Rendering protein-based particles transiently insoluble for therapeutic applications

Article abstract of DOI:10.1021/ja302363r

Herein, we report the fabrication of protein (bovine serum albumin, BSA) particles which were rendered transiently insoluble using a novel, reductively labile disulfide-based cross-linker. After being cross-linked, the protein particles retain their integrity in aqueous solution and dissolve preferentially under a reducing environment. Our data demonstrates that cleavage of the cross-linker leaves no chemical residue on the reactive amino group. Delivery of a self-replicating RNA was achieved via the transiently insoluble PRINT protein particles. These protein particles can provide new opportunities for drug and gene delivery.

Full text of DOI:10.1021/ja302363r

Communication
pubs.acs.org/JACS
Rendering Protein-Based Particles Transiently Insoluble for
Therapeutic Applications
Jing Xu,†,‡ Jin Wang,†,§,‡,∥ J. Christopher Luft,†,§,⊥,⊗ Shaomin Tian,†,§,⊥,⊗ Gary Owens, Jr.,#
Ashish A. Pandya,†,⊗ Peter Berglund,# Patrick Pohlhaus,# Benjamin W. Maynor,# Jonathan Smith,#
Bolyn Hubby,# Mary E. Napier,†,§,⊥,⊗,× and Joseph M. DeSimone*,†,+,§,¶,⊥,⊗,∇,◆
+
§
¶
†Department of Chemistry, Department of Pharmacology, Carolina Center of Cancer Nanotechnology Excellence, Institute for
Advanced Materials, Institute for Nanomedicine, ⊗Lineberger Comprehensive Cancer Center, and Department of Biochemistry
⊥
×
and Biophysics, University of North Carolina, Chapel Hill, North Carolina 27599, United States
#Liquidia Technologies Inc., Morrisville, North Carolina 27560, United States
∇Department of Chemical and Biomolecular Engineering, North Carolina State University, Raleigh, North Carolina 27695, United
States
◆Sloan-Kettering Institute for Cancer Research, Memorial Sloan-Kettering Cancer Center, New York, New York 10021, United
States
S
* Supporting Information
do not allow control over particle size or shape.9 PRINT is a
ABSTRACT: Herein, we report the fabrication of protein
(bovine serum albumin, BSA) particles which were
rendered transiently insoluble using a novel, reductively
labile disulfide-based cross-linker. After being cross-linked,
the protein particles retain their integrity in aqueous
solution and dissolve preferentially under a reducing
environment. Our data demonstrates that cleavage of the
cross-linker leaves no chemical residue on the reactive
amino group. Delivery of a self-replicating RNA was
achieved via the transiently insoluble PRINT protein
particles. These protein particles can provide new
opportunities for drug and gene delivery.
platform technology that is an off-shoot of soft lithography that
enables the molding of micro- and nanoparticles having
precisely controlled size, shape, chemical composition and
surface functionality.10,11 PRINT has been transitioned to a
continuous roll-to-roll fabrication technique that can enable the
scale-up of particle production to practical levels for
applications in the clinic.
Dry microspheres or nanospheres composed of proteins are
usually instantaneously soluble when placed into aqueous
solutions.9 A couple of strategies have been reported that
maintain the stability of protein-based particles: (i) thermal
cross-linking, which causes the formation of intermolecular
disulfide bridges between free thiol groups;12 (ii) the use of
nonreversible chemical cross-linkers, such as 1-ethyl-3-(3-
dimethylaminopropyl)carbodiimide (EDC), glutaraldehyde,
formaldehyde, among others;7,13 and (iii) the use of reversible
cross-linkers like Lomant’s reagent, dithiobis[succinimidyl
propionate] (DSP), which can be cleaved upon exposure to
certain biologic conditions.5 Thermal cross-linking involves the
thermal denaturation of a given protein at high temperature and
cannot be applied to the delivery of fragile cargos. The use of
nonreversible chemical cross-linkers introduces permanent
cross-linkages between individual protein molecules which
limit the release of encapsulated cargos. Reversible cross-linkers
can be cleaved upon exposure to certain biologic conditions but
leaves a chemical residue, a potential neo-epitope, on the
protein after cleavage of the disulfide bond (Scheme 1a).14 If
the released protein from the particles has molecular pendants
attached, it may elicit undesirable immune responses toward
foreign antigens, which may induce adverse health effects.15,16
Inspired by the chemistry Wender et al. developed to release
drugs in their original state from a pro-drug form,17,18 we
designed a “traceless” reversible cross-linker that leaves no
pendant chemical residues on the molecule it reacts with after
elivering promising biological therapeutics to the desired
D
location in the body in a safe and effective fashion is one
of the key challenges in medicine. Proteins as the matrices for
drug delivery particles have many advantages including
biodegradability, biocompatibility, and amenability to surface
modification.1 Drug carriers using proteins as matrices for the
delivery of small molecule drugs and biological cargos, such as
plasmid DNA and siRNA, are being extensively studied.2−4
Each of these applications would benefit from having protein-
based particles that dissolve slowly in a controlled and desirable
manner.5 Herein, we report the synthesis of size- and shape-
specific protein micro- and nanoparticles using a top-down
particle fabrication technique called PRINT. Our approach
involves the design and synthesis of a novel “traceless” cross-
linker that renders protein-based particles transiently insoluble
in aqueous solutions. We also report the cytoplasmic delivery of
a self-replicating RNA via PRINT protein particles stabilized by
this novel cross-linker.
Protein particles are often made through costly and
complicated processes which include wet-milling, spray-freeze-
drying, microemulsion, microencapsulation, or supercritical
fluid methods.6−8 Frequently, these procedures result in highly
heterogeneous polydispersed spherical or granular particles and
Received: March 9, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja302363r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
dissolution of our protein particles. Wender et al. developed a
disulfide based pro-drug linker, which contains a carbonate
group instead of conventionally used ester linkage, to release
the drug in its original state.17 This chemistry has also been
applied to develop a fluorogenic probe for thiol detection and a
pro-drug for intracellular delivery.21−23 Inspired by previous
work, we developed a novel cross-linker dithio-bis(ethyl 1H-
imidazole-1-carboxylate) (DIC) for this study. Compared to
DSP, DIC has several advantages. Imidazoles were introduced
as the leaving groups in DIC to replace the highly reactive NHS
as in DSP in order to better control the rate of the cross-linking
reaction and the cross-linking density on the particle surface.
Furthermore, DIC is a “traceless” reversible cross-linker, which
does not leave any molecular pendants after disulfide cleavage
(Scheme 1b). As a control, a nondisulfide nondegradable cross-
linker, 2,2′-oxybis(ethane-2,1-diyl) bis(1H-imidazole-1-carbox-
ylate) (OEDIC) was also synthesized (Scheme 1c).
Scheme 1. (a) Structure of DSP and Reaction Scheme for
Protein Cross-Linking and Dissolution; (b) Structure of DIC
and Reaction Scheme for Protein Cross-Linking and
Dissolution; (c) Structure of OEDIC
To demonstrate the ability of DIC to release the amino
group in its original form after cleavage of the disulfide, we
utilized tyramine as a model, a small molecule with only one
amino group. Two tyramine molecules were reacted with DIC
in isopropyl alcohol, which completely simulates the cross-
linking conditions for protein-based particles (Scheme S2). To
prove this unique property of DIC, the commercially available
disulfide cross-linker DSP was used as a control in this study.
The dimer products were denoted as tyramine-DIC and
tyramine-DSP, respectively. Both compounds were treated with
50 mM of dithiothreitol (DTT) in phosphate buffer saline
(PBS) to cleave the disulfide bond. Gas chromatography mass
spectrum (GC−MS) results indicated that after cleavage of the
disulfide bond, tyramine was regenerated from tyramine-DIC.
No peak of tyramine was observed with tyramine-DSP (Figure
S4). In addition, 1H NMR and high-resolution mass
spectrometry confirmed the structure of tyramine recovered
after DTT treatment of tyramine-DIC (Supporting Informa-
tion).
The PRINT albumin particles were cross-linked with DIC
and OEDIC at different cross-linker concentrations (based on a
constant particle concentration) and a quantitative study of
particle dissolution was performed (Figure 1). From these data,
it was apparent that the particles cross-linked with the disulfide
linker DIC retained their stability in PBS and preferentially
dissolved under reducing conditions. The rate of particle
dissolution can be effectively modulated by changing the cross-
linker concentration.
To demonstrate the capability of the particles to deliver
biological cargos to a reducing cellular compartment, we
selected an RNA molecule capable of self-replication and
encoding Chloramphenicol Acetyl Transferase (CAT) protein
exclusively in the cytoplasm. Successful delivery of this RNA
into the cytoplasm of Vero cells would yield expression of CAT
protein which can be detected by enzyme-linked immunosorb-
ent assay (ELISA) using antigen-specific antibodies. Particles
were fabricated using the PRINT process containing the
Replicon RNA cargo (1 wt %), BSA, lactose, and glycerol and
were stabilized using the DIC cross-linker. The isoelectric point
of BSA is 4.75, resulting in cross-linked particles with a negative
ζ-potential. On the basis of the previous studies, cells
preferentially internalize positively charged particles.11 Trans-
IT-mRNA transfection reagent (TransIT) was mixed with BSA
particles in order to introduce positive charges to BSA particle
surface to enhance cell uptake and endosomal escape (Table
S3). As a result of internalization of the particles coated with
cleavage of the linker. We have applied the use of such
transient, traceless chemical cross-linkers to achieve stabiliza-
tion of protein particles fabricated using the PRINT
technology.
On the basis of our previous efforts,9 a melt-solidification
strategy was employed in the fabrication of PRINT protein
particles (Figure S1). Lactose and glycerol (for use as
plasticizers) are mixed with the protein of choice, in this case
bovine serum albumin (BSA), to form the preparticle material
that readily “flows” when heated. Serum albumin was chosen in
this study for two reasons: (1) as the most abundant blood
plasma protein, it is essential for the transport of many
physiological molecules, and (2) it also has the advantage of
being readily available.1 In particular, BSA was used in this
study due to its easy accessibility and cost effectiveness for our
proof-of-concept study. The processing temperature used to
fabricate protein-based particles in this study can be as low as
60 °C, which avoids the potential degradation of the delicate
biological therapeutics we are investigating.
Taking advantage of the PRINT process, a series of BSA
particles were fabricated in the size range of 200 nm to several
micrometers (Figure S2). In this study, cylindrical particles with
both diameter and height as 1 μm were fabricated with a
preparticle composition containing 37.5 wt % of BSA, 37.5 wt
% of α-D-lactose, and 25.0 wt % of glycerol. To determine the
final composition after the purification step, particles were
analyzed by high-performance liquid chromatography (HPLC)
coupled with evaporative light scattering detection (ELSD).
The dry particles were found to contain 86.7 1.1 wt % of
BSA, 10.2 1.7 wt % of lactose, and 3.1 0.7 wt % of glycerol
(Table S1). The protein particles, at this stage, are fully soluble
when brought into contact with water (Figure S3), which
indicated the necessity for a cross-linker to transiently stabilize
the albumin particles.
To utilize protein-based particles for therapeutic applications,
they are usually stabilized with cross-linkers, which can be
cleaved under certain physiological stimuli.19 The cytoplasm of
cells is known for its high concentration of reduced glutathione
(GSH) compared to the extracellular environment (GSH
concentration differs by 1000 folds intracellularly and
extracellularly).20 In this study, we take advantage of the
reducing environment in the cytoplasm of cells by introducing a
disulfide-based cross-linker that should trigger the intracellular
B
dx.doi.org/10.1021/ja302363r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
complexing of extracellular RNA and TransIT reagents, DIC-
cross-linked BSA/RNA particles were incubated in PBS for 4 h
at 37 °C, particles were pelleted by centrifugation, and the
supernatant was dosed to cells with TransIT. Absence of CAT
protein expression was observed confirming particle mediated
delivery of RNA. Analysis using confocal microscopy was also
carried out to visually confirm the generation of CAT protein
(Figure S6).
In summary, a novel cross-linker strategy was able to
effectively render the PRINT protein particles transiently
insoluble in aqueous solutions. The cross-linked particles
preferentially dissolved under reducing conditions and the
rate of particle dissolution can be controlled by adjusting the
cross-linker concentration. By coating the RNA-loaded particles
with TransIT, CAT protein was expressed via delivery of
PRINT particles. These precisely engineered protein particles
showed great potential for drug and gene delivery.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures, synthetic schemes, characterization;
the fabrication process for protein-based PRINT particles
(Figure S1); SEM of protein particles (Figure S2); particle
composition (Table S1); BSA particle dissolution by
microscopy image (Figure S3); synthesis scheme of cross-
linker DIC and OEDIC (Scheme S1); synthesis scheme of
tyramine-DIC and tyramine-DSP (Scheme S2); GC−MS
characterization of tyramine-DIC and tyramine-DSP after
treatment with DTT (Figure S4); characterization of cross-
linked BSA particles (Table S2); particle dissolution by
microscopy image at 5-h time point (Figure S5); charracteriza-
tion of cross-linked BSA particles with and without TransIT
(Table S3); confocal image of CAT protein expression (Figure
S6); cytotoxicity of particles coated with TransIT (Figure S7);
study of the depth of cross-linking reaction. This material is
available free of charge via the Internet at http://pubs.acs.org.
Figure 1. (a) SEM image of BSA particles (cross-linked at 4.4 mM of
DIC) after incubation with water, scale bar represents 10 μm. (b)
Dissolution profile of cross-linked BSA particles in PBS containing 5
mM GSH (GSH, squares with solid lines) and PBS only (PBS,
triangles with dotted lines); A, B and C, particles cross-linked at 4.4
mM, 6.6 mM, and 9.9 mM of DIC, respectively; D, particles cross-
linked at 4.4 mM of OEDIC. The error bars stand for the standard
deviation calculated from triplicates.
TransIT into Vero cells, CAT protein was generated (Figure 2).
CAT protein production via delivery of PRINT particles was
competitively comparable to the same amount of RNA directly
transfected by TransIT. As a negative control, blank particles
(no RNA encapsulated) did not induce any protein expression.
Furthermore, to rule out CAT protein expression induced by
AUTHOR INFORMATION
■
Corresponding Author
desimone@unc.edu
Present Address
∥Department of Pharmacology, Baylor College of Medicine,
Huston, Texas, 77030.
Author Contributions
‡These authors contributed equally.
Notes
The authors declare the following competing financial
interest(s): The research reported in this paper received partial
financial support from Liquidia Technologies (www.liquidia.-
com), a company co-founded by Professor Joseph DeSimone.
Prof. Joseph M. DeSimone is on the board of directors and has
personal financial interest in Liquidia Technologies.
ACKNOWLEDGMENTS
■
We thank Tammy Shen for help with SEM images in Figure S2.
The work is supported in part by National Institutes of Health
Director’s Pioneer Award (1DP1OD006432) and R01
(R01EB009565), University of North Carolina Cancer
Research Fund, the Chancellor’s Eminent Professorship of
Chemistry at the University of North Carolina at Chapel Hill,
and a sponsored research agreement with Liquidia Technolo-
gies.
Figure 2. CAT protein generated from Vero cells (2 × 104 cells per
well) measured by ELISA. Black, CAT RNA standards with TransIT;
purple, blank particles with TransIT; orange, CAT RNA containing
BSA particles with TransIT, 2, 1, and 0.5 μg/mL of particles
correspond to 30, 15, and 7.5 ng/mL of RNA (Supporting
Information); blue, supernatant from particles incubated in PBS for
4 h at 37 °C with TransIT. Error bars represent standard deviation
calculated from four wells.
C
dx.doi.org/10.1021/ja302363r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
REFERENCES
■
(1) Kratz, F. J. Controlled Release 2008, 132, 171.
(2) Hawkins, M. J.; Soon-Shiong, P.; Desai, N. Adv. Drug Delivery
Rev. 2008, 60, 876.
(3) Rhaese, S.; von Briesen, H.; Rubsamen-Waigmann, H.; Kreuter,
̈
J.; Langer, K. J. Controlled Release 2003, 92, 199.
(4) Abbasi, S.; Paul, A.; Prakash, S. Cell Biochem. Biophys. 2011, 61,
277.
(5) Yu, M.; Ng, B. C.; Rome, L. H.; Tolbert, S. H.; Monbouquette,
H. G. Nano Lett. 2008, 8, 3510.
(6) Maa, Y. F.; Nguyen, P. A.; Sweeney, T.; Shire, S. J.; Hsu, C. C.
Pharm. Res. 1999, 16, 249.
(7) Arshady, R. J. Controlled Release 1990, 14, 111.
(8) Carrasquillo, K. G.; Carro, J. C. A.; Alejandro, A.; Toro, D. D.;
Griebenow, K. J. Pharm. Pharmacol. 2001, 53, 115.
(9) Kelly, J. Y.; DeSimone, J. M. J. Am. Chem. Soc. 2008, 130, 5438.
(10) Wang, J.; Tian, S.; Petros, R. A.; Napier, M. E.; Desimone, J. M.
J. Am. Chem. Soc. 2010, 132, 11306.
(11) Gratton, S. E. A.; Ropp, P. A.; Pohlhaus, P. D.; Luft, J. C.;
Madden, V. J.; Napier, M. E.; DeSimone, J. M. Proc. Natl. Acad. Sci.
U.S.A. 2008, 105, 11613.
(12) Chatterjee, J.; Haik, Y.; Chen, C.-J. Colloid Polym. Sci. 2001, 279,
1073.
(13) Patil, G. V. Drug Dev. Res. 2003, 58, 219.
(14) Johnson, R. P.; Craig, S. W. J. Biol. Chem. 2000, 275, 95.
(15) Jenkins, N. Cytotechnology 2007, 53, 121.
(16) Zhang, W.; Czupryn, M. J. Biotechnol. Prog. 2002, 18, 509.
(17) Dubikovskaya, E. A.; Thorne, S. H.; Pillow, T. H.; Contag, C.
H.; Wender, P. A. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 12128.
(18) Jones, L. R.; Goun, E. A.; Shinde, R.; Rothbard, J. B.; Contag, C.
H.; Wender, P. A. J. Am. Chem. Soc. 2006, 128, 6526.
(19) Yan, M.; Du, J.; Gu, Z.; Liang, M.; Hu, Y.; Zhang, W.; Priceman,
S.; Wu, L.; Zhou, Z. H.; Liu, Z.; Segura, T.; Tang, Y.; Lu, Y. Nat.
Nanotechnol. 2010, 5, 48.
(20) Saito, G.; Swanson, J. A.; Lee, K.-D. Adv. Drug Delivery Rev.
2003, 55, 199.
(21) Namanja, H. A.; Emmert, D.; Davis, D. A.; Campos, C.; Miller,
D. S.; Hrycyna, C. A.; Chmielewski, J. J. Am. Chem. Soc. 2012, 134,
2976.
(22) Pires, M. M.; Chmielewski, J. Org. Lett. 2008, 5, 837.
(23) Li, C.; Wu, T.; Hong, C.; Zhang, G.; Liu, S. Angew. Chem., Int.
Ed. 2012, 51, 455.
D
dx.doi.org/10.1021/ja302363r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX