Non-covalent delivery of proteins into mammalian cells

Article abstract of DOI:10.1039/b809006h

Substances that mediate the import of proteins into cells, "carriers", have many potential applications. The most potentially useful carriers do not have to be covalently linked to their protein cargoes. However, a common problem with all carrier molecules is that they tend to deposit the cargo proteins into endosomes; diffuse distribution in the cytosol is the desired outcome. This paper describes the import of four different labeled (Alexa Fluor 488) proteins (avidin, recombinant streptavidin, bovine serum albumin, and β-galactosidase), with the well-known non-covalent carrier called pep-1 (also known as Chariot ), with R₈ (a molecule that is not widely appreciated to import protein cargoes via a non-covalent mode of action), and with a new molecule called azo-R₈. The data collected from fluorescence microscopy and flow cytometry indicate that all three non-covalent carriers can facilitate transport. At 37 °C, import into endocytic compartments dominates, but at 4 °C weak, diffuse fluorescence is observed in the cytosol, indicative of a favorable mode of action.

Full text of DOI:10.1039/b809006h

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Non-covalent delivery of proteins into mammalian cells†
Aurore Loudet,a Junyan Han,a Rola Barhoumi,b Jean-Philippe Pellois,c Robert C. Burghardtb and
Kevin Burgess*a
Received 29th May 2008, Accepted 19th August 2008
First published as an Advance Article on the web 30th October 2008
DOI: 10.1039/b809006h
Substances that mediate the import of proteins into cells, “carriers”, have many potential applications.
The most potentially useful carriers do not have to be covalently linked to their protein cargoes.
However, a common problem with all carrier molecules is that they tend to deposit the cargo proteins
into endosomes; diffuse distribution in the cytosol is the desired outcome. This paper describes the
R
import of four different labeled (Alexa Fluorꢀ 488) proteins (avidin, recombinant streptavidin, bovine
serum albumin, and b-galactosidase), with the well-known non-covalent carrier called pep-1 (also
known as ChariotTM ), with R8 (a molecule that is not widely appreciated to import protein cargoes via a
non-covalent mode of action), and with a new molecule called azo-R8. The data collected from
fluorescence microscopy and flow cytometry indicate that all three non-covalent carriers can facilitate
transport. At 37 ◦C, import into endocytic compartments dominates, but at 4 ◦C weak, diffuse
fluorescence is observed in the cytosol, indicative of a favorable mode of action.
et al. indicates that the macromolecular mechanism of import
Introduction
involves macropinocytosis,10 and the liberation of protein cargo–
Tat conjugates from endosomes could be achieved by adding the
N-terminal 20 amino acids of the influenza virus hemagglutinin
protein, HA-2,11 to the carrier sequence.6
The import of proteins into cells is an important problem that is
frequently encountered in many aspects of biological chemistry.
One of the best-studied approaches is to covalently attach a
peptide carrier, either chemically or genetically, to the protein
of interest. Perhaps the most commonly used carriers of this
type are the short peptide segments derived from HIV-1 Tat
and Drosophila antennapedia homeodomain proteins.1,2 Use of
the HIV-1 Tat carrier, in particular, has motivated researchers
to look for simpler peptides for intracellular delivery. In living
cells, both R8 and R16 have been reported to facilitate significant
cellular uptake, while R4 gave relatively little internalization.3,4
A major drawback to the use of these peptide delivery agents,
however, is that cargo proteins imported into the cells tend to be
concentrated in vesicular structures.5,6 These vesicles are widely
assumed to correspond to endosomes, hence the cargo proteins
are not released in the cytoplasm. This has inspired many groups
to investigate the mechanism of import, and to look for ways of
liberating the cargo proteins from the endosomes. For instance,
Wender and co-workers have suggested that the oligo-guanidine
moieties can form an ideal hydrogen bonding network with the cell
surface phosphates, and this facilitates the import on a molecular
level.7–9 Meanwhile, an elegant series of experiments by Dowdy
Genetic or chemical methods for covalently attaching carrier
peptide sequences to cargo proteins are experimentally inconve-
nient, time consuming, and restrictive with respect to the scope
of the experiments that can be performed. Conversely, carrier
vehicles that can import proteins without being covalently attached
have the potential to circumvent all these disadvantages. “Non-
covalent carriers” include “pep-1” (also known as ChariotTM),12
the cationic pyridinium amphiphile and a helper lipid, SAINT-
R
PhD,13 BPQ24 BioPORTERꢀ QuikEaseTM (a protein delivery
kit of unspecified composition),14 and a few used systems like
the peptide K16SP15 and the somewhat cytotoxic polymer,16
polyethyleneimine (PEI) (Fig. 1).17 However, the issue of whether
or not the imported cargo proteins are trapped in endosomes blurs
the true utility of these systems. It is clear from some reports in the
literature that fluorescently labeled proteins imported using these
systems become concentrated in cytoplasmic vesicles; this was our
experience in previous studies.18,19 However, some papers claim
imported fluorescently labeled proteins appear to be free in the
cytoplasm, and there have been papers wherein import of proteins
using these systems is thought to give a predictable functional
response.20
aDepartment of Chemistry, Texas A & M University, Box 30012, College
Station, TX 77841. E-mail: burgess@tamu.edu
There were no reports of simple Arg-oligomers being useful
for non-covalent import until recent work by Lee et al.21,22 They
described experiments in which high concentrations (600 mM) of
R9 mediated import of various fluorescent proteins (e.g. GFP) and
b-galactosidase into plant (onion root tip) and animal (MCF7)
cells. It was claimed that this produced diffuse fluorescence in the
cytosol and in the nucleus, however, only low resolution images
were shown, and in these cells the fluorescence appears as small
green aggregates. Furthermore, the b-galactosidase was imaged
bDepartment of Veterinary Integrative Biosciences, Texas A & M University,
College Station, TX 77843. E-mail: rburghardt@cvm.tamu.edu
cDepartment of Biochemistry and Biophysics, Texas A & M University,
College Station, TX 77841. E-mail: pellois@tamu.edu
† Electronic supplementary information (ESI) available: Synthesis of azo-
R8 and R8; delivery of BSA-F*, b-gal-F* and rec. streptavidin-F* at
4
◦C; delivery of avidin-F* at 37 ◦C mediated by R8, azo-R8 and pep-
1; R8 mediated delivery of avidin at 4 ◦C; flow cytometry data for the
cellular uptake of avidin-F*, BSA-F* and b-gal-F* at 37 ◦C. See DOI:
10.1039/b809006h
4516 | Org. Biomol. Chem., 2008, 6, 4516–4522
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Results and discussion
Azo-R8: design and synthesis
At the onset of this project, we had hypothesized that mimics of
pep-1 could be made by fusing well-known “promiscuous binders”
(molecules that seem to bind many proteins in high throughput
screens for drug leads)24,25 to a cell-penetrating warhead. The idea
was that the promiscuous binder parts of several pep-1 analogs
might non-covalently attach to the protein, coating it with entities
that promote cell penetration. Thus, an azo-compound was chosen
for the promiscuous binder part, and R8 for the cell-penetrating
unit; these were synthetically joined to give azo-R8 (Fig. 2).
However, azo-R8 was found to behave in much the same way as
R8, but with some minor differences as indicated below.
Fig. 1 Structures of the “non-covalent carriers” (A) pep-1 (also known
as ChariotTM), (B) polyethyleneimine (PEI) and (C) K16SP.
after fixing the cells, and this is known to give different results
relative to live cells.21,22
Fig. 2 Structure of azo-R8.
Data presented in this paper deal with the import of four
R
proteins labeled with Alexa Fluorꢀ 488 (F*):23 specifically, avidin,
Formation of carrier : cargo complexes
bovine serum albumin (BSA), b-galactosidase (b-gal), and a
recombinant streptavidin. These cargoes were chosen to represent
proteins with different pI values and sizes (Table 1). The potential
carriers examined were pep-1, R8, and a novel system that was
synthesized “in house”, azo-R8, all of which were not covalently
attached to the protein. The key observations are that: (i) at 37 ◦C,
avidin was imported by all three carriers (pep-1, R8, and azo-R8)
but in each case the labeled protein primarily accumulated in
vesicles that co-localized with the endosomal marker FM 4–64;
however, (ii) at 4 ◦C weak, diffuse fluorescence was observed within
the cytoplasm with little evidence of punctate vesicle formation
for all four proteins. These observations indicate a temperature
dependence of carrier-mediated protein delivery that was similar
for three chemically different carriers.
Pep-1 and azo-R8 are amphipathic peptides, while R8 is cationic.
Pep-1 is known to associate with protein cargoes through non-
covalent electrostatic or hydrophobic interactions and form stable
complexes.12,26–28 The formation of the carrier : cargo complex for
azo-R8 was easily monitored by fluorescence spectroscopy. The
azo compound can act as a quencher of the label on the protein,
proving that the two are in close contact. Fig. 3 shows that the
intense fluorescence of BSA-F* (2 mM in DMEM, Dulbecco’s
Modified Eagle’s Medium) was greatly quenched when azo-R8 was
added (1.0 : 10 mol ratio protein : carrier). In a control experiment,
a solution of fluorescein (0.1 mM) was mixed with 1 mM azo-R8
under almost identical conditions and no quenching was observed.
Delivery at 37 ◦C: uptake into punctate vesicular structures
Cellular uptake of avidin-F* into COS-7 at 37 ◦C was studied
in the first phase of this work. All three carriers, pep-1, R8,
and azo-R8, were used at 10 : 1 carrier : cargo mol : mol
ratios, and similar results were observed in all cases. Fig. 4 for
avidin uptake is illustrative. After 1 h of incubation, followed by
Table 1 Proteins studied for the (Arg)8 mediated cellular uptake
Molecular
weight/kDa
Protein
Size (a.a.)
pI (unlabeled protein)
R
a 15 min incubation period with FM 4–64, the Alexa Fluorꢀ
Avidin
BSA
66–68
66
512
583
10–10.5
4.7
488-labeled protein accumulated as green, punctate cytoplasmic
vesicles. These green vesicles were localized in endosomes as
suggested by co-loading with the endosomal marker FM 4–64
Streptavidin, rec.
b-Galactosidase
52
540
560
1171
7.4–7.7
4.8
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 4516–4522 | 4517
©

View Article Online
Fig. 3 Azo-R8 forms a non-covalent complex with BSA-F*. The fluores-
cence of BSA-F* upon excitation at 488 nm is compared to that of a 1 : 10
molar ratio protein : carrier mixture.
(Fig. 4C).29 When the cells were co-incubated with avidin-F*
and FM 4–64 for 1 h at 37 ◦C, little to no co-localization was
observed with the endosomes, as the mitochondria were labeled
under those conditions and not the endosomes (see ESI†). The
perinuclear distribution of these vesicles suggests potential sorting
and transport to Golgi and lysosomes. This was confirmed by co-
R
loading the cells with the Golgi marker BODIPYꢀ TR ceramide
complexed to BSA (Fig. 4D).23 After extended incubation times
(up to 24 h) the vesicles persisted, and no significant dispersed
fluorescence was observed in the cytosol. These observations are
consistent with the imported protein being trapped in endosomes,
even after long periods of time.
Consideration of the images from import mediated by azo-R8
shows that this consistently directs more of the labeled protein
into the cellular membrane compared with R8 and pep-1. This
observation implies some protein–azo-R8 complex may be trapped
in the membrane in these experiments. It was also observed that
when confluent cell cultures were used, more membrane staining
was observed for all three carriers, but more so for azo-R8.
Delivery at 4 ◦C: diffuse cytosolic fluorescence
Punctate vesicle formation was largely suppressed when the
experiments described in the previous section were repeated at
4 ◦C. As an added precaution against surface binding,10 the cells
were treated with heparin (3 ¥ 5 min, 1 mg mL-1 of PBS) after the
PBS washes. Nevertheless, a weak, diffuse fluorescence signal was
observed (see ESI†). Further, fluorescence deconvolution imaging
of the cells showed the same diffuse fluorescence pattern. These
observations indicate that at 4 ◦C, the protein is indeed being
imported inside the cytoplasm by all the carriers.
Experiments were performed to increase the fluorescence signal
in the cytosol by increasing the concentration of the protein from
0.5 mM to 2 and 5 mM (carrier : protein = 10 : 1 mol : mol,
as before). At 2 mM, the fluorescence intensity in the cytosol
was increased but some additional binding to the cell membrane
was observed. At 5 mM, the intensity of membrane labeling was
considerably brighter, however, the cytoplasmic signal was re-
tained (Fig. 5).
Fig. 4 Delivery of avidin-F* in COS-7 cells at 37 ◦C. (A) Non-covalent
protein delivery mediated by R8; (B) non-covalent protein delivery
mediated by azo-R8; (C) endosomal co-localization of avidin-F* and FM
4–64 (fluorescent general endosomal marker); (D) Golgi co-localization
of avidin-F* and BODIPYꢀR TR ceramide complexed to BSA (fluorescent
marker for Golgi); (E) cell autofluorescence; (F) non-mediated protein
delivery. Throughout, the carrier (5.0 mM), avidin-F* (0.5 mM) and
the COS-7 cells were incubated at 37 ◦C for 1 h; the cells were then
washed with PBS and analyzed by fluorescence microscopy. For endosomal
co-localization, the cells were incubated with FM 4–64 at 37 ◦C for 15 min,
then washed with PBS before imaging. For Golgi co-localization, the
cells were incubated for 30 min at 4 ◦C in DMEM containing 5 mM
R
BODIPYꢀ TR ceramide complexed to BSA, washed several times with
ice-cold medium and incubated in fresh medium for 30 min at 37 ◦C.
(a) Overlaid images of the avidin-F* (green) and the nuclei (blue, Hoechst
33342 marker). (b) Differential interference contrast (DIC) images.
Similar experiments on BSA-F*, b-gal-F*, and recombinant
streptavidin-F* also gave diffuse cytoplasmic fluorescence
(see ESI†). The uptake of b-gal-F* was significantly lower than
4518 | Org. Biomol. Chem., 2008, 6, 4516–4522
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Fig. 5 Delivery of avidin-F* in COS-7 cells at 4 ◦C mediated by R8.
(A) 0.5 mM of avidin–F*. (B) 2 mM avidin–F*. (C) 5 mM avidin–F*.
(D) No carrier used. Throughout, COS-7 cells were incubated for 1 h at
4 ◦C with R8 and avidin-F* (10 : 1.0 mol : mol), then washed 4 ¥ with PBS
buffer. Images shown are after deconvolution; top images are fluorescence
images of avidin-F* and Hoechst 33342, bottom images are DIC.
that of the avidin and BSA dye-conjugates, even when a
higher ratio of carrier : protein (20 : 1) was used. This difference
was confirmed in the flow cytometry experiments described in the
next section.
Comparison of uptake levels for different carriers via
flow cytometry
Flow cytometry was used to analyze and compare the non-
covalent protein transduction (Fig. 6). Before the analyses, the
cells were washed with PBS, treated with trypsin, then washed with
heparin (3 ¥ 5 min, 0.5 mg mL-1 PBS) to minimize the possibilities
for surface binding. In each case, the uptake measured by flow
cytometry was greater when the carrier molecule was included
(0.5 mM protein; carrier : protein 10 : 1.0 for avidin-F* and
BSA-F*, and 20 : 1.0 for b-gal-F*). The import of avidin-F* was
significant, with pep-1 and R8 being more effective than azo-R8 at
Fig. 6 (A) Flow cytometric analysis of the uptake of the Alexa FluorꢀR
488 labeled proteins, avidin-F*, BSA-F* and b-gal-F* at 4 ◦C relative to
FITC quantum bead standards (shaded histograms). Each histogram for
avidin-F* and BSA-F* represents 20 000 to 24 000 cells and 20 000 beads.
For b-gal-F*, each histogram represents 10 000 cells and 20 000 beads. The
FITC quantum bead peaks represent 8534, 25 857, 79 264, and 21 4887
MESF units each. The scale for the x-axis is MESF units. (B) Summary
of the flow cytometric data for each protein with the carriers tested.
Results are presented as the median MESF units for each protein–carrier
combination.
◦
4 C (Fig. 6A). However, for BSA-F* the reverse was true: azo-
R8 was far more effective than the other two carriers. The largest
protein–dye conjugate, b-gal-F*, was the least well imported of
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 4516–4522 | 4519
©

View Article Online
the three proteins that were tested, with pep-1 and azo-R8 being
the most effective carriers.
into the cytosol, prevails at 4 ◦C. This observation is parallel and
consistent with work reported by Futaki et al.5 They studied the
mechanism of translocation of R8-Texas Red at 37 and 4 ◦C,
without cargo proteins and observed vesicular staining at 37 ◦C
and diffuse fluorescence in the cytosol at 4 ◦C. The simplest
explanation for the reduced efficiency of import of b-gal-F*
relative to the other proteins is that it is approximately eight
times larger. Similarly, the simplest explanation for the observation
that three chemically different carriers facilitate import of three
Evaluation of the cytotoxicity of the carriers
Cell viability during the non-covalent protein internalization me-
diated by R8 and azo-R8 was accessed using ethidium homodimer.
Thus COS-7 cells were treated with a l.0 : 10 mol ratio protein :
carrier (2 mM protein) for 1 h at 4 ◦C, washed with PBS and treated
with ethidium homodimer (2 mg mL-1). Fig. 7A illustrates that cells
treated with R8 and azo-R8 peptides for 1 h at 4 ◦C did not result
in any cytotoxicity. Incubation at 37 ◦C for another 16 h gave no
cytotoxicity (the fluorescence was now mostly concentrated into
vesicles, probably lysosomes, Fig. 7B). After a further incubation
period of 24 h at 37 ◦C however, all the cells were dead (Fig. 7C).
◦
different proteins at 4 C via an energy independent pathway is
that they form pores in◦the cells membrane that allow leakage
into the cells, even at 4 C. This would explain the fact that the
levels of diffuse fluorescence observed are weak, and that the
larger protein was the one least effectively imported. At 37 ◦C,
other mechanisms, perhaps involving macropinocytosis (an energy
dependant process), become dominant.
Experimental
Preparation of R8 and azo-R8
See the ESI† data for protocols and characterization information.
Material and methods
R
Pep-1 was obtained from Active Motif. Avidin–Alexa Fluorꢀ
R
488 conjugate, BSA–Alexa Fluorꢀ 488 conjugate, FM 4–64,
R
BODIPYꢀ TR ceramide complexed to BSA and Hoechst 33342
were purchased from Invitrogen. b-Galactosidase and the recom-
binant streptavidin were purchased from Calbiochem and Roche,
R
respectively and labeled with Alexa Fluorꢀ 488 5 SDP ester
(purchased from Invitrogen) according to the procedure provided
by Invitrogen.
Fluorescence quenching experiment
The fluorescence intensity of a solution of BSA-F* (1 mM in
DMEM) was compared to the fluorescence intensity of a solution
of azo-R8 : BSA-F* complex (1 mM BSA-F* : 10 mM azo-R8
in DMEM). Both solution were excited at 488 nm. As a control
experiment, the fluorescence quenching of a solution of fluorescein
(0.1 mM) with azo-R8 (1 mM) was also studied.
Fig. 7 Viability assay. Delivery of avidin-F* (2 mM) in COS-7 cells at
4
◦C mediated by (a) R8 and (b) azo-R8. (A) COS-7 cells were incubated
for 1 h at 4 ◦C with R8 or azo-R8 and avidin-F* (10 : 1.0 mol : mol), then
washed 4 ¥ with PBS buffer and treated with ethidium homodimer for
30 min. (B) The same cells were incubated at 37 ◦C for 16 h. (C) The medium
was removed, fresh PBS was added and the same cells were incubated at
37 ◦C for another 24 h. Before imaging, 2 mL of ethidium homodimer were
added.
Cell culture
COS-7 cells were purchased from the American Type Culture
Collection (ATCC) and cultured on 75 cm2 culture flasks in
DMEM supplemented with 10% fetal bovine serum (FBS) in
a humidified incubator at 37 ◦C with 5% CO2. Cells grown to
subconfluence were plated 2–3 days prior to the experiments in
Lab-Tek two well chambered coverglass slides (Nunc).
Conclusions
Import into cells at 37 ◦C could be regulated via energy-dependent
or independent pathways, but at 4 ◦C it is generally accepted
that only energy independent pathways are operative. The data
Fluorescence microscopy
The subcellular protein localization was measured in living COS-
7 cells using a Zeiss Stallion dual detector imaging system
consisting of an Axiovert 200M inverted fluorescence microscope,
CoolSnap HQ digital cameras and Intelligent Imaging Innova-
tions (3I) software. Digital images of Alexa Fluorꢀ 488-tagged
proteins, FM 4–64 labeled membranes and endosomes, BODIPYꢀ
◦
accumulated here indicate that at 37 C import into endosomes
is prevalent and significant diffuse fluorescence in the cytosol is
not observed. However, the relative brightness of the vesicular
staining may obscure low level cytoplasmic fluorescence. Thus,
at least two different pathways appear to be operative for three
different carrier molecules, and the desired one, diffuse import
R
R
TR ceramide complexed to BSA labeled Golgi, and Hoechst
4520 | Org. Biomol. Chem., 2008, 6, 4516–4522
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
33342-labeled nuclear DNA were captured with a C-APO 63X/
1.2 W CORR D = 0.28M27 objective with the following filter
sets: exciter BP470/20, dichroic FT 493, emission BP 505–530 for
compromised. Viability assessment was conducted by incubating
cells at 4 ◦C for 1 h with a 1 : 10 mol ratio of protein : carrier,
followed by washing cells with PBS with Ca2+ before imaging of
proteins. Following image capture of the proteins, EthD-1 (1 mM
final concentration) was added and images were recorded at 5,
10, 20 and 30 min using the 630/75 BP filter and revealed no
EthD-1 fluorescence. Cultures were returned to the incubator at
37 ◦C for 17 h before the viability was again evaluated by EthD-1
staining. The morphology of cells monitored by DIC and Hoechst
33342 staining was normal and no EthD-1 uptake into cell nuclei
was detected. Therefore, no cytotoxic effects of protein : carrier
combinations were detected for up to 24 h of cell treatment.
R
Alexa Fluorꢀ 488; exciter BP560/40, dichroic FT 585, emission
R
BP 630/75 for FM 4–64 and BODIPYꢀ TR ceramide complexed
to BSA; and exciter G 365, dichroic FT 395, emission BP 445/50
for Hoechst 33342. Sequential optical sections (Z-stacks) from the
basal-to-apical surfaces of the cell were acquired. Digital image
acquisition was initiated approximately 1 mm below the basal
surface of the cells and optical slices were collected at 0.5 mm
steps through the apical surface of the cells with a high numerical
objective lens (C-APO 63X/1.2 W CORR D = 0.28M27).
These wide-field images were subjected to deconvolution using
3I software.
Flow cytometry
The protein : carrier complexes were pre-formed at room
temperature for 30 min by mixing (in a mol : mol ratio) the
protein and the carrier in DMEM supplemented with 10% fetal
bovine serum (FBS). A 1 : 10 mol : mol ratio of protein : carrier
was used for avidin-F*, BSA-F* and rec. streptavidin-F*, while
a 1 : 20 mol : mol ratio was used for b-gal-F*. To study the
cellular uptake of the proteins, the culture medium was removed,
the preformed protein : carrier complex was added, and the cells
were incubated for another hour at 4 or 37 ◦C. After the incubation
period, the cells were washed with phosphate-buffered saline (PBS,
pH 7.4) several times before imaging. For experiments at 4 ◦C,
cells were preincubated at 4 ◦C for 30 min before being incubated
with the protein solution for 1 h. The cells were also co-stained
with Hoescht (2 mg mL-1), a nuclear marker, and FM 4–64, an
endosome marker.
The protein internalization was measured by flow cytometry on
living cells. COS-7 cells were cultured as subconfluent monolayers
on25cm2 cellcultureplatewithventcaps inDMEMsupplemented
with 10% fetal bovine serum (FBS) in a humidified incubator at
37 ◦C with 5% CO2. Carrier–protein complexes were formed in 1
mL DMEM at a molar protein : carrier ratio of 1 : 10 for avidin
and BSA and 1 : 20 for b-galactosidase and incubated for 30 min
at room temperature. Cells grown to 60–70% confluency were then
overlaid with the preformed complexes and incubated for 1 h at
37 or 4 ◦C. For experiments at 4 ◦C, the cells were preincubated at
4 ◦C for 30 min before addition of the protein–carrier complex.
After 1 h incubation, the cells were washed with PBS, and treated
with trypsin (2 min) and heparin (0.5 mg mL-1 in PBS, 3 ¥ 5 min) to
remove extracellular bound protein. Samples were resuspended in
500 mL PBS and transferred to sterile tubes. Cells were analyzed on
a FACSCalibur (Becton Dickinson Immunocytometry Systems,
San Jose, CA) flow cytometer, equipped with a 15 mW air-cooled
argon laser, using CellQuest (Becton Dickinson) acquisition
Endosomal co-localization
COS-7 cells were incubated with the protein : carrier complex for
1 h at 37 ◦C. After the cells were washed with PBS, FM 4–64 (5◦mg
mL-1) was added and the cells were incubated for 15 min at 37 C.
The cells were washed again with PBS before imaging.
Note: when the cells were co-incubated with the labeled protein
and FM 4–64 for 1 h at 37 ◦C, predominant labeling of the
mitochondria was observed.
R
software. Green fluorescence from Alexa Fluorꢀ 488 or fluorescein
was collected through a 530/30-nm bandpass filter. List mode data
were acquired on a minimum of 10 000 cells or beads defined by
forward and side light scattering light scatter properties. Data
analysis was performed in FlowJo (Treestar, Inc., Ashland, OR),
using forward and side light scatter to gate on cells or single
beads. The Calibrated Parameter platform of FlowJo was used to
determine the molecules of equivalent soluble fluorescein (MESF).
Data are expressed as the median MESF.
Golgi co-localization
After the protein was loaded into the COS-7 cells, the cells were
washed. Then fresh DMEM medium was added, and the cells
R
were incubated with 5 mM of BODIPYꢀ TR ceramide complexed
Acknowledgements
to BSA for 30 min at 4 ◦C. The cells were then rinsed several times
with ice-cold medium and incubated in fresh medium for 30 min
at 37 ◦C. Finally, the cells were washed with PBS before imaging.
We thank the National Institutes of Health (GM72041), the
Robert A. Welch Foundation for financial support, the members
of the TAMU/LBMS-Applications Laboratory directed by Dr
Shane Tichy for assistance with mass spectrometry, and Dr. Roger
Smith (Flow Cytometry Core Laboratory, College of Veterinary
Medicine, Texas A & M University) for the flow cytometry assays.
Viability assay
Theviability of thecells was evaluated bysearchingfor anychanges
in cellular morphology using Nomarski differential interference
contrast (DIC) microscopy during and following analysis of the
cellular uptake of the proteins. Parallel cultures were also evaluated
using DIC microscopy along with fluorescence analysis of the cell-
impermeant viability indicator ethidium homodimer-1 (EthD-1)
(Invitrogen). This high-affinity nucleic acid stain is weakly flu-
orescent until bound to DNA. Unlike Hoechst 33342, EthD-1
can only penetrate cells in which the plasma membrane is
References
1 M. Magzoub and A. Graeslund, Q. Rev. Biophys., 2004, 37, 147–195.
2 H. Brooks, B. Lebleu and E. Vives, Adv. Drug Deliv. Rev., 2005, 57,
559–577.
3 S. Futaki, Adv. Drug Deliv. Rev., 2005, 57, 547–558.
4 M. Kosuge, T. Takeuchi, I. Nakase, A. T. Jones and S. Futaki,
Bioconjugate Chem., 2008, 19, 656–664.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 4516–4522 | 4521
©

View Article Online
5 I. Nakase, M. Niwa, T. Takeuchi, K. Sonomura, N. Kawabata, Y.
Koike, M. Takehashi, S. Tanaka, K. Ueda, J. C. Simpson, A. T. Jones,
Y. Sugiura and S. Futaki, Mol. Ther., 2004, 10, 1011–1022.
6 J. S. Wadia, R. V. Stan and S. F. Dowdy, Nat. Med. (N. Y.), 2004, 10,
310–315.
7 J. B. Rothbard, T. C. Jessop, R. S. Lewis, B. A. Murray and P. A.
Wender, J. Am. Chem. Soc., 2004, 126, 9506–9507.
8 J. B. Rothbard, T. C. Jessop and P. A. Wender, Adv. Drug Deliv. Rev.,
2005, 57, 495–504.
18 R. Bandichhor, A. D. Petrescu, A. Vespa, A. B. Kier, F.
Schroeder and K. Burgess, J. Am. Chem. Soc., 2006, 128, 10688–
10689.
19 R. Bandichhor, A. D. Petrescu, A. Vespa, B. Kier Ann, F. Schroeder
and K. Burgess, Bioconjugate Chem., 2006, 17, 1219–1225.
20 E. Gros, S. Deshayes, M. C. Morris, G. Aldrian-Herrada, J. Depollier,
F. Heitz and G. Divita, Biochim. Biophys. Acta, Biomembr., 2006, 1758,
384–393.
21 Y.-H. Wang, C.-P. Chen, M.-H. Chan, M. Chang, Y.-W. Hou,
H.-H. Chen, H.-R. Hsu, K. Liu and H.-J. Lee, Biochem. Biophys. Res.
Commun., 2006, 346, 758–767.
9 E. A. Goun, T. H. Pillow, L. R. Jones, J. B. Rothbard and P. A. Wender,
ChemBioChem, 2006, 7, 1497–1515.
10 I. M. Kaplan, J. S. Wadia and S. F. Dowdy, J. Controlled Release, 2005,
102, 247–253.
11 E. Wagner, C. Plank, K. Zatloukal, M. Cotten and M. L. Birnstiel,
Proc. Natl. Acad. Sci. U. S. A., 1992, 89, 7934–7938.
12 M. C. Morris, J. Depollier, J. Mery, F. Heitz and G. Divita, Nat.
Biotechnol., 2001, 19, 1173–1176.
13 www.Synvoluxproducts.com, Saint PhD protein delivery kit.
14 www.sigmaaldrich.com, Bio Porter protein delivery kit.
15 E. Mahlum, D. Mandal, C. Halder, A. Maran, M. J. Yaszemski, R.
B. Jenkins, M. E. Bolander and G. Sarkar, Anal. Biochem., 2007, 365,
215–221.
22 M. Chang, J.-C. Chou, C.-P. Chen, B. R. Liu and H.-J. Lee, New Phytol.,
2007, 174, 46–56.
23 http://probes.invitrogen.com, Invitrogen, in Molecular Probes, Invit-
rogen Corporation, 2006.
24 S. L. McGovern and B. K. Shoichet, J. Med. Chem., 2003, 46, 1478–
1483.
25 S. L. McGovern, E. Caselli, N. Grigorieff and B. K. Shoichet, J. Med.
Chem., 2002, 45, 1712–1722.
26 S. Deshayes, M. Morris, F. Heitz and G. Divita, Adv. Drug Deliv. Rev.,
2008, 60, 537–547.
27 M. C. Morris, P. Vidal, L. Chaloin, F. Heitz and G. Divita, Nucleic
Acids Res., 1997, 25, 2730–2736.
28 M. C. Morris, L. Chaloin, J. Mery, F. Heitz and G. Divita, Nucleic
Acids Res., 1999, 27, 3510–3517.
29 T. A. Vida and S. D. Emr, J. Cell Biol., 1995, 128, 779–792.
16 S. V. Vinogradov, E. V. Batrakova, S. Li and A. V. Kabanov, J. Drug
Targeting, 2004, 12, 517–526.
17 V. V. Didenko, H. Ngo and D. S. Baskin, Anal. Biochem., 2005, 344,
168–173.
4522 | Org. Biomol. Chem., 2008, 6, 4516–4522
This journal is
The Royal Society of Chemistry 2008
©