Nanoscale Biodegradable Organic–Inorganic Hybrids for Efficient Cell Penetration and Drug Delivery

Article abstract of DOI:10.1002/anie.201606065

We report a comprehensive study on novel, highly efficient, and biodegradable hybrid molecular transporters. To this end, we designed a series of cell-penetrating, cube-octameric silsesquioxanes (COSS), and investigated cellular uptake by confocal microscopy and flow cytometry. A COSS with dense spatial arrangement of guanidinium groups displayed fast uptake kinetics and cell permeation at nanomolar concentrations in living HeLa cells. Efficient uptake was also observed in bacteria, yeasts, and archaea. The COSS-based carrier was significantly more potent than cell-penetrating peptides (CPPs) and displayed low toxicity. It efficiently delivered a covalently attached cytotoxic drug, doxorubicin, to living tumor cells. As the uptake of fluorescently labeled carrier remained in the presence of serum, the system could be considered particularly attractive for the in vivo delivery of therapeutics.

Full text of DOI:10.1002/anie.201606065

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201606065
German Edition: DOI: 10.1002/ange.201606065
Drug Delivery
Nanoscale Biodegradable Organic–Inorganic Hybrids for Efficient Cell
Penetration and Drug Delivery
Sebastian Hçrner+, Sascha Knauer+, Christina Uth+, Marina Jçst, Volker Schmidts,
Holm Frauendorf, Christina Marie Thiele, Olga Avrutina, and Harald Kolmar*
Abstract: We report a comprehensive study on novel, highly
efficient, and biodegradable hybrid molecular transporters. To
this end, we designed a series of cell-penetrating, cube-
octameric silsesquioxanes (COSS), and investigated cellular
uptake by confocal microscopy and flow cytometry. A COSS
with dense spatial arrangement of guanidinium groups dis-
played fast uptake kinetics and cell permeation at nanomolar
concentrations in living HeLa cells. Efficient uptake was also
observed in bacteria, yeasts, and archaea. The COSS-based
carrier was significantly more potent than cell-penetrating
peptides (CPPs) and displayed low toxicity. It efficiently
delivered a covalently attached cytotoxic drug, doxorubicin, to
living tumor cells. As the uptake of fluorescently labeled carrier
remained in the presence of serum, the system could be
considered particularly attractive for the in vivo delivery of
therapeutics.
were thoroughly investigated and improved.[3] However,
several issues associated with toxicity, stability, and efficacy
of cellular uptake still require work. As the peptidic structure
of CPPs intrinsically limits the scope of improvements, recent
efforts are focused on nanoparticles or small non-peptidic
molecular scaffolds.[4] These simple, uniform molecular archi-
tectures can be easily tailored, leading to cell-penetrating
molecules with entirely new properties. In contrast to the
macromolecular delivery systems, such as (bio)polymers,
dendrimers, lipid-based or viral-like carriers, some of which
are actually on the market or under clinical trials,[4c] the next-
generation molecular transporters still require optimization.
General strategies towards the improvement of cellular
uptake include the reduction of conformational freedom by
backbone cyclization of cell-penetrating peptides or by the
usage of scaffolds which induce spatial organization of the
uptake-mediating functional groups.[3a,4d,5] Interestingly, the
proximity of the charged groups to the backbone was found to
influence the efficiency of cell uptake as well.[3c,d,5c,6]
Herein, we chose the cube-octameric silsesquioxane
scaffold (COSS) as the starting point for the development
of new-generation cell-penetrating compounds. COSS are
highly ordered organic–inorganic hybrid molecules with
a cage-like core of alternating silicon and oxygen atoms
surrounded by eight pendant organic residues. Such an
architecture with charged groups located at the flanking
arms tethered to a compact (0.7 nm)[7] core ensures a compact,
rigid, and symmetric construct. Generally, COSS are used in
certain medical fields, for example, tissue engineering, or for
the oligomerization of bioactive ligands, among them pep-
tides and carbohydrates.[7,8] They are considered non-toxic
and the hydrolytic degradation of the inorganic core under
physiological conditions has been thoroughly investigated.[7]
COSS bearing seven ammonium groups were found to
penetrate cells.[7,9] We have previously shown that these
molecules enable the delivery of a functional peptidic cargo
into living HeLa cells.[10] To improve this drug delivery
system, we synthesized a series of COSS-based molecular
transporters and investigated the uptake efficacy of a cova-
lently attached cytotoxic drug.
S
ince Linus Paulingꢀs groundbreaking publication in Sci-
ence[1] in 1949, achievements in the rapidly advancing field of
molecular medicine and related areas are very impressive.
However, while today a vast arsenal of potent and selective
drugs is available, an efficient strategy to deliver these
therapeutic compounds inside the cell, in particular, in the
cell nucleus, has become as important as the design and
optimization of the pharmacophore itself. Considering that
promising newly developed potential drug candidates, such as
peptides and proteins, are water-soluble, a bottleneck in their
application in living systems is the passage across the cellular
membrane. As a consequence, drug delivery has emerged as
one of the major fields in biomedical research. In 1994, the
first cell-penetrating peptide (CPP) penetratin was described
as a vehicle for cargo delivery into cells.[2] Since then, CPPs
[*] S. Hçrner,[+] S. Knauer,[+] C. Uth,[+] M. Jçst, Dr. O. Avrutina,
Prof. Dr. H. Kolmar
Clemens-Schçpf-Institut fꢀr Organische Chemie und Biochemie
Technische Universitꢁt Darmstadt
Alarich-Weiss-Strasse 4, 64287 Darmstadt (Germany)
E-mail: kolmar@biochemie-tud.de
Dr. V. Schmidts, Prof. Dr. C. M. Thiele
Clemens-Schçpf-Institut fꢀr Organische Chemie und Biochemie
Technische Universitꢁt Darmstadt
Compounds 2–7 were synthesized following a two-step
procedure (Scheme 1). Thus, inexpensive octaammonium
COSS hydrochloride 1 was functionalized with a) guanidi-
nium groups positively charged under physiological condi-
tions or b) permanent positive charges installed by quaternary
amines. Additionally, we investigated the influence of the
flanking armꢀs length on cellular uptake. To visualize the
constructs in cell assays, tetramethylrhodamine (TAMRA)
was attached to a single corner of COSS 1 in a stoichiometri-
Alarich-Weiss-Strasse 16, 64287 Darmstadt (Germany)
Dr. H. Frauendorf
Institut fꢀr Organische und Biomolekulare Chemie
Georg-August Universitꢁt Gçttingen
Tammannstrasse 2, 37077 Gçttingen (Germany)
[+] These authors contributed equally to this work.
Supporting information for this article can be found under:
http://dx.doi.org/10.1002/anie.201606065.
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Scheme 1. Synthesis of fluorescently labeled cell-penetrating COSS
derivatives 2–7 equipped with cationic functional groups separated
from the core by spacers of different lengths. Counterions are excluded
for clarity. Sequences of CPPs are shown in Supporting Informa-
tion 5.2.
Figure 1. Cell uptake of TAMRA-GuCOSS (4). a) Live-cell laser scan-
ning confocal microscopy imaging of HeLa cells incubated with 4 at
378C; b)–d) fluorescence microscopy imaging of cells incubated with
4; b) Gram-negative bacterium (E. coli); c) yeast (Saccharomyces cerevi-
siae), both incubated at 378C; d) archaeon (Sulfolobus islandicus)
incubated at 808C.
cally controlled reaction leading to TAMRA-aminoCOSS (2).
Subsequently, the remaining seven amine functionalities of 2
were converted into the corresponding quaternary amines by
N-methylation (TAMRA-quartCOSS 3) or guanidinylation
(TAMRA-GuCOSS 4). Alternatively, for the introduction of
a linker separating the siloxane core and the charged
elements, 4-aminobutyric, 4-trimethylaminobutyric, or 4-gua-
nidinobutyric acids were installed via amide coupling leading
to, respectively, TAMRA-aminoCOSS-L (5), TAMRA-quart-
COSS-L (6), and TAMRA-GuCOSS-L (7) (Scheme 1). To
investigate toxicity of molecular transporters we synthesized
amino-GuCOSS (8) lacking a fluorescent label (Supporting
Information 5.2). Fluorescein-TAMRA-GuCOSS (9) was
designed to assess biodegradation of the carriers (Figure 3a).
The integrity of the cage-like siloxane core was confirmed by
NMR spectroscopy (compounds 2–7; Supporting Informa-
tion 5.3).
Fluorescently labeled derivatives 2–7 were investigated
for their ability to penetrate living cells. To qualitatively
estimate the uptake, we performed live-cell imaging using
confocal laser scanning microscopy. Thus, HeLa cells were
incubated with compounds 2–7 at a concentration of 20 mm in
serum-free Dulbeccoꢀs modified eagle medium (DMEM) for
30 min. The cells were washed three times with phosphate
buffered saline (PBS) and imaged in DMEM with fetal
bovine serum (FBS). Compounds 2–4 bearing shorter linkers
(Figure 1a, Figure S1) demonstrated enhanced cellular
uptake and prominent accumulation inside the nucleus, the
nucleoli, and the cytoplasm, whereas an increase of the
spacerꢀs length led to reduced cellular uptake and primarily
cytoplasmic localization (compounds 5–7, Figure S1). In view
of predominant accumulation in the nucleus, molecular
transporters 2–4 are particularly attractive for the delivery
of drugs addressing this cellular compartment.
These observations are of particular interest in view of the
exceptional membrane composition of archaea.[11] As guani-
dinylated COSS could be of interest for the delivery of
antibiotics, we investigated its uptake in E. coli in detail
(Figure S2c).
To quantify the cellular uptake in eukaryotic cells, we
performed comprehensive flow-cytometric experiments with
COSS derivatives 2–7 and TAMRA-labeled cell-penetrating
peptides: TAT (10), penetratin (11), heptaarginine (12), and
decaarginine (13).[12] We incubated HeLa cells with these
compounds at a final concentration of 20 mm in serum-free
DMEM at 378C up to 60 min. To remove surface-bound
carrier molecules, cells were trypsinized (Supporting Infor-
mation 5.1). Assays were performed in triplicate, the results
were verified in three independent experiments (Figure S3a–
g) with carrier 4 having shown the best cellular uptake.
Indeed, the intensity of the fluorescence signal was found
155 times higher than that for the TAT peptide 10. Interest-
ingly, compound 4 carrying seven guanidinium groups dis-
played a 78-fold higher cellular uptake than heptaarginine
(12; Figure 2a). In agreement with our microscopy studies,
shorter linkers correlated with enhanced fluorescence inten-
sity (Figure S4). Similar results were obtained for HEK 293
and CHO cells (Figure S5a–c). This higher uptake of 4 can be
attributed to its more compact arrangement, hence increased
density of uptake-mediating functional groups. Indeed, cyc-
lization of CPPs, leading to more constrained and rigid
structures, is an efficient strategy to improve cellular upta-
ke.[5c]
To evaluate whether the cellular uptake is energy-
dependent, we compared the fluorescence intensity of HeLa
cells incubated with 4 at 378C and at 48C (Figure 2d). As the
uptake was only negligibly decreased at 48C, an energy-
independent mechanism was assumed. Time-resolved flow
cytometry indicated fast uptake kinetics with a first shift of
the population within an incubation time of 1 min at 378C
(Figure S6). To determine the minimal internalization thresh-
old, cells were incubated with 4 at different concentrations.
Even at the lowest concentration (80 nm) a shift in fluores-
As the guanidinylated carrier TAMRA-GuCOSS (4)
exhibited the highest uptake in HeLa cells, its ability to
penetrate cells from all three domains of life was further
studied. The microscopic images obtained suggest that it is
able to enter both eukaryotic and prokaryotic cells, among
them yeast (S. cerevisiae) and mammalian cells (HeLa), as
well as bacteria (E. coli) and archaea (S. islandicus, S. toko-
daii, Halobacterium salinarum) (Figure 1b–d, Figure S2).
2
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
toxicity of amino-GuCOSS (8) lacking a fluorescent label
(Supporting Information 5.2) in HeLa cells using an XTT cell
viability assay. Thereby the LC50 was determined to be 84 mm
(Figure 3e), which is comparable to that of polyarginines
(76 mm), TAT (86.6% viability at 50 mm), and penetratin
(88.2% at 50 mm).[17] This low toxicity may be caused by
biodegradation of the inorganic core under physiological
conditions. Indeed, the pH-dependent degradation of poly-
hedral silsesquioxanes resulting in primary siloxanes is well
established.[7]
Figure 2. Flow-cytometric experiments using HeLa cells. a) mean fluo-
rescence intensity of selected COSS derivatives and cell-penetrating
peptides. b) representative histogram of TAMRA-GuCOSS (4) and cell-
penetrating peptides (petrol-green: TAT (10); yellow-green: heptaargi-
nine (12); orange: penetratin (11), light blue: decaarginine (13); red:
TAMRA-GuCOSS (4). c) HeLa cells incubated with 4 in the presence
and the absence of 2% FBS in DMEM. d) Cellular uptake of 4 in HeLa
cells incubated in serum-free DMEM for 10 min with 20 mm 4 at 378C
or 48C.
Figure 3. GuCOSS degradation studies. a) Guanidinylated COSS con-
struct bearing the two fluorophores, fluorescein and TAMRA (Fluores-
cein-TAMRA-GuCOSS (9)). b) Decrease of the quenched fraction of
fluorescein in the in vitro degradation studies in living HeLa cells.
c) 50% hydrolysis of TAMRA-GuCOSS (4) as a function of the pH in
PBS. d) Kinetics of the degradation of 4 at pH 7.0 in PBS analyzed by
HPLC. e) XTT assay of the in vitro toxicity of guanidinylated COSS (8)
bearing no fluorescent label.
cence intensity was observed and 1 mm 4 was needed to reach
full shift of the population after an incubation time of 10 min
at 378C (Figure S7). In contrast to cell-penetrating peptides,
no minimal internalization threshold within the investigated
concentrations was observed.[13] As recent studies suggest that
both the silsesquioxane core and guanidine groups promote
clustering,[7,14] it could be supposed that high local concen-
tration of 4 (as a result of its assembly on the cell surface)
ensures cell penetration even at low concentrations. Our
observations point to an energy-independent direct cell
translocation, in accordance with the uptake mechanism of
other polyguanidines.[13]
The uptake of cell-penetrating peptides is generally
retarded in serum-containing media.[15] Impaired CPP-medi-
ated permeation efficacy has been reported for polyarginines
both in serum-containing media and in in vivo experiments,
presumably owing to their aggregation with serum proteins.
Therefore, we imaged 4 in the presence of 10% FBS in
DMEM (Figure S8) and found that 40% of the mean
fluorescence intensity was retained, compared to that for
serum-free media.[3c,15a,16] This tolerance of serum proteins,
combined with fast and effective cellular uptake, makes 4
a promising carrier for in vivo applications.
For characterization of pH-dependent degradation of 4,
we monitored decomposition of the inorganic core by RP-
HPLC. To that end, 4 was incubated in PBS at pH 6.5–9.0
(378C) and time-resolved HPLC traces were recorded within
12 h at 554 nm (absorption maximum of TAMRA). Hydrol-
ysis intermediates and the degradation products were quanti-
fied by determination of the peak areas. Although hydroxyl-
bearing intermediates were eluted earlier from the column,
the completely hydrolyzed TAMRA-decorated siloxanes had
longer retention times. Typical HPLC traces at pH 7.0 are
shown at Figure S9, and respective half-life values at pH 6.5–
9.0 in Figure 3c. Thereby, the t1/2 at pH 7.4 (PBS) was
determined to be 252 min. The degradation at neutral pH is
shown in Figure 3d.
To assess the biodegradation of the GuCOSS-based
carriers, we synthesized derivative 9 carrying two fluorescent
markers, fluorescein and TAMRA (Figure 3a). As both dyes
are attached to the same siloxane core in spatial proximity, the
fluorescence of fluorescein is quenched upon Fçrster reso-
nance energy transfer (FRET; Figure S10). As the inorganic
core loses its integrity during breakdown, either fluorescein or
TAMRA separates from the siloxane cage, and the fluores-
cence of fluorescein is restored. Along with excitation at
Since many cell-penetrating compounds were found to be
toxic above a certain concentration, we investigated the
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
488 nm the increase of the emission at 520 nm was used to
quantify the fraction of partially disassembled carriers. Based
on the fluorescence recovery of fluorescein, a half-life of
186 min was found in human serum, indicating sufficient
stability for in vivo applications (Figure S11). As HPLC
analysis allowed the erosion process to be monitored up to
the final degradation products, the half-life determined using
this approach was significantly longer (252 min at pH 7.4). In
former studies, the half-life of cell-penetrating peptides in the
presence of serum was found to be in the range of minutes
(e.g. t1/2 = 5 min for penetratin).[18] This fast decay is most
likely caused by proteolysis. It is clear that the hybrid
GuCOSS construct does not serve as a substrate for
proteolytic enzymes. The enhanced half-life of derivative 9
in serum is an additional indication that the degradation of
the COSS core is predominantly pH dependent.
Figure 4. Cell viability assay (MTT) of free doxorubicin (DOX, 14) and
the doxorubicin-GuCOSS conjugate (15). Error bars indicate standard
deviations from three independent measurements.
To better understand how the degradation proceeds in
living cells, we monitored the recovery of fluorescein
fluorescence by live-cell confocal laser scanning microscopy
upon internalization of construct 9 in HeLa cells (Figure 3b,
Figure S12). The fluorescent markers were disconnected in
half of the starting material within 149 min, and complete
disassembly (reappearance of fluorescence) occurred in 11 h.
The half-lives obtained in these experiments are on the same
order of magnitude as those observed upon incubation in
human serum (t1/2 = 186 min).
Applicability of GuCOSS as a molecular transporter was
examined upon cellular delivery of a cytotoxic cargo doxor-
ubicin (DOX, 14)—a widely applied antitumor drug.[19] Being
able to intercalate DNA, this agent induces apoptosis in
cancer cells by activating the intrinsic death pathway.[20]
Therefore, it is clear that the therapeutic effect of DOX
could be exerted only if the drug gains access to the cell
nucleus. However, because only passive diffusion ensures its
penetration into tumor tissues, the efficiency of DOX is
strongly compromised, which represents the major limitation
of this highly potent compound.[21] In a model construct, we
connected the GuCOSS delivery module to a DOX functional
cargo via a disulfide yielding the conjugate 15 (Supporting
Information 5.2). This bond is rapidly reduced in the reduc-
tive environment of cytosol, enabling drug release inside the
cell.[22]
Cellular delivery was studied in HeLa cells. First, an
incubation time-dependent cell assay with the free antibiotic
was performed (Supporting Information 4.0, 5.1). Because of
the slow uptake of free doxorubicin, the number of cancer
cells killed correlated with the duration of incubation (Fig-
ure S13). The conjugate 15 as well as the controls (free
doxorubicin (14) and untreated cells) were incubated at the
same concentrations for 1 h (Supporting Information 3.0).
Then the cells were washed with DMEM, and after 18 h an
MTTassay was performed (Figure 4). The results clearly show
that the hybrid construct 15 had an enhanced cytotoxic effect
compared to the free drug 14.
Although it is one of the cornerstones of cancer therapy,
free doxorubicin causes irreversible cardiac damage.[23]
Whereas its liposomal formulation[24] mitigates the toxic
side effects,[23] the potency is unaltered compared to the
conventional drug.[25] While the recently reported formulation
based on liposomal coencapsulation of DOX with Listerio-
lysin O enables enhancement of nuclear targeting in certain
carcinoma cell lines, it is supposed to be highly immuno-
genic.[26] Therefore, we believe that our delivery platform
combining highly efficient cell penetration with small size,
low toxicity, and biodegradability might provide clear bene-
fits. Nevertheless, further validation by animal studies is
required, which particularly addresses cardiotoxic effects in
comparison to liposomal doxorubicin formulations. It may
also be interesting to investigate whether the cellular uptake
of other cytotoxins with particularly low cell-penetrating
efficacy, such as for example, hygromycin,[27] can be enhanced
upon COSS conjugation.
To summarize, we developed new-generation hybrid cell-
penetrating compounds based on the cube-octameric silses-
quioxane scaffold. Thus, the guanidinylated fluorescent COSS
derivative was found to efficiently penetrate cells from all
three domains of life with a 155-fold enhanced, compared to
the cell-penetrating peptide TAT, cellular uptake in HeLa
cells. The carrier has fast uptake kinetics and penetrates cells
at double-digit nanomolar concentrations. It has low toxicity
and decomposes under physiological conditions within 11 h.
This novel molecular transporter retains its activity in the
presence of serum, which makes it a promising candidate for
in vivo delivery of drugs. Taking into consideration that these
organic–inorganic hybrids are very small and compact, no or
weak immune response could be assumed. We believe that
our delivery platform may enrich the toolbox of low-toxic and
highly efficient molecular systems needed for the develop-
ment of future-oriented therapeutics.
Acknowledgements
We thank Dr. Anna-Lena Krause (Nationales Centrum fꢁr
Tumorerkrankungen, DKFZ, Heidelberg, Germany) for
assistance during flow cytometry interpretation, Thomas
Pirzer for providing HeLa cells, and PD Dr. Arnulf Kletzin
(Fachbereich Biologie, Technische Universitꢂt Darmstadt)
for providing sulfobacter and archaea.
4
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
[9] C. McCusker, J. B. Carroll, V. M. Rotello, Chem. Commun. 2005,
996 – 998.
[10] S. Hçrner, S. Fabritz, H. D. Herce, O. Avrutina, C. Dietz, R. W.
Stark, M. C. Cardoso, H. Kolmar, Org. Biomol. Chem. 2013, 11,
2258 – 2265.
Keywords: cell penetration · cell-penetrating compound ·
COSS · drug delivery · silsesquioxanes
[11] M. De Rosa, A. Gambacorta, A. Gliozzi, Microbiol. Rev. 1986,
50, 70 – 80.
[1] L. Pauling, H. A. Itano, S. J. Singer, I. C. Wells, Science 1949, 110,
543 – 548.
[2] D. Derossi, M. H. Joliot, G. Chassaing, M. Prochiantz, J. Biol.
Chem. 1994, 269, 10444 – 10450.
[3] a) A. Ho, S. R. Schwarze, S. J. Mermelstein, G. Waksman, S. F.
Dowdy, Cancer Res. 2001, 61, 474 – 477; b) A. Nasrolahi Shirazi,
R. Tiwari, B. S. Chhikara, D. Mandal, K. Parang, Mol. Pharm.
2013, 10, 488 – 499; c) P. A. Wender, D. J. Mitchell, K. Pattabira-
man, E. T. Pelkey, L. Steinman, J. B. Rothbard, Proc. Natl. Acad.
Sci. USA 2000, 97, 13003 – 13008; d) P. A. Wender, J. B. Roth-
bard, T. C. Jessop, E. L. Kreider, B. L. Wylie, J. Am. Chem. Soc.
2002, 124, 13382 – 13383.
[12] a) D. Derossi, S. Calvet, A. Trembleau, A. Brunissen, G.
Chassaing, A. Prochiantz, J. Biol. Chem. 1996, 271, 18188 –
18193; b) S. Futaki, T. Suzuki, W. Ohashi, T. Yagami, S.
Tanaka, K. Ueda, Y. Sugiura, J. Biol. Chem. 2001, 276, 5836 –
5840; c) E. Vivꢃs, P. Brodin, B. Lebleu, J. Biol. Chem. 1997, 272,
16010 – 16017.
[13] F. Duchardt, M. Fotin-Mleczek, H. Schwarz, R. Fischer, R.
Brock, Traffic 2007, 8, 848 – 866.
[14] M. Vazdar, E. Wernersson, M. Khabiri, L. Cwiklik, P. Jurkiewicz,
M. Hof, E. Mann, S. Kolusheva, R. Jelinek, P. Jungwirth, J. Phys.
Chem. B 2013, 117, 11530 – 11540.
[15] a) M. Kosuge, T. Takeuchi, I. Nakase, A. T. Jones, S. Futaki,
Bioconjugate Chem. 2008, 19, 656 – 664; b) I. Nakase, Y. Konishi,
M. Ueda, H. Saji, S. Futaki, J. Controlled Release 2012, 159, 181 –
188; c) K. Park, J. Controlled Release 2012, 159, 153.
[16] A. F. Saleh, A. Arzumanov, R. Abes, D. Owen, B. Lebleu, M. J.
Gait, Bioconjugate Chem. 2010, 21, 1902 – 1911.
[17] S. W. Jones, R. Christison, K. Bundell, C. J. Voyce, S. M.
Brockbank, P. Newham, M. A. Lindsay, Br. J. Pharmacol. 2005,
145, 1093 – 1102.
[4] a) V. Bagnacani, V. Franceschi, M. Bassi, M. Lomazzi, G.
Donofrio, F. Sansone, A. Casnati, R. Ungaro, Nat. Commun.
2013, 4, 1721; b) S. Bersani, S. Salmaso, F. Mastrotto, E.
Ravazzolo, A. Semenzato, P. Caliceti, Bioconjugate Chem.
2012, 23, 1415 – 1425; c) Z. G. Chen, Trends Mol. Med. 2010,
16, 594 – 602; d) N. W. Luedtke, P. Carmichael, Y. Tor, J. Am.
Chem. Soc. 2003, 125, 12374 – 12375; e) M. Okuyama, H. Laman,
S. R. Kingsbury, C. Visintin, E. Leo, K. L. Eward, K. Stoeber, C.
Boshoff, G. H. Williams, D. L. Selwood, Nat. Methods 2007, 4,
153 – 159; f) J. Valero, M. Van Gool, R. Perez-Fernandez, P.
Castreno, J. Sanchez-Quesada, P. Prados, J. de Mendoza, Org.
Biomol. Chem. 2012, 10, 5417 – 5430.
[5] a) Y. A. Fillon, J. P. Anderson, j. Chmielewski, J. Am. Chem. Soc.
2005, 127, 11798 – 11803; b) N. P. Gabrielson, H. Lu, L. Yin, D.
Li, F. Wang, J. Cheng, Angew. Chem. Int. Ed. 2012, 51, 1143 –
1147; Angew. Chem. 2012, 124, 1169 – 1173; c) G. Lꢂttig-Tꢁnne-
mann, M. Prinz, D. Hoffmann, J. Behlke, C. Palm-Apergi, I.
Morano, H. D. Herce, M. C. Cardoso, Nat. Commun. 2011, 2,
453; d) D. Mandal, A. Nasrolahi Shirazi, K. Parang, Angew.
Chem. Int. Ed. 2011, 50, 9633 – 9637; Angew. Chem. 2011, 123,
9807 – 9811; e) A. Nasrolahi Shirazi, R. K. Tiwari, D. Oh, B.
Sullivan, K. McCaffrey, D. Mandal, K. Parang, Mol. Pharm.
2013, 10, 3137 – 3151; f) H. Tang, L. Yin, K. H. Kim, J. Cheng,
Chem. Sci. 2013, 4, 3839.
[6] K. K. Maiti, W. S. Lee, T. Takeuchi, C. Watkins, M. Fretz, D. C.
Kim, S. Futaki, A. Jones, K. T. Kim, S. K. Chung, Angew. Chem.
Int. Ed. 2007, 46, 5880 – 5884; Angew. Chem. 2007, 119, 5984 –
5988.
[7] S. Fabritz, S. Hçrner, O. Avrutina, H. Kolmar, Org. Biomol.
Chem. 2013, 11, 2224 – 2236.
[8] a) S. Fabritz, S. Hçrner, D. Konning, M. Empting, M. Reinwarth,
C. Dietz, B. Glotzbach, H. Frauendorf, H. Kolmar, O. Avrutina,
Org. Biomol. Chem. 2012, 10, 6287 – 6293; b) D. Heyl, E.
Rikowski, R. C. Hoffmann, J. J. Schneider, W. D. Fessner,
Chem. Eur. J. 2010, 16, 5544 – 5548.
[18] M. Hꢂllbrink, J. Oehlke, G. Papsdorf, M. Bienert, Biochim.
Biophys. Acta Biomembr. 2004, 1667, 222 – 228.
[19] A. M. Meredith, C. R. Dass, J. Pharm. Pharmacol. 2016, 68, 729.
[20] a) Y. Sun, X. Yan, T. Yuan, J. Liang, Y. Fan, Z. Gu, X. Zhang,
Biomaterials 2010, 31, 7124 – 7131; b) S. Wesselborg, I. H.
Engels, E. Rossmann, M. Los, K. Schulze-Osthoff, Blood 1999,
9, 3053 – 3063.
[21] A. I. Minchinton, I. F. Tannock, Nat. Rev. Cancer 2006, 6, 583 –
592.
[22] G. Saito, J. A. swanson, K. D. Lee, Adv. Drug Delivery Rev. 2003,
55, 199 – 215.
[23] T. Safra, Oncologist 2003, 8, 17 – 24.
[24] A. L. B. Seynhaeve, B. M. Dicheva, S. Hoving, G. A. Koning,
T. L. M. ten Hagen, J. Controlled Release 2013, 172, 330 – 340.
[25] a) L. Harris, G. Batist, R. Belt, D. Rovira, R. Navari, N. Azarnia,
K. Welles, E. Winer, T. D.-S. Grp, Cancer 2002, 94, 25 – 36; b) E.
Rivera, Oncologist 2003, 8, 3 – 9.
[26] a) J. A. Carrero, H. Vivanco-Cid, E. R. Unanue, Plos One 2012,
7, e32310; b) Z. F. Walls, H. Gong, R. J. Wilson, Mol. Pharm.
2016, 13, 1185 – 1190.
[27] J. C. Lacal, L. Carrasco, Antimicrob. Agents Chemother. 1983, 24,
273 – 275.
Received: June 22, 2016
Revised: September 27, 2016
Published online: && &&, &&&&
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Drug Delivery
S. Hçrner, S. Knauer, C. Uth, M. Jçst,
V. Schmidts, H. Frauendorf, C. M. Thiele,
O. Avrutina, H. Kolmar*
&&&& — &&&&
Nanoscale Biodegradable Organic–
Inorganic Hybrids for Efficient Cell
Penetration and Drug Delivery
COSS and effect: New-generation
carriers are biodegradable, low-toxic, and
show efficient uptake in living cells of all
three domains of life.
molecular transporters are based on cell-
penetrating cube-octameric silsesquiox-
anes (COSS). These nanoscale hybrid
6
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!