Adsorbents with high selectivity for uremic middle molecular peptides containing the Asp-Phe-Leu-Ala-Glu sequence

Article abstract of DOI:10.1021/la904272e

Asp-Phe-Leu-Ala-Glu (DE5) is a frequent sequence of many toxic middle molecular peptides that accumulate in uremic patients. To eliminate these peptides by hemoperfusion, three adsorbents (CP1-Zn²⁺, CP2-Zn ²⁺, and CP3-Zn²⁺) were designed on the basis of coordination and hydrophobic interactions. Adsorption experiments indicated that CP2-Zn²⁺ had the highest affinity for DE5 among these three adsorbents. Also, the adsorption capacity of CP2-Zn²⁺ in DE5 and DE5-containing peptides was about 2-6 times higher than that of peptides without the DE5 sequence. Linear polymers bearing the same functional groups of the adsorbents were used as models to study the adsorption mechanism via isothermal titration calorimetry (ITC) and computer-aided analyses. The results indicated that coordination and hydrophobic interactions played the most important roles in their affinity. When two carboxyl moieties on Asp and Glu residues coordinated to CP2-Zn²⁺, the hydrophobic interaction took place by the aggregation of the hydrophobic amino acid residues with phenyl group on CP2-Zn²⁺. The optimal collaboration of these interactions led to the tight binding and selective adsorption of DE5-containing peptides onto CP2-Zn²⁺. These results may provide new insight into the design of affinity adsorbents for peptides containing DE5-like sequences.

Full text of DOI:10.1021/la904272e

pubs.acs.org/Langmuir
© 2010 American Chemical Society
Adsorbents with High Selectivity for Uremic Middle Molecular Peptides
Containing the Asp-Phe-Leu-Ala-Glu Sequence
Yitao Qiao, Jianxin Zhao, Pinglin Li, Jun Wang, Jing Feng, Wei Wang, Hongwei Sun, Yi Ma, and
Zhi Yuan*
Key Laboratory of Functional Polymer Materials, Ministry of Education, College of Chemistry, Nankai
University, Tianjin 300071, PR China
Received November 11, 2009. Revised Manuscript Received February 1, 2010
Asp-Phe-Leu-Ala-Glu (DE5) is a frequent sequence of many toxic middle molecular peptides that accumulate in
uremic patients. To eliminate these peptides by hemoperfusion, three adsorbents (CP1-Zn2þ, CP2-Zn2þ, and CP3-Zn2þ
)
were designed on the basis of coordination and hydrophobic interactions. Adsorption experiments indicated that CP2-
Zn2þ had the highest affinity for DE5 among these three adsorbents. Also, the adsorption capacity of CP2-Zn2þ in DE5
and DE5-containing peptides was about 2-6 times higher than that of peptides without the DE5 sequence. Linear
polymers bearing the same functional groups of the adsorbents were used as models to study the adsorption mechanism
via isothermal titration calorimetry (ITC) and computer-aided analyses. The results indicated that coordination and
hydrophobic interactions played the most important roles in their affinity. When two carboxyl moieties on Asp and Glu
residues coordinated to CP2-Zn2þ, the hydrophobic interaction took place by the aggregation of the hydrophobic amino
acid residues with phenyl group on CP2-Zn2þ. The optimal collaboration of these interactions led to the tight binding
and selective adsorption of DE5-containing peptides onto CP2-Zn2þ. These results may provide new insight into the
design of affinity adsorbents for peptides containing DE5-like sequences.
1. Introduction
a more efficient ligand as reported by Labrou and Liu.7,8
Molecular recognition theory has also been used to direct the
design of affinity adsorbents to improve adsorption selectivity.9
The accumulation of metabolites in the human body is thought
to be toxic and contributes to many pathological changes in
patients with organ failure.10,11 These toxins could be eliminated
by hemodialysis or hemoperfusion; for example, low-density
lipoprotein could be removed by adsorbents with sulfonic and
cholesterol groups.12 According to the middle molecule hypothe-
sis,13 middle-molecular-weight toxins ranging from 500 to 5000
Da are the main toxic substances that accumulate in uremic
patients, and middle molecular peptides have been assumed to be
one type of major uremic toxin. Kaplan et al. reported that of the
51 middle molecular peptides isolated from the serums of uremic
patients 38 contain DE5 sequences.14 It seems that DE5 is related
to the uremia, so if there are efficient adsorbents that can
eliminate DE5-containing peptides by hemoperfusion, this will
be very helpful in the treatment of uremia.
Adsorption technology has become increasingly important
owing to its wide range of applications in the separation of toxic
compounds from environmental and industrial waste. This tech-
nology has also been used for the elimination of toxins that
accumulate in the human body.1-3 Affinity adsorbents, with
specific adsorption and high elimination efficiency, have attracted
much attention in this field,4-6 and researchers are now focused
on the design of affinity adsorbents for the specific adsorption of
target toxins.
There are two main types of affinity adsorbent ligands:
biomolecules and synthetic ligands. Bioactive ligand adsorbents
have good selectivity and high capacity, but they are usually
inconvenient because of the low structural stability and high cost
of biomolecules. The advantages of synthetic ligands are stability
and low cost, and they could be promising for the preparation of
affinity adsorbents. Traditionally, to identify a proper affinity
adsorbent, different synthetic ligands are immobilized onto
matrices and the more efficient adsorbents are screened out by
adsorption experiments; however, this approach is time-consum-
ing and labor-intensive. Recently, the design of affinity adsor-
bents for specific targets has attracted much interest and has been
extensively developed. Computer simulation has been used to find
In this work, the frequent DE5 sequence was used as the tar-
get peptide. On the basis of the structural features of DE5, three
adsorbents, CP1-Zn2þ, CP2-Zn2þ, and CP3-Zn2þ, were designed
and synthesized on the basis of the coordination and hydro-
phobic interactions whereas CP4-Zn2þ and CP5-Zn2þ, bearing
only coordination groups or hydrophobic groups, were used as
the controls. Static adsorption experiments were performed to
*Corresponding author. Tel: þ86-22-23501164. Fax: þ86-22-23503510.
E-mail: zhiy@nankai.edu.cn.
(1) Davankov, V.; Tsyurupa, M.; Ilyin, M.; Pavlova, L. J. Chromatogr., A 2002,
956, 65–73.
(2) Hughes, R. D. Int. J. Artif. Organs 2002, 25, 911–917.
(3) Rozga, J. Xenotransplantation 2006, 13, 380–389.
(4) Duarte, I. S.; Zollner, R. L.; Bueno, S. M. A. Artif. Organs 2006, 30, 606–614.
(5) Tani, N. Artif. Organs 1996, 20, 922–929.
(9) Platis, D.; Sotriffer, C. A .; Clonis, Y.; Labrou, N. E. J. Chromatogr., A 2006,
1128, 138–151.
(10) Vanholder, R.; De Smet, R. J. Am. Soc. Nephrol. 1999, 10, 1815–1823.
(11) Mitzner, S. R.; Stange, J.; Klammt, S.; Peszynski, P.; Schmidt, R.; Noldge-
Schomburg, G. J. Am. Soc. Nephrol. 2001, 12, S75–S82.
(12) Wang, S. Q.; Yu, Y. T.; Cui, T.; Cheng, Y. Biomaterials 2003, 24, 2799–
2802.
(6) Zhong, W.; Huang, W.; Li, J. H.; Hou, G. H.; Fang, J.; Yuan, Z.
J. Chromatogr., B 2007, 852, 288–292.
(7) Labrou, N. E.; Eliopoulos, E.; Clonis, Y. D. Biotechnol. Bioeng. 1999, 63,
322–332.
(13) Babb, A. L.; Farrell, P. C.; Uvelli, D. A.; Scribner, B. H. ASAIO Trans.
1972, 18, 98–105.
(14) Kaplan, B.; Cojocaru, M.; Unsworth, E.; Knecht, A.; Martin, B. M.
J. Chromatogr., B 2003, 796, 141–153.
(8) Liu, F. F.; Wang, T.;Dong, X. Y.; Sun, Y. J. Chromatogr., A 2007, 1146, 41–50.
Langmuir 2010, 26(10), 7181–7187
Published on Web 03/04/2010
DOI: 10.1021/la904272e 7181

Article
Qiao et al.
examine their adsorption ability with respect to DE5 and DE5-
containing peptides. Soluble linear polymers bearing the same
functional groups were used as models of the adsorbents to
determine the adsorption mechanism using isothermal titration
calorimetry (ITC) and computer simulation.
(3.49 mL, 23.75 mmol) was added. After that, acryloyl chloride
(1.93 mL, 23.75 mmol) was added dropwise. The mixture was
stirred at -10 °C for another 12 h. After being filtered to remove
the deposited material, the solution was decolorized with acti-
vated charcoal for 6 h at room temperature. The solution was
concentrated under reduced pressure and purified by column
chromatography (5:1 CH2Cl2/EtOAc) to give M4 as a white solid
(5.26 g, 79%). 1H NMR (400 MHz, CDCl3) δ: 7.31-7.48(m, 5H),
6.53-6.56 (m, 1H), 6.39 (d, 1H), 5.84 (d, 1H), 3.75 (s, 2H). ESI-
MS: [M þ H]þ m/z = 281.08.
2. Experimental Section
2.1. Materials. Tetrahydrofuran (THF) was dried over sodium/
benzophenone and distilled immediately before use. Acrylamide
(AM) was recrystallized from acetone, 2,20-azobisisobutyronitrile
(AIBN) was recrystallizedfrom ethanol, and 2,6-diaminopyridine
was recrystallized from toluene. Phenylacetyl chloride and 3-phe-
nylpropanoyl chloride were prepared by the reaction of thionyl
chloride with the corresponding carboxylic acid. Acryloyl chlo-
ride was prepared by the reaction of benzoyl chloride with acrylic
acid. Peptides were purchased from GL Biochem (Shanghai) Ltd.
2-Aminopyridine was purchased from Shanghai Chemical Co.
and recrystallized in petroleum ether. The other reagents and
solvents were commercially available and used without purifica-
tion.
2.2. General Procedure for the Synthesis of Functional
Monomers. 2.2.1. Synthesis of 2-Acryloylamido-6-ben-
zoylamindo-pyridine (M1). M1, M2, and M3 were synthesized
from the corresponding acylchloride by reaction with 2,6-diami-
nopyridine in tetrahydrofuran (THF) at -10 °C ((1 °C) under a
nitrogen atmosphere.15 Typical synthesis procedures are as fol-
lows.
2.2.5. Syntheses of 2-Acryloylamidopyridine (M5). Com-
pound M5 was prepared from acryloyl chloride and amidopy-
ridine according to the literature.16 Yield: 4.38 g, 61%. 1H NMR
(400 MHz, CDCl3) δ: 8.56 (s, 1H), 8.29 (d, 1H), 7.72 (t, 1H),
7.05(d, 1H), 6.46 (d, 1H), 6.25 (m, 1H), 5.81 (d, 1H). ESI-MS:
[M þ H]þ m/z = 149.28.
2.3. Preparation of Adsorbents (CP1, CP2, CP3, CP4,
and CP5). The functional monomer (80% molar fraction),
acrylamide(10% molar fraction), cross-linkingagentN,N0-methy-
lenebisacrylamide (10% molar fraction), AIBN (w = 3%), and
THF (concentration of monomer, 0.1 g/mL) were placed in a
round-bottomed flask under a nitrogen atmosphere with a mag-
netic stir bar. The reaction was carried out at 60 °C in an oil bath
for 24 h. The filtered precipitate was extracted in a Soxhlet
extractor successively by acetone and distilled water for 48 h,
respectively. The final adsorbents were ground and treated with
standard sieves (80-100 meshes).
2.4. Preparation of Linear Polymers (P1, P2, P3, P4, and
P5). The linear polymers were synthesized and purified accord-
ing to our previous work.15 Briefly, the functional monomer
(10% molar fraction), acrylamide (90% molar fraction), AIBN
(w = 3%), and 1:1 v/v THF/H2O (concentration of monomer,
0.1 g/mL) wereplaced ina round-bottomed flask under a nitrogen
atmosphere with a magnetic stir bar. The reaction was carried out
at 60 °C in an oil bath for 24 h. The crude linear polymer was
obtained by precipitation in acetone. Then, the crude product was
purified by reprecipitation in acetone three times until no mono-
mer remained as analyzed by 1H NMR. The mole fractions of the
functional monomer in P1, P2, P3, P4, and P5 were 4.2, 2.7, 3.0,
3.6, and 7.7%, respectively, as determined by 1H NMR.
2.5. ITC Experiments. ITC titration was performed on a
VP-ITC microcalorimeter (MicoCal Inc., Northampton, MA). In
general, a solution of the peptide (1 mM) was injected in portions
(7 μL ꢀ 39 times) into a solution (cell volume = 1.4685 mL) of the
polymer (0.05 mM, calculated by the number of functional
monomer repeat units) or polymer (0.05 mM)-Zn2þ (0.165 mM)
under different conditions. The heat flow was recorded, plotted
against time, and converted into enthalpy (ΔH) by integration of
the appropriate reaction peaks. Dilution effects were eliminated
by the subtraction of the data from a blank experiment. All of the
data were calculatedwith Origin ITC data-analysis software using
one set of site models.
Benzoyl chloride (3.05 mL, 26.25 mmol) in tetrahydrofuran
(100 mL) was added dropwise (finished within 5 h) to a solution
of 2,6-diaminopyridine (2.73 g, 25 mmol) and triethylamine
(3.85 mL, 26.25 mmol) in tetrahydrofuran (150 mL) at -10 °C
((1 °C) under a nitrogen atmosphere. After the mixture was
stirred overnight, triethylamine (3.49 mL, 23.75 mmol) was added
to the mixture and acryloyl chloride (1.93 mL, 23.75 mmol) was
then added dropwise; the mixture was again stirred overnight.
After being filtered to remove the deposited material, the mixture
was decolored by acticarbon for 6 h at room temperature. The
solutionwasconcentrated togiveareddish-brown oil. Theresidue
was purified by column chromatography, eluting with 8:1 di-
chloromethane/acetic ether to give 2-acryloylamido-6-benzoyla-
mindo-pyridine (M1, 3.55 g, 53%) as a white solid. 1H NMR (400
MHz, CDCl3) δ: 5.81 (d, 1H), 6.25 (m, 1H), 6.46 (d, 1H), 7.49
(t, 2H), 7.57(t, 1H), 7.78(t, 1H), 7.90(d, 2H), 7.99(d, 1H), 8.08(d,
1H). ESI-MS: [M þ H]þ m/z = 268.34.
2.2.2. Synthesis of 2-Acryloylamido-6-phenylacetamido-
pyridine (M2). Compound M2 was prepared from phenylacetyl
chloride according to the preparation procedure of M1. Yield:
41%. The mobile phase of column chromatography was 10:1
1
dichloromethane/acetone. H NMR (400 MHz, CDCl3) δ: 3.74
(s, 2H), 5.76(d, 1H), 6.21(d, 1H), 6.44(d, 1H), 7.33-7.92(m, 8H).
ESI-MS: [M þ H]þ m/z = 282.11.
2.2.3. Synthesis of 2-Acryloylamido-6-hydrocinnamamido-
pyridine (M3). Compound M3 was prepared from 3-phenylpro-
panoyl chloride according to the preparation procedure of M1.
Yield: 48%. The mobile phase of column chromatography was
12:1 dichloromethane/acetic ether. 1H NMR (400 MHz, CDCl3)
δ: 2.67 (t, 2H), 3.04 (t, 2H), 5.77 (d, 1H), 6.23 (d, 1H), 6.41 (d, 1H),
7.18-7.31 (m, 5H), 7.70 (t, 1H), 7.89-7.97 (m, 2H). ESI-MS:
[M þ H]þ m/z = 296.20.
2.6. Adsorption Experiments. Dry adsorbent (0.1 g) was
incubated with ZnCl2 solution (0.66 mM, 10 mL) and shaken for
24 h at 30 °C. The peptide (0.2 mM) was then added with further
shaking. The peptide absorption was determined by a Unico UV-
4802 spectrometer at 220 nm. The control experiments were
carried out under the same conditions. The adsorption capacity
was calculated from the equation
2.2.4. Synthesis of N-(6-Phenylacetylaminobenzen-2-yl)-
acrylamide (M4). Phenylacetyl chloride (3.48 mL, 26.25 mmol)
in THF (100 mL) was added dropwise to a solution of 2,6-
diaminobenzene (2.70 g, 25 mmol) and triethylamine (3.85 mL,
26.25 mmol) in THF (150 mL) within 4.5 h at -10 °C. The
mixture was stirred at -10 °C for 12 h, and then triethylamine
ðc1 - c2ÞMV
Ac ¼
m
where Ac stands for the adsorption capacity, c1 and c2 stand for
the peptide concentrations before and after adsorption, respec-
tively, V stands for the volume of solution used in adsorption,
(15) Zhao, J. X.; Qiao, Y. T.; Feng, J.; Luo, Z. F.; Yuan, Z. Chem. J. Chin. Univ.
2008, 29, 658–660.
(16) Qiao, Y. T.; Wei, Z.; Feng, J.; Chen, Y. C.; Li, P. L.; Wang, W.; Ma, Y.;
Yuan, Z. J. Sep. Sci. 2009, 32, 2462–2468.
7182 DOI: 10.1021/la904272e
Langmuir 2010, 26(10), 7181–7187

Qiao et al.
Article
M stands for the peptide molar mass, and m stands for the mass
of the dry adsorbent. All of the adsorption experiments were
carried out in triplicate, and data are expressed as the mean ( SD.
2.7. Molecular Modeling Studies. Hartree-Fock(HF) and
density functional theory (DFT) calculations were performed
with the Gaussian 03 suite of programs.17 N-(6-Formylaminopyr-
idin-2-yl)-2-phenylacetamide (FPP) was selected as the modeling
complexes for the adsorbent. The complex structure of FPP-Zn2þ
was built and optimized at the HF/6-31G (D) level.
A Silicon Graphics SGI Indigo 2 workstation was used for the
simulation of the adsorption. Molecular building, geometry
optimization, and conformational searching were carried out
using molecular modeling package Sybyl, version 6.6. In the
calculations, the molecular mechanics and molecular dynamics
methods were used.18
3. Results and Discussion
Figure 1. Adsorption capacities of DE5 onto adsorbents at 30 °C
(n = 3).
3.1. Design of Functional Monomers. To our knowledge,
the adsorption of proteins and peptides onto adsorbents is usually
based on various noncovalent interactions, including electrostatic
forces, coordination interactions, hydrophobic interactions, and
hydrogen bonding. Hence, the introduction of functional groups
bearing these noncovalent interaction binding sites to adsorbents
may improve their adsorption ability. To achieve high affinity,
the collaboration of multiple interactins is also a promising
strategy.19
Scheme 1. Chemical Structures of the Functional Monomers
In this work, target peptide DE5 has a special structure with
hydrophobic amino acid residues in the middle (Phe-Leu) and an
acidic amino acid residue (Asp and Glu) at each terminus.
Considering the structural feature of DE5, the collaboration of
coordination and hydrophobic interactions was expected to have
a substantial effect on peptide adsorption. Metal-coordination
interactions show a high affinity for electron-enriched groups
such as phosphate and carbonate on peptides20-22 whereas
hydrophobic interactions are also an important driving force in
the adsorption of hydrophobic peptides in aqueous solution.
Moreover, these interactions could be efficiently achieved in
aqueous solution. It may therefore be beneficial to utilize these
interactions in improving the adsorption of DE5 in aqueous
solution. Until now, the coordination interaction between pyridyl
group and divalent transition-metal ions has been widely applied
in binding studies,23 and the phenyl group has been involved in
many hydrophobic interactions owing to its hydrophobicity.
Thus, functional monomers bearing the coordination interaction
group (M5), hydrophobic group (M4), and both coordination
and hydrophobic groups (M1, M2, and M3) were designed and
synthesized (Scheme 1). To facilitate geometrical matching, M1,
M2, and M3 differed only in the distance between the terminal
phenyl group and the pyridyl groups.
3.2. Adsorption Capacity of DE5. Polyacrylamide (PAM)
has been extensively used in recent years for protein separation
owing to its good biocompatibility and environmental stability.
Moreover, hydrophilic PAM on the adsorbent can decrease the
nonspecific adsorption of peptides.24 Previous work has con-
firmed that no obvious interaction occurs between PAM and
DE5.15 Therefore, the corresponding adsorbents were synthesized
by the copolymerization of functional monomers, acrylamide,
and cross linker. We evaluated the adsorption capacities of CP1-
Zn2þ, CP2-Zn2þ, CP3-Zn2þ, CP4-Zn2þ, and CP5-Zn2þ for DE5
at 30 °C by static adsorption experiments. As shown in Figure 1,
all static adsorption experiments reached an equilibrium state
within 1.75 h. As expected, CP1-Zn2þ, CP2-Zn2þ, and CP3-Zn2þ
bearing both coordination and hydrophobic groups exhibited
better adsorption to DE5. In particular, CP2-Zn2þ had an opti-
mal adsorption capacity for DE5, which was almost 8 times
greater than that of CP4-Zn2þ or CP5-Zn2þ. It has been found
that cross-linked polyacylamide showed neglected adsorption to
DE5 (<0.1 mg/g), so the adsorption of DE5 onto adsorbents
should be attributed to the binding of DE5 with the functional
groups of the adsorbents. Elemental analyses showed that CP1,
CP2, CP3, and CP4 had similar functional group contents, which
were 29.2, 25.3, 27.7, and 22.7%, respectively. As for CP5, the
content of functional groups was 40.1%, slightly higher than for
other adsorbents. Compared with the great differences in their
adsorption capacities, the difference in functional group content
(17) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.;
Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.;
Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.;
Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.;
Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev,
O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;
Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V.
G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck,
A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;
Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong,
M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision C.02; Gaussian, Inc.:
Wallingford, CT, 2004.Frisch, M. J.; Trucks, G. W.; Schlegel, H. B. et al. . Gaussian 03,
Revision C.02, Gaussian, Inc., Wallingford CT, 2004.
(18) Vaidyanathan, J.; Vaidyanathan, T. K.; Yadav, P.; Linaras, C. E. Bioma-
terials 2001, 22, 2911–2920.
(19) Peczuh, M. W.; Hamilton, A. D. Chem. Rev. 2000, 100, 2479–2494.
(20) Yamaguchi, S.; Yoshimura, I.; Kohira, T.; Tamaru, S.; Hamachi, I. J. Am.
Chem. Soc. 2005, 127, 11835–11841.
(21) Ojida, A.; Mito-oka, Y.; Inoue, M.; Hamachi, I. J. Am. Chem. Soc. 2002,
124, 6256–6260.
(22) Imai, H.; Munakata, H.; Uemori, Y.; Sakura, N. Inorg. Chem. 2004, 43,
1211–1213.
(23) Kruppa, M.; Konig, B. Chem. Rev. 2006, 106, 3520–3560.
(24) Li, Y.; Lee, M. L. J. Sep. Sci. 2009, 32, 3369–3378.
Langmuir 2010, 26(10), 7181–7187
DOI: 10.1021/la904272e 7183

Article
Qiao et al.
Table 1. Amino Acid Sequence of Peptides
multiple acidic amino acid residues
peptide
amino acid sequence
Asp-Phe-Leu-Ala-Glu
hydrophobic amino acid residues
DE5
DG8
EV11
SH5
WF4
GF6
yes
yes
yes
yes
no
yes
yes
yes
no
yes
yes
Asp-Phe-Leu-Ala-Glu-Gly-Gly-Gly
Glu-Gly-Asp-Phe-Leu-Ala-Glu-Gly-Gly-Gly-Val
Ser-Glu-Ala-Asp-His
Trp-Met-Asp-Phe
Gly-Arg- Trp-Met-Asp-Phe
no
Figure 2. Adsorption capacity of CP2-Zn2þ to different peptides
(n = 3).
was much smaller and should not be the factor leading to the
differences in adsorption capacity. Moreover, all adsorption ex-
periments were repeated three times, which excluded the possible
effect of heterogeneous adsorbents on adsorption. Therefore, the
differences in the adsorption capacities should not be attributed to
the specific area of the adsorbents but rather to the binding affinity
of the functional groups to DE5. It seems that the collaboration of
coordination and hydrophobic interactions led to the good ad-
sorption ability of CP1-Zn2þ, CP2-Zn2þ, and CP3-Zn2þ. The
functional groups of CP1-Zn2þ, CP2-Zn2þ, and CP3-Zn2þ differed
only in the number of CH2 groups. Thus, the distance between the
hydrophobic phenyl group and the pyridyl group was therefore
considered to be the only factor that contributes to the difference in
the adsorption capacity; indeed, the best geometrical matching led
to the best adsorption among the three adsorbents.
3.3. Adsorption Selectivity towards DE5 and DE5-Con-
taining Peptides. To examine the adsorption selectivity of CP2-
Zn2þ to peptides containing the DE5 sequence, another five
peptides were selected (Table 1). DE8 and EV11 were middle
molecular peptides containing the DE5 sequence found in uremic
serum.14 SH5 is another penta-peptide with double acidic residues
but without the hydrophobic residue. WF4 and GF6 have multi-
ple hydrophobic residues but only one acidic residue. We tested
the adsorption capacity of CP2-Zn2þ to these peptides by static
adsorption experiments. As shown in Figure 2, CP2-Zn2þ ex-
hibited good adsorption ability to peptides containing the DE5
sequence (DE5, DE8, and EV11). In marked contrast to these
peptides, CP2-Zn2þ exhibited a much weaker affinity to peptides
without the DE5 sequence (SH5, WF4, and GF6), about one-
sixth to one-half the capacity of peptides containing DE5. The
adsorption results indicated that CP2-Zn2þ could provide selec-
tive adsorption to DE5-containing peptides.
Figure 3. Microcalorimetric thermogram of injecting DE5 (1.0 mM)
into a cell filled with P2 (0.05 mM) and Zn2þ (0.165 mM). (Upper
panel) Raw data for the sequential injection of a fixed volume (7 μL)
of DE5 solution into P2-Zn2þ solution. (Lower panel) Integrated
binding heat after subtracting the dilution control; the data were fit by
one set-of-sites model.
tional groups can be used as adsorbent models to investigate the
adsorption mechanism.25 In the present work, the linear copoly-
mers of functional monomers and acrylamide were used as the
adsorbent models; they have similar compositions to the corre-
sponding adsorbents except for the cross linker.
ITC is a routine method for the generation of thermodynamic
data relating to molecular association, and it has been used to
examine the binding of protein and peptide with polymers.26-28
ITC was also used in our previous study to examine the binding of
oligopeptides with linear polymers.29 Accordingly, in this study,
ITC was used to quantify the interaction between these models
and DE5 to investigate the adsorption mechanism. The heat flow
of each injection was measured when DE5 was injected into the
cell containing polymer and Zn2þ; the best Zn2þ concentration
had been determined in the previous study.15 All data were
analyzed using ITC analysis software after subtracting the re-
ference obtained by injecting DE5 into water (Figure 3). The
3.4. Adsorption Affinity from Model Study. As our earlier
study has demonstrated, linear polymers bearing the same func-
(26) Kimhi, O.; Bianco-Peled, H. Langmuir 2007, 23, 6329–6335.
(27) Asadi, A.; Saboury, A. A.; Moosavi-Movahedi, A. A.; Divsalar, A.;
Sarbolouki, M. N. Int. J. Biol. Macromol. 2008, 43, 262–270.
(28) Goncalves, E.; Kitas, E.; Seelig, J. Biochemistry 2006, 45, 3086–3094.
(29) Li, J. H.; Feng, J.; Dang, Q. Q.; Qiao, Y. T.; Zhao, J. X.; Zhang, S. H.; Sun,
H. W.; Wen, X.; Yuan, Z. Polymer 2009, 50, 1602–1608.
(25) Li, G. H.; Li, J. H.; Wang, W.; Yang, M.; Zhang, Y. W.; Sun, P. C.; Yuan,
Z.; He, B. L.; Yu, Y. T. Biomacromolecules 2006, 7, 1811–1818.
7184 DOI: 10.1021/la904272e
Langmuir 2010, 26(10), 7181–7187

Qiao et al.
Article
Table 2. Affinity Constants for the Interaction of P-Zn2þ with DE5 Derived from ITC at 30 °C
complex
P1-Zn-DE5
P2-Zn-DE5
P3-Zn-DE5
P4-Zn-DE5
P5-Zn-DE5
K/104 (M-1
)
8.68 ( 0.538
10.4 ( 0.510
7.80 ( 0.460
0.693 ( 0.319
0.864 ( 0.219
Table 3. Thermodynamic Parameters ΔH and TΔS for the Complexation of P1-Zn-DE5, P2-Zn-DE5, and P3-Zn-DE5 at Different Temperatures
as Derived from ITC
P1-Zn-DE5
ΔH (kcal mol-1 TΔS (kcal mol-1
P2-Zn-DE5
ΔH (kcal mol-1 TΔS (kcal mol-1
P3-Zn-DE5
ΔH (kcal mol-1 TΔS (kcal mol-1
)
temperature (°C)
)
)
)
)
)
10
20
30
50
-3.44 ( 0.0960
-4.56 ( 0.223
-5.97 ( 0.372
-6.56 ( 0.836
3.10
2.12
0.879
0.0650
-2.59 ( 0.191
-2.84 ( 0.320
-6.07 ( 0.325
-7.82 ( 0.634
4.08
3.93
0.885
-0.801
-3.59 ( 0.183
-4.16 ( 0.316
-4.41 ( 0.296
-5.55 ( 0.524
2.87
2.53
2.37
1.15
obtained affinity constant K values are summarized in Table 2.
Consistent with the adsorption results, P1-Zn2þ, P2-Zn2þ, and
P3-Zn2þ all showed a high affinity for peptide DE5, whereas P4-
Zn2þ and P5-Zn2þ exhibited a much weaker binding affinity
under the same conditions. The agreement of the adsorption
results with the affinity constant indicated that it was feasible to
investigate the adsorption interaction using linear models by ITC.
3.5. Investigation of the Adsorption Mechanism. Although
the coordination and hydrophobic interactions were the main
factors considered in the design of adsorbents and the adsorption
experiments have confirmed the efficiency of adsorbent CP1-
Zn2þ, CP2-Zn2þ, and CP3-Zn2þ, the actual weak interactions
involved still caught our attention.
To determine the adsorption mechanism, ITC experiments in
which DE5 was injected into P1-Zn2þ, P2-Zn2þ, and P3-Zn2þ
were performed at 10, 20, 30, and 50 °C. When the temperature
was increased from 10 to 50 °C, all ΔH values increased markedly,
as shown in Table 3. By plotting ΔH versus the tempera-
ture change, we calculated that the change in the heat capacity
(ΔCp) of P1-Zn-DE5, P2-Zn-DE5, and P3-Zn-DE5 was -88.83,
-134.1, and -47.92 cal mol-1 K-1, respectively; these values were
very negative and indicative of a classical strong hydrophobic
interaction.30-32 Furthermore, linear enthalpy-entropy compen-
sation was also observed for P1-Zn-DE5, P2-Zn-DE5, and P3-
Zn-DE5. It was found that all of the points for TΔS ≈ ΔH of the
three complexations lay on a regression line with slope and
intercept values of 0.942 and 6.452 (Figure 4). The enthalpy-
entropy compensation indicated that similar weak interactions
were formed in the binding of DE5 to P1-Zn2þ, P2-Zn2þ, and P3-
Zn2þ. Moreover, good enthalpy-entropy compensation oc-
curred in the presence of hydrophobic interactions.33 The slope
(0.942) demonstrated that only 6.16% of the entropy loss was
reflected in complex stability owing to compensation from the
gain in enthalpy.34 All of these analyses revealed that hydropho-
bic interactions were an important driving force in the adsorption
process
Figure 4. Enthalpy-entropy compensation plots for the com-
plexation of P1-Zn-DE5, P2-Zn-DE5, and P3-Zn-DE5. All points
lie on the regression line of TΔS = (0.942 ( 0.025)ΔH þ (6.452 (
0.127) (R = 0.996, P < 0.0001).
Figure 5 shows a plot of heat flow per mole of titrant versus the
molar ratio of P1-Zn2þ, P2-Zn2þ, P3-Zn2þ, P2, and P2-Zn2þ
/
Figure 5. Different ITC curves for the titration of DE5 into P1-
Zn2þ, P2-Zn2þ, P3-Zn2þ, P2, and P2-Zn2þ/EDTA after the sub-
traction of the reference titration.
EDTA (with concentration equal to that of Zn2þ) at 30 °C after
subtracting the reference titration. Negative deflection from zero
upon the addition of DE5 to P1-Zn2þ, P2-Zn2þ, and P3-Zn2þ
indicated the exothermic binding events, and the best fitting by
one set-of-sites model gave affinity constants of 8.68 ꢀ 104, 1.04 ꢀ
105, and 7.80 ꢀ 104 M-1. However, no obvious heat change was
observed without the addition of Zn2þ to the P2 interaction
system. The same phenomenon took place when adding equal
concentrations of EDTA to the cell filled with P2-Zn2þ, where
EDTA coordinated to Zn2þ and prevented the interaction of
Zn2þ with CP2 and DE5. Without the participation of Zn2þ, the
coordination interaction cannot form and led to a decrease in the
(30) Arouri, A.; Garidel, P.; Kliche, W.; Blume, A. Eur Biophys. J. 2007, 36, 647–
660.
(31) Sharrow, S. D.; Novotny, M. V.; Stone, M. J. Biochemistry 2003, 42, 6302–
6309.
(32) Srimal, S.; Surolia, N.; Balasubramanian, S.; Surolia, A. Biochem. J. 1996,
315, 679–686.
(33) Liu, L.; Yang, C.; Guo, Q. X. Bull. Chem. Soc. Jpn. 2001, 74, 2311–2314.
(34) Inoue, Y.; Hakushi, T.; Liu, Y.; Tong, L. H.; Shen, B. J.; Jin, D. S. J. Am.
Chem. Soc. 1993, 115, 475–481.
Langmuir 2010, 26(10), 7181–7187
DOI: 10.1021/la904272e 7185

Article
Qiao et al.
Table 4. Thermodynamic Parameters K, ΔH, and ΔS of P1-Zn-DE5, P2-Zn-DE5, and P3-Zn-DE5 at 30 °C at Different NaCl Concentrations
P1-Zn-DE5
P2-Zn-DE5
P3-Zn-DE5
NaCl
(mM) K/104 (M-1
ΔH
ΔS (cal
ΔH (kcal
ΔS (cal
ΔH (kcal
ΔS (cal
)
(kcal mol-1
)
mol-1 K-1
)
K/105 (M-1
)
mol-1
)
mol-1 K-1
)
K/104(M-1
)
mol-1
)
mol-1 K-1
)
0
50
100
150
8.68 ( 0.538 -5.97 ( 0.372
7.72 ( 0.238 -2.21 ( 0.520
4.19 ( 0.192 -1.14 ( 0.444
3.95 ( 0.254 -0.249 ( 0.0752
2.90
15.1
17.4
20.2
1.04 ( 0.0510 -6.07 ( 0.325
0.900 ( 0.137 -2.84 ( 0.398
0.855 ( 0.122 -2.20 ( 0.294
0.850 ( 0.160 -1.73 ( 0.202
2.92
13.3
15.3
16.8
7.80 ( 0.460 -4.41 ( 0.296
6.66 ( 0.678 -3.99 ( 0.271
3.15 ( 0.422 -1.88 ( 0.301
2.78 ( 0.292 -0.528 ( 0.107
7.83
8.90
14.4
18.6
Table 5. Binding Energy of CP1-Zn-DE5, CP2-Zn-DE5, and CP3-
Zn-DE5 by Molecular Docking
amount of heat released. Moreover, the only difference between
CP2 and CP4 was whether there was a coordination site N on the
adsorbent; the much higher adsorption capacity of CP2-Zn2þ
could be attributed to the coordination interaction. Similar
phenomena were also observed with the addition of equal con-
centrations of EDTA to the cell filled with P1-Zn2þ and P3-Zn2þ
or without the addition of Zn2þ to the interaction system of P1
and P3, and the binding heat sharply decreased to close to zero
just like for P2 (data not shown). These results indicated that the
coordination interaction played an important role in the adsorp-
electrostatic
energy (kcal mol-1
steric energy
(kcal mol-1
total energy
(kcal mol-1
)
complex
)
)
CP1-Zn-DE5
CP2-Zn-DE5
CP3-Zn-DE5
-22.32
-28.44
-21.34
-2.391
-4.502
2.030
-24.71
-32.92
-19.31
tion of DE5 onto CP1-Zn2þ, CP2-Zn2þ, and CP3-Zn2þ
.
Usually, the salt concentration has a substantial effect on the
adsorption capacity; high ionic strength would be expected to
interrupt the coordination interactions and affect hydrophobic
interactions. In this part, ITC experiments were performed at
various NaCl concentrations. Table 4 summarizes the changes in
the thermodynamic parameters. As shown in Table 4, for P1-Zn-
DE5, P2-Zn-DE5, and P3-Zn-DE5, ΔH decreased with increas-
ing NaCl concentration. As a contrast, ΔS obviously increased.
On the basis of the thermodynamic equation ΔG = -RT ln
K = ΔH - TΔS, the contribution of enthalpy to affinity binding
decreased but the contribution of entropy increased. These
changes may be attributed to the interference of Naþ and Cl-
in the solution, which may have disturbed the formation of the
coordination interactions and enhanced the hydrophobic inter-
action by increasing the polarity of the solution.35,36 Moreover, it
can be seen that the ionic strength caused a varying decrease in the
binding affinity, which was 54.5, 18.3, and 64.4% for P1-Zn-DE5,
P2-Zn-DE5, and P3-Zn-DE5, respectively. The decrease for P2-
Zn-DE5 was much less than that for P1-Zn-DE5 and P3-Zn-DE5.
The single difference among these three functional groups was
the interval between the hydrophobic and coordination interac-
tion sites. Thus, the degree of collaboration of these two weak
interactions was the crucial factor in the change in binding
affinity. The slightest decrease in binding affinity indicated that
P2-Zn-DE5 had the optimal degree of collaboration and P2-Zn2þ
could provide optimal matching with DE5 in the complexation.
Indeed, this optimal matching made CP2-Zn2þ have the best
adsorption capacity and adsorption affinity to DE5.
According to the analysis above, both coordination and
hydrophobic interactions took place in the adsorption of DE5
onto CP2-Zn2þ and the collaboration of these interactions re-
sulted in the high affinity constant, consistent with the rationale of
the preliminary design. Similar interactions should also occur
during the adsorption of DE8 and EV11. By forming coordina-
tion and hydrophobic interactions with the DE5 sequence on
these peptides, CP2-Zn2þ displayed selective adsorption toward
DE8 and EV11. SH5 has two acidic residues, but it lacks the
hydrophobic resides. The resulting low adsorption capacity indi-
cated that it was difficult to form stable binding by coordination
interactions only. WF4 and GF6, which contain multiple hydro-
Figure 6. Interaction conformation of CP2-Zn2þ with DE5 by
molecular docking (A) and schematic diagram for the adsorption
of DE5-containing peptides onto CP2-Zn2þ (B).
phobic residues, have the ability to form hydrophobic interac-
tions; however, only hydrophobic interactions cannot form
strong enough bonds, and their adsorption capacity on CP2-
Zn2þ was also lower than that of peptides containing the DE5
sequence. In summary, the optimal collaboration of multiple
interactions led to the tight binding of DE5-containing peptides
for CP2-Zn2þ adsorption.
The results of the adsorption mechanism study demonstrated
the effect of the coordination interaction of N (pyridyl) on the
functional groups. To understand the coordination structure of
the complex further, Hartree-Fock (HF) and density functional
theory (DFT) calculations were performed using the Gaussian 03
suite of programs. The results revealed that O (carbonyl on the
functional group) also participated in the coordination interac-
tions according to the optimized geometries by B3LYP/6-31G*;
this is similar to the reports from Tai et al. and Riihimki et al.37,38
On the basis of the results from HF and DFT analyses, molecular
docking analyses using a Silicon Graphics SGI INDIGO 2
workstation were carried out. As shown in Table 5, all of the
binding energy of CP1-Zn-DE5, CP2-Zn-DE5, and CP3-Zn-DE5
was negative, indicating favorable complexation. Among these,
CP2-Zn-DE5 exhibited the lowest binding energy, followed by
CP1-Zn-DE5 and CP3-Zn-DE5, and this trend was quite con-
sist with the results of the adsorption capacity and adsorption
affinity. The optimal matching of CP2-Zn2þ with DE5 led to the
excellent adsorption properties of CP2-Zn2þ to DE5 (Figure 6A).
A final schematic diagram of the interaction between CP2-Zn2þ
and DE5-containing peptides is shown in Figure 6B. When
(35) Rekharsky, M.; Inoue, Y.; Tobey, S.; Metzger, A.; Anslyn, E. J. Am. Chem.
Soc. 2002, 124, 14959–14967.
(36) Lin, F. Y.; Chen, W. Y.; Hearn, M. T. W. Anal. Chem. 2001, 73, 3875–3883.
(37) Tai, H. C.; Lim, C. J. Phys. Chem. A 2006, 110, 452–462.
(38) Riihima1ki, E. S.; Kloo, L. Inorg. Chem. 2006, 45, 8509–8516.
7186 DOI: 10.1021/la904272e
Langmuir 2010, 26(10), 7181–7187

Qiao et al.
Article
DE5-containingpeptides were adsorbed onto CP2-Zn2þ, multiple
interactions were formed between the DE5 sequence and CP2-
Zn2þ. Hydrophobic interactions occurred because of the aggre-
gation of hydrophobic amino acid residues and the phenyl group
of CP2-Zn2þ, and two carboxyl moieties of Asp and Glu were
bound to CP2-Zn2þ by coordination interactions. The collabora-
tion of these interactions and the geometrical matching of DE5
with CP2-Zn2þ led to the good selectivity and adsorption capacity
of CP2-Zn2þ to DE5-containing peptides.
phobic groups, respectively. In particular, CP2-Zn2þ showed the
optimal adsorption capacity for DE5, which was almost 8 times
that of CP4-Zn2þ or CP5-Zn2þ. Additionally, CP2-Zn2þ showed
selective adsorption toward DE5-containing peptides, with its
adsorption capacity being almost 2-6 times higher than that for
peptides without the DE5 sequence. Further mechanism investi-
gations using ITC and computer-aided analysis indicated that it
was the optimal collaboration of coordination and hydrophobic
interactions that contributed to the higher adsorption capacity
and selectivity of CP2-Zn2þ. Our results indicate a possible
strategy not only for the design of efficient affinity adsorbents
but also for understanding the adsorption mechanisms.
4. Conclusions
To eliminate the uremic middle molecular peptides containing
the DE5 sequence, five affinity adsorbents (CP1-Zn2þ, CP2-Zn2þ
,
CP3-Zn2þ, CP4-Zn2þ, and CP5-Zn2þ) were designed and pre-
pared. As expected, CP1-Zn2þ, CP2-Zn2þ, and CP3-Zn2þ bearing
both coordination and hydrophobic groups exhibited a better
adsorption ability with respect to DE5, compared with CP4-
Zn2þand CP5-Zn2þ bearing only coordination groups or hydro-
Acknowledgment. Y.Q. and J.Z. contributed equally to this
work. We are grateful to Dr. Zhaofeng Luo at the University of
Science and Technology of China for ITC testing. This work was
supported by the National Natural Science Foundation of China
(nos. 20634030 and 50573034).
Langmuir 2010, 26(10), 7181–7187
DOI: 10.1021/la904272e 7187