju- for imaging, and the other with the ability to emit UV light

gated payload through a photochemical mechanism.^[

5g,6a–d,12]

(340–360 nm) for controlled drug release. The modular

The effectiveness of such modular applications of UCNs has nanostructure also carries a drug payload through covalent

been demonstrated in the generation of reactive oxygen spe- attachment to the UCN core, rather than to the dendrimer

[

6b,c]

cies (ROS) for photodynamic therapy (PDT),

as well shell—the latter of which has been used in our earlier

small 2015, 11, No. 45, 6078–6090

www.small-journal.com 6079

full papers

www.MaterialsViews.com

methods for dendrimer-based drug delivery.^[⁹^,¹⁸^a^,²⁵^]The approach.^[⁴^a^,⁶^e^–^h^,²⁶^]We applied a hot injection method^[⁷^b^]

rationale for this site of drug conjugation relates to receptor for synthesizing two types of UCN (x = 1, 2) which display

x

binding avidity and drug release kinetics which are described distinct upconversion properties (Figure 2). While each of

below. However, FA was attached to the dendrimer shell these is excited by NIR light absorption (λ_a_b_s= 980 nm),

because of its well validated platform capacity for tight and the ﬁrst type of UCN, UCN , was designed to emit light in

1

[

18c]

selective FAR binding via multivalency,

and also because the vis range, while the second type of UCN, UCN , was

2

of the greater surface accessibility of FAR to the FA on the designed to emit light at shorter UV–vis wavelengths. First,

dendrimer as compared to presentation on the inner UCN we synthesized UCN type 1 (UCN = NaYF : 20% Yb/2%

1

4

surface. This study evaluates the design principles involved Er) by the cothermolysis (290–330 °C) of CF COONa and

3

in generating and optimizing a series of UCN@(G5FA) and (CF COO) Re, (Re = Y, Yb, Er; Y:Yb:Er = 1.0:0.26:0.026)

3

UCN@(ONB-Dox)(G5FA) nanostructures, and demon- with oleic acid and sodium oleate as mixed ligands as

[

7b]

strates their effectiveness with regards to FAR-speciﬁc cell described elsewhere. This method also allowed us to vary

binding, NIR-based optical detection, and light controlled the diameter of UCN by adjusting the amount of sodium

1

cytotoxicity. The ﬁndings of these studies provide important oleate added in the thermolysis process. Second, we prepared

insights into the modular design of UCN-based core/shell UCN type 2 (UCN = NaYF : 25% Yb/0.3% Tm) in a similar

2

4

nanostructures and their relationship with the biological manner, except a different (CF COO) Re (Re = Y, Yb, Tm)

3

activity of the nanoparticle, allowing for rational optimiza- mixture and ratio (Y:Yb:Tm = 1.0:0.33:0.04) was used. After

tion of design components. preparation of the UCNs, selected UCN and UCN , each

1

2

with d = 54 or 50 nm, respectively, were surface modiﬁed by

coating with a thin amine-terminated silica layer (NH )SiO .

2

. Results and Discussion

This modiﬁcation was performed following a nonionic water-

[

27]

in-oil microemulsion method by treatment with (3-amino-

propyl)triethoxysilane (APTES) via a silanization reaction.

The size and distribution of UCNs were determined by

2.1. Synthesis of UCNs

A number of methods have been developed for UCN syn- performing transmission electron microscopy (TEM). TEM

thesis, each with variations in the reaction condition or images acquired for a representative NaYF :Yb/Er sample

4

Figure 2. a) Synthetic scheme for upconversion nanocrystals UCN (type x = 1 or 2) and (NH )SiO -coated UCN : i) RE(CF CO ) (RE = Y, Yb and

x

2

x

3

2 3

Er (for UCN ) or Tm (for UCN )), CF CO Na, oleic acid sodium salt, 1-octadecene, 330 °C, 20 min; ii) (3-aminopropyl)triethoxysilane, ammonia,

1

2

3

2

polyoxyethylene nonylphenylether (M = 441); b,c) Transmission electron microscope (TEM) images of representative UCN (NaYF :20% Yb/2%

n

1

4

Er) and UCN₂(NaYF :25%Yb/0.3% Tm). Right: Selected area electron diffraction (SAED) patterns; d) Upconversion luminescence spectra of

4

UCN acquired by excitation at 980 nm are compared as a function of UCN diameter: black (diameter = d = 54 nm), red (d = 30 nm), and blue

1

(

d = 20 nm), each synthesized individually; e) An upconversion luminescence spectrum of UCN (d = 50 nm) by excitation at 980 nm. Inset: a photograph

2

−1

showing emission of green light (c) or faint blue light (e) by exposure of UCN dispersed in chloroform (2 mg mL ) to NIR laser light (980 nm).

x

6080 www.small-journal.com

small 2015, 11, No. 45, 6078–6090

www.MaterialsViews.com

showed that UCN₁is comprised of nanocrystals with an ratio. The major γ-isomer 1 was fully analyzed for its identity

1

average diameter of ≈54 nm and a narrow size distribution by standard analytical characterization methods, including H

(range = 45–60; Figure 2b). Some of these nanocrystals were NMR and mass spectrometry (Figure S1, Supporting Infor-

hexagonal in shape, but others appeared less deﬁned. How- mation), and the linker N-Boc group was deprotected by

ever, selected area electron diffraction (SAED) patterns of TFA treatment, yielding FA-linker 2.

NaYF :Yb/Er (right) showed diffraction rings corresponding

4

to the (100), (110), (200), and (201) planes, indicating a hex-

agonal lattice. TEM analysis performed for NaYF :Yb/Tm 2.3. Synthesis of Photocaged Doxorubicin (ONB-Dox)

4

(

UCN ) also showed hexagonal-shaped nanocrystals with an

2

average diameter of ≈50 nm with a very narrow distribution The drug module in the nanocomposite consists of a

Figure 2c). photocaged doxorubicin molecule. Photocaging refers to the

We measured the upconversion emission proper- temporary inactivation of a drug or ligand molecule with a

(

[

31]

ties of NaYF :Yb/Er nanocrystals (UCN ; d = 54 nm) by molecular cage which is cleavable by light exposure. This

4

1

NIR excitation at 980 nm (Figure 2d). These nanocrystals technology serves as an effective means to release or acti-

showed two major emission peaks including a green band vate drug/imaging molecules through an actively controlled

4

3+

[22c,32]

(

max = 542 nm; S → I transition for Er ) and a red mechanism

as opposed to passive release mecha-

which are controlled by tumor-associated factors

3

/2

15/2

band (max = 651 nm; F₉→ ⁴I₁₅_/₂transition for Er ). Fluo- nisms

4

3+

[33]

/2

rescence intensity at the green band was greater than at the such as lower pH or overexpressed proteolytic enzymes. We

red band. Such emission peaks and relative intensities for have demonstrated the effectiveness of this drug release

UCN are consistent with the values determined for other mechanism using a dendrimer platform in which photocaged

1

[

28]

[9]

[34]

Er, Yb-doped NaYF₄nanocrystals in previous studies.

doxorubicin (Dox) or methotrexate carried by an FAR-

In addition, it is notable that the size of the UCN has a targeted G5 dendrimer is activated and released from the

1

signiﬁcant impact on the efﬁciency of its upconversion dendrimer in a controlled manner using UV light as the

[

6f,29]

luminescence.

A decrease in the diameter from ≈50 to trigger. Photocaged Dox 3 was prepared by derivatization of

3

0 or 20 nm led to a 10–15-fold reduction in emission inten- Dox with an ortho-nitrobenzyl (ONB) group at its daunosa-

[

9]

sity. Such size variation, however, did not change the frac- mine moiety and fully characterized as described elsewhere

tional intensity (f ) between the green and red bands, which (Scheme 1, Scheme S2, Supporting Information). Removal of

g/r

remained largely similar. Thus the UCN with a diameter of the linker N-Boc group resulted in ONB-Dox 4 which con-

1

5

4 nm was used for the subsequent integration steps. The tains a free amine moiety at the linker to be used for covalent

upconversion luminescence properties of NaYF :Yb/Tm attachment to the dendrimer or UCN.

4

nanocrystals UCN (d = 50 nm) were measured by excitation

2

at 980 nm, which gave three emission peaks containing a UV

[

30]

band (340–360 nm), a blue band (450–475 nm) and a broad 2.4. Synthesis of G5(FA)(ONB-Dox) Conjugate

peak at 800 nm (Figure 2e). Thus, UCN emits a combina-

2

tion of vis and UV light that serves as the mechanism for UV Glutaric acid (GA)-modiﬁed dendrimer 6 G5(GA)^[⁹^]was

light-controlled drug release.

employed for conjugation with FA-linker 2 alone or in

combination with photocaged Dox 4 following a N-(3-

dimethylaminopropyl)-N′-ethylcarbodiimide

amide coupling method as described elsewhere (Scheme 1).

(EDC)-based

[9]

2.2. Synthesis of FA-Linker

First, conjugate 7 G5(FA) (n

= 6) was prepared by cova-

n

mean

Multivalent conjugation of FA to G5 dendrimers serves as lent attachment of 2 and puriﬁed by dialysis to remove unre-

an effective strategy for tight binding of the conjugate to acted ligand and reagents present in the reaction mixture

[

18a,b,d]

the FARs overexpressed (+) on certain tumor cells.

Our method for FA conjugation involves prederivatization homogeneity ≥94%) and structural identity were fully veri-

of FA with a linker terminated with a primary amine as a ﬁed by ¹H NMR, UV–vis spectroscopy (λ

= 354 nm,

using membrane tubing (MWCO 10 kDa). Its purity (polymer

max

−1

chemical handle for amide coupling to a carboxylic acid-ter- ε = 62 800 m cm ; FA), matrix-assisted laser desorption

minated dendrimer molecule. We chose 2,2′-(ethylenedioxy)- ionization time-of-ﬂight (MALDI-TOF) mass spectrometry

bis(ethylamine) as a linker as it has a medium length that (M = 38 600 g mol ) and gel permeation chromatography

−1

r

−1

allows for conformational ﬂexibility. The FA derivatization (GPC; M = 42 000 g mol ) (Table S1, Figures S2–S4, Sup-

p

consisted of two steps in sequence: preactivation of FA by porting Information). Second, conjugate 8 G5(FA) (ONB-

n

treatment with PyBOP ((benzotriazol-1-yloxy)tripyrrolidi- Dox)_mwas prepared similarly by coattachment of

2

nophosphonium hexaﬂuorophosphate) and 1-hydroxybenzo- and 4 to dendrimer G5(GA). The resulting conjugate 8

triazole (1-HOBt), and subsequent reaction with the diamine G5(FA) (ONB-Dox)_m(n_m_e_a_n= 6; m

in situ (Scheme 1). After the linker derivatization, the pri- dialysis (≥96%) and characterized by H NMR, UV–vis spec-

= 6.9) was puriﬁed by

n

mean

1

−1

mary amine left at the opposite end of the linker was tem- troscopy (λ_m_a_x= 355 nm, ε = 163 300 m cm (FA); 490 nm,

−1

porarily protected as an N-Boc group for the convenience of ε = 121 000 m cm (Dox)), MALDI-TOF spectrometry

−1

puriﬁcation by ﬂash column chromatography which enabled (M = 40 400 g mol ), and other methods as noted above.

r

the separation of two regioisomeric adducts of the linker at

Values of the valencies of the ligand (n) and the caged

l-Glu, γ-adduct 1 (19.3%), and α-adduct (6.6%), at a ≈2.9:1 drug (m) were determined by Beer–Lambert analysis of

small 2015, 11, No. 45, 6078–6090

www.small-journal.com 6081

full papers

www.MaterialsViews.com

Scheme 1. Synthesis of a) folic acid (FA) derivatized with linker 2, b) doxorubicin (Dox) caged with an ortho-nitrobenzyl (ONB) group 4, and c) G5

a)

dendrimer conjugates 7 and 8. Reagents and conditions: i) PyBOP, HOBt, (i-Pr) EtN, DMSO, 1.5 h; then 2,2′-(ethylenedioxy)diethylamine (3 mol

2

equiv), 18 h; ii) (Boc) O (7 mol equiv), 6 h; iii) CF CO H, CH Cl , 0.5 h; iv) Glutaric anhydride, Et N, MeOH; v) EDC, NHS, 4-dimethylaminopyridine,

2

3

2

3

DMF, 12 h; vi) 2, 7 h; vii) 2, 4, 12 h. Each reaction was performed at room temp.

UV–vis absorption (A) at 354 nm (FA) or 490 nm (Dox) Information) which is attributable to the cleavage of the

relative to a standard calibration plot prepared by free FA ONB linker and the formation of nitrosobenzaldehyde as

[

9]

or Dox and by the analysis of NMR peak integration as sum- suggested previously. HPLC analysis allowed quantitative

marized in Table S1, Supporting Information. The two sets of evaluation of the release kinetics of Dox from 3. Figure 3

values were consistent with narrow deviations lying within shows HPLC traces acquired for the photolysed solutions as

±2% of their mean value.

a function of exposure time. Very shortly after irradiation,

a new peak appeared with a retention time (t ) of 7.8 min

R

which was identical to that of free Dox. The peak intensity

increases rapidly as a function of UVA exposure time while

the peak for ONB-Dox 3 (t_R= 9.8 min) decreases. Area

2.5. Light-Controlled Release of Doxorubicin

Our earlier studies showed that photochemical cleavage of under curve (AUC) analysis of the HPLC traces indicated

−

1

−1

the ortho-nitrobenzyl (ONB; λ_m_a_x= 340 nm, ε = 2750 m cm ) that Dox release occurs efﬁciently with ≥85% being released

cage occurs rapidly upon exposure to UVA light (365 nm) within 5 min. The rapid decay of ONB-Dox had a ﬁrst-order

[

9,34,35]

−3 −1

with qunatum efﬁciency (Φ) of 0.29.

We validated the rate constant of 7.5 × 10 s .

An important factor to consider is the attachment site

light-triggered Dox release from ONB-Dox 3 by monitoring

its progress with UV–vis spectroscopy and high-performance of photocaged Dox to the UCN-dendrimer system. Dox-

liquid chromatography (HPLC) (Figure 3). Drug release ONB can either be directly attached to the UCN surface

was evidenced by changes in the UV–vis spectral traces of or attached through conjugation to the FA-conjugated

the light-exposed solutions (0.103 × 10⁻m in aq methanol) dendrimer. We ﬁrst investigated whether the kinetics of

compared to unexposed solutions. They showed a gradual Dox release from ONB-Dox is affected by dendrimer con-

increase in absorption at 230–280 nm (Figure S8, Supporting jugation. Light-controlled Dox release from conjugate 8

3

6082 www.small-journal.com

small 2015, 11, No. 45, 6078–6090

www.MaterialsViews.com

Figure 3. a) Light-controlled release of doxorubicin (Dox) from 3 ONB-Dox (top) or 8 G5 dendrimer (bottom) through the cleavage of the ortho-

−

3

nitrobenzyl (ONB) cage. b) HPLC traces of 3 (0.103 × 10 M in 10% aq MeOH) after UVA exposure (365 nm), and c) AUC analysis of Dox release (%)

−

6

vs exposure time. d) HPLC and e) GPC traces of 8 (24.8 × 10 M in water) after UVA treatment as a function of exposure time (t = 0, 10, 30 min).

inset: %AUC (t = exposure time) of each peak relative to AUC (t = 0).

G5(FA) (ONB-Dox) in water was monitored with UV–vis shifted to faster retention times over the course of UV expo-

6

6.9

spectrometry and HPLC analysis (Figure 3). Changes in the sure (Figure 3d). However, AUC analysis of Dox release was

UV–vis spectral traces (Figure S8, Supporting Information) not attempted for these HPLC traces since the peak for free

are evident, showing an increase in absorbance at 280 nm as Dox (t_R= 7.8 min) was entirely buried within the broad den-

seen with 3 ONB-Dox. Overlaid HPLC traces, each acquired drimer peak, and the kinetics of Dox release could not be

after a variable length of UV exposure, showed that the determined for direct comparison with that of free ONB-Dox

broad peaks corresponding to the dendrimer species are (Figure 3b).

small 2015, 11, No. 45, 6078–6090

www.small-journal.com 6083

full papers

www.MaterialsViews.com

Table 1. Kinetic parameters and equilibrium dissociation constants of FA, 7 G5(FA) and 8 G5(FA) (ONB-Dox) to a CM5 sensor chip immobilized

6

6.9

with folate binding protein (FBP).

k_a^a⁾[M⁻¹s⁻¹

1.5 × 10³

1.5 × 10⁵

1.7 × 10⁵

]

k [s⁻¹

a)

]

K_D^b⁾[M]

β^c⁾

1.0

Dendrimer

FA

d

9.0 (±2.9) × 10^–³

8.5 (±5.3) × 10^–⁴

2.8 (±0.36) × 10^–³

5.9 × 10^–⁶

5.9 × 10^–⁹

1.7 × 10^–⁸

7

8

G5(FA)₆

1,000 (167^d⁾)

347 (58^d⁾)

G5(FA) (ONB-Dox)

6

6.9

^a⁾Mean value (±SD) from serially diluted concentrations (n ≥ 4); ^b⁾KD = kd/ka; ^c⁾Multivalent enhancement factor = KD (FA)/KD (dendrimer); ^d⁾Valency corrected = multivalent enhancement factor/6

= FA valency).

(

We expect that Dox release from conjugate 8 results in multivalent receptor-ligand interactions that occur on the cell

[

18c,37]

a change in the distribution of the dendrimer species. We surface.

examined this by using a novel method based on GPC analysis Folate binding protein (FBP) was immobilized on the

of the photolysed solutions. Elution proﬁles of the dendrimer sensor chip surface as a model receptor for FAR overex-

[

18c,19c,25]

solution before and after exposure are shown in Figure 3e. pressed on a tumor cell.

The FBP model surface

The GPC trace for 8 shows a characteristically broad peak was prepared following a standard EDC/NHS-based amide

[

18c]

−2

shape (width at half height = w = 7.0 min), which is indica- coupling method

at a protein density of 13.1 ng mm

h

tive of a wide dendrimer distribution. This is consistent with (response unit (RU) = 13090; Figure S6, Supporting Infor-

[

36]

a prediction by Poissonian simulation

performed for 8 mation). This FBP density is equivalent to a number den-

G5(FA) (ONB-Dox)_m(n_m_e_a_n= 6; m_m_e_a_n= 6.9). This Poisson sity of ≈2.6 × 10¹¹FBP molecules per mm , comparable to

2

n

analysis suggests that the conjugate solution should consist the density of FAR overexpressed in several malignant

[

38]

of a large distribution of dendrimer species with variable cell lines including ovarian and endometrial cancers

10

11

2 [18c]

conjugation ratios of the ligand and drug valency (n = 1–12; (≈10 –10 receptor molecules per mm ).

m = 2–13; Figure S4, Supporting Information). GPC traces of the immobilized FBP to FA was determined by acquiring

after UV exposure for 10 and 30 min resulted in a signiﬁ- dose-dependent sensorgrams (Figure S7a, Supporting Infor-

The afﬁnity of

8

cant decrease in the w to 5.9 min and 5.5 min, respectively. mation). Each of the sensorgrams was ﬁt according to a global

h

This narrower distribution was expected, as the heteroge- curve ﬁtting analysis based on a Langmuir model (1:1 binding

neity would be reduced upon release of the attached Dox mode) to determine the two rate constants: k (association)

a

[

39]

molecules (m_m_e_a_n= 6.9 → 0). In addition, the GPC trace after and k (dissociation). Using these two kinetic parameters,

d

3

0 min exposure shows a similarity in the peak shape and t_Rwe calculated an equilibrium dissociation constant K

D

−

6

when compared to 7 G5(FA) , indicating the efﬁcient release (= k k ) of 5.9 × 10 m (Table 1), comparable to those

of Dox (m_m_e_a_n≈ 0). AUC analysis of the GPC traces (inset, reported in the literature (K_D= 2–5 × 10 m).

6

d a

−

6

[18c]

Figure 3e) was performed and the AUC decay as calculated

SPR experiments performed with 7 G5(FA)₆showed

as a ﬁrst-order rate constant (2.8 × 10⁻⁴s ) occurs at a rate dose-dependent adsorption of the dendrimer to the FBP

−1

≈

27-fold lower than free ONB-Dox by HPLC.

These results demonstrate that UV light serves as an was detectable at ≤0.16 × 10 m, a concentration at which no

surface (Figure S7b, Supporting Information). Adsorption

−

6

effective trigger for the active control of Dox release from the binding is detectable by free FA, indicating tighter binding

ONB photocaged form. Interestingly, Dox release occurred of the dendrimer conjugate than free FA alone. Further-

faster from ONB-Dox 3 than from the FAR-targeted den- more, this conjugate dissociates much more slowly than FA.

drimer 8 carrying the photocaged Dox. These observed dif- Such markedly slow desorption is consistent with earlier

ferences have not been reported in literature and may be studies that show slow dissociation kinetics of other multiva-

due to the presence of FA coattached to the dendrimer. The lent G5(FA) and G5(MTX) (n ≥ 3; MTX = methotrexate)

n

[

18c,24c]

pteridine chromophore in FA has signiﬁcant UV absorptivity conjugates.

(

Global ﬁtting analysis gave a k value cor-

d

λ_m_a_x= 354 nm; ε = 62 800 m⁻cm ) which can effectively responding to a ≈10-fold slower rate of desorption by 7 than

1

−1

compete with the ONB linker for light absorption. Thus, FA (Table 1). This slow dissociation is a kinetic feature that

this release study suggests that attachment of ONB-Dox contributes to the high avidity binding commonly associated

[

17b,17c]

to the UCN is more preferable than to the FA-conjugated with multivalent systems.

dendrimer.

From this SPR analysis, the

K_Dvalue for 7 was determined as 5.9 × 10⁻m, reﬂecting high

9

binding avidity with a multivalent enhancement factor of 167

(

valency corrected).

2.6. Binding Avidity of G5(FA) Conjugates

Finally, the SPR experiments were performed with 8

G5(FA) (ONB-Dox) , which is identical to 7 G5(FA) in its

6

6.9

6

In our UCN-based core/shell nanostructure, the outer shell FA valency but in addition carries an ONB-Dox payload.This

layer is comprised of the FA-conjugated dendrimer module conjugate showed faster apparent dissociation than 7.Its mean

which confers FAR-targeting properties to the nanostructure. fractional desorption (= (RU_d_e_s_o_r_p_t_i_o_n/RU_a_d_s_o_r_p_t_i_o_n) × 100%;

We determined the multivalent binding avidity (K ) of FA- see Figure S7, Supporting Information) calculated at the end

D

conjugated dendrimers 7, and 8 to the receptor on the surface of the dissociation phase (t = 600 s; 88 ± 2%) was larger than

using SPR spectroscopy. SPR is highly suited for studying that of 7 G5(FA) (t = 600 s; ≈37 ± 12%; Figure S7, Supporting

6

6084 www.small-journal.com

small 2015, 11, No. 45, 6078–6090

www.MaterialsViews.com

Information). Such greater desorption of 8

is largely explained by its ≈3-fold higher k_d

value which leads to lower avidity with a

K_Dvalue of 1.7 × 10⁻m. We believe this

reduction in avidity is caused primarily by

8

[

19b,40]

an unfavorable steric effect

in which

the extended dendritic branches with

the ONB-Dox might interfere with the

optimal multivalent association between

the multiple FA ligands and the receptors

on the surface.

In summary, this SPR binding study

demonstrated high avidity binding by

FA-conjugated dendrimers to FBP recep-

tors on the cell surface. However it was

also evident that the copresence of a drug

payload carried by the ligand-conjugated

dendrimer compromises the targeting

capability of the dendrimer carrier. These

SPR results along with the slower rate

of drug release observed for the dual-

functional module as described above led

us to rationally identify a better design

approach for the UCN core/shell, in which

the two separate modules, 7 G5(FA) and

6

ONB-Dox are directly attached to the

UCN core independently.

Scheme 2. Synthesis of various UCN-based core/shell nanostructures, each conjugated with

2

.7. Covalent Integration of UCNs with

a)

G5(FA) , Cy5 dye and/or ONB-Dox. a) 10 UCN1@(Cy5)(G5FA); b) 13, 14 UCN @(ONB-Dox)

6

x

b),c)

d)

G5(FA) and ONB-Dox

(G5FA) (x = 1, 2); c,d) UV–vis absorption spectra

c; inset) of selected nanostructures. Reagents and conditions: i) 7 G5(FA) , EDC, NHS, DMF,

and ﬂuorescence emission spectrum

a)

(

6

rt, 13 h; ii) Sulfo-Cy5 NHS ester, DMSO, rt; iii) epibromohydrin, MeOH, rt; then 4 ONB-Dox,

Dendrimer conjugation to the UCN was

performed by amide coupling between a

b)

−1

c)

−1

rt to 45 °C. [9 or 10] = 1 mg mL water, [7] = 62.5 µg mL water; [12 or 14] = 1 mg mL

d)

−1

MeOH; [10] = 0.25 mg mL MeOH; λ_e_x= 600 nm, λ_e_m= 667 nm.

carboxylic acid residue in 7 G5(FA) and

6

an amine residue present in (NH )SiO -

2

coated UCN (type x = 1, 2). It is illustrated by an EDC- attached to the UCN by analysis of the ONB-Dox solution

x

based covalent attachment of 7 to UCN₁in the synthesis (λ_m_a_x= 495 nm) before and after the conjugation reaction.

of 9 UCN @(G5FA) (Scheme 2a). This dendrimer-coated The ﬁndings suggested that the weight fraction of ONB-Dox

1

UCN was further modiﬁed for imaging purposes by conjuga- to UCN (w/w) was approximately 3.5% (11) and 6.5% (12)

tion with a sulfo-Cy5 to yield 10 UCN @(Cy5)(G5FA). We (the Supporting Information). In the next step, 7 G5(FA) was

1

6

attached the Cy5 dye (λ_e_x= 600 nm, λ_e_m= 667 nm) to the attached by EDC-based amide coupling to 11 or 12 UCN @

x

nanostructure as a secondary probe for vis imaging in addi- (ONB-Dox), which resulted in 13 and 14 UCN @(ONB-Dox)

x

tion to the primary NIR imaging properties of the UCN core (G5FA) (x = 1, 2), respectively (TEM images of 13 and 14

in order to measure cellular uptake. The UV–vis spectra of provided in Figure S5, Supporting Information). The efﬁ-

these UCN suspensions show an increase in the absorption ciency of this dendrimer conjugation was estimated by UV–

range (≈240–400 nm) which is attributable to FA absorption vis analysis, indicating that ≈9% of the dendrimer added in

(

Scheme 2c). In addition, the presence of the sulfo-Cy5 label large excess in the reaction was covalently attached. UV–vis

on 10 is clearly indicated by the characteristic ﬂuorescence spectra overlaid in Scheme 2d show corresponding changes

emission spectrum (λ_e_x= 600 nm, λ_e_m= 667 nm; inset). in the absorption λ_m_a_xvalues and intensities indicative of sur-

Coattachment of 7 G5(FA) and 4 ONB-Dox to UCN face modiﬁcation of UCN with ONB-Dox (λ_m_a_x= 498 nm)

6

was performed by a slightly modiﬁed method because of the and G5(FA) ( λ

= 200–220 nm).

6

max

need for a surface modiﬁcation that allows crosslinking with

amine-containing 4 (Scheme 2b). Thus, (NH )SiO -coated

2

UCN was treated with epibromohydrin which converts some 2.8. Confocal Imaging Study of UCN@(G5FA) in KB Cells

of its surface amine residues to amine-reactive epoxide resi- In Vitro

dues. This modiﬁed UCN reacted with 4 ONB-Dox, affording

drug-crosslinked 11 and 12 UCN@(ONB-Dox). UV–vis spec- To examine the cellular binding and uptake of the UCN

trometry was used to determine the fraction of ONB-Dox nanostructures, we treated FAR-overexpressing (+) KB

small 2015, 11, No. 45, 6078–6090

www.small-journal.com 6085

full papers

www.MaterialsViews.com

Figure 4. Cellular binding and uptake of 9 UCN @(G5FA) and 10

1

UCN @(Cy5)(G5FA) in FAR(+) KB cells imaged by confocal microscopy.

1

a–d) Images of KB cells treated with 10 or 9 taken in Cy5 mode (λ_e_x

=

6

40 nm; λ_e_m= 670 nm). e–h) Images of KB cells treated with 10 or 9

taken in both NIR mode for imaging the UCN core (λ_e_x= 980 nm; λ_e_m

5

Cy5 (green).

=

Figure 5. a,b) Flow cytometric analysis of binding and uptake of

10 UCN1@(Cy5)(G5FA) by FAR(+) KB cells and FAR(−) B16 F10 cells.

c) Fraction of cells with bound or internalized UCN, determined by the

percentage of cells with Cy5 ﬂuorescence (FL3) greater than that of cells

50 nm, 650 nm) and Cy5 mode. DAPI (blue: nuclei); UCN core (red),

−

1

tumor cells with increasing concentrations (50–500 µg mL ) alone (gate M1). Each data point represents a mean value of duplicate

measurements (±SD).

of 9 UCN @(G5FA) or 10 UCN @(Cy5)(G5FA) for 2 h,

1

and imaged the cells with confocal microscopy using two

detection modes: (i) Cy5 mode (λ_e_x= 640 nm; λ_e_x= 670 nm) the absence of the Cy5 label (Figure 4g,h). Thus, visible

and (ii) UCN NIR mode (λ = 980 nm; λ_e_x= 550, 650 nm). light-emitting UCN serves as an effective platform for label

1

ex

1

As shown in Figure 4, confocal images of the cells with 10 free cellular imaging with NIR excitation.

clearly indicated dose-dependent binding and uptake by

KB cells (Figure 4a–c). No signal was detected through this

Cy5 channel in cells treated with 9, an identical nanoparticle 2.9. Flow Cytometric Analysis of UCN@(G5FA) Binding

without the Cy5 attachment, conﬁrming that the ﬂuorescence and Uptake in KB Cells In Vitro

signal is from the Cy5 (Figure 4d).

We then imaged the treated cells in NIR mode with exci- Cellular association of 10 UCN @(Cy5)(G5FA) was further

1

tation at 980 nm and detection at 550 or 650 nm (Figure 4). investigated by ﬂow cytometry. Two cell lines which are dis-

The localization of FAR-targeted UCNs on the cell surface tinct in their FAR density were used to determine the speci-

as well as intracellularly was observed under both the NIR ﬁcity of UCN in FAR targeting: FAR(+) KB cells and FAR(−)

and Cy5 imaging modes in cells treated with 10 (200 or B16-F10 cells (a mouse melanoma cell line). Cells were incu-

−

1

−1

5

00 µg mL ). Merged images showed colocalization of the bated with variable concentrations of 10 (0–50 µg mL ) for

NIR and Cy5 signals, demonstrating the ability to image 1 h in FA-free media at 37 °C and analyzed. Histograms and

UCNs with either mode of detection (Figure 4e,f). Further- ﬂuorescence intensities (FL3 for Cy5) of gated live cells are

more, UCN 9 could be detected under the NIR mode despite shown in Figure 5. The fraction of KB cells with bound 10

6086 www.small-journal.com

small 2015, 11, No. 45, 6078–6090

www.MaterialsViews.com

increased in a dose-dependent manner. In contrast, FAR(−)

B16 cells showed only a small fraction of cells with bound

UCN under the same conditions (Figure 5b). The frac-

tion of bound cells determined at the highest concentration

−

1

(

50 µg mL ) was 44 ± 19% for FAR(+) KB cells compared to

only 11 ± 3% for FAR(−) B16 cells. These results support the

speciﬁc binding and uptake of 10 UCN @(Cy5)(G5FA) by a

1

FAR-dependent mechanism. As evidence further supporting

−

1

FAR speciﬁcity, cells were incubated with 10 (10 µg mL

−6

UCN equivalent to ≈0.02 × 10 m FA) in the presence of

free FA (50 × 10 m) to verify whether FA could competi-

−

6

tively inhibit nanoparticle adsorption to KB cells. Addition of

−

6

5

0 × 10 m FA led to a decrease in the fraction of cells with

bound UCN from 44% to 31%. This competitive inhibition is

supportive of FAR-speciﬁc cell binding, though such a small

effect is perhaps due to the weaker µM afﬁnity of FA relative

to the higher afﬁnity of the multivalent dendrimer attached

to the UCN.

FAR-speciﬁc cell binding by 10 is consistent with ear-

lier results observed with FAR-targeted dendrimers such

[

9,18c,24c]

as G5(FA)₆_.₄(ONB-Dox)₅_.₉and G5(MTX)₁₀.

The

binding avidity of 10 to a FAR model surface was not meas-

urable by SPR because of its suspension-like nature (due to

low solubility in water), however we believe that such cell

speciﬁcity is primarily attributable to the tight multivalent

interactions between multiple G5(FA)₆dendrimers in the

shell layer and the cell surface receptors as suggested by the

SPR study above (K_D= 5.9 × 10⁻m; Table 1).

9

2.10. Light-Controlled Cytotoxicity of UCNs Conjugated

with ONB-Dox

Lastly, we evaluated the efﬁcacy of two different types of

light excitation for the controlled induction of cytotoxicity

by 13 UCN @(ONB-Dox)(G5FA) or 14 UCN @(ONB-Dox)

1

2

(

G5FA) in FAR(+) KB cancer cells (Figure 6) . 13 consists of

3

.5% (w/w) 4 ONB-Dox:UCN conjugate and 14 consists of

6

.5% (w/w) 4 ONB-Dox:UCN conjugate. We employed this

cell-based assay primarily to compare the overall cytotoxicity

of each UCN conjugate triggered by UV and NIR irradia-

tion. This assay allows direct determination of the functional

activity of decaged (activated) drug molecules by addressing

both the release efﬁciency and penetration capability of

each light source through cellular membranes. UV light was

ﬁrst validated as a reference light source for Dox release. Figure 6. Light-controlled cytotoxicity of 13 UCN @(ONB-Dox)(G5FA)

1

KB cells were treated with 13 or 14 and the cell viability and 14 UCN2@(ONB-Dox)(G5FA) in FAR(+) KB cancer cells. (a, b) Viability

(

1

%) of cells treated with 13 or 14 and exposed to a UV lamp (365 nm;

was determined under two different exposure conditions

Figure S9, Supporting Information):“pre-exposed 0.5 h” and

5 W; output = ≈3 × 10¹⁹photons s ) under two different conditions

−1

(

as deﬁned in the text. Each data point refers to a mean value (±SD;

n ≥ 3). c) Viability (%) of cells treated with 14 ([UCN] = 750 µg mL ),

each with or without exposure to a continuous wave NIR laser (980 nm;

“

0

1 h exposure on cells.” In the ﬁrst condition (“pre-exposed

.5 h”), the test solution alone in PBS/1% BSA (PBSB) was

−1

exposed to UV light (365 nm) for 0.5 h prior to its addition 1 W; output = ∼5 × 10¹⁸photons s ) or a UV lamp for 1 h exposure on

to and incubation with the cells for 1.5 hat 37 °C. In the other cells. Each bar represents a mean value (±SD; n ≥ 5).

condition (“1 h exposure on cells”), the cells were incubated

−1

with each test solution in PBSB and the entire mixture of replacing the conjugate containing media with fresh media.

cells with UCN was exposed to UV light for 1 h followed by Viability was measured after 4 d of incubation at 37 °C in

incubation at 37 °C for an additional 0.5 h. Fresh media was this fresh media.

added to all of the treated cells to dilute the conjugate by a

As shown in Figure 6a,b, The KB cells were treated with

factor of four, and the cells were incubated for 24 h prior to 13 or 14 suspended at two concentrations: 500 µg mL :

−1

small 2015, 11, No. 45, 6078–6090

www.small-journal.com 6087

full papers

www.MaterialsViews.com

equivalent [ONB-Dox] = 20 × 10 m (13), 30 × 10⁻⁶m (14); components into a single nanostructure could be achieved

−6

−

1

−6

7

50 µg mL : [Dox-ONB] = 37 × 10 m (13), 56 × 10 m (14). through a variety of conceivable combinations. Many design

Treatment at the lower concentration was ineffective regard- aspects fundamental to this type of UCN integration (ex.

less of light exposure. In contrast, treatment at the higher drug or targeting ligand attachment sites) have not been sys-

[

2,5b,6a–e]

concentration led to a marked decrease in viability and was tematically evaluated in previous studies.

highly dependent on the conditions of the light exposure. The rationale for designing such composite UCN nano-

In particular, the “exposure on cells” condition in which the structures lies in the unique optical properties conferred by

cells were incubated with the UCN prior to being exposed the UCN cores, in which the UCN core selectively absorbs

[

5g,6a,b,d]

to UV light was the most effective in killing the cells. Simi- focused NIR (980 nm) light and emits UV–vis light,

larly, the KB cells were treated with 4 ONB-Dox and 8 thus serving as a light emitting nanodevice. Two UCN types

G5(FA) (Dox) as positive controls under various condi- were designed: UCN ( λ = 542, 651 nm) suitable for vis

6

6.9

1

em

tions of light exposure, as summarized in Figure S10, Sup- imaging and UCN₂( λ_e_m= 340–360, 450–475 nm) suitable

porting Information. This study validated the light control of for light-controlled drug release. UCN @(G5FA) enabled

1

Dox release from the UCN conjugates, and also allowed us to the use of NIR imaging for sensitive, label-free detection of

identify a threshold nanoparticle dose for inducing cytotox- FAR(+) carcinoma cells. Furthermore, UCN integration into

2

icity in KB cells in a light dependent manner.

the UCN @(ONB-Dox)(G5FA) nanocomposite allowed for

2

In contrast to conjugate 13 which contains a UCN core effective targeted drug delivery triggered by either UV or

1

that only emits in the visible range, conjugate 14 consists of NIR irradiation. Our approach for the controlled release of

a UCN core that emits UV light (λ = 345, 362 nm) upon Dox is based on its temporary inactivation with a UV light-

2

em

excitation with NIR (980 nm) (Figure 2). We thus investi- cleavable cage, an ortho-nitrobenzyl (ONB) group. UVA

gated whether excitation of 14 UCN @(ONB-Dox)(G5FA) light penetrates through the skin, reaching cells in the layers

2

[

41]

with NIR could trigger drug release and subsequently of the epidermis and dermis (depth ≤0.15 mm), providing

result in cytotoxicity. FAR(+) KB cells were treated with a potentially spatial mechanism of drug release to tumor cells

50 µg mL the UCN conjugate and the cells were exposed localized to the skin and vascular endothelial cells. However,

−

1

7

to NIR radiation (980 nm; power output = 1 W) for 1 h (“1 h the use of UV-triggered photorelease for targeting more

exposure on cells”). Figure 6c compares the effect of expo- interiorly located tumors is limited by the penetration depth

sure to NIR and UV light to dark controls. Cells showed no of UV excitation. Integration of the photocaged drug with

signiﬁcant decrease in viability in the presence or absence UCNs, however, allows for the use of a NIR excitation light

of 14 without any exposure to light excitation. Exposure to source in place of UV to trigger drug release using the same

UV or NIR light for the treatment period also did not have UV-cleavable photocage due to the UV emission of the UCN

any signiﬁcant impact on cell viability. However, exposure of core upon excitation with NIR. This ability to trigger drug

cells treated with 14 to NIR light led to a signiﬁcant decrease delivery with NIR enables the use of photocontrolled drug

in cell viability comparable to that induced by the direct delivery to target more interiorly localized cells, as NIR light

cleavage of the photocaged Dox by UV triggered release.

penetrates deep into tissues (depth ≥2–3 mm) without a sig-

[

41]

Cytotoxicity as triggered by NIR, however, showed a niﬁcant loss of transmission from absorption by water. We

wider variation (±SD) in activity. We believe this experi- believe that these novel NIR-excited UCN nanocomposite

mental error is attributable to a number of technical chal- structures have great potential for extending the scope of

lenges facing the NIR exposure. First, unlike the UV release imaging and drug release from UV-accessible skin cells to

which was performed under a UV lamp, NIR release was other tissues and inner organs of interest. Furthermore, the

performed by irradiation with a continuous wave (CW) NIR lower energy of NIR makes it more desirable as a treatment

laser. Thus the wide variation between replicates could have light source relative to UVA, which while less damaging than

been due to the lack of homogeneous excitation across each UVB, still has the potential to damage bystander cells upon

sample well due to the relatively small size of the laser beam. prolonged exposure.

Second, NIR release requires an extended period of irradi-

Two modes for structural integration of photocaged doxo-

ation (1 h) primarily due to the low quantum yield of UV rubicin ONB-Dox were examined: conjugation to G5(FA)₆

emission from UCN as illustrated by the low intensity of its or direct conjugation to the UCN. Our ﬁndings suggested

2

UV emission peaks relative to the intensity of UCN visible potentially greater beneﬁts through the direct attachment of

1

emission (Figure 2). Despite such technical challenges that photocaged drug to the UCN surface. Coconjugation of ONB-

need to be addressed, we believe that this NIR release study Dox and FA to the PAMAM dendrimer G5(FA) (ONB-Dox)

6

supports the use of the UCN core as an effective way to con- resulted in lower binding avidity to the model surface for a

trol drug release by NIR radiation.

FAR(+) cell compared to G5(FA) perhaps due to steric inter-

6

ference from the bulky Dox molecule blocking FAR access

to the FA ligand molecule. In the UV-mediated drug release

study, Dox release occurred faster from ONB-Dox than from

3

. Conclusion

the dendrimer conjugated form G5(FA) (ONB-Dox). Thus,

6

In summary, in this study, we employed a bottom-up approach cell receptor targeting avidity and speciﬁcity as well as drug

toward the modular integration of UCN with a FAR-targeted release kinetics from UCN nanocomposites appear to be

multivalent dendrimer G5(FA) and photocaged doxorubicin highly dependent on the site of conjugation of the targeting

6

ONB-Dox. Integration of these three individually functional ligand and photocaged drug payload. Considering all of these

6088 www.small-journal.com

small 2015, 11, No. 45, 6078–6090

www.MaterialsViews.com

data together, this study provides implications of signiﬁcant

importance to the design and screening of multifunctional

UCN nanostructures for surface receptor-targeted imaging

and treatment of tumor cells.

P. Smith, O. Kaso, K. Sinniah, S. K. Choi, J. Phys. Chem. B 2014,

18, 2872; g) B. Yan, J.-C. Boyer, D. Habault, N. R. Branda,

Y. Z h a o , J. Am. Chem. Soc. 2012, 134, 16558.

1

[

6] a) Y.-H. Chien, Y.-L. Chou, S.-W. Wang, S.-T. Hung, M.-C. Liau,

Y.-J. Chao, C.-H. Su, C.-S. Yeh, ACS Nano 2013, 7, 8516; b) S. Cui,

D. Yin, Y. Chen, Y. Di, H. Chen, Y. Ma, S. Achilefu, Y. Gu, ACS Nano

2

012, 7, 676; c) N. M. Idris, M. K. Gnanasammandhan, J. Zhang,

P. C. Ho, R. Mahendran, Y. Zhang, Nat. Med. (N. Y., NY, U. S.) 2012,

8, 1580; d) M. K. G. Jayakumar, N. M. Idris, Y. Zhang, Proc. Natl.

4

. Experimental Section

1

Methods for the synthesis of UCNs, G5 PAMAM dendrimer and

photocaged Dox, and analytical methods (TEM, HPLC, GPC, ¹

Acad. Sci. USA 2012, 109, 8483; e) Y. Sun, X. Zhu, J. Peng, F. Li,

ACS Nano 2013, 7, 11290; f) F. Wang, X. Liu, Acc. Chem. Res.

H

2

014, 47, 1378; g) Y.-F. Wang, G.-Y. Liu, L.-D. Sun, J.-W. Xiao,

NMR, UV–vis spectrometry, MALDI-TOF mass spectrometry) are

described in detail in the Supporting Information and references

cited therein. Full details for other materials and methods (drug

release kinetics, SPR spectroscopy, cell culture, ﬂow cytom-

etry, confocal microscopy) are also provided in the Supporting

Information.

J.-C. Zhou, C.-H. Yan, ACS Nano 2013, 7, 7200; h) S. Han, R. Deng,

X. Xie, X. Liu, Angew. Chem. Int. Ed. 2014, 53, 11702; i) Y. Dai,

H. Xiao, J. Liu, Q. Yuan, P. a. Ma, D. Yang, C. Li, Z. Cheng, Z. Hou,

P. Yang, J. Lin, J. Am. Chem. Soc. 2013, 135, 18920; j) S. Gai,

P. Yang, C. Li, W. Wang, Y. Dai, N. Niu, J. Lin, Adv. Funct. Mater.

2010, 20, 1166.

[7] a) Y. Sui, K. Tao, Q. Tian, K. Sun, J. Phys. Chem. C 2011, 116,

1

2

732; b) Q. Tian, K. Tao, W. Li, K. Sun, J. Phys. Chem. C 2011, 115,

2886.

[

8] D. A. Tomalia, A. M. Naylor, I. William, A. Goddard, Angew. Chem.

Int. Ed. 1990, 29, 138.

9] S. K. Choi, T. Thomas, M. Li, A. Kotlyar, A. Desai, J. R. Baker Jr.,

Chem. Commun. 2010, 46, 2632.

Supporting Information

Supporting Information is available from the Wiley Online Library [10] a) J.-C. Boyer, F. Vetrone, L. A. Cuccia, J. A. Capobianco, J. Am.

Chem. Soc. 2006, 128, 7444; b) G. Chen, H. Qiu, P. N. Prasad,

X. Chen, Chem. Rev. 2014, 114, 5161; c) M. Haase, H. Schäfer,

Angew. Chem. Int. Ed. 2011, 50, 5808; d) F. Wang, X. Liu, Chem.

Soc. Rev. 2009, 38, 976; e) J. Shen, G. Chen, A.-M. Vu, W. Fan,

O. S. Bilsel, C.-C. Chang, G. Han, Adv. Opt. Mater. 2013, 1, 644.

11] Y. Yang, Q. Shao, R. Deng, C. Wang, X. Teng, K. Cheng, Z. Cheng,

L. Huang, Z. Liu, X. Liu, B. Xing, Angew. Chem. Int. Ed. 2012, 51,

3125.

12] a) J. Shen, G. Chen, T. Y. Ohulchanskyy, S. J. Kesseli, S. Buchholz,

Z. Li, P. N. Prasad, G. Han, Small 2013, 9, 3213; b) Y. Yang,

B. Velmurugan, X. Liu, B. Xing, Small 2013, 9, 2937.

13] M. M. Dcona, Q. Yu, J. A. Capobianco, M. C. T. Hartman, Chem.

Commun. 2015, 51, 8477.

or from the author.

Acknowledgements

[

This study was supported by the Michigan Nanotechnology Insti-

tute for Medicine and Biological Sciences (S.K.C.), and by the Uni-

versity of Michigan Ofﬁce of the Vice President for Research, and

Shanghai Jiao Tong University (K.S., S.K.C.), and its publication is

supported by British Council and Department for Business Inno-

vation & Skills through a Global Innovation Initiative (GII) project

(K.S., K.T., S.K.C.).

[14] Y. Min, J. Li, F. Liu, E. K. L. Yeow, B. Xing, Angew. Chem. Int. Ed.

014, 53, 1012.

2

[

15] Y. Yang, F. Liu, X. Liu, B. Xing, Nanoscale 2013, 5, 231.

16] D. A. Tomalia, H. Baker, J. Dewald, M. Hall, G. Kallos, S. Martin,

J. Roeck, J. Ryder, P. Smith, Polym. J. 1985, 17, 117.

[

1] a) V. Biju, Chem. Soc. Rev. 2014, 43, 744; b) L. Cheng, C. Wang,

L. Feng, K. Yang, Z. Liu, Chem. Rev. 2014, 114, 10869; c) M. Dahl, [17] a) C. Fasting, C. A. Schalley, M. Weber, O. Seitz, S. Hecht,

Y. Liu, Y. Yin, Chem. Rev. 2014, 114, 9853; d) N. L. Rosi,

C. A. Mirkin, Chem. Rev. 2005, 105, 1547; e) X. Wu, H. Liu,

J. Liu, K. N. Haley, J. A. Treadway, J. P. Larson, N. Ge, F. Peale,

M. P. Bruchez, Nat. Biotechnol. 2003, 21, 41.

B. Koksch, J. Dernedde, C. Graf, E.-W. Knapp, R. Haag,

Angew. Chem. Int. Ed. 2012, 51, 10472; b) L. L. Kiessling,

J. E. Gestwicki, L. E. Strong, Angew. Chem. Int. Ed. 2006, 45, 2348;

c) M. Mammen, S. K. Choi, G. M. Whitesides, Angew. Chem. Int.

Ed. 1998, 37, 2754.

[

2] M. Wang, C.-C. Mi, W.-X. Wang, C.-H. Liu, Y.-F. Wu, Z.-R. Xu,

C.-B. Mao, S.-K. Xu, ACS Nano 2009, 3, 1580.

3] L. A. Ponomarenko, F. Schedin, M. I. Katsnelson, R. Yang,

E. W. Hill, K. S. Novoselov, A. K. Geim, Science 2008, 320, 356.

4] a) S. Heer, K. Kömpe, H. U. Güdel, M. Haase, Adv. Mater. 2004,

[18] a) J. F. Kukowska-Latallo, K. A. Candido, Z. Cao, S. S. Nigavekar,

I. J. Majoros, T. P. Thomas, L. P. Balogh, M. K. Khan, J. R. Baker Jr.,

Cancer Res. 2005, 65, 5317; b) I. J. Majoros, C. R. Williams,

A. Becker, J. R. Baker Jr., Wiley Interdiscip. Rev.: Nanomed. Nano-

biotechnol. 2009, 1, 502; c) J. E. Silpe, M. Sumit, T. P. Thomas,

B. Huang, A. Kotlyar, M. A. van Dongen, M. M. Banaszak Holl,

B. G. Orr, S. K. Choi, ACS Chem. Biol. 2013, 8, 2063; d) J. Zhu,

X. Shi, J. Mater. Chem. B 2013, 1, 4199.

1

2

1

6, 2102; b) X. Liu, C.-H. Yan, J. A. Capobianco, Chem. Soc. Rev.

015, 44, 1299; c) S. Gai, C. Li, P. Yang, J. Lin, Chem. Rev. 2014,

14, 2343; d) D. Yang, P. a. Ma, Z. Hou, Z. Cheng, C. Li, J. Lin,

Chem. Soc. Rev. 2015, 44, 1416.

[

5] a) A. Agarwal, M. A. Mackey, M. A. El-Sayed, R. V. Bellamkonda, [19] a) T. P. Thomas, S. K. Choi, M.-H. Li, A. Kotlyar, J. R. Baker Jr.,

ACS Nano 2011, 5, 4919; b) J. Jin, Y.-J. Gu, C. W.-Y. Man, J. Cheng,

Z. Xu, Y. Zhang, H. Wang, V. H.-Y. Lee, S. H. Cheng, W.-T. Wong,

ACS Nano 2011, 5, 7838; c) X. Li, H. Zhou, L. Yang, G. Du,

A. S. Pai-Panandiker, X. Huang, B. Yan, Biomaterials 2011, 32,

Bioorg. Med. Chem. Lett. 2010, 20, 5191; b) A. B. Witte,

C .M . Timme r, J .J . Gam S, .K . Ch o, iM . M. Banaszak Holl, B. G. Orr,

J. R. Baker, K. Sinniah, Biomacromolecules 2012, 13, 507;

c) P. T. Wong, K. Tang, A. Coulter, S. Tang, J. R. Baker, S. K. Choi,

Biomacromolecules 2014, 15, 4134.

2

540; d) P. a. Ma, H. Xiao, X. Li, C. Li, Y. Dai, Z. Cheng, X. Jing, J. Lin,

Adv. Mater. 2013, 25, 4898; e) J. L. Vivero-Escoto, I. I. Slowing, [20] a) C. A. Boswell, P. K. Eck, C. A. S. Regino, M. Bernardo, K. J. Wong,

C.-W. Wu, V. S. Y. Lin, J. Am. Chem. Soc. 2009, 131, 3462;

f) A. B. Witte, A. N. Leistra, P. T. Wong, S. Bharathi, K. Reﬁor,

D. E. Milenic, P. L. Choyke, M. W. Brechbiel, Mol. Pharmaceutics

2008, 5, 527; b) L. Kong, C. S. Alves, W. Hou, J. Qiu, H. Möhwald,

small 2015, 11, No. 45, 6078–6090

www.small-journal.com 6089

full papers

www.MaterialsViews.com

H. Tomás, X. Shi, ACS Appl. Mater. Interfaces 2015, 7, 4833; [31] J. W. Walker, J. A. McCray, G. P. Hess, Biochemistry 1986, 25, 1799.

c) R. Shukla, T. P. Thomas, J. Peters, A. Kotlyar, A. Myc, J. R. Baker Jr., [32] a) S. Impellizzeri, B. McCaughan, J. F. Callan, F. M. Raymo, J.

Chem. Commun. 2005, 5739.

Am. Chem. Soc. 2012, 134, 2276; b) A. V. Karginov, Y. Zou,

D. Shirvanyants, P. Kota, N. V. Dokholyan, D. D. Young, K. M. Hahn,

A. Deiters, J. Am. Chem. Soc. 2010, 133, 420; c) E. A. Lemke,

D. Summerer, B. H. Geierstanger, S. M. Brittain, P. G. Schultz, Nat.

Chem. Biol. 2007, 3, 769; d) G. Mayer, A. Heckel, Angew. Chem.

Int. Ed. 2006, 45, 4900; e) P. T. Wong, S. K. Choi, Chem. Rev.

2015, 115, 3388; f) C. Bao, L. Zhu, Q. Lin, H. Tian, Adv. Mater.

2015, 27, 1647.

[

21] a) B. Huang, J. Otis, M. Joice, A. Kotlyar, T. P. Thomas,

Biomacromolecules 2014, 15, 915; b) S. H. Medina,

M. E. H. El-Sayed, Chem. Rev. 2009, 109, 3141.

22] a) W. T. Al-Jamal, K. Kostarelos, Acc. Chem. Res. 2011, 44, 1094;

b) S. M. Janib, A. S. Moses, J. A. MacKay, Adv. Drug Delivery Rev.

2

010, 62, 1052; c) M. R. Molla, P. Rangadurai, G. M. Pavan,

S. Thayumanavan, Nanoscale 2015, 7, 3817; d) D. Peer,

J. M. Karp, S. Hong, O. C. Farokhzad, R. Margalit, R. Langer, Nat. [33] L. Bildstein, C. Dubernet, P. Couvreur, Adv. Drug Delivery Rev.

Nanotechnol. 2007, 2, 751. 2011, 63, 3.

23] T. Xiao, W. Hou, X. Cao, S. Wen, M. Shen, X. Shi, Biomater. Sci. [34] a) S. K. Choi, T. P. Thomas, M.-H. Li, A. Desai, A. Kotlyar,

[

2

013, 1, 1172.

J. R. Baker, Photochem. Photobiol. Sci. 2012, 11, 653;

b) S. K. Choi, M. Verma, J. Silpe, R. E. Moody, K. Tang, J. J. Hanson,

J. R. Baker Jr., Bioorg. Med. Chem. 2012, 20, 1281.

24] a) E. N. Cline, M.-H. Li, S. K. Choi, J. F. Herbstman, N. Kaul,

E. Meyhöfer, G. Skiniotis, J. R. Baker, R. G. Larson, N. G. Walter,

Biomacromolecules 2013, 14, 654; b) I. J. Majoros, A. Myc, [35] M. A. Inlay, V. Choe, S. Bharathi, N. B. Fernhoff, J. R. Baker Jr.,

T. Thomas, C. B. Mehta, J. R. Baker, Biomacromolecules 2006, 7, I. L. Weissman, S. K. Choi, Chem. Commun. 2013, 49, 4971.

72; c) T. P. Thomas, B. Huang, S. K. Choi, J. E. Silpe, A. Kotlyar, [36] D. G. Mullen, M. Fang, A. Desai, J. R. Baker Jr., B. G. Orr,

5

A. M. Desai, J. Gam, M. Joice, J. R. Baker Jr., Mol. Pharmaceutics

M. M. Banaszak Holl, ACS Nano 2010, 4, 657.

2

012, 9, 2669.

[37] a) S. K. Choi, A. Myc, J. E. Silpe, M. Sumit, P. T. Wong, K. McCarthy,

A. M. Desai, T. P. Thomas, A. Kotlyar, M. M. Banaszak Holl,

B. G. Orr, J. R. Baker, ACS Nano 2013, 7, 214; b) J. Rao, L. Yan,

B. Xu, G. M. Whitesides, J. Am. Chem. Soc. 1999, 121, 2629.

[

25] M.-H. Li, S. K. Choi, T. P. Thomas, A. Desai, K.-H. Lee, A. Kotlyar,

M. M. Banaszak Holl, J. R. Baker Jr., Eur. J. Med. Chem. 2012, 47,

5

60.

26] a) H. Schäfer, C. Hess, H. Tobergte, A. Volf, S. Ichilmann, [38] a) N. Parker, M. J. Turk, E. Westrick, J. D. Lewis, P. S. Low,

H. Eickmeier, B. Voss, N. Kashaev, J. Nordmann, W. Akram,

B. Hartmann-Azanza, M. Steinhart, Small 2015, 11, 931;

b) H. Schäfer, P. Ptacek, H. Eickmeier, M. Haase, Adv. Funct. Mater.

C. P. Leamon, Anal. Biochem. 2005, 338, 284; b) S. D. Weitman,

R. H. Lark, L. R. Coney, D. W. Fort, V. Frasca, V. R. Zurawski,

B. A. Kamen, Cancer Res. 1992, 52, 3396.

2

009, 19, 3091.

27] F. J. Arriagada, K. Osseo-Asare, J. Colloid Interface Sci. 1999, 211,

10.

[39] N. J. de Mol, M. J. E. Fischer, Handbook of Surface Plasmon Reso-

nance, Vol. 5, The Royal Society of Chemistry, Cambridge, UK,

2008, p. 123.

[

2

28] F. Chen, S. Zhang, W. Bu, Y. Chen, Q. Xiao, J. Liu, H. Xing, L. Zhou, [40] W. S. Hlavacek, R. G. Posner, A. S. Perelson, Biophys. J. 1999, 76, 3031.

W. Peng, J. Shi, Chem. - Eur. J. 2012, 18, 7082.

29] S. Schietinger, L. d. S. Menezes, B. Lauritzen, O. Benson, Nano

Lett. 2009, 9, 2477.

[41] a) R. R. Anderson, J. A. Parrish, J. Invest. Dermatol. 1981, 77,

13; b) L. Fodor, M. Elman, Y. Ullmann, Aesthetic Applications of

Intense Pulsed Light, Springer-Verlag, London, 2011, p. 11.

30] a) N. M. Idris, M. K. G. Jayakumar, A. Bansal, Y. Zhang, Chem.

Soc. Rev. 2015, 44, 1449; b) X. Liu, M. Zheng, X. Kong, Y. Zhang,

Q . Z e n g , Z . S u n , W . J . B u m a , H . Z h a n g , Chem. Commun. 2013, 49,

Received: June 3, 2015

Revised: July 13, 2015

3

224.

Published online: October 13, 2015

6090 www.small-journal.com

small 2015, 11, No. 45, 6078–6090

Article Doi

Modular Integration of Upconverting Nanocrystal-Dendrimer Composites for Folate Receptor-Specific NIR Imaging and Light-Triggered Drug Release

DOI: 10.1002/smll.201501575

Source and publish data:

Authors:

Article abstract of DOI:10.1002/smll.201501575

Full text of DOI:10.1002/smll.201501575

Products guided by the article

R&D Labs maybe for 1227683-25-1

Relevant to this article

Hot Product