Water-soluble lipophilic MR contrast agents for cell membrane labeling

Article abstract of DOI:10.1007/s00775-015-1280-4

Long-term cell tracking using MR imaging necessitates the development of contrast agents that both label and are retained by cells. One promising strategy for long-term cell labeling is the development of lipophilic Gd(III)-based contrast agents that anch

Full text of DOI:10.1007/s00775-015-1280-4

J Biol Inorg Chem
DOI 10.1007/s00775-015-1280-4
ORIGINAL PAPER
Water‑soluble lipophilic MR contrast agents for cell membrane
labeling
Christiane E. Carney1 · Keith W. MacRenaris1 · Thomas J. Meade1
Received: 18 May 2015 / Accepted: 27 June 2015
© SBIC 2015
Abstract Long-term cell tracking using MR imaging
necessitates the development of contrast agents that both
label and are retained by cells. One promising strategy for
long-term cell labeling is the development of lipophilic
Gd(III)-based contrast agents that anchor into the cell mem-
brane. We have previously reported the efficacy of mono-
meric and multimeric lipophilic agents and showed that
the monomeric agents have improved labeling and contrast
enhancement of cell populations. Here, we report on the
synthesis, characterization, and in vitro testing of a series
of monomeric lipophilic contrast agents with varied alkyl
chain compositions. We show that these agents disperse in
water, localize to the cell membrane, and label HeLa and
MCF7 cells effectively. Additionally, these agents have up
to tenfold improved retention in cells compared to clini-
cally available ProHance®.
Introduction
Cell transplantation has been implicated as a potential ther-
apeutic strategy for the treatment of conditions such as can-
cer [1], cardiovascular [2, 3], and degenerative diseases [4,
5]. One significant challenge with cell transplantation ther-
apy is the need to monitor the location and distribution of
transplanted cells over time [6, 7]. MRI is ideally suited for
serial imaging of transplanted cells since it does not require
ionizing radiation or radiotracers like CT, PET, and SPECT.
Additionally, MRI provides high spatial and temporal reso-
lution with unlimited penetration depth. Intrinsic MR con-
trast can be enhanced using paramagnetic contrast agents
with Gd(III)-based complexes the most commonly used in
the clinic [8]. These agents generate positive (bright) image
contrast by decreasing the proton spin lattice relaxation
time (T1) of surrounding water protons [9]. The efficacy
of a contrast agent is defined by its relaxivity (r1) which
reflects its ability to shorten the T1 of water protons.
Keywords MRI · Contrast agents · Gadolinium · Cell
membrane · Cell labeling
For Gd(III)-based contrast agents to be used in longitu-
dinal cell tracking studies, new probes must be developed
that label cells with high levels of Gd(III) and remain asso-
ciated with cells for long periods of time. Agents developed
for this application have included small molecules [10],
peptides [11–13], polymers [14–16], and nanoparticles
[17]. All of these contrast agents are internalized by cells
which can result in endosomal entrapment that in turn can
lead to relaxivity quenching and contrast agent degradation
[18, 19]. This may limit the amount of time labeled cells
remain detectable by MRI.
Electronic supplementary material The online version of this
article (doi:10.1007/s00775-015-1280-4) contains supplementary
material, which is available to authorized users.
* Thomas J. Meade
tmeade@northwestern.edu
An alternate strategy for long-term cell labeling is the
development of lipophilic contrast agents that anchor
into the cell membrane without cellular internalization.
This strategy has been used for decades in optical imag-
ing with dyes such as DiO and DiA. The first reported
1
Department of Chemistry, Molecular Biosciences,
Neurobiology, Biomedical Engineering, and Radiology,
Northwestern University, 2145 Sheridan Road, Evanston, IL
60208-3113, USA
1 3

J Biol Inorg Chem
membrane-anchored MR contrast agent involved conjugat-
ing alkyl chains to a Gd(III) complex [20]. The resulting
agent labeled cells at low incubation concentrations and
showed potential for long-term imaging. Our lab continued
the development of lipophilic contrast agents by compar-
ing the labeling and retention of monomeric (contains one
Gd(III) complex) and multimeric (contains three Gd(III)
complexes) agents [21]. We showed that the monomeric
agent had the surprising ability to label cells more effec-
tively and produce more significant contrast enhancement
than the multimeric counterpart. Further, the monomeric
agent had improved solubility and did not require incuba-
tion with detergent in biological media, a limitation of the
multimeric design. Herein, we further develop the design of
monomeric lipophilic Gd(III) contrast agents that contain a
thermodynamically and kinetically stable macrocyclic che-
late. We vary the alkyl chain composition of the chelate and
investigate the effect on cell labeling and retention.
lowest concentration that caused a blue shift in emission
of the Nile red. Concentrations of 1–5 were verified using
ICP-MS.
ICP‑MS
The Gd(III) content of relaxivity solutions, logP solu-
tions, and cell suspensions was determined using ICP-MS
according to an established procedure [21].
Dynamic light scattering (DLS)
Solutions of 1–5 were prepared at 1 mM and filtered
through 0.2 μm filters into SARSTEDT clear polystyrene
10 × 10 × 45-mm cuvettes. Data were acquired on a Mal-
vern Instruments Zetasizer Nano Series Nano-ZS equipped
with Dispersion Technology Software v5.03 (Worcester-
shire, United Kingdom).
Cell culture methods
Materials and methods
Synthetic methods
HeLa (ATCC® CCL-2™) and MCF7 (ATCC® HTB-22)
were purchased from the American Type Culture Collec-
tion (Manassas, VA, USA). HeLa cells were cultured in
phenol red-free minimum essential media (MEM) supple-
mented with 10 % fetal bovine serum (FBS). MCF7 cells
were cultured in RPMI medium 1640 supplemented with
10 % FBS. Cells were plated and incubated for 24 h before
all experiments. Doses were filtered through 0.2 μm sterile
filters prior to administration. Cells were harvested using
0.25 % TrypLE.
Alkyne-modified Gd(III) chelate (8) [22] and compounds 6
[23], 9 [24], 11 [25], and 12 [25] were synthesized accord-
ing to the literature procedures. See the Supporting Infor-
mation for synthetic procedures of new compounds.
LogP measurements
Approximately 1 mg of complexes 1–5 were dissolved in
1 mL of 1:1 water:octanol. The samples were vortexed
and placed on a rotator for 14 h of mixing. Samples were
removed from the rotator and allowed to equilibrate for
12 h. An aliquot was removed from each layer and ana-
lyzed for Gd(III) content by ICP-MS. The partition coef-
ficient was calculated from the following equation: log10
P = log10(Co/Cw), Co is the concentration of Gd(III) in the
octanol layer and Cw is the concentration of Gd(III) in the
water layer.
Cell counting and viability
A Guava EasyCyte Mini Personal Cell Analyzer (EMD
Millipore, Billerica, MA) was used to count cells and
determine viability after labeling experiments. Briefly,
cells were harvested and an aliquot of the cell suspension
was mixed with Guava ViaCount reagent to reach a total
volume of 200 μL. The viability and cell count was deter-
mined using ViaCount module software. Cell viability was
confirmed using a CellTiter 96® AQueous Non-Radioactive
Cell Proliferation Assay (Promega, Madison, WI) where
cells were plated in 96-well plates at a density of 5000 cells
per well. Cells were incubated with various concentrations
of 1–5 for 24 h. The assay was then carried out according to
the manufacturer’s protocol. IC50 values were determined
using GraphPad Prism software (La Jolla, CA, USA).
Determination of CMC
A Nile red assay was used to determine CMC. Solutions
of 1–5 ranging 0–30 μM were prepared in DPBS. To
each was added 2 μL of Nile red stock solution (150 μM
in ethanol) to yield a total volume of 1 mL. Fluorescence
emission spectra were recorded using a Hitachi F-45000
Fluorescence Spectrophotometer with an excitation wave-
length of 550 nm. The excitation slit width, emission slit
width and photomultiplier voltages were 10, 10 nm, and
700 V, respectively. The CMC was determined to be the
Cellular labeling studies
Labeling studies were performed with HeLa and MCF7
cell lines with 25,000–30,000 cells plated in each well
1 3

J Biol Inorg Chem
O
O
O
O
N
of a 24-well plate. For concentration-dependent labeling
studies, complexes 1–5 were dissolved into media at con-
centrations of 0–120 μM for 1 and 3–5 while complex
2 was dissolved at concentrations of 0–40 μM (180 μL
dose). For time-dependent labeling studies, cells were
incubated with 35 μM of 1–5 for 1, 2, 4, 8, and 24 h.
Cells were harvested according to previously described
procedures [21].
O
O
N
O
N
O
R
Gd
O
N
N
H
N
Gd
N
N
O
N
N
N
N
O
O
O
H
ProHance
O
1 R =
2 R =
3 R =
Mechanism of cell labeling
Mechanism of labeling studies was performed with HeLa
cells with 35,000 cells plated in each well of a 24-well
plate. Cells were incubated with one of the following inhib-
itors or media (control): 0.00275 % poly-l-lysine, 250 μM
amiloride, 25 μM chlorpromazine, 5 μM filipin (180 μL
dose). After a 30-min incubation, 20 μL of a 10X solution
of 1–5 or ProHance was added to each well and incubated
an additional 4 h. Cells were harvested as described in the
cell labeling section. Statistical significance was deter-
mined using GraphPad Prism software.
O
O
4 R =
5 R =
O
O
Fig. 1 Lipophilic MR contrast agents 1–5 contain alkyl chains of
various lengths and branching. Complexes were synthesized with
click chemistry and contain the same Gd(III) chelate. ProHance®, a
clinically approved contrast agent, was used as a control
Cell retention
Results and discussion
Retention was determined in HeLa and MCF7 cells with
50,000–60,000 cells plated in each well of a 12-well plate.
Cells were incubated with various concentrations of 1–5
and ProHance® chosen to equalize cell labeling for 24 h
(600 μL dose). Cells were harvested according to previ-
ously described procedures [21].
Synthesis and characterization of complexes
A series of lipophilic MR contrast agents was synthesized to
contain various alkyl chains conjugated to the same Gd(III)
chelate via ‘click’ chemistry (Fig. 1). Complexes 1 and 2
were synthesized according to Scheme S1. Brominated alkyl
chains were converted to the azide and clicked to alkyne-
t
Low‑field relaxivity (r1)
modified chelate (8) in 2:1 BuOH:H2O with CuSO4 and
sodium ascorbate. The final products contained alkyl tails of
14-carbons (1) and 18-carbons (2). Complex 3 was synthe-
sized to contain an unsaturated 18-carbon tail according to
Scheme S2. Specifically, oleyl alcohol was brominated using
CBr4 and PPh3 in DCM at 0 °C. Subsequently, the prod-
uct was converted to the azide and clicked to 8 to afford 3.
Complex 4 was synthesized with a PEG spacer according
to Scheme S3. Specifically, tetraethylene glycol was mono-
tosylated and converted to the azide. The PEG spacer was
conjugated to 1-bromotetradecane using NaH and KI in DMF
Relaxivity at 1.41 T was determined using a Bruker mq60
minispec NMR spectrometer (Bruker Canada; Milton,
Ontario, Canada) using solutions of 1–5 prepared in DPBS
at concentrations of 1 mM and serially diluted four times.
Cell pellet and solution MR imaging
Cell pellet images and high-field relaxivity were deter-
mined at 7 T according to previously described methods
using a Bruker Pharmscan 7 T imaging spectrometer [21].
Briefly, a rapid-acquisition rapid-echo (RARE-VTR) T1-
map pulse sequence with static TE (11 ms), variable TR
(150, 250, 500, 750, 1000, 2000, 4000, 6000, 8000, and
10,000 ms) values, field of view (FOV) = 25 × 25 mm2,
matrix size (MTX) = 256 × 256, number of axial
slices = 4, slice thickness (SI) = 1.0 mm, and averages
(NEX) = 3 was used.
t
at 80 °C. This product was clicked to 8 in 2:1 BuOH:H2O
with CuSO4 and sodium ascorbate to afford 4. Complex 5
was synthesized according to Scheme S4. 2-Hexyldecanol
was brominated with N-bromosuccinimide, converted to the
azide, and clicked to 8 to afford the final product.
The lipophilicity of complexes 1–5 was determined by
octanol–water partition coefficient (logP) measurements
(Table 1). Complex 4 was the least lipophilic with a logP
of −0.01
0.03 while complexes 1 and 5 had similar
1 3

J Biol Inorg Chem
Table 1 Characterization
of complexes 1–5 including
logP, relaxivity, and size
measurements
LogP
r1
r1
Size (nm)
CMC (μM)
1.41 T, 37 °C
7 T, 25 °C
(mM−1 s−1
)
(mM−1 s−1
)
1
2
3
4
5
0.96 0.03
18
19
18
11
14
1
1
1
2
2
4.3 0.1
5.2 0.1
5.2 0.1
5.1 0.3
4.7 0.2
5.6 0.6
6.5 0.2
4.3 0.6
7.6 0.8
6.4 0.5
21
n/aa
5
1
1.20 0.01
1.22 0.06
−0.01 0.03
0.94 0.06
1
14
22
1
1
All measurements were made above the CMC of each complex
a
The CMC was too low to be measured with the Nile red assay (<1 μM)
logP values of 0.96 0.03 and 0.94 0.06, respectively.
The most lipophilic complexes were 2 and 3 with values
of 1.20 0.01 and 1.22 0.06, respectively. Despite the
range of logP values obtained, all complexes were dis-
persed in water without detergent, an improvement from
our previous generation of multimeric lipophilic agents
[21].
Critical micelle concentrations (CMC) were measured
using a Nile red assay where complexes 1–5 were co-incu-
bated with Nile red dye. Below the CMC of the lipophilic
contrast agent, the Nile red emission will match that of the
dye alone. Above the CMC, the emission is blue shifted
allowing for facile determination of micelle formation [26].
Complexes 2 and 3 had the lowest CMC values of 5 μM
and below while complexes 1, 4, and 5 had CMC values
between 14 and 22 μM (Table 1). Relaxivity measurements
and subsequent cell studies were performed at concentra-
tions above the CMC of each complex (except for con-
centration-dependent uptake). Relaxivity was measured at
both low (1.41 T) and high (7 T) magnetic field strengths
(Table 1; see Table S1 for r2 values). At 1.41 T, complex 4
had the lowest relaxivity of 10.5 1.6 mM−1 s−1 followed
by 5 with a relaxivity of 14.4 1.8 mM−1 s−1. Complexes
the remaining complexes can be incubated at 100–120 μM.
This is consistent with other studies that have found a cor-
relation between aliphatic tail length and cytotoxicity [27,
28]. It is likely that the high labeling of 2 perturbs the cell
membrane, and results in a lower IC50 value compared to
the other agents.
In HeLa cells, complexes 1, 3, and 4 achieve the same
labeling at all concentrations examined. Complex 5 with
the 6 and 10 carbon alkyl chains labels cells the least effec-
tively achieving a maximum of 2.6 fmol Gd(III)/cell. In
MCF7 cells, 3 labels cells with a maximum of 4.6 fmol
Gd(III)/cell followed by 1 (3.6 fmol Gd(III)/cell), 4 (2.1
fmol Gd(III)/cell), and 5 (1.7 fmol Gd(III)/cell). Overall,
these results show that 2 is the most effective complex for
cell labeling though it is limited by high cytotoxicity. Com-
plexes 1, 3, and 4 have similar labeling and lower cytotox-
icity than complex 2 making them candidates for further
evaluation. Complex 5 is the least effective for cell labeling
in both HeLa and MCF7 cell lines likely because the 6- and
10-carbon tails are too short to effectively intercalate into
the cell membrane.
Mechanism of cellular labeling
1, 2, and 3 had relaxivities of approximately 18 mM−1 s−1
.
At 7 T, relaxivities drop to 4.3–5.2 mM−1 s−1. These relax-
ivity values are attributed to aggregation as measured by
DLS. All the agents form micelles in solution ranging from
4 to 8 nm in diameter (Table 1).
The mechanism of cell labeling for complexes 1–5
was investigated in HeLa cells and compared to that of
Table 2 IC50 values for 1–5 were determined in HeLa cells using an
MTS assay
Cell labeling and toxicity of lipophilic agents
IC50 (mM)
Cellular labeling of complexes 1–5 was investigated in
HeLa and MCF7 cells with incubation concentrations that
maintained ≥90 % cell viability (see Table 2 for IC50 val-
ues) for 24 h (Figure S11). In both cell lines, complex 2
with the 18-carbon alkyl tail achieves high labeling (6 fmol
Gd(III)/cell for HeLa and 3.7 fmol Gd(III)/cell for MCF7)
at 30 μM incubations indicating that it is the most effective
complex at low concentrations (Fig. 2). The cytotoxicity
of 2 prevents incubations at higher concentrations though
1
2
3
4
5
0.159 0.022
0.076 0.008
0.176 0.012
0.136 0.03
1.22 0.003
Error bars represent the standard deviation of the mean of duplicate
experiments. The MTS value from each experiment was determined
by fitting data from cells treated with 8 concentrations of each com-
plex in triplicate (for a total of 24 wells per experiment)
1 3

J Biol Inorg Chem
Fig. 2 Concentration-dependent cell labeling of complexes 1–5 in a.
HeLa and b. MCF7 cells. These results show that 2 is the most effec-
tive complex for cell labeling at low concentrations. Complex 5 is the
least effective in both cell lines. The remaining complexes (1, 3, 4)
label cells to a similar degree. Error bars standard deviation of the
mean of triplicate experiments
Table 3 Mechanism of cellular uptake of 1–5 and ProHance® was investigated by co-incubation with various inhibitors in HeLa cells
Inhibitor
Inhibitor of
1
2
3
4
5
ProHance
Poly-l-lysine
Amiloride
Cell membrane
Macropinocytosis
Clathrin
*
*
*
***
ns
*
ns
*
ns
ns
ns
***
ns
ns
ns
**
ns
ns
ns
***
ns
ns
**
n/aa
Chlorpromazine
Filipin
ns
ns
**
**
Caveolae
ns
Temperature (4 °C)
Cell membrane fluidity
***
Data show that the lipophilic complexes label the cell membrane while ProHance® is taken up via macropinocytosis and caveolae-mediated
endocytosis. Statistical significance from controls was determined using an unpaired t test
ns Not significant
a
Cell death was observed in cells treated with 5 at 4 °C. This effect was not observed for other complexes or cells treated with 5 and incubated
at 37 °C
* P < 0.05, ** P < 0.01, *** P < 0.001
ProHance®, a clinically available contrast agent known
to be internalized by cells in cell culture (Table 3) [18].
Experiments were performed according to the literature
procedures [29]. To assess the contribution of various
endocytotic pathways to labeling, cells were treated with
the following inhibitors: poly-l-lysine (disruption of cell
membrane associations), amiloride (inhibitor of macropi-
nocytosis), chlorpromazine (inhibitor of clathrin-mediated
endocytosis), filipin (disruption of caveolae-mediated endo-
cytosis), and low temperature (disrupts energy-depend-
ent processes and reduces fluidity of cell membrane) for
30 min prior to the addition of 1–5 and ProHance®. Cells
were then incubated an additional 4 h to allow for contrast
agent labeling. The effect of the inhibitor was determined
by comparing the labeling to cells treated with 1–5 and
ProHance® alone. Poly-l-lysine and low temperature (4 °C)
were the only inhibitors to cause a statistically significant
decrease in labeling for all the lipophilic contrast agents
indicating that the complexes label the cell membrane and
are not internalized into cells. However, filipin decreased
the labeling of 5 indicating at least partial internalization
by caveolae-mediated endocytosis. Conversely, the labeling
of ProHance® was decreased by amiloride, filipin, and low
temperature. This implicates macropinocytosis and caveo-
lae-mediated endocytosis as the mechanisms responsible
for cell labeling. This result is consistent with other studies
that have found the uptake of ProHance® occurs via macro-
pinocytosis [18, 19].
MR imaging of cell pellets
To assess the ability of 1–5 to produce MR contrast
enhancement in labeled cells, T1-weighted images of HeLa
cell pellets were acquired at 7 T (Fig. 3). Complexes 2 and
3 produced the greatest contrast enhancement with a 58 %
reduction in T1 compared to untreated cells followed closely
by 1 with a 51 % reduction in T1. Complexes 4 and 5 per-
formed less effectively with a 32 and 20 % reduction in
T1, respectively. No significant enhancement was observed
in cells treated with ProHance®. Surprisingly, 3 produces
1 3

J Biol Inorg Chem
Fig. 3 T1-weighted HeLa cell pellet images acquired at 7 T of 1–5 and ProHance®. Scale bar 1 mm. Error bars standard deviation of the
mean of 4 slices. These images show that 2 and 3 produce the greatest reduction in T1
Fig. 4 Cellular proliferation and retention of 1–5 and ProHance®
in HeLa and MCF7 cells 72 h post-labeling. a Cellular proliferation
was determined by calculating the fold increase in cell count between
t = 0 and 72 h. Data show that complexes 1–5 do not slow prolif-
eration. b Cellular retention was determined by calculating the fold
decrease in Gd(III) per cell between t = 0 and 72 h. These data show
that retention is a cell line-dependent property with the lipophilic
complexes outperforming ProHance® in HeLa cells but not in MCF7
cells
the greatest image contrast despite having labeling of only
1.17 fmol Gd(III) per cell. Cells labeled with 2 have higher
Gd(III) content (2.44 fmol Gd(III) per cell) and the same T1
as cells labeled with 3 but do not produce as significant T1-
weighted contrast. This may be attributed to some T2 short-
ening due to the higher labeling. We believe that this effect
is unlikely to be the result of relaxivity changes upon inter-
calation into the cell membrane because previously lipo-
philic agents were shown to exhibit the same relaxivity in
solution and bound to nanoparticle-cell membrane mimics
(called Nanodiscs) [30]. Overall, these data show that while
all the lipophilic complexes produce significant contrast
compared to cells incubated with ProHance®, complex 3 is
the most promising for high-field imaging.
with various concentrations of the agents to equalize
Gd(III) per cell labeling. Labeled cells were re-plated in
fresh (contrast agent free) media at t = 0 and allowed to
proliferate for 72 h. Cell count and Gd(III) labeling were
determined at t = 0 and t = 72 h. Cells treated with 1–5
showed the same proliferative ability as cells treated with
ProHance® (Fig. 4a).
Cellular retention was determined by calculating the fold
decrease in cell labeling between t = 0 and 72 h. In HeLa
cells, 1–5 all have improved cellular retention compared
to ProHance® (Fig. 4b). Complexes 1, 4, and 5 have the
greatest retention with an approximately tenfold improve-
ment compared to ProHance®. Complexes 2 and 3 have a
fourfold and fivefold improvement, respectively, compared
to ProHance® indicating that the superior cell labeling
and MR contrast produced by these complexes does not
directly correlate to enhanced cellular retention. In MCF7
cells, complexes 3, 4, 5 and ProHance® have similar cellu-
lar retention. Surprisingly, the retention of 1 and 2 is lower
Cellular retention and proliferation
Cellular proliferation and retention of 1–5 and ProHance®
were determined in HeLa and MCF7 cells by incubation
1 3

J Biol Inorg Chem
than ProHance® suggesting that these complexes are not
suited for long-term cell tracking studies in all cell lines.
These data show that cellular retention is dependent
upon cell line selection. While the lipophilic complexes
provide enhanced retention in HeLa cells, there is no sig-
nificant enhancement in MCF7 cells. Additionally, com-
plexes 4 and 5 have the most consistent cellular retention in
both cell lines despite the lower labeling attained with these
agents in 24-h labeling experiments (Fig. 2). This suggests
that it may be necessary to sacrifice high cell labeling for
improved cellular retention.
Cantalapiedra A, Fernandez J, Gutierrez O, Sanchez PL, Her-
nandez C, Sanz R, Garcia-Sancho J, Sanchez A (2004) Circ Res
95;742–748
3. Semsarian C (2002) Intern Med J 32:259–265
4. Petit-Zeman S (2001) Nat Biotechnol 19:201–206
5. Marr RA, Thomas RM, Peterson DA (2010) Future Neurol
5:527–541
6. Webb GA (2006) Modern magnetic resonance. Springer,
Dordrecht
7. Rogers WJ, Meyer CH, Kramer CM (2006) Nat Clin Pract Card
3:554–562
8. Caravan P, Ellison JJ, McMurry TJ, Lauffer RB (1999) Chem
Rev 99:2293–2352
9. Merbach AE, Tóth E (2001) The chemistry of contrast agents in
medical magnetic resonance imaging. Wiley, Chichester
10. Daldrup-Link HE, Rudelius M, Metz S, Piontek G, Pichler B,
Settles M, Heinzmann U, Schlegel J, Oostendorp RAJ, Rum-
meny EJ (2004) Eur J Nucl Med Mol Imaging 31:1312–1321
11. Allen MJ, MacRenaris KW, Venkatasubramanian PN, Meade TJ
(2004) Chem Biol 11:301–307
Conclusion
We have synthesized five lipophilic Gd(III)-based contrast
agents with varied alkyl chain compositions including satu-
rated single alkyl chains (1 and 2), unsaturated alkyl chains
(3), hydrophilic PEG spacer (4), and double alkyl chains
(5). We show that each complex disperses in water, forms
micelles in solution, and labels the cell membrane in vitro.
Further, we show that there is an inverse relationship
between cell labeling and retention where the complexes
that label cells most effectively do not have enhanced cellu-
lar retention and the complexes that label cells least effec-
tively have the most enhanced cellular retention. These
experiments provide guidance for design principles of lipo-
philic Gd(III)-based MRI contrast agents. To achieve high
cell labeling, agents should be synthesized with long alkyl
chains and highly positive logP values. For improved cel-
lular retention, agents should possess short alkyl chains or
hydrophilic spacers. In future work, the mechanisms for
label dilution over time and the potential for lipid exchange
to cause cell-to-cell jumping of lipophilic contrast agents
must be investigated before these agents can be used for
long-term cell tracking applications.
12. Endres PJ, MacRenaris KW, Vogt S, Meade TJ (2008) Bioconju-
gate Chem 19:2049–2059
13. Wunderbaldinger P, Josephson L, Weissleder R (2002) Bioconju-
gate Chem 13:264–268
14. Modo M, Cash D, Mellodew K, Williams SCR, Fraser SE,
Meade TJ, Price J, Hodges H (2002) Neuroimage 17:803–811
15. Modo M, Mellodew K, Cash D, Fraser SE, Meade TJ, Price J,
Williams SCR (2004) Neuroimage 21:311–317
16. LiuY, He ZJ, Xu B, Wu QZ, Liu G, Zhu HY, Zhong Q, Deng DY,
Ai H, Yue Q, Wei Y, Jun S, Zhou GQ, Gong QY (2011) Brain
Res 1391:24–35
17. Hsiao JK, Tsai CP, Chung TH, Hung Y, Yao M, Liu HM, Mou
CY, Yang CS, Chen YC, Huang DM (2008) Small 4:1445–1452
18. Terreno E, Crich SG, Belfiore S, Biancone L, Cabella C,
Esposito G, Manazza AD, Aime S (2006) Magn Reson Med
55:491–497
19. Di Gregorio E, Gianolio E, Stefania R, Barutello G, Digilio G,
Aime S (2013) Anal Chem 85:5627–5631
20. Zheng Q, Dai HQ, Merritt ME, Malloy C, Pan CY, Li WH
(2005) J Am Chem Soc 127:16178–16188
21. Carney CE, MacRenaris KW, Mastarone DJ, Kasjanski DR,
Hung AH, Meade TJ (2014) Bioconjugate Chem 25:945–954
22. Song Y, Kohlmeir EK, Meade TJ (2008) J Am Chem Soc
130;6662–6663
23. Pajk S, Garvas M, Strancar J, Pecar S (2011) Org Biomol Chem
9:4150–4159
24. Basabe P, de Roman M, Marcos IS, Diez D, Blanco A, Bodero
O, Mollinedo F, Sierra BG, Urones JG (2010) Eur J Med Chem
45:4258–4269
25. Chalker JM, Gunnoo SB, Boutureira O, Gerstberger SC, Fernan-
dez-Gonzalez M, Bernardes GJL, Griffin L, Hailu H, Schofield
CJ, Davis BG (2011) Chem Sci 2:1666–1676
Acknowledgments This work was supported by the National Insti-
tutes of Health (NIH Grants R01EB005866 and P01HL108795) and
by a National Science Foundation Graduate Research Fellowship (C.
C.). Imaging was performed at the Northwestern University Center
for Advanced Molecular Imaging generously supported by NCI
CCSG P30 CA060553 awarded to the Robert H Lurie Comprehensive
Cancer Center. MRI was performed on the 7 T Bruker Pharmscan sys-
tem purchased with the support of NCRR 1S10RR025624-01. Metal
analysis was performed at the Northwestern University Quantitative
Bioelemental Imaging Center generously supported by NASA Ames
Research Center NNA06CB93G.
26. Kurniasih IN, Liang H, Mohr PC, Khot G, Rabe JP, Mohr A
(2015) Langmuir 31:2639–2648
27. Zheng YR, Suntharalingam K, Johnstone TC, Yoo H, Lin W,
Brooks JG, Lippard SJ (2014) J Am Chem Soc 136:8790–8798
28. Li XS, Turanek J, Knotigova P, Kudlackova H, Masek J, Parkin
S, Rankin SE, Knutson BL, Lehmler HL (2009) Colloid Surf B
73;65–74
29. New EJ, Congreve A, Parker D (2010) Chem Sci 1:111–118
30. Carney CE, Lenov IL, Baker CJ, MacRenaris KW, Ecker-
mann AL, Sligar SG, Meade TJ (2015) Bioconjugate Chem
26:899–905
References
1. Valone FH, Small E, MacKenzie M, Burch P, Lacy M, Peshwa
MV, Laus R (2001) Cancer J 7:S53–S61
2. Fernandez-Aviles F, San Roman JA, Garcia-Frade J, Fernan-
dez ME, Penarrubia MJ, de la Fuente L, Gomez-Bueno M,
1 3