Albumin-binding prodrugs via reversible iminoboronate forming nanoparticles for cancer drug delivery

Article abstract of DOI:10.1016/j.jconrel.2020.12.035

Albumin-based nanomedicines are important nanoplatforms for cancer drug delivery. The drugs are either physically encapsulated or covalently conjugated to albumin or albumin-based nanosystems. Physical encapsulation is advantageous due to requiring no chemical modification of drug molecules, but many drugs, for instance, camptothecin (CPT) and curcumin (CCM), though very hydrophobic, can't be loaded in or form nanoformulations with albumin. Herein, we demonstrate prodrugs readily binding to proteins via iminoboronates and forming nanoparticles for cancer drug delivery. CPT and CCM were functionalized with 2-acetylphenylboronic acid (2-APBA) to produce prodrugs CPT-SS-APBA and CCM- APBA. The prodrugs bound to bovine serum albumin (BSA) via formation of iminoboronates and the produced BSA/prodrug readily self-assembled into well-defined nanoparticles with high loading efficiency, improved colloidal stability, and much-improved pharmacokinetics. The nanoparticles effectively released drugs in the intracellular acidic environment or the cytosol rich in glutathione (GSH). In vivo, the nanoparticles showed enhanced anticancer efficacy compared with clinically used irinotecan or sorafenib in subcutaneous 4 T1 or HepG2 tumor models. This work demonstrates a versatile protein-binding prodrug platform applicable to protein-based drug formulations and even antibody-drug conjugates.

Full text of DOI:10.1016/j.jconrel.2020.12.035

Journal of Controlled Release 330 (2021) 362–371
Contents lists available at ScienceDirect
Journal of Controlled Release
journal homepage: www.elsevier.com/locate/jconrel
Albumin-binding prodrugs via reversible iminoboronate forming
nanoparticles for cancer drug delivery
,
,
,
,
Lingqiao Haoa 1, Quan Zhoua 1, Ying Piaoa b, Zhuxian Zhoua b, Jianbin Tanga,
,b,*
Youqing Shena
a Center for Bionanoengineering and Key Laboratory of Biomass Chemical Engineering of Ministry of Education, College of Chemical and Biological Engineering, Zhejiang
University, Zheda Road 38, Hangzhou 310007, China
b ZJU-Hangzhou Global Scientific and Technological Innovation Center, Hangzhou, Zhejiang 311215, China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Albumin-based nanomedicines are important nanoplatforms for cancer drug delivery. The drugs are either
physically encapsulated or covalently conjugated to albumin or albumin-based nanosystems. Physical encap-
sulation is advantageous due to requiring no chemical modification of drug molecules, but many drugs, for
instance, camptothecin (CPT) and curcumin (CCM), though very hydrophobic, can’t be loaded in or form
nanoformulations with albumin. Herein, we demonstrate prodrugs readily binding to proteins via iminoboro-
nates and forming nanoparticles for cancer drug delivery. CPT and CCM were functionalized with 2-acetylphe-
nylboronic acid (2-APBA) to produce prodrugs CPT-SS-APBA and CCM- APBA. The prodrugs bound to bovine
serum albumin (BSA) via formation of iminoboronates and the produced BSA/prodrug readily self-assembled into
well-defined nanoparticles with high loading efficiency, improved colloidal stability, and much-improved
pharmacokinetics. The nanoparticles effectively released drugs in the intracellular acidic environment or the
cytosol rich in glutathione (GSH). In vivo, the nanoparticles showed enhanced anticancer efficacy compared with
clinically used irinotecan or sorafenib in subcutaneous 4 T1 or HepG2 tumor models. This work demonstrates a
versatile protein-binding prodrug platform applicable to protein-based drug formulations and even antibody-
drug conjugates.
Albumin-based nanomedicine
Prodrug
Iminoboronate chemistry
Acid and GSH-responsive release
Camptothecin
Curcumin
1. Introduction
the FDA in 2005 and used widely in clinics for the first-line treatment of
locally advanced or metastatic cancer [9]. It shows obvious clinical
Albumin has been extensively explored as drug delivery carriers
owing to its advantages, including high biocompatibility, non-
antigenicity, good biodegradability, and easy surface modification
[1–3]. To date, several albumin-based nanosized drug delivery systems
have successfully clinical translated, which significantly improved the
pharmacokinetics and tumor accumulation of therapeutics, and thus
improved drugs’ therapeutic efficiency and minimized side effects [2,4].
Albumin is inherently capable of hosting hydrophobic molecules in
its hydrophobic interiors, and thus physical encapsulation of drugs is
preferable due to avoiding chemical modifications [5,6]. Some albumin-
based nanomedicines have already been approved for clinical use or in
clinical trials [3,7,8]. A most successful example is Abraxane, a human
albumin nanoparticle encapsulated with paclitaxel (PTX) approved by
advantages over the first-generation Cremophor EL-based PTX (Taxol)
[10]. Many hydrophobic anticancer drugs, however, cannot be loaded
into albumin at acceptable drug loading contents. For example, physical
encapsulation of very hydrophobic camptothecin (CPT) or curcumin
(CCM) into the albumin or other proteins is still unsuccessful due to their
intrinsically planar structure, moderate polarity, and extremely fast
crystallization tendency [11,12]. Covalent conjugation is an alternative
to load these drugs to albumin [13], including drug-albumin conjugate
(MTX-HSA) [14] and albumin-binding prodrugs such as (6-mal-
eimidocaproyl)hydrazone derivative of doxorubicin [15], monomethyl
auristatin E [16]. The critical design is the intracellular cleavable linkers
to release the carried drugs [17,18].
Iminoboronates have been extensively exploited to the biorthogonal
* Corresponding author at: Center for Bionanoengineering and Key Laboratory of Biomass Chemical Engineering of Ministry of Education, College of Chemical and
Biological Engineering, Zhejiang University, Zheda Road 38, Hangzhou 310007, China.
E-mail address: shenyq@zju.edu.cn (Y. Shen).
1
ǂEqual Contributors
https://doi.org/10.1016/j.jconrel.2020.12.035
Received 26 October 2020; Received in revised form 16 December 2020; Accepted 20 December 2020
Available online 24 December 2020
0168-3659/© 2020 Elsevier B.V. All rights reserved.

L. Hao et al.
Journal of Controlled Release 330 (2021) 362–371
Scheme 1. Iminoboronate-forming prodrugs for BSA-binding. 2-Acetylphenylboronic acid (2-APBA) conjugated prodrugs (CPT-SS-APBA and CCM-APBA) rapidly
bind to albumin via the iminoboronate. The hydrophobized BSA self-assembled into ~100 nm nanoparticles with high drug loading efficacy and improved colloidal
stability. At the lysosomal pH or in the presence of GSH, the iminoboronate was quickly cleaved to release the drug. The control prodrugs with acetylphenyl (AP)
group (CPT-SS-AP and CCM-AP) can’t bind with BSA and the hydrophobic interactions among the free drug molecules dominate the self-assembly process to produce
precipitates.
chemistry in recent years [19,20] as the reaction is regarded as a
controllable and reversible “click reaction” and allows labeling biolog-
ical amines at low millimolar concentrations with good thermodynamic
stability [21,22]. Importantly, iminoboronates are very labile by acid or
GSH due to cleavage of the B-N bond [23]. Thus, iminoboronates were
used for efficient cyclization of peptides [24], reversible lysine modifi-
cation [25], and cancer cell targeting [26].
(Xi’an, China). Albumin, IgG, and RNase A were obtained from Sangon
Biotech Co., Ltd. (Shanghai, China). The anti-PD1 antibody (aPD1) was
purchased from Innovent Biologics Co., Ltd. (Suzhou, China). Curcumin
was purchased from Energy Chemical Reagent Co., Ltd. (Shanghai,
China). All the chemicals were used as received. Buffer solution (pH =
7.4) used here was prepared from Tris⋅HCl (0.1 M) and NaCl (15 mM).
Motivated by the formation of stable iminoboronates at pH of 6–9
2.2. Characterizations
between lysine ε-amino group and 2-acetylphenylboronic acid (2-APBA)
1H NMR, 13C NMR, and 11B NMR spectra were obtained on a 400
MHz Varian Gemini NMR spectrometer in deuterated solvents as noted.
[22], we herein demonstrate iminoboronate-forming prodrugs for in-situ
albumin binding (Scheme 1). 2-APBA was conjugated to CPT or CCM
(denoted as CPT-SS-APBA and CCM-APBA). We proposed that both CPT-
SS-APBA and CCM-APBA readily bound to BSA via the iminoboronate
chemistry and induced in situ self-assembly to form well-defined nano-
particles with high loading efficiency and improved colloidal stability.
The loaded drugs are effectively unloaded in response to the lysosomal
pH or intracellular GSH. As a result, the BSA/CPT-SS-APBA nano-
particles showed enhanced anticancer efficacy than the clinically used
irinotecan (CPT-11) and sorafenib in subcutaneous 4 T1 and HepG2
tumor models. Apart from CPT and CCM, the current strategy may be
extended to other hydrophobic drugs, rendering it a versatile and robust
protein-binding prodrug platform applicable for protein-based drug
formulations and even antibody-drug conjugates.
◦
The sizes and zeta potentials of samples were measured at 25 C on a
Zetasizer Nano-ZS (Malvern Instruments, UK) with 632.8 nm laser light
set at a scattering angle of 173◦. HPLC analyses were performed using a
Waters HPLC system consisting of a Waters 1525 binary pump, a Waters
2475 fluorescence detector, and a symmetry C18 column (5 μm, 4.6 ×
250 mm Column) in a 1500 column heater. For detecting CPT, methyl
alcohol-water (80:20) was used as the mobile phase at 35 ◦C at a flow
rate of 1 mL/min. The fluorescence detector was set at 360 nm for
excitation and 540 nm for emission. For detecting CCM and CCM-APBA,
the mobile phase consists of acetonitrile and water (containing 0.1%
TFA) in a gradient mode at a flow rate of 1 mL/min at 35 ◦C; the content
of acetonitrile linearly increased from 50% to 80% within 30 min. The
UV detection wavelength was set at 420 nm. Data were processed using
Breeze software. Linear calibration curves for concentrations in the
2. Materials and methods
range of 0.001–10 μg/mL were constructed using the peak areas by
2.1. Materials
linear regression analysis.
All the materials other than indicated were purchased from Sino-
pharm Chemical Reagent Co., Ltd. (Shanghai, China). Dichloromethane
(CH2Cl2, DCM), N,N-dimethylformamide (DMF), pyridine (Py), aceto-
nitrile (MeCN), and triethylamine (TEA) were distilled over calcium
hydride (CaH2) or treated with a 4 Å molecular sieve to remove the
residue water. Camptothecin (CPT) and irinotecan hydrochloride tri-
hydrate (CPT-11) were purchased from Xindifu Pharmaceutical Co., Ltd.
2.3. Preparation of BSA/CPT-SS-APBA nanoparticles
The preparation of BSA/CPT-SS-APBA nanoparticles (mBSA/mCPT-SS-
APBA = 5) is as follows: CPT-SS-APBA (1 mg) was dissolved in 100 μL THF
and added dropwise to 1 mL of BSA solution (5 mg/mL) with vigorous
stirring. After stirring at r.t. for 10 min, THF was removed by rotary
evaporator to give the BSA/CPT-SS-APBA solution. Other nanoparticles
363

L. Hao et al.
Journal of Controlled Release 330 (2021) 362–371
were prepared in the same way, and the concentrations of RNase A,
aPD1, and IgG were 2.5 mg/mL, 1 mg/mL, and 1 mg/mL, respectively.
The Z-averaged diameters and zeta potentials of the prepared nano-
particles were determined by dynamic light scattering (DLS).
2.9. In vivo pharmacokinetics
BSA/CPT-SS-APBA nanoparticles, CPT-SS-APBA dispersed in
PEG400 (30%) or CPT-11 (5 mg CPT or SN38-eq./kg) were injected via
the tail vein (n = 3 for each group). Blood samples (50 μL) were collected
from the retro-orbital plexus of the mouse eye at timed intervals and
2.4. Stability assay of BSA/CPT-SS-APBA nanoparticles
then added with 0.1 N NaOH (50 μL). The mixtures were incubated at
37 ◦C overnight to release CPT or SN38 followed by adding with 1 mL of
acetonitrile. The solution was sonicated and centrifuged at 5000 rpm for
The BSA/CPT-SS-APBA nanoparticles were incubated in 1640 cell
culture medium (with 10% FBS) at 37 ◦C with shaking. The nanoparticle
sizes and zeta-potentials were measured by DLS at different times.
5 min to obtain the supernatant, 200 μL of which was mixed with 200 μL
0.1 N HCl aqueous solution for HPLC analysis. The CPT or SN38 content
was calculated according to the standard curve. Pharmacokinetic pa-
rameters were obtained using DAS by fitting a two-compartment model.
2.5. Transmission Electron Microscope (TEM) observation
2.10. In vivo biodistribution
The prepared BSA/CPT-SS-APBA nanoparticle solution (CPT-SS-
APBA equivalent concentration 0.5 mg/mL) was applied onto a 150-
mesh carbon‑copper grid. After wicking off the excess solution with
filter paper, images were recorded using a transmission electron mi-
croscope (JEM-1200EX, TEM) operated at a voltage of 80 kV. For the
lyophilized nanoparticles, the powder was redissolved in water (CPT-SS-
APBA concentration, 0.5 mg/mL), and their sizes, zeta potentials, and
morphology of the lyophilized nanoparticles were determined by DLS
and TEM.
4 T1 tumor-bearing mice (tumor volume of ~200 mm3) were
intravenously injected with BSA/CPT-SS-APBA at 5 mg/kg CPT-
equivalent (CPT-eq.) dose via the tail vein. Mice were sacrificed at 6 h
after administration, and the tumor, heart, liver, spleen, lung, and kid-
ney were collected. Weighted tissues were homogenized in 400 μL PBS
and then added to 0.1 N NaOH (50 μL). The mixtures were incubated at
37 ◦C overnight followed by adding with 1 mL acetonitrile. The solution
was sonicated and centrifuged at 5000 rpm for 5 min to obtain the su-
pernatant, 200 μL of which was mixed with 200 μL 0.1 N HCl aqueous
2.6. In vitro CPT release
solution for HPLC analysis. The CPT content was calculated according to
the standard curve.
Nanoparticle solution (CPT-SS-APBA equivalent concentration 1
mg/mL, 1 mL) was loaded into a dialysis bag (MWCO, 3.5 kDa) against
45 mL of PBS (pH 7.4 or 5.0) with or without 10 mM of GSH shaken at
37 ◦C at 200 rpm. At timed intervals, 0.1 mL of the PBS buffer solution
was sampled and replaced with 0.1 mL fresh PBS buffer solution. After
mixed with 0.9 mL acetonitrile, the released CPT content was measured
by HPLC. Meanwhile, the particle sizes in the dialysis bag containing 10
mM GSH at pH 7.4 were detected using DLS at 0 h, 12 h, 24 h, and 48 h,
respectively. The release kinetics of CCM from BSA/CCM-APBA nano-
particles at pH 7.4 or 5.0 were performed in the same way.
2.11. In vivo antitumor activity
Subcutaneous 4 T1 tumor model: 4 T1 tumor-bearing mice (6–8
weeks old, female) were established by subcutaneous injection of a
suspension of 5 × 105 4 T1 cells in PBS (100
μL). After 7 days post-
inoculation, the mice were randomly divided into 3 groups (n = 6),
and injected via the tail vein with 200 μL of PBS, CPT-11(5 mg/kg SN38-
eq. dose) or BSA/CPT-SS-APBA (5 mg/kg CPT-eq. dose) for 6 times.
Tumor volume and bodyweight of the mice were monitored every 2
days. When the tumor volume exceeded 1000 mm3, mice were sacri-
ficed, and the tumors were collected and weighed.
2.7. Cell lines and animals
Subcutaneous HepG2 tumor model: HepG2 tumor-bearing nude
mice (6–8 weeks old, female) were established by subcutaneous injec-
All cell lines were obtained from the American Type Culture
Collection (ATCC, Manassas, VA) and cultured in 1640 medium with
10% Fetal Bovine Serum (FBS, Gibco, Grand Island, USA), 1% penicillin/
streptomycin (Sigma-Aldrich, St. Louis, USA).
tion of a suspension of 4 × 106 HepG2 cells in PBS (100
μL). After 10
days post-inoculation, the mice were randomly divided into 3 groups (n
= 5): PBS, sorafenib (oral administration of 10 mg/kg every two days for
6 times), and BSA/CPT-SS-APBA nanoparticles (intravenously admin-
istration of 5 mg/kg CPT-eq. dose every two days for 6 times). Tumor
volume and mouse weight were monitored every 2 days. When the
tumor volume exceeded 1000 mm3, mice were sacrificed and the tumors
were collected and weighed. Tumors and other organs were harvested
and then fixed with 4% paraformaldehyde, embedded, sliced, and
stained for H&E examination and TUNEL assay.
Female BALB/c mice and ICR mice (6–8 weeks) were purchased from
the Animal Center of Zhejiang University and maintained under stan-
dard conditions. Animal use and care were approved by the Animal
Ethics Committee of Zhejiang University and were carried out in
accordance with the institutional guidelines.
2.8. Cell viability assay
2.12. H&E staining
The cytotoxicity of BSA/CPT-SS-APBA nanoparticles (or BSA/CCM-
APBA nanoparticles) to different cell lines was determined by the
typical MTT (3-(4, 5-dimethyl-2-thiazolyl)-2, 5-diphenyl-2H-tetrazo-
lium bromide) assay. In brief, cells were planted in 96-well plates with a
density of 5000 cells per well and incubated for 24 h. The medium was
replaced with a fresh medium containing the drugs at a series of
The freshly collected HepG2 tumor samples and some other organs
(heart, liver, spleen, lung, and kidney) were washed in PBS, fixed with
4% neutral buffered paraformaldehyde, dehydrated by gradient con-
centration of ethanol and p-xylene, embedded in paraffin, and cross-
sectioned at a thickness of 5
μm. The sections were stained with
hematoxylin-eosin (H&E, Beyotime, China) and observed under light
designated concentrations. After 48 h incubation, 20
μL MTT (5 mg/mL)
in PBS was added to each well, followed by incubation for another 4 h.
microscopy.
Finally, the medium was carefully removed and replaced with 100 μL
DMSO to dissolve the formazan crystals. The absorbance intensity of
each well was measured in a Molecular Devices microplate reader at
562/620 nm. Each drug concentration was tested in triplicate and in
three independent experiments.
2.13. TUNEL immunofluorescence assay
The freshly collected subcutaneous HepG2 tumors were sectioned
with a thickness of 4 μm, and the TUNEL immunofluorescence staining
364

L. Hao et al.
Journal of Controlled Release 330 (2021) 362–371
Scheme 2. Synthesis of 2-acetylphenylboronic acid (2-APBA) conjugated prodrugs and controls. a,h, tert-butyl bromoacetate, K2CO3, DMF, r.t., overnight; b,l, tri-
fluoromethanesulfonic anhydride, pyridine, DCM, r.t., 24 h; c,m, bis(pinacolato)diboron, Pd(dppf)Cl2dppf, potassium acetate, dioxane, 90 ◦C, 12 h; d,i,n,r, TFA, DCM,
6 h; e, CPT, triphosgene, 4-dimethylaminopyridine (DMAP), DCM/THF, r.t., 24 h; f,j,o,s, N,N′-dicyclohexylcarbodiimide(DCC)/DMAP, DCM, r.t., 24 h; g,p, boronic
acid resin, CH3CN: HCl (1 M) (9: 1), r.t., 24 h; k,q, tert-butyl 4-bromobutanoate, K2CO3, DMF, r.t., overnight.
was performed following the manufacturer’s instructions. Briefly, the
slices were fixed with cold acetone for 10 min, dried on slide rack for 30
min, rinsed with PBS (0.1 M, pH = 7.4) for 3 times, blocked with 10%
goat serum for 60 min at r.t. and then incubated with the TUNEL anti-
body (Beyotime, 1:500) at 4 ◦C overnight. The sections were rinsed three
times with PBS and further incubated with FAM-labeled second anti-
body (BD, 1:200) for 1 h at room temperature in the dark. The slices
were rinsed three times with PBS and the nuclei were stained with DAPI
(Beyotime, C1006) staining buffer for 15 min. A confocal microscope
acquired fluorescent images.
2.14. Statistical analysis
Statistical analysis was done using Excel, Origin, and GraphPad
Prism. Comparisons were made using a two-tailed, unpaired Student’s t-
test. P < 0.05 was considered statistically significant and all the results
are expressed as mean ± s.d..
3. Results and discussion
The syntheses of CPT-SS-APBA prodrug is shown in Scheme 2. 2,4-
Dihydroxyacetophenone was selectively alkylated with tert-butyl bro-
moacetate to yield compound 1 in a 79% yield. The phenol group in
365

L. Hao et al.
Journal of Controlled Release 330 (2021) 362–371
Fig. 1. 2-APBA-functionalized prodrugs complexing with albumin and forming nanoparticles. Particle size distributions (PSD in volume distribution, %) as obtained
from DLS and correlation functions of the BSA solution (5 mg/mL) after adding the prodrugs or CPT or CCM control; the BSA/(pro)drug ratio was 5.0. (a) CPT or
prodrugs; (b) CCM or prodrugs.
compound 1 was converted to a triflate (compound 2) in an alkaline
environment, followed by a typical Miyaura borylation reaction with bis
(pinocolato)diboron to afford compound 3 as a white solid. The disulfide
linker was connected to CPT as reported [27] and subsequently conju-
gated with compound 4 to obtain compound 6. The pinacol group was
then removed by boronic acid resin to produce CPT-SS-APBA. Similarly,
the control prodrug with acetylphenyl (AP) group, CPT-SS-AP, was
synthesized using the 4′-hydroxyacetophenone as the starting materials
(Scheme 2i). The curcumin prodrugs, CCM-APBA and CCM-AP, were
prepared similarly except using tert-butyl 4-bromobutanoate to avoid
the steric hindrance (Scheme 2ii).
[29] or BSA forming adducts. The planar structures of CPT and CPT-SS-
AP made them easily crystallize rather than assemble with BSA. As a
result, adding CPT or control prodrug CPT-SS-AP to the BSA solution
immediately produced large precipitates (Fig. 1a), giving very low drug
loading contents (< 1%) and drug loading efficiency (< 10%) in BSA
(Table S1). In contrast, the CPT-SS-APBA bound to BSA via forming
iminoboronates and thus produced more hydrophobic CPT-SS-APBA/
BSA adducts, which self-assembled adduct nanoparticles. Therefore,
adding CPT-SS-APBA into the BSA solution with stirring gave a clear
blueish solution. DLS measurements showed that the solution contained
nanoparticles with sizes of ~100 nm (Fig. 1a). The CPT-equivalent
(CPT-eq.)-loading contents could be controlled in a range of 2.20% to
9.55% (wt%) by tuning the BSA/CPT-SS-APBA ratios (Fig. S18,
Table S2). The loading efficiencies were over 98% at the BSA to CPT-SS-
APBA (w/w) ratios higher than 5.0 (Table S2). Very interestingly, the
formed nanoparticle sizes were independent of the ratios, all about 125
nm (Table S2). BSA/CPT-SS-APBA at a ratio of 5.0 produced nano-
particles with a 7.68% CPT-eq.-loading content and a drug loading ef-
ficiency of 98.76% and was selected for further evaluations.
1
The resulting compounds were characterized and confirmed by H
NMR, 13C NMR spectra, and ESI-MS spectra (Fig. S1–S17). It should be
pointed out that in mass spectrometry under electrospray ionization
conditions, arylboronic acids show unique properties due to the special
gas-phase chemistry of boronic acid [28]. The CPT-SS-APBA and CCM-
APBA peaks appeared as [M+28+Na]+ because the boronic acid reac-
ted with methanol to give the boronate esters (Fig. S10 & S15).
Once a hydrophobic drug is added to the BSA solution, drug mole-
cules competitively assemble with themselves forming precipitation
Similarly, CCM-APBA complexed with the BSA and produced
366

L. Hao et al.
Journal of Controlled Release 330 (2021) 362–371
Fig. 2. Iminoboronate formation detected by 11B NMR and 1H NMR spectra. (a) 11B NMR spectra of 2-APBA, and its mixture with lysine at the molar ratios of 1: 10 or
1:1 or with BSA at an equimolar ratio of BSA’s primary amines in deuterated PBS buffer (pH = 7.4) calibrated against the BF.3 Et2O (0 ppm). (b) 1H NMR spectra of 2-
APBA, its mixture with lysine (1:10), lysine, acetophenone, and acetophenone mixture with lysine (1:10).
Fig. 3. Stability and drug release of the BSA/prodrug nanoparticles. (a) The DLS patterns of BSA/CPT-SS-APBA nanoparticles in the cell growth medium with 10%
FBS for different times. (b,c) The lyophilization effects: the DLS patterns (b) and representative TEM images (c) of BSA/CPT-SS-APBA nanoparticles before and after
lyophilization. (d) The release kinetic of CCM from BSA/CCM-APBA nanoparticles with or without GSH (10 mM) at pH 7.4 and 5.0. (e) Scheme of CPT-release from
BSA/CPT-SS-APBA nanoparticles in response to GSH and acidity. (f) The release kinetics of CPT from BSA/CPT-SS-APBA nanoparticles with or without GSH (10 mM)
at pH 7.4 and 5.0.
nanoparticles of sizes about 130 nm, whereas adding CCM or CCM-AP
into the BSA solution also precipitated (Fig. 1b). These results indi-
cated that 2-APBA-modified prodrugs efficiently complexed with BSA,
forming BSA-drug adducts that self-assembled into ~100 nm NPs.
The 2-APBA-functionalized prodrugs could also bind other proteins,
such as RNase A, IgG, and aPD1. CPT-SS-APBA or CCM-APBA-bound
proteins self-assembled into nanoparticles (Fig. S19-S21 and
Table S1). Interestingly, CPT-SS-APBA could also form uniform nano-
particles once adding in mouse serum, indicating that they bound the
BSA in the serum (Fig. S22).
The iminoboronate formation between the prodrug and BSA’s pri-
mary amines was detected using 2-APBA and lysine as the model
367

L. Hao et al.
Journal of Controlled Release 330 (2021) 362–371
Fig. 4. Cytotoxicity assays. (a) The MTT assay curves of free CPT, CPT-SS-APBA, BSA/CPT-SS-APBA, and CPT-11 against HepG2 and 4 T1 cells and their IC50 values.
(b) The MTT assay curves of free CCM, BSA/CCM-APBA, and CCM-APBA against HepG2 and 4T1cells and their calculated IC50 values; 48 h incubation; n = 3.
compounds with 11B NMR and 1H NMR in deuterated PBS (Fig. 2). 2-
APBA alone had a typical peak of aryl boronic acid/ester at 18–20
ppm (BF3⋅Et2O as the internal reference). Mixing 2-APBA with lysine
produced a new peak at 8–10 ppm, which is characteristic of the anionic
boron in iminoboronates [30]. The peak also appeared when 2-APBA
was mixed with BSA at an equimolar ratio of BSA’s primary amines,
indicating the formation of iminoboromates, as further verified by the
1H NMR spectra (Fig. 2b). In contrast, acetophenone did not react with
the Lys amines.
and zeta-potentials (Fig. 3b, Fig. S24), which were further verified by
the TEM results (Fig. 3c, Fig. S25).
The drug release profiles of the formed nanoparticles were first
evaluated using BSA/CCM-APBA nanoparticles. BSA/CCM-APBA nano-
particles were stable and free of burst release; only ~5% CCM was
released after 48 h at the physiological pH. GSH triggered the hydrolysis
of the iminoboronate by disrupting the B-N bond, but the hydrolysis of
CCM-APBA to CCM at physiological pH was slow and enhanced at pH
5.0 (Fig. 3d). However, the hydrolysis of CCM-APBA was incomplete,
only ~60%, even after 48 h incubation at pH 5.0 with GSH. In this re-
gard, the disulfide linker was introduced into the CPT-SS-APBA prodrug
to accelerate the prodrug’s conversion to CPT by responding to the
intracellular glutathione (Fig. 3e). Indeed, once adding GSH to the so-
lution of BSA/CPT-SS-APBA at pH 7.4 or 5.0, micrometer-sized pre-
cipitates were observed, and the original 100 nm peak disappeared
(Fig. S26), indicating complete disassembly of the nanoparticles and the
The serum stability and lyophilization of the BSA/CPT-SS-APBA
nanoparticles were further investigated. As depicted in Fig. 3a and
Fig. S23, both the size (in the presence of 10% FBS) and zeta potential (in
the presence of 100% mouse plasma) of the BSA/CPT-SS-APBA main-
tained constant over 48 h, indicating great colloidal stability of the
nanoparticles. After lyophilization, the reconstructed BSA/CPT-SS-
APBA and BSA/CCM-APBA nanoparticles maintained the same sizes
Fig. 5. Pharmacokinetic profiles and parameters of CPT-11, free CPT-SS-APBA, and BSA/CPT-SS-APBA nanoparticles. (a) Blood clearance profiles and (b) phar-
macokinetic parameters of CPT-11, free CPT-SS-APBA, and BSA/CPT-SS-APBA nanoparticles at a CPT (or SN38)-eq. dose of 5 mg/kg. T1/2 alpha: half-life time of
distribution phase. T1/2 beta: half-life time of elimination phases. AUC0-t: area under the curve. MRT: mean residence time. CL: clearance. Data are depicted as mean
± SD.
368

L. Hao et al.
Journal of Controlled Release 330 (2021) 362–371
Fig. 6. In vivo antitumor activity assays of BSA/CPT-SS-APBA nanoparticles. (a–c) The BALB/c mice bearing subcutaneous 4 T1 tumors (n = 6) were intravenously
treated with PBS, CPT-11 (5 mg/kg) and BSA/CPT-SS-APBA nanoparticles (CPT-eq. dose 5 mg/kg); (d–f) The BALB/c nude mice bearing subcutaneous HepG2 tumors
(n = 5) were treated with PBS, sorafenib (oral administration, 10 mg/kg) and BSA/CPT-SS-APBA (i.v., CPT-eq. dose 5 mg/kg). Dosing schedules are indicated by red
arrows. (a,d) Tumor volumes as a function of time. (b,e) The average tumor weight of each group at the end of the antitumor experiment. (c,f) Body weights change of
the mice during the treatment. (g,h) Representative histological features (g) and TUNEL immunofluorescence staining (h) of sections of the subcutaneous HepG2
tumor resected after the mice were sacrificed. Scale bars = 40 μm. Error bars represent the standard deviation of means; Statistical significance: P values were
obtained using the Student’s t-test. *P < 0.05, **P < 0.01, ***P < 0.001. (For interpretation of the references to colour in this figure legend, the reader is referred to
the web version of this article.)
precipitation of CPT. The CPT-release from BSA/CPT-SS-APBA nano-
particles was monitored by a dialysis method (Fig. 3f). BSA/CPT-SS-
APBA nanoparticles were very stable; less than 10% CPT was released
after 48 h incubation in PBS at the physiological pH. Both the GSH and
acidity dramatically accelerated the CPT release from the particles. The
CPT- release rate was further accelerated at pH 5.0 in the presence of
GSH. Thus, the BSA/CPT-SS-APBA nanoparticles would keep stable in
the bloodstream while selectively release CPT in the acidic lysosomes
and the tumor cell cytosol rich in GSH.
APBA can be effectively released from the nanoparticles upon inter-
nalized by the tumor cells (Fig. 4b).
The pharmacokinetics of BSA/CPT-SS-APBA nanoparticles was sub-
sequently estimated, and the small molecular drug CPT-11 was used as
control. Not surprisingly, CPT-11 was quickly cleared, and the plasma
concentrations of SN38 were below 0.1 μg/mL in minutes. The BSA/
CPT-SS-APBA underwent a slower clearance. The serum CPT level in
the mice treated with BSA/CPT-SS-APBA nanoparticles was still
approximately 0.1
μg/mL after 6 h. Thus, the BSA/CPT-SS-APBA
The in vitro cytotoxicity of BSA/CPT-SS-APBA nanoparticles against
HepG2 and 4 T1 cells was determined by the MTT assay (Fig. 4a). CPT-
SS-APBA, BSA/CPT-SS-APBA, CPT-11, and free CPT all showed dose-
dependent cytotoxicity against both cell lines. The IC50 values of BSA/
nanoparticles had an AUC ~96 fold larger than CPT-11 (Fig. 5a–b).
Therefore, the BSA/CPT-SS-APBA nanoparticles could accumulate in the
tumor tissue through the passive enhanced permeability and retention
(EPR) [32] effect. Notably, CPT-SS-APBA also had much better phar-
macokinetics than CPT-11 because it could rapidly bind to the endog-
enous serum proteins. However, the severe acute side effect was
observed when free CPT-SS-APBA was i.v. injected because of its hy-
drophobic nature of CPT-SS-APBA and uncontrolled binding to key
proteins. The biodistribution of BSA/CPT-SS-APBA was conducted on
the subcutaneous 4 T1 xenograft tumor model (Fig. S27). BSA/CPT-SS-
APBA had very low levels in the heart, spleen, and lung. Notably, BSA/
CPT-SS-APBA against HepG2 and 4 T1 cells were 0.17 and 1.50 μg/
mL, respectively, comparable to those of free CPT and CPT-SS-APBA,
indicating efficient CPT-release from the nanoparticles in the cyto-
plasm. In contrast, the IC50 values of CPT-11 were higher than 10 μg/mL
to the cells due to the low conversation rate of CPT-11 [31]. Similarly,
CCM, CCM-APBA, and BSA/CCM-APBA nanoparticles showed similar
cytotoxicity against HepG2 and 4 T1 cells, further suggesting the CCM-
369

L. Hao et al.
Journal of Controlled Release 330 (2021) 362–371
[2] A.O. Elzoghby, W.M. Samy, N.A. Elgindy, Albumin-based nanoparticles as
potential controlled release drug delivery systems, J. Control. Release 157 (2012)
168–182.
CPT-SS-APBA accumulated significantly more in the kidney, liver, and
tumors. These biodistribution profiles indicate an effective tumor-
targeting capability of BSA/CPT-SS-APBA.
[3] F. Kratz, A clinical update of using albumin as a drug vehicle - a commentary,
J. Control. Release 190 (2014) 331–336.
Encouraged by the excellent in vitro cytotoxicity and the improved
pharmacokinetics, BSA/CPT-SS-APBA nanoparticles were assessed for
tumor therapy in a 4 T1 tumor-bearing xenograft model with CPT-11 as
control (Fig. 6a). BSA/CPT-SS-APBA nanoparticles exhibited improved
antitumor effects and delayed tumor progression compared with PBS (P
= 0.00004) and CPT-11 (P = 0.00006). The tumor inhibition rate
calculated from the tumor weight was 55.5% for BSA/CPT-SS-APBA
nanoparticles compared with 29.1% for CPT-11 (Fig. 6b). No signifi-
cant body weight loss was observed during the treatment period
(Fig. 6c).
[4] F.F. An, X.H. Zhang, Strategies for preparing albumin-based nanoparticles for
multifunctional bioimaging and drug delivery, Theranostics 7 (2017) 3667–3689.
[5] Q. Chen, C. Liang, C. Wang, Z. Liu, An imagable and photothermal "Abraxane-like"
nanodrug for combination cancer therapy to treat subcutaneous and metastatic
breast tumors, Adv. Mater. 27 (2015) 903–910.
[6] T.H. Kim, H.H. Jiang, Y.S. Youn, C.W. Park, K.K. Tak, S. Lee, H. Kim, S. Jon,
X. Chen, K.C. Lee, Preparation and characterization of water-soluble albumin-
bound curcumin nanoparticles with improved antitumor activity, Int. J. Pharm.
403 (2011) 285–291.
[7] B. Elsadek, F. Kratz, Impact of albumin on drug delivery–new applications on the
horizon, J. Control. Release 157 (2012) 4–28.
[8] Y. Min, J.M. Caster, M.J. Eblan, A.Z. Wang, Clinical translation of nanomedicine,
Chem. Rev. 115 (2015) 11147–11190.
The BSA/CPT-SS-APBA against the subcutaneous HepG2 tumor
model was further evaluated and compared with a clinical used anti-
hepatoma drug sorafenib (Fig. 6d-f). The BSA/CPT-SS-APBA nano-
particles again showed an enhanced tumor inhibition activity compared
to PBS (P = 0.0005) and sorafenib (P = 0.006) with a higher tumor
inhibition rate than sorafenib (65.5% vs 29.1%, P = 0.001). The body
weights of the mice treated with BSA/CPT-SS-APBA nanoparticles
decreased slightly during the treatment but remained stable once the
treatment was terminated (Fig. 6f). The hematoxylin and eosin (H & E)
staining showed no obvious organ toxicity after BSA/CPT-SS-APBA
nanoparticle treatment (Fig. S28). The H&E staining of the tumors
showed extensive tumor cell vacuolization and nuclear condensation in
the BSA/CPT-SS-APBA treated tumors compared to the tightly-packed
tumor cells in the PBS and sorafenib treated group, suggesting that
BSA/CPT-SS-APBA induced remarkable cell apoptosis. The TUNEL
immunofluorescence assays also verified these results (Fig. 6g,h).
[9] A.M. Sofias, M. Dunne, G. Storm, C. Allen, The battle of “nano” paclitaxel, Adv.
Drug Deliv. Rev. 122 (2017) 20–30.
[10] E. Bernabeu, M. Cagel, E. Lagomarsino, M. Moretton, D.A. Chiappetta, Paclitaxel:
what has been done and the challenges remain ahead, Int. J. Pharm. 526 (2017)
474–495.
[11] S. Hu, E. Lee, C. Wang, J. Wang, Z. Zhou, Y. Li, X. Li, J. Tang, D.H. Lee, X. Liu,
Y. Shen, Amphiphilic drugs as surfactants to fabricate excipient-free stable
nanodispersions of hydrophobic drugs for cancer chemotherapy, J. Control.
Release 220 (2015) 175–179.
[12] X. Yang, X. Xue, Y. Luo, T.Y. Lin, H. Zhang, D. Lac, K. Xiao, Y. He, B. Jia, K.S. Lam,
Y. Li, Sub-100nm, long tumor retention SN-38-loaded photonic micelles for tri-
modal cancer therapy, J. Control. Release 261 (2017) 297–306.
[13] W. Um, J. Park, A. Youn, H. Cho, S. Lim, J.W. Lee, H.Y. Yoon, D.K. Lim, J.H. Park,
K. Kim, A comparative study on albumin-binding molecules for targeted tumor
delivery through covalent and noncovalent approach, Bioconjug. Chem. 30 (2019)
3107–3118.
[14] D. Pilati, K.A. Howard, Albumin-based drug designs for pharmacokinetic
modulation, Expert Opin. Drug Metab. Toxicol. 16 (2020) 783–795.
[15] J. Delahousse, C. Skarbek, A. Paci, Prodrugs as drug delivery system in oncology,
Cancer Chemother. Pharmacol. 84 (2019) 937–958.
[16] X. Liu, R.P. Mohanty, E.Y. Maier, X. Peng, S. Wulfe, A.P. Looney, K.L. Aung,
D. Ghosh, Controlled loading of albumin-drug conjugates ex vivo for enhanced
drug delivery and antitumor efficacy, J. Control. Release 328 (2020) 1–12.
[17] F. Kratz, G. Ehing, H.M. Kauffmann, C. Unger, Acute and repeat-dose toxicity
studies of the (6-maleimidocaproyl)hydrazone derivative of doxorubicin (DOXO-
EMCH), an albuminbinding prodrug of the anticancer agent doxorubicin, Hum.
Exp. Toxicol. 26 (2007) 19–35.
4. Conclusion
In summary, 2-acetylphenylboronic acid-functionalized CPT and
CCM prodrugs can efficiently bind proteins by forming iminoboronates
[18] D. Lebrecht, A. Geist, U.P. Ketelsen, J. Haberstroh, B. Setzer, F. Kratz, U.A. Walker,
The 6-maleimidocaproyl hydrazone derivative of doxorubicin (DOXO-EMCH) is
superior to free doxorubicin with respect to cardiotoxicity and mitochondrial
damage, Int. J. Cancer 120 (2007) 927–934.
with lysine
ε-amines and induce the protein-drug conjugates self-
assemble into nanoparticles with high loading efficiency and
improved colloidal stability. The acidity or GSH-labile iminoboronates
renders the burst-free nanoparticles fast-drug release. As a result, BSA/
CPT-SS-APBA nanoparticles showed enhanced anticancer efficacy in
subcutaneous 4 T1 and HepG2 tumor models than clinical used CPT-11
and sorafenib. Therefore, this work demonstrates a versatile protein-
binding prodrug platform applicable for protein-based drug formula-
tions and even antibody-drug conjugates.
[19] B. Akgun, D.G. Hall, Boronic acids as bioorthogonal probes for site-selective
labeling of proteins, Angew. Chem. Int. Ed. Engl. 57 (2018) 13028–13044.
[20] S. Cambray, J. Gao, Versatile bioconjugation chemistries of ortho-boronyl aryl
ketones and aldehydes, Acc. Chem. Res. 51 (2018) 2198–2206.
[21] H. Faustino, M. Silva, L.F. Veiros, G.J.L. Bernardes, P.M.P. Gois, Iminoboronates
are efficient intermediates for selective, rapid and reversible N-terminal cysteine
functionalisation, Chem. Sci. 7 (2016) 5052–5058.
[22] P.M. Cal, J.B. Vicente, E. Pires, A.V. Coelho, L.F. Veiros, C. Cordeiro, P.M. Gois,
Iminoboronates: a new strategy for reversible protein modification, J. Am. Chem.
Soc. 134 (2012) 10299–10305.
[23] R. Ma, C. Zhang, Y. Liu, C. Li, Y. Xu, B. Li, Y. Zhang, Y. An, L. Shi, Iminoboronate-
based dual-responsive micelles via subcomponent self-assembly for hydrophilic
1,2-diol-containing drug delivery, RSC Adv. 7 (2017) 21328–21335.
[24] A. Bandyopadhyay, J. Gao, Iminoboronate-based peptide cyclization that responds
to pH, oxidation, and small molecule modulators, J. Am. Chem. Soc. 138 (2016)
2098–2101.
Declaration of Competing Interest
All the authors declare no conflicting interests.
Acknowledgment
[25] P. Cal, R.F.M. Frade, C. Cordeiro, P.M.P. Gois, Reversible lysine modification on
proteins by using functionalized boronic acids, Chem-Eur. J. 21 (2015) 8182–8187.
[26] P.M. Cal, R.F. Frade, V. Chudasama, C. Cordeiro, S. Caddick, P.M. Gois, Targeting
cancer cells with folic acid-iminoboronate fluorescent conjugates, Chem. Commun.
50 (2014) 5261–5263.
We thank the National Natural Science Foundation of China
(51833008, 21704090, 51522304), the National Key Research and
Development Program (2016YFA0200301), and the China postdoctoral
Science Foundation (NO. 2019M662059) for financial support.
[27] Q. Zhou, S.Q. Shao, J.Q. Wang, C.H. Xu, J.J. Xiang, Y. Piao, Z.X. Zhou, Q.S. Yu, J.
B. Tang, X.R. Liu, Z.H. Gan, R. Mo, Z. Gu, Y.Q. Shen, Enzyme-activatable polymer-
drug conjugate augments tumour penetration and treatment efficacy, Nat.
Nanotech. 14 (2019) 799–809.
[28] L.F. Wang, C.F. Dai, S.K. Burroughs, S.M.L. Wang, B.H. Wang, Arylboronic acid
chemistry under electrospray conditions, Chem-Eur. J. 19 (2013) 7587–7594.
[29] S.X. Lv, Y.C. Wu, K.M. Cai, H. He, Y.J. Li, M. Lan, X.S. Chen, J.J. Cheng, L.C. Yin,
High drug loading and sub-quantitative loading efficiency of polymeric micelles
driven by donor-receptor coordination interactions, J. Am. Chem. Soc. 140 (2018)
1235–1238.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.jconrel.2020.12.035.
References
[1] Z. Liu, X. Chen, Simple bioconjugate chemistry serves great clinical advances:
albumin as a versatile platform for diagnosis and precision therapy, Chem. Soc.
Rev. 45 (2016) 1432–1456.
370

L. Hao et al.
Journal of Controlled Release 330 (2021) 362–371
[30] A. Bandyopadhyay, J.M. Gao, Iminoboronate formation leads to fast and reversible
conjugation chemistry of alpha-nucleophiles at neutral pH, Chem-Eur. J. 21 (2015)
14748–14752.
[31] P.D. Senter, K.S. Beam, B. Mixan, A.F. Wahl, Identification and activities of human
carboxylesterases for the activation of CPT-11, a clinically approved anticancer
drug, Bioconjug. Chem. 12 (2001) 1074–1080.
[32] J. Shi, P.W. Kantoff, R. Wooster, O.C. Farokhzad, Cancer nanomedicine: progress,
challenges and opportunities, Nat. Rev. Cancer 17 (2017) 20–37.
371