Host-rotaxanes with oligomeric axles are intracellular transport agents

Article abstract of DOI:10.1016/j.bmcl.2008.11.053

Polymeric macromolecules are promising drug delivery devices with endocytotic properties that need to be resolved. Host-rotaxanes (HRs) also deliver materials into cells but require improved in vivo targeting capacity. Combining the targeting properties o

Full text of DOI:10.1016/j.bmcl.2008.11.053

Bioorganic & Medicinal Chemistry Letters 19 (2009) 520–523
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Host-rotaxanes with oligomeric axles are intracellular transport agents
Jing Zhu a, Molly McFarland-Mancini b, Angela F. Drew b, David B. Smithrud a,
*
a Department of Chemistry, University of Cincinnati, PO Box 210172, Cincinnati, OH 45221, USA
b Department of Cancer and Cell Biology, University of Cincinnati, Cincinnati, OH 45267, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Polymeric macromolecules are promising drug delivery devices with endocytotic properties that need to
be resolved. Host-rotaxanes (HRs) also deliver materials into cells but require improved in vivo targeting
capacity. Combining the targeting properties of nanoparticles with the transport function of HRs may
improve drug efficacy. Our prototype HR (HR 1) has a short axle and is an efficient transporter. Here,
we have constructed HRs that contain an oligo(ethylene glycol) (HR 2) or an oligoalkyl (HR 3) axle with
the future goal of combining them with nanoparticles. HR 2 more efficiently delivers Fl-peptides into
ovarian cancer cells than HR 3 and, in most cases, than HR 1. HR 2 appears to possess the appropriate
balance between water solubility and lipophilicity to be an efficient transporter along with a suitable
structure for incorporation into a larger nanoparticle.
Received 12 October 2008
Revised 6 November 2008
Accepted 10 November 2008
Available online 20 November 2008
Keywords:
Transporters
Intracellular delivery
Host–guest chemistry
Published by Elsevier Ltd.
The cellular membrane blocks the passage of most materials
into cells. Although necessary, this barrier severely limits the
development of drugs and therapeutic methods. The challenge is
to develop materials that selectively enter targeted cells without
compromising their membrane and inducing cell death. We have
developed a new class of intracellular transport agents, for exam-
ple, host-rotaxane 1 (HR 1), that have low toxicities in the concen-
tration range used for delivery and apparently pass through cell
membranes unaided.1 Host-rotaxanes are compounds with an
interlocked wheel and axle with a blocking group and a synthetic
host on the ends of the axle to keep the wheel threaded
(Fig. 1).2,3 Even though HRs are promising delivery devices, they
lack a targeting mechanism for a particular cell type and so far
have not been tested extensively in animals.
covalent incorporation with polymers. This chain was chosen since
polymeric ethers tend to be water-soluble, nontoxic, and nonim-
munogenic.13 Oligo(ethylene glycol) segments have been grafted
onto long alkyl chains14,15 or attached to the surface of a nanopar-
ticle16 and can be designed to release their payload at targeted
sites.17–20 The rest of the rotaxane was not changed to maintain
efficient transport properties, including a cleft containing phenolic
rings, which is used to house an aromatic guest, and a wheel con-
taining arginine groups to promote cell surface binding.
The presence of an oligo(ethylene glycol) chain could greatly
affect intracellular delivery. HR 2 when fully extended (4 nm) would
span across half of a membrane (membrane thickness 6–9 nm). The
full length of transporter HR 1 is approximately 2.5 nm. Long
rotaxanes may undergo drastic conformational changes or become
To provide a targeting mechanism and promote biostability, we
proposed that the HRs can be combined with polymeric nanopar-
ticles. Nanosized polymers can traverse many of the barriers
imposed in living systems while protecting and improving the
pharmacokinetic profile of an encapsulated drug. Although nano-
particles perform well in delivering drugs to cells, endolysosomal
trafficking of the materials can degrade their payload. Cell-pene-
trating peptides (CPPs) have been combined with drug delivery
materials to provide an alternative route for cellular entry.4–8
Unfortunately, CPPs appear to enter cells via endocytosis.9–12 HRs
combined with polymeric nanoparticles may result in a more effi-
cient delivery method.
We created a host-rotaxane with an oligo(ethylene glycol) chain
(HR 2) to better equip the host-rotaxanes for noncovalent or
* Corresponding author.
E-mail address: david.smithrud@uc.edu (D.B. Smithrud).
Figure 1. Host-rotaxanes used in this study and their components.
0960-894X/$ - see front matter Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2008.11.053

J. Zhu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 520–523
521
permanently embedded within a membrane and halt transport.
Additionally, changes made to the structure of the axle may affect
properties such as permeability, guest recognition, or solubility of
a rotaxane.
be diminished. We determined the binding enthalpy and entropy
to explore the relationship between the wheel and pocket that ex-
ists in the host–guest complexes.24
The thermodynamic parameters (see Supporting Information)
for the HRs bound to fluorescein and to a fluoresceinated peptide,
Fl-AVWAL, in buffered water (PBS, pH 7.3) and DMSO were deter-
mined (Table 1). These guests were chosen on the basis of their
commonality with potential drugs and their extensive character-
ization in previous rotaxane studies.23,24 The temperature depen-
dence on the free energy was used to derive the binding
enthalpies, according to the van’t Hoff relationship. Knowing
We were most concerned with the increase in the hydrophilic-
ity of the rotaxane, which could decrease the ability of the HR to
penetrate cellular membranes. A more lipophilic variant of HR 2
was created (HR 3) to improve the lipophilicity of long axle rotax-
anes if needed. Its axle is made of repeating –CH2– groups, which
increase the membrane permeability of HR 3, but decrease its
water solubility. In this study, we show that HRs 2 and 3 bind
and transport fluoresceinated guests into ovarian cancer cells. Sur-
prisingly, HR 2 outperformed HR 1 and HR 3 for most guests in
transport function; a process that was independent of endocytosis.
Combining HRs with nanoparticles could prove to be a way to
overcome the problem of endocytosis associated with the delivery
of drugs by nanoparticles.
The synthesis of HR 2 and 3 started with the aromatic cleft com-
ponent that was used in the synthesis of HR 1 (Scheme 1). It was
oxidized to an acid to provide an attachment site for the oligoalkyl
or oligo(ethylene glycol) strand. Synthesis of the oligo(ethylene
glycol) strand followed the procedure of Martell.21 The oligoalkyl
strand was made from undecanedioic acid, which was amidated
and then reduced. Once one amine of a strand was coupled to
the cleft, the other end was exposed to a DCC–rotaxane.22 DMSO
was needed in the coupling reaction of the oligo(ethylene glycol)
strand to improve its solubility. The amount of rotaxane formed
with the oligoalkyl strand is lower (42%) than the yield usually ob-
tained for rotaxane formation (around 70%). The wheels were
deprotected and attached to Boc protected arginines, which were
subsequently deprotected to produce the rotaxanes.
We have previously shown that efficient transport requires
movement of the wheel along the axis. 2D NMR analysis demon-
strated that dominant conformations exist for HR 1 bound to fluo-
rescein in polar and apolar environments with the wheel residing
near and far from the cleft pocket, respectively.1 Additionally, we
found that the likelihood of an HR to deliver a compound into cells
is proportional to the strength of the complex that is formed
between them, especially in DMSO.23 Changing the length or the
constituents of an axle could alter the spatial relationship between
the wheel and pocket. This could result in restricted motion of the
wheel or weak host–guest complexes. In either case, delivery could
D
G0s and H0s, the binding entropies were obtained by solving
D
the Gibbs free energy equation. A similar magnitude is seen in
the binding free energies for the complexes formed between HRs
1–3 and the guests. The highly stable complexes in DMSO indi-
cated that HRs 2 and 3 would be transporters. The results also
show that the strength of a complex is insensitive to length and
nature of the axle, at least for these axles.
On the other hand, the length and constituents of an axle
greatly affects the magnitude of the enthalpy and entropy of bind-
ing. A wide rangÀe1of values is seen for
D
H0 and
S0 = 56 J molÀ1 KÀ1
and in DMSO
dD
S0 = 82 J molÀ1 KÀ1). The long axle
D
S0 in water
(dD
H0 = 12 kJ mol and
H0 = 23 kJ molÀ1 and
dD
)
(dD
rotaxanes HR 2 and 3 bind fluorescein in water with greater favor-
Table 1
Thermodynamic energies for HR complexesa
HR
Guest
Waterb
H0
DMSO
H0
D
G0
D
D
S0
D
G0
D
D
S0
1c
2
Fl
À27
À30
À28
À22
À26
À24
À13
À6
47
81
80
31
76
25
À33
À33
À31
À32
À31
À31
À11
À11
À33
À26
À10
À16
74
74
À8
18
71
51
Fl-AVWAL
Fl
Fl-AVWAL
Fl
À5
À12
À4
3
Fl-AVWAL
À16d
a
D
G
0 = ÀRTlnK, K from fluorescence quenching assays, uncertainty in K’s 6 5%,
D
G0 calculated for 25 °C, kJ molÀ1
uncertainty in
;
D
H0 kJ molÀ1
, from van’t Hoff’s analysis,
D
H
0 < 10%;
DS
0 < 10%.
D
S
0 = (
D
G0
À
D ,
H0)/T, calculated for 25 °C, J molÀ1 KÀ1
uncertainty in
b
1 mM phosphate pH 7.4.
Ref. 24.
From the integrated van’t Hoff equation.
c
d
Scheme 1. Reagents and conditions: (a) Na2Cr2O7, H2SO4, H2O (88%); (b) i—ClCOCOCl, Et3N, CH2Cl2; ii—H2N(CH2)11NH2, DMSO or H2N(CH2CH2O)3CH2CH2NH2, CH2Cl2 (33%,
73%); (c) i—CHCl3 (65%, 42%); ii—TFA, CH2Cl2 (96%, 84%); (d) i—(Boc)3ArgOH, HOBt, DCC (68%, 56%); ii—TFA, CH2Cl2 (96%, 90%).

522
J. Zhu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 520–523
100
80
60
40
20
0
60
room temp.
low temp.
low ATP
HR 1
50
HR 2
40
30
20
10
0
HR 3
Fl-AVWAL Fl-KKALR Fl-AQSAV Fl-AQEAV
Fl-KKALR
Fl-AQSAV
Fl-AQEAV
Figure 2. Delivery of Fl-peptides into CaOV3 cells by the HRs (standard deviations
less than 10%).
Figure 3. The delivery of Fl-peptides by HR 2 at different conditions to explore the
transport mechanism (standard deviations less than 10%).
able binding entropy than HR 1. This could be the result of a great-
er number of water molecules being released from the axle as the
wheel slides a greater distance. On the other hand, the binding of
the larger Fl-AVWAL by HR 2 and 3 is more enthalpically driven
than for its complex with HR 1 in water. In DMSO, HR 1 and HR
3, which have axles containing alkyl groups, bind the guests with
repeated at 4 °C or by using an established ATP-depleting cocktail
of 2-deoxyglucose and NaN3 to deplete the cellular energy.12
Fl-AVWAL was omitted as we have previously observed apparent
precipitation in the assay solutions at 4 °C, most likely caused by
the low solubility of the HR–Fl-AVWAL complex. Similar levels of
transport were observed for each Fl-peptide under the energy-de-
pleted conditions and in the physiologically relevant solution
(Fig. 3). High levels of viable cells were measured in the assay
performed at 4 °C and with depleted ATP (90–98% live cells, as
indicated by calcein blue AM, and less than 7% dead cells, as indi-
cated by PI). A larger number of dead cells were observed with a
depleted level of ATP than at 4 °C. The flow cytometric results show
that endocytosis is not the major pathway for the delivery of mate-
rials into cells. A cell-passive, rotaxane-dependent mechanism is
more likely followed.
similar energies and
D D
H0 and S0 both contribute to complex sta-
bility. The HR 2 complexes are driven through a favorable change
in enthalpy. Thus, the source of the free energy does depend on
the length or nature of the axle. A
for these complexes, which lessens the differences in the free
energies.
D
H0 À
D
S0 compensation exists
The ability of the HRs to transport materials into CaOV3 cells
was investigated (see Supporting Information). Fluorescein was
not used as a guest due to a high background level of fluorescence
at the concentration needed for efficient delivery. Flow cytometry
was used to quantify the relative level of Fl-AVWAL within the cells.
The background level of fluorescence was set at 5%. Fl-AVWAL
In summary, we found that the length or nature of an HR axle
can increase or diminish the amount of a Fl-peptide that is deliv-
ered into cells. The results emphasize the importance of water sol-
ubility and lipophilicity, which are both needed by a host-rotaxane
to be a transporter. HR 2, which has a oligo(ethylene glycol) as the
axle, performed the best overall. It is highly soluble in water, forms
strong complexes with Fl-peptides, and efficiently transports
(10 lM) was delivered to a similarly high degree by HR 1 and HR
2 and approximately 50% less by HR 3 (Fig. 2). Since HR 3 is the least
soluble in water of the HRs studied, the poorer performance could
be caused by HR 3 or the HR 3ÁFl-AVWAL complex precipitating
during the assay. To enhance the water solubility of the HR com-
plexes, the highly water-soluble Fl-KKALR, Fl-AQSAV, and Fl-AQEAV
highly charged Fl-peptides into cells. A wide range of
S0 values is seen for the HR–guest complexes. Notwithstanding
the limited data set, these values do not correlate with transport
efficiency, leaving
G0 as the best indicator of transport.23 More
D
H0 and
D
were tested as guests at 15 lM. The observed trend in delivery is
consistent with previous studies of HRs.25 HR 2 outperformed HR
1 in the delivery of these Fl-peptides, whereas HR 3 still delivered
them less efficiently (Fig. 2). Since the multiple –CH2– groups of
the axle should enhance the membrane permeability of HR 3, its
poor water solubility most likely causes its less efficient transport.
The cellular assay solutions also contained calcein blue AM and
propidium iodide (PI) to determine the number of viable and dead
cells, respectively. A high proportion of calcein blue positive cells
(90–98%) was observed in the assays, and less than 3% of the cells
were dead, according to the level of PI within the cells. Calcein
blue/PI and Fl-peptide fluorescence were independent variables.
Membrane integrity was verified by measuring the amount of en-
zyme released from the cells during the assay. The level of lactate
dehydrogenase (CytoTox-One Integrity Assay, Promega) released
into the solution was 7–13% for cells exposed to the various re-
agents and 6% for untreated cells. These results demonstrate that
the HRs and the Fl-peptides are minimally or not toxic at these
concentrations.
D
importantly, we found that HRs can be modified to enable nano-
particle attachment without interfering with efficient intracellular
transport. Our ongoing studies focus on the mechanism of trans-
port incorporation of long axle HRs with nanoparticles to improve
the delivery of materials into cells.
Acknowledgments
The work was supported by the National Science Foundation
under Grant No. CHE-0400539 and the American Cancer Society,
Ohio Division, Inc. (AFD).
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2008.11.053.
Polymeric nanoparticles, as discussed above, and some highly
argininated peptides enter cells through endocytosis. We previ-
ously observed that HR 1 delivers materials into cells in an energy
independent process.1 To determine whether endocytosis is the
major pathway for cellular entry for HRs 2 and 3, the assays were
References and notes
1. Bao, X.; Isaacsohn, I.; Drew, A. F.; Smithrud, D. B. J. Am. Chem. Soc. 2006, 128,
12229.

J. Zhu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 520–523
523
2. Sauvage, J.-P.; Dietrich-Buchecker, C. O., Eds.; In Molecular Catenanes, Rotaxanes
14. Lutz, J. F. J. Polym. Sci. A Polym. Chem. 2008, 46, 3459.
15. Lutz, J. F.; Borner, H. G. Prog. Polym. Sci. 2008, 33, 1.
16. Hong, R.; Fischer, N. O.; Verma, A.; Goodman, C. M.; Emrick, T.; Rotello, V. M. J.
Am. Chem. Soc. 2004, 126, 739.
17. Luten, J.; van Nostruin, C. F.; De Smedt, S. C.; Hennink, W. E. J. Control. Release
2008, 126, 97.
18. Panyam, J.; Labhasetwar, V. Adv. Drug Deliv. Rev. 2003, 55, 329.
19. Hans, M. L.; Lowman, A. M. Curr. Opin. Solid State Mater. Sci. 2002, 6, 319.
20. Soppimath, K. S.; Aminabhavi, T. M.; Kulkarni, A. R.; Rudzinski, W. E. J. Control.
Release 2001, 70, 1.
21. Sun, Y. Z.; Martell, A. E.; Motekaitis, R. J. Inorg. Chem. 1985, 24, 4343.
22. Zehnder, D. W.; Smithrud, D. B. Org. Lett. 2001, 3, 2485.
23. Bao, X.; Isaacsohn, I.; Drew, A. F.; Smithrud, D. B. J. Org. Chem. 2007, 72,
3988.
24. Zhu, J.; Smithrud, D. B. Org. Biomol. Chem. 2007, 5, 2992.
25. Wang, X.; Bao, X.; McFarland-Mancini, M.; Isaacsohn, I.; Drew, A. F.; Smithrud,
D. B. J. Am. Chem. 2007, 129, 7284.
and Knots: A Journey Through the World of Molecular Topology; Wiley-VCH:
Weinheim, Germany, 1999.
3. Smukste, I.; House, B. E.; Smithrud, D. B. J. Org. Chem. 2003, 68, 2559.
4. Deshayes, S.; Morris, M.; Heitz, F.; Divita, G. Adv. Drug Deliv. Rev. 2008, 60, 537.
5. Moon, C.; Kwon, Y. M.; Lee, W. K.; Park, Y. J.; Yang, V. C. J. Control. Release 2007,
124, 43.
6. Juliano, R. Biochem. Soc. Trans. 2007, 35, 41.
7. Gupta, B.; Levchenko, T. S.; Torchilin, V. P. Adv. Drug Deliv. Rev. 2005, 57, 637.
8. Wadia, J. S.; Dowdy, S. F. Curr. Opin. Biotechnol. 2002, 13, 52.
9. Duchardt, F.; Fotin-Mleczek, M.; Schwarz, H.; Fischer, R.; Brock, R. Traffic 2007,
8, 848.
10. Futaki, S. Adv. Drug Deliv. Rev. 2005, 57, 547.
11. Fuchs, S. M.; Raines, R. Biochemistry 2004, 43, 2438.
12. Richard, J. P.; Melikov, K.; Vives, E.; Ramos, C.; Verbeure, B.; Gait, M. J.; Leblue,
B. J. Biol. Chem. 2003, 278, 585.
13. Greenwald, R. B. J. Control. Release 2001, 74, 159.