Full text of DOI:10.1007/BF02879808

Vol. 44 No. 4
SCIENCE IN CHINA (Series B)
August 2001
Liver-targeting macromolecular MRI contrast agents
YAN Guoping (鄢国平)1, ZHUO Renxi (卓仁禧)1, XU Mianyi (徐勉懿)1,
LI Liyun (李丽云)2 & TANG Yuefeng (唐岳枫)2
1. Laboratory of Biomedical Polymer Materials of the Ministry of Education, Department of Chemistry, Wuhan Univer-
sity, Wuhan 430072, China;
2. State Key Laboratory of Magnetic Resonance and Atomic and Molecular Physics, Wuhan Institute of Physics and
Mathematics, Chinese Academy of Sciences, Wuhan 430071, China
Correspondence should be addressed to Zhuo Renxi (email: bmp@whu.edu.cn)
Received March 12, 2001
Abstract Macromolecular ligands with liver-targeting group (pyridoxamine, PM) PHEA-DTPA-PM
and PAEA-DTPA-PM were prepared by the incorporation of different amount of diethylenetria-
minepentaacetic acid monopyridoxamine group (DTPA-PM) into poly-α, β-[N-(2-hydroxyethyl)-L-
aspartamide] (PHEA) and poly-α, β-[N-(2-aminoethyl)-L-aspartamide] (PAEA). The macromolecu-
lar ligands thus obtained were further complexed with gadolinium chloride to give macromolecular
MRI contrast agents with different Gd(Ⅲ) contents. These macromolecular ligands and their gado-
1
linium complexes were characterized by H NMR, IR, UV and elementary analysis. Relaxivity
studies showed that these polyaspartamide gadolinium complexes possess higher relaxation ef-
fectiveness than that of the clinically used Gd-DTPA. Magnetic resonance imaging of the liver in
rats and experimental data of biodistribution in mice indicate that these macromolecular MRI con-
trast agents containing pyridoxamine exhibit liver-targeting property.
Keywords: polyaspartamide, gadolinium complexes, magnetic resonance imaging (MRI), relaxivity, liver-targeting.
Macromolecular magnetic resonance imaging (MRI) contrast agent can be prepared by the
incorporation of a paramagnetic metal chelated complex to a macromolecule. The macromolecular
paramagnetic metal complex thus obtained usually exhibits more effective relaxation than that of
the metal complex alone since an increase in rotational correlation time can improve relaxivity. In
addition, when an organ-targeting group is attached to this macromolecular metal complex, it
could be endowed with organ-targeting property. On the other hand, macromolecular MRI contrast
agents may show prolonged intravascular retention due to its bulky molecular volume, and it can
be used clinically as a blood pool contrast agent[1, 2]
.
Polyaspartamide is a water-soluble, biologically well tolerated synthetic polymer that has
been proposed as a plasma extender and a drug carrier. It is nontoxic, nonantigenic and biode-
gradable in living systems, and can be further modified via the reactive pendant group[3].
In this work, pyridoxamine (PM) was chosen as a liver-targeting group[4]. PM-containing
DTPA mono(N-hydroxysuccinimide) ester (SuO-DTPA-PM) was prepared by reacting PM with
DTPA bis(N-hydroxysuccinimide) ester. The PM-containing DTPA active ester thus obtained was
further incorporated into poly-α, β-[N-(2-hydroxyethy1)-L-aspartamide] (PHEA) and poly-α, β-

No. 4
LIVER-TARGETING MACROMOLECULAR MRI CONTRAST AGENTS
345
[N-(2-aminoethy1)-L-aspartamide] (PAEA) to give liver-targeting macromolecular ligands PHEA-
DTPA-PM and PAEA-DTPA-PM. Finally, by the metalation of the ligands with Gd(III), two kinds
of macromolecular gadolinium complexes were synthesized. The synthetic process is shown as
scheme 1.
Scheme 1. Synthetic route and molecular structures for MRI contrast agents.
1
Experimental
1.1 Instruments and reagents
1
IR spectra were recorded on a Nicolet-170SX FT-IR spectrophotometer. H NMR data were
obtained on a JEOL FX-90Q NMR (90 MHz) spectrometer. The elementary analysis was per-
formed on a Carlo Erba 1106 analyzer. The concentrations of the paramagnetic species [Gd3+]
were measured by ICP Atomscan-2000 spectrometer. UV spectra were recorded on a Shimadzu
UV-240 spectrophotometer. The molecular weight and polydispersity of poly(amino acid) were
measured by GPC (waters 2960D separation module, ultrahydrogel 2000 column, poly(ethylene

346
SCIENCE IN CHINA (Series B)
Vol. 44
glycol ) standard, 0.1 mol·L−1 KH2PO4 solvent, 0.500 mL·min−1 flow, 35℃ column tempera-
ture and 30℃ detector temperature ). Biodistribution in mice was reported by an SN-695 gamma
counter RIA program Ver 5.33 and Kunming mice weighing (20±2) g were used. MR images
were performed using a BIOSPEC 47/30 system (Bruker Instruments, 4.7 T) and SD rats weigh-
ing 150 g were used. Longitudinal relaxation time (T1) for paramagnetic metal complexes was
determined by a Bruker WP-80SY NMR (80 MHz，1.88 T) spectrometer. The optical densities
(OD570) were measured with a DG-3022A ELISA-Reader. HeLa cells and L-02 cells were pro-
vided by China Center for Type Culture Collection of Wuhan University.
L-aspartic acid, N-hydroxysuccinimide and poly-L-lysine (PLys, Mw: 3×104—7×104) were
biochemical reagents. Diethylenetriaminepentaacetic acid, ethylenediamine, aminoethanol and
gadolinium chloride were analytical reagents. The growth medium was the RPMI-1640 media
(10% fetal bovine serum (Gibco. Co. USA), 100 units/mL penicillium, 100 μg/mL streptomycin).
DTPA bis(N-hydroxysuccinimide) ester (SuO-DTPA-OSu) was prepared according to ref. [5].
Pyridoxamine (PM) was prepared according to ref. [6]. All the other chemicals and solvents were
analytical reagents.
1.2 Synthesis of macromolecular gadolinium complexes
1.2.1 Preparation of PHEA and PAEA. Polysuccinimide (PSI), poly-α, β-[N-(2-hydroxy-
ethy1)-L-aspartamide] (PHEA) and poly-α, β-[N-(2-aminoethy1)-L-aspartamide] (PAEA) were
synthesized according to ref. [7]. The molecular weight and polydispersity of poly(amino acid) are
shown in table 1.
Table 1 Mn, Mw and polydispersity of poly(amino acid)
Mn ×104
2.61
Mw×104
3.77
Poly(amino acid)
PHEA
Polydispersity
1.44
PAEA
1.16
1.91
1.65
1.2.2 Synthesis of SuO-DTPA-PM. 0.57 g (3.4 mmol) of pyridoxamine (PM) in 30 mL of
DMF-H2O (v/v=3︰5) was added slowly to a solution of SuO-DTPA-OSu (2.0 g, 3.4 mmol) in
DMF (20 mL) with rapidly stirring at −15—−20℃. The reaction mixture was stirred for 4 h at
the same temperature and for 6 h at 20℃, then precipitated by pouring into 600 mL ethanol-ether
(v/v=1/6) and filtered. The precipitate was dried for 2 h at room temperature under vacuum to
yield 1.74 g (80%) product, m. p. 103—104℃.
IR(KBr,cm−1): 3400(OH), 3062(C＝C), 2962(CH2CH2), 1725(COOH), 1656(CONH), 1390
1
(
N ). H NMR (DMSO-d6, δ ): 8.5 (s, 1H), 4.5(s, 2H), 3.9—4.2 (s, 2H), 3.5 (m, 8H), 3.0 (s,
3H), 2.85 (s, 6H),2.7 (s, 4H), 2.3—2.5 (t, 4H). Anal: calcd. for (C26H36N6O13)(%): C 48.75, H 5.63,
N 13.13; found(%): C 48.41, H 5.66, N 13.03.
1.2.3 Synthesis of macromolecular ligands PHEA-DTPA-PM(L1 4). 0.20 g (0.316 mmol) of
—

No. 4
LIVER-TARGETING MACROMOLECULAR MRI CONTRAST AGENTS
347
SuO-DTPA-PM (mole ratio of SuO-DTPA-PM to PHEA repeat units:10%) was dissolved in DMF
(20 mL), and then added into a PHEA (0.5g, 3.16 mmol) in DMF (20 mL) solution with rapidly
stirring at 20℃. The reaction mixture was stirred for 48 h at the same temperature and precipitated
by pouring into ethanol (300 mL). After filtration, the precipitate was recrystallized from
DMF-ethanol (v/v=1/6) and dried for 6 h at 60℃ under vacuum to give product PHEA-DTPA-PM
(L1)(0.36 g, 54.5%).
IR(KBr, cm−1): 3316(OH), 3069(C＝C), 1662(COOH), 1645, 1547(CONH). 1H NMR (D2O,
δ ): 7.9 (
), 3.9 (CCH ), 3.6—3.3 (CH CH ), 3.0 (CH ), 2.82 (NCH CO). UV(H O, nm):
2 2 2 3 2 2
H
N
282. PHEA-DTPA-PM(L2 4)(mole ratio of SuO-DTPA-PM to PHEA repeat units: 20%, 60%,
—
100%) were similarly synthesized.
1.2.4 Synthesis of macromolecular ligands PAEA-DTPA-PM(L5 7). 0.33 g (0.51 mmol) of
—
SuO-DTPA-PM (mole ratio of SuO-DTPA-PM to PAEA repeat units: 20%) was dissolved in DMF
(20 mL), and then added into a PAEA (0.40 g, 2.55 mmol) in DMF-H2O (v/v=1/5, 30 mL) solution
with rapidly stirring at 0℃. The reaction mixture was stirred for 8 h at the same temperature and
for 24 h at 20℃. The reaction mixture was evaporated and the residue solution was precipitated by
pouring into 500 mL ethanol-ether (v/v=1/4). The precipitate was collected and dialysed against
distilled water. After lyophilization, the product PAEA-DTPA- PM(L5) was obtained (0.58 g,
86.4%).
IR(KBr, cm−1): 3281(OH), 3062(C＝C), 1665(COOH), 1652, 1540(CONH). 1H NMR (D2O,
δ ): 7.9 (
), 3.9 (CCH ), 3.6—3.2 (CH CH ), 3.0 (CH ), 2.82 (NCH CO). UV(H O, nm):
2 2 2 3 2 2
H
N
282. PAEA-DTPA-PM(L6, L7) (mole ratio of SuO-DTPA-PM to PAEA repeat units: 60%, 100%)
were similarly synthesized.
1.2.5 Sythesis of macromolecular gadolinium complexes. 5 mmol of PHEA-DTPA-PM(L1
)
4
—
or PAEA-DTPA-PM(L5 7) in distilled water was added into a solution of GdCl3 (7.5 mmol) in
—
distilled water with stirring at room temperature. The reaction mixture was stirred for 1 h, adjusted
the reaction solution to pH = 5 with dilute NaOH solution, and continually stirred for 12 h at room
temperature. After dialysis and evaporation of water under vacuum, macromolecular gadolinium
complexes were obtained (table 2).
PHEA-Gd-DTPA-PM: IR(KBr, cm−1): 3468—3436 (OH), 2938 (C==C), 1644 (COO), 1638,
1536 (CONH).
PAEA-Gd-DTPA-PM: IR(KBr, cm−1): 3436 (NH2), 2935 (C==C), 1661 (COO), 1645, 1596
(CONH).
1.3 Relaxivity of metal complexes
The proton longitudinal relaxation times (T1) for PHEA-Gd-DTPA-PM, PAEA-Gd-DTPA-

348
SCIENCE IN CHINA (Series B)
Vol. 44
PM and Gd-DTPA in distilled water at 25℃ were measured by a Bruker WP-80SY NMR(80
MHz, 1.88 T) spectrometer. The experimental data are listed in table 3.
Table 2 Elementary analysis of macromolecular gadolinium complexes
Polymer metal com-
Mole ratio of SuO-DTPA-PM/Polymeric
repeat units (%)
Elem. anal.(%)
Gd a)
C
N
H
plexes
PHEA-Gd-DTPA-PM
M1
M2
M3
M4
10
40
60
1.70
2.24
2.65
2.90
49.80
45.62
45.46
45.37
18.10
16.53
16.40
16.33
6.19
5.66
5.63
5.62
100
PAEA-Gd-DTPA-PM
M5
M6
M7
20
60
100
10.76
11.04
13.91
42.77
42.65
41.51
16.32
16.17
14.70
4.92
4.80
4.77
a) Measured by ICP Atomscan-2000 spectrometer.
Table 3 Experimental data of relaxivity
[Gd3+]/mmol·L−1
R1/(mmol·L−1·s)−1
Metal complexes
T1 obsd/s
Gd-DTPA
3.3066
0.05323
5.58
PHEA-Gd-DTPA-PM
M1
M2
M3
M4
1.8708
2.3630
1.4206
1.8708
0.03801
0.02816
0.04266
0.02723
13.88
14.88
16.26
19.45
PAEA-Gd-DTPA-PM
M5
M6
M7
1.2410
2.7710
1.8390
0.07890
0.03365
0.03740
9.94
10.61
14.35
Temp: 25℃, NMR frequency: 80 MHz, T1d =2.95 s.
1.4 In vitro cytotoxicity assay of polyaminoacid
HeLa cells were plated in 24 well plates in the growth medium. The cells grew for 24 h and
then the growth medium was removed and replaced with growth medium containing polyamino-
acid. After 15 h incubation, the cells were washed with the Hank’s solution without Ca2+ or Mg2+
and digested by the trypsin solution, then collected by centrifugalization and dyed for 5 min using
Trypan blue assay. So the dead cells were dyed by Trypan blue and the active cells were not dyed
by Trypan blue. The optical density was measured at microscope and expressed as a percent rela-
tive to control.
1.5 In vitro cytotoxicity assay of macromolecular gadolinium complexes
L-02 cells (2×105/mL) were plated in 96-well plates in the growth medium and the number
of cells in each well is 2×104. The cells were incubated for 24 h in incubator (37℃, 5% CO2) and
100 μL of the growth medium containing gadolinium complex was added. After 72 h incubation,
the cells were washed with the growth medium, 20 μL of the MTT solution (1.0 mg/mL) was
added and incubated for 4 h, then washed again with 3% fetal calf serum and phosphate buffer
saline solution (3% FCS-PBS), 100 μL of DMSO added and shaken for 30 min at room tempera-
ture. The optical density (OD570) was measured at 570 nm with a DG-3022A ELISA-Reader and
expressed as a percent relative to control.

No. 4
LIVER-TARGETING MACROMOLECULAR MRI CONTRAST AGENTS
349
1.6 MR imaging
MR imaging experiments were performed on a BIOSPEC 47/30 system (Bruker instruments,
4.7 T). MR imaging of the liver was carried out on the rat under anesthesia. SD rats (150 g) were
anesthetized with urethane (20%, 1 g/kg), positioned pronely and fixed to a polystyrene cradle
with adhesive tape to minimize respiratory motion. Coronal images of the liver were performed
before SD rats had received the injection of gadolinium complexes. Gd-DTPA (0.1 mmol/kg,
Magnevist), PHEA-Gd-DTPA-PM (M3, 0.06 mmol/kg) or PAEA-Gd-DTPA-PM (M6, 0.08
mmol/kg) were injected via the tail. Coronal images of the liver were obtained with a T1-weighted
spin-echo sequence (TR=500 ms, TE=15 ms, the field of view is 40 mm, with an image matrix of
256×256. Slice thickness is 2 mm, 8 multislices, average 8, with a l mm interslice gap).
1.7 In vivo biodistribution
In vivo biodistribution of mice was measured on an SN-695 gamma counter RIA program Ver
5.33. PAEA-DTPA-PM (L7) was labeled with 99m TcO−4 and reduced with fresh tin (II) tartrate
and vitamin C at room temperature (radiochemical purity >95%). Forty-five mice (Kunming, (20
±2) g) were divided into nine groups, and each mouse was injected with 99mTc-labeled PAEA-
DTPA-PM solution via the tail and sacrificed at different time intervals postadministration. Sam-
ples of blood, lung, liver, spleen, kidney, muscle, heart, bone, and small intestine were taken. Tis-
sues were blotted to remove excess blood and weighed, analyzed by an SN-695 gamma counter
RIA program Ver 5.33. Percentage of injected dose per organ (%ID/g) was calculated.
2
Results and discussion
2.1 Synthesis and characterization
The elementary analysis of macromolecular gadolinium complexes is shown in table 2. In
this work, PHEA-Gd-DTPA-PM and PAEA-Gd-DTPA-PM in aqueous solution after dialysis show
absorption at 282 nm, however, PHEA and PAEA show no UV absorption at the above-mentioned
wavelengths. Thus the UV data in aqueous solution give the evidence of the incorporation of
DTPA-PM into PHEA and PAEA.
2.2 Relaxivity of metal complexes
In the absence of solute-solute interactions, the solvent relaxation rates are linearly dependent
on the concentration of the paramagnetic species ([M]). Relaxivity, R1, is defined as the slope of
this dependence[2]:
(1/T1)obsd=(1/T1)d+R1(M),
where (1/T1)obsd is the observed solvent relaxation rate in the presence of a paramagnetic species,
and (1/T1)d is the solvent relaxation rate in the absence of a paramagnatic species. In this experi-
ment, the concentrations of the paramagnetic species [Gd3+] were measured by ICP Atom-
scan-2000 spectrometer. Thus R1 for paramagnetic metal complexes in distilled water could be
worked out (table 3).

350
SCIENCE IN CHINA (Series B)
Vol. 44
Table 3 illustrates that all polyaspartamide gadolinium complexes possess higher relaxation
effectiveness than that of the corresponding small molecular complex: Gd-DTPA.
2.3 Cytotoxicity assay of polyaminoacid
The effect of PHEA and PAEA on
cell growth and metabolism of HeLa cells
in vitro was determined as a function of
polymer concentration and compared to
polylysine (fig. 1). The preliminary results
show that over the concentration range
tested, the cells incubated with PHEA and
PAEA retain 55.1% and 61.9% viability at
100 μg/mL, while in the presence of
poly-L-lysine (PLys), HeLa cells show no
viability under the same concentration. At
a higher concentration (200 μg/mL) of
PHEA and PAEA, the cells still retain
Fig. 1. Cytotoxicity assay of PLys, PHEA and PAEA in HeLa cells.
1, PHEA; 2, PAEA; 3, PLys.
62.4% and 49.8% viability respectively, relative to control.
2.4 Cytotoxicity assay of metal complexes
The effect of metal complexes on cell growth and metabolism of L-02 cells in vitro is shown
in fig. 2. At the concentration (25600 μg/mL) of gadolinium complexes in the growth medium, the
L-02 cells incubated with Gd-DTPA and
PHEA-Gd-DTPA-PM (M3) retain respec-
tively above 4.0% and 88.7% viability re-
spectively relative to control, illustrating
that PHEA-Gd-DTPA-PM (M3) possesses
obviously lower cytotoxicity to L-02 cells
than that of Gd-DTPA. However, L-02
cells incubated with PAEA-Gd-DTPA- PM
(M6) over the same concentration range
had lower viability because the cells were
covered by PAEA-Gd-DTPA-PM (M6) and
Fig. 2. Cytotoxicity assay of metal complexes in L-02 cells. 1,
could not grow up.
PHEA-Gd-DTPA-PM (M3); 2, Gd-DTPA; 3, PAEA-Gd-
DTPA-PM (M6).
2.5 MR imaging
Compared to the signal intensity (SI) of the liver in the rat injected with Gd-DTPA (0.1
mmol/kg), it demonstrates that SI of the liver in the rat injected with PHEA-Gd- DTPA-PM (M3,
0.06 mmol/kg) or PAEA-Gd- DTPA-PM (M6, 0.08 mmol/kg) was obviously enhanced, the irradi-
ated portion of the liver was brighter and the demarcation became clearer (fig.3).

No. 4
LIVER-TARGETING MACROMOLECULAR MRI CONTRAST AGENTS
351
Fig. 3. N1, N2 and N3 are the T1-weighted images of the rat received no injection of MRI contrast agent; A1 and B1 are the
T1-weighted images of the rat received injection of Gd-DTPA (0.1 mmol/kg, Magnevist) after 6 and 44 min; A2 and B2 are the
T1-weighted images of the rat received injection of PHEA-Gd-DTPA-PM (M3, 0.06 mmol/kg) after 15 min and 46 min; A3 and B3
are the T1-weighted images of the rat received injection of PAEA-Gd-DTPA-PM (M6, 0.08 mmol/kg) after 20 and 46 min.
2.6 In vivo biodistribution
Fig.4 shows that the rapid decrease of
the macromolecular MRI contrast agent in
blood, heart and spleen correlates with its
increasing capture by the liver, indicating
that these macromolecular MRI contrast
agents containing pyridoxamine were taken
up specifically by the liver.
3
Conclusion
Fig. 4. Organ distribution at the different times after injection.
1, Liver; 2, kidney; 3, spleen; 4, lung; 5, blood; 6, heart.
Polyaspartamide gadolinium chelated
complexes containing pyridoxamine moiety

352
SCIENCE IN CHINA (Series B)
Vol. 44
can be targeted to specific organs such as liver and spleen. They possess obviously higher
relaxation effectiveness than that of Gd-DTPA. MR images show that the signal intensity (SI) of
the liver in rat injected PHEA-Gd-DTPA-PM or PAEA-Gd-DTPA-PM is obviously enhanced.
Acknowledgements The authors would like to thank Mr. Cao Guoqiang and Mr. Cao Wei (Department of Nuclear
Medicine of Wuhan Union Hospital) for their assistance in the biodistribution experiment, Profs. Zheng Congyi and Mr. Qu
Sanfu (College of Life Sciences, Wuhan University) for their assistance in the cytotoxicity assay. This work was supported by the
National Natural Science Foundation of China (Grant No. 29874028).
References
1. Lauffer, R. B., Paramagnetic metal complexes as water proton relaxation agents for NMR imaging: theory and design,
Chem. Rev., 1987, 87: 901—927.
2. Caravan, P., Ellison, J. J., Mcmurry, T. J. et al., Gadolinium (III) chelates as MRI contrast agents: structure, dynamics, and
applications, Chem. Rev., 1999, 99: 2293—2352.
3. Saudek, V., Drobnik, J., Havranova, M. et al., High molecular weight derivative of Trypsin-Kallikrein inhibitor for poten-
tial medical use: preparation and characterization, Macromol. Chem., 1982, 183: 1473—1484.
4. Gibby, W. A., Puttagunta, N. R., Aminosaccharide contrast agents for magnetic resonance images, United States Patent,
5,330,743 (Jul. 19, 1994).
5. Fu, Y. J., Zhuo, R. X., Studies on D-Galactose-containing DTPA bisamide gadolinium complexes as liver-targeting MRI
contrast agents, Chem. J. Chin. Univ., 1997,18 (7): 1072—1079.
6. Wei, J. F., Zhuo, R. X., Studies on the novel liver-targeting MRI contrast agents containing vitamin B6, Chin. Scien. Bull.,
1997，42(14): 1519—1524.
7. Neuse, E. W., Perlwitz, A. G., Schmitt, S., Water-soluble polyamides as potential drug carriers, Macromol., 1991, 192:
35—50.