Switching biological functionalities of biointerfaces via dynamic covalent bonds

Article abstract of DOI:10.1039/c5tb02072g

The development of biointerfaces with switchable properties is of growing interest for fabricating advanced biomaterials since the adjustment of surface properties can be made on demand. Herein, we report a highly versatile approach for the preparation of

Full text of DOI:10.1039/c5tb02072g

Journal of
Materials Chemistry B
View Article Online
View Journal
PAPER
Switching biological functionalities of
biointerfaces via dynamic covalent bonds†
Cite this:DOI: 10.1039/c5tb02072g
Jie Deng,a Xinyue Liu,a Lang Ma,a Chong Cheng,*ab Shudong Suna and
Changsheng Zhao*a
The development of biointerfaces with switchable properties is of growing interest for fabricating
advanced biomaterials since the adjustment of surface properties can be made on demand. Herein, we
report a highly versatile approach for the preparation of a switchable biointerface through a dynamic
covalent bond (DCB). The switchability can be achieved via the reversible attaching/detaching of
aldehyde end-functionalized biomacromolecules onto/from an acylhydrazide anchored substrate surface. By
applying the DCB protocol, three types of well-designed aldehyde end-functionalized biomacromolecules
including aldehyde-poly(styrenesulfonate)-co-poly(ethylene glycol)methyl ether methacrylate (Ald-PSP, blood
compatible), aldehyde-poly([2-(meth acryloyloxy)ethyl]trimethylammonium chloride) (Ald-PMT, antibacterial),
and aldehyde-poly([2-(meth acryloyloxy)ethyl]trimethylammonium chloride-co-poly(ethylene glycol)methyl
ether methacrylate) (Ald-PMP, combined antifouling and antibacterial) could be reversibly and alterably
immobilized on the substrate surface by switching the pH conditions. As a result, we succeeded in altering
the biointerface performances; excellent blood compatibility, antibacterial ability, or combined antifouling and
antibacterial capabilities could be alterably achieved on the biointerface, which made the obtained material
interfaces more adaptable and capable of satisfying different biofunctional requirements. Moreover, many
other properties with specific biofunctions of interests can also be achieved via designing specific
aldehyde-terminated molecules.
Received 6th October 2015,
Accepted 17th December 2015
DOI: 10.1039/c5tb02072g
www.rsc.org/MaterialsB
(such as host–guest interactions,8,9) and dynamic covalent
bonds (DCBs, such as benzoic imine bonds,10 and Diels–Alder
Introduction
Stimuli responsive biointerfaces have made rapid advances in bonds11) have been employed to reversibly immobilize specific
recent years due to their prospective applications in the design functional groups or polymer brushes onto substrate surfaces.
of smart and tunable biomaterials.1–7 Traditionally, to achieve Thus, the surface performances such as switchable biofunction,
stable interface functionalization, functional groups are covalently tunable wettability, and reversible cell adhesion could be dra-
linked to biointerfaces, which usually need multi-steps of chemical matically altered. However, among the above methods, for the
synthesis to approach the functionalization. Moreover, once the host–guest interaction, the surface grafting density should be
functional groups have been covalently anchored onto the lower than that for the method by grafting to via DCBs due to
biointerfaces, the surface properties may no longer be altered the relatively large size of cyclodextrin or cucurbit[8]uril. Moreover,
on demand. Thus, the conventional methodologies for the surface the Diels–Alder bond based dynamic switchable surface system
modification have adversely affected the reusability and application usually requires multiple synthesis steps, which are not efficient
potentials in multiple fields of these biomaterials due to their enough. Meanwhile, the breaking of Diels–Alder bonds usually
untunable interface properties. Recently, to achieve a more needs high temperature conditions, which is not convenient or
switchable biointerface that could endow the biointerface suitable for designing the switchable surface in biomedical
with switchable properties, various noncovalent interactions applications.11,12 The benzoic-imine based DCB can undergo
an excellent pH responsive manner in water. However, the
a College of Polymer Science and Engineering, State Key Laboratory of Polymer
Materials and Engineering, Sichuan University, Chengdu 610065, China.
E-mail: sagecheng@163.com, chong.cheng@fu-berlin.de, zhaochsh70@163.com,
zhaochsh70@scu.edu.cn; Fax: +86-28-85405402; Tel: +86-28-85400453
b Department of Chemistry and Biochemistry, Freie Universitat Berlin, Takustr. 3,
14195 Berlin, Germany
linkage of the benzoic imine bond based DCB has been found
to be not stable enough in a physiological environment.13,14
More recently, the acylhydrazone bond has been known as a
prominent candidate for constructing a dynamic system, which
can be reversibly changed between hydrazone and its component
aldehyde and hydrazide by responding to a pH stimulus.15–18 In
† Electronic supplementary information (ESI) available: XPS spectra for the sub-
strates, 1H NMR for Ald-PMP, and GPC measurements. See DOI: 10.1039/c5tb02072g contrast to imines, hydrazones are more stable, allowing for their
This journal is ©The Royal Society of Chemistry 2015
J. Mater. Chem. B

View Article Online
Paper
Journal of Materials Chemistry B
isolation and analysis in aqueous solutions at around neutral
pH or under physiological conditions; however, which will undergo
hydrolysis with the trigger of acid.19,20 Using the hydrazone-based
DCBs, Mahon et al.21 have established polymer-scaffolded dynamic
combinatorial libraries in aqueous solution by reversible con-
jugation of different acylhydrazide residues onto aldehyde-
functionalized dimethylacrylamide-based polymer scaffolds. Li
et al.19 have also fine-tuned the properties of thermoresponsive
polymers by reversible conjugation of functional groups via this
dynamic acylhydrazone linkage. Moreover, because of the stable
and reversible characteristics, the acylhydrazone bond has been
widely employed as a dynamic linkage to construct self-healing
hydrogels.22,23 These hydrogels can undergo acylhydrazone
exchange to repair the damage in an acidic environment. Thus,
we hypothesize that the acylhydrazone bond may be an ideal
choice for constructing a stimuli-switchable biointerface that
allows for reversible and alterable immobilization of specific
functional polymer brushes.
To the best of our knowledge, only a few strategies have been
reported to immobilize some biotin ligands via an acylhydra-
zone bond to achieve functionalization.24–26 However, none of
these reports focused on the construction of smart biointerfaces
with switchable properties. Herein, we present a new switchable
functionalization strategy based on the DCB of an acylhydra-
zone bond that allows for reversible and alterable attaching/
detaching of desired biofunctional polymer brushes. Thus, the
surface performance would be dramatically altered to meet
different applications on demand. In this study, adipic acid
dihydrazide (AD) was anchored onto a 3-glycidoxypropyl trimethoxy-
silane (GPS) immobilized silicon wafer substrate (SW-GPS) to create
an acylhydrazide functionalized surface (SW-AD). Three different
kinds of benzaldehyde-terminated functional polymers including
aldehyde-poly(styrenesulfonate-co-poly(ethylene glycol)methyl ether
methacrylate (Ald-PSP), aldehyde-poly([2-(meth acryloyloxy)ethyl]tri-
methylammonium chloride) (Ald-PMT), and aldehyde-poly([2-(meth
acryloyloxy)ethyl]trimethylammonium chloride-co-poly(ethylene
glycol)methyl ether methacrylate) (Ald-PMP) were then alternately
immobilized onto the SW-AD surface to achieve the switchable
biofunctional interfaces. In detail, the sulfonic groups in Ald-PSP
could mimic the biofunctionalities of heparin (such as antic-
oagulant and antithrombotic capabilities);27 while the positively
charged Ald-PMT with quaternary ammonium groups has been
widely used to construct antibacterial surfaces.28 To improve the
antifouling property of the Ald-PMT immobilized surface, poly-
(ethylene glycol)methyl ether methacrylate) (PEGMA) was used
to copolymerize with [2-(meth acryloyloxy)ethyl]trimethylammonium
chloride to obtain dual-biofunctional Ald-PMP.29 Thus, the
biofunctionalities of the interfaces could be facilely regulated
by alternate immobilization of these three types of Ald-polymers
(Scheme 1).
Scheme 1 Schematic depiction of pH-triggered alternate and reversible
assembly of Ald-PSP, Ald-PMT, and Ald-PMP onto SW-AD.
activation of these functionalized surfaces were evaluated in
detail by plasma protein adsorption, platelet adhesion, human
platelet factor 4, clotting time, coagulation activation, and
blood-related complement activation. Escherichia coli (E. coli)
and Staphylococcus aureus (S. aureus) were selected as the model
bacteria to evaluate the bactericidal capability of these interfaces.
Antifouling capability of these functionalized surfaces was also
investigated by combining the results from protein adsorption,
platelet adhesion and bacterial adhesion measurements. The
proposed advanced methodology may offer a dynamically program-
mable design criterion for the functionalization of biomaterials.
Results and discussion
Synthesis of Ald-Br and Ald-polymers (Scheme 2)
The aldehyde functionalized initiator (Ald-Br) was synthesized
by the esterification reaction of 4-hydroxybenzaldehyde and
1
a-bromoisobutyryl bromide (BIBB). Fig. 1A shows the H NMR
spectrum of the Ald-Br. The peak at about 2.07 ppm was
ascribed to the methyl of BIBB, while the peak at about 10 ppm
was assigned to the aldehyde of benzaldehyde. The results were
consistent with the previous report,30 and the peak integral ratios
were consistent with the chemical structure of the target product.
Generally, active polymerization is an efficient method to
synthesize specific functional group terminated polymers. In
this study, Ald-polymers are synthesized via atom transfer
radical polymerization (ATRP) from Ald-Br. Thus, the aldehyde
terminated polymer brushes can be immobilized onto an
acylhydrazide functionalized surface via acylhydrazone bonds.
1
Fig. 1B–D show the H NMR spectra of Ald-PSP, Ald-PMT, and
Ald-PMP, respectively, which indicate that the Ald terminated
polymers have been successfully synthesized.
To this end, we have illustrated a new strategy to construct a
biointerface with switchable biological functionalities; then their
Fabrication of an acylhydrazide anchored surface
propensities to undergo a reversible and alterable immobilization First, pristine silicon wafers were treated with piranha solution
manner were evaluated by means of atomic force microscopy (H2SO4/H2O2 7 : 3) to generate a hydroxylated silicon surface.
(AFM), X-ray Photoelectron Spectroscopy (XPS), and water contact After the treatment via piranha solution, the surface became
angle measurements. Moreover, the blood compatibility and highly hydrophilic and the water contact angle (WCA) decreased
J. Mater. Chem. B
This journal is ©The Royal Society of Chemistry 2015

View Article Online
Journal of Materials Chemistry B
Paper
Acylhydrazone linkages are more stable in water than the corres-
ponding imines, and considered to be stable at neutral pH. Over
the past several years, acylhydrazones have been extensively
utilized to construct polymer-scaffolded dynamic combinatorial
libraries, but with very limited exploration reported on reversible
and alterable surface immobilization. Thus, in this study, three
types of Ald-terminated polymers (Ald-PSP, Ald-PMT, and Ald-
PMP) were alternately immobilized onto an acylhydrazide
functionalized surface to achieve a switchable surface with
changing biofunctionalities.
AFM was first utilized to observe the surface morphology
variations at different modification stages. As shown in Fig. 2,
before immobilization, the surface of SW-AD is flat and even.
However, with the immobilization of PSP (SW-PSP), the surface
became rougher, and some spherule-like structures could be
obviously observed, which indicated that the Ald-PSP brushes
had been successfully anchored onto the SW-AD surface. After
detachment of the PSP, the surface returns to a flat and smooth
appearance. The surface anchored with PMT (SW-PMT) or PMP
(SW-PMP) showed similar results as that of SW-PSP. Moreover, as
shown in Fig. 2H, after anchoring the Ald-polymers, the surface
root-mean-square (Rms) roughness dramatically increased when
compared to that of the SW-AD and Ald-polymer detached surface.
Moreover, XPS analysis was performed to detect the chemical
composition variations of the surfaces at different modification
stages. As shown in Fig. S1 (ESI†), no N element can be detected
for SW-GPS; while for other substrates, the signals for N can be
obviously observed in the N 1s spectra. However, as shown in
the N 1s spectrum for SW-AD in Fig. S1B (ESI†), the peak
of protonated amino (binding energy: 402 eV) can be observed
and the peak intensity (H3N+–) is comparative with the peak
Scheme 2 Synthetic routes for Ald-Br, Ald-PSP, Ald-PMT, and Ald-PMP.
Fig. 1 1H NMR spectra for Ald-Br (A) in CD3Cl, Ald-PSP (B) in D2O,
Ald-PMT (C) in D2O, and Ald-PMP (D) in D2O.
to about 81. With the anchor of GPS, the WCA for the surface
increased dramatically. Afterward, AD was anchored onto the
surface via the reaction between amino and epoxy groups. Then,
the new peak of N for the SW-AD in the XPS N 1s spectrum could
be obviously observed compared with that for SW-GPS (Fig. S1,
ESI†). Moreover, after the immobilization of AD, the WCA also
dramatically decreased (Fig. 3). Thus, it is confirmed that the AD
has been successfully anchored onto the silicon surface.
Reversible immobilization of Ald-polymers and surface
characterization
The acylhydrazone bond has been identified as a promising
dynamic covalent bond to reversibly conjugate residues onto an
appropriately functionalized polymer scaffold.31 Under mildly
acidic conditions, acylhydrazone linkages can be generated by
condensation of aldehydes with acylhydrazides; while, when the
Fig. 2 AFM images of the substrates for SW-AD (A), SW-PSP (B), Ald-PSP
detached surface (SW-DPSP, (C)), SW-PMT (D), Ald-PMT detached surface
(SW-DPMT, (E)), SW-PMP (F), and Ald-PMP detached surface (SW-DPMP, (G));
pH is lower than 4, these linkages would undergo hydrolysis.19 Rms roughnesses for the substrates at different modification stages (H).
This journal is ©The Royal Society of Chemistry 2015
J. Mater. Chem. B

View Article Online
Paper
Journal of Materials Chemistry B
of N–C (binding energy: 400 eV).9 After the immobilization of the switching ability of blood compatibility of these as-prepared
Ald-PSP, the peak intensity for the protonated amino (H3N+–) samples, the platelet adhesions on the functionalized surfaces
decreases compared to that for the N–C (Fig. S1C, ESI†). After were evaluated and quantified by SEM. As shown in Fig. 4, for
detaching the Ald-PSP, the curve of the N 1s spectrum becomes the surfaces of SW-AD and SW-PMT, substantive platelets
similar to that for SW-AD. Moreover, as shown in Fig. S2 (ESI†), accumulated and aggregated on the surfaces. For the pristine
the S element for the SW-PSP can be obviously observed. Thus, SW surface, although there were a large number of platelets
the Ald-PSP has been confirmed to show a reversible attaching/ that adhered on the surface, many of the platelets maintained
detaching manner on the SW-AD surface. The procedures for rounded shape, indicating that moderate platelet activation
attaching and detaching of the Ald-PMT show similar results as had occurred. While, the platelet adhesion on the surface of
that of the Ald-PSP except the dramatic increase of N+–C intensity the SW-PSP had been successfully suppressed; no obvious
(binding energy: 402.2 eV). Similarly, with the attaching of pseudopodia could be observed, which should be attributed
Ald-PMP, the peak for N+–C (binding energy: 402.2 eV) becomes to the introduction of the antifouling PEGMA segments in
more obvious compared to that of the Ald-PMT detached surface.32 Ald-PSP. When the surface was switched to SW-PMT, the
In addition, compared to SW-GPS, the relative content for the number of adhered platelets on the surface dramatically
CQO bond (binding energy: 288.2 eV) obviously increased (Fig. S3, increased and the platelets on the surface were activated with
ESI†).9 Thus, from the XPS analysis results, we can conclude that deformed shapes. The quaternary ammonium enriched surface
the Ald-polymers have been successfully attached/detached onto/ has been previously demonstrated to have a deleterious effect
from the acylhydrazide functionalized surface.
on platelets.27 However, with the substituent of Ald-PMT by
Ald-PMP, the number of adhered platelets on the surface of
SW-PMP dramatically decreased again. Thus, the antifouling
capability for the SW-PMP has been improved compared to that
of SW-PMT. The platelet adhesion property on the surface has
been successfully regulated via the reversible immobilization
of these three types of Ald-polymers. To further evaluate the
platelet activation of the samples, human platelet factor 4 (PF4)
concentrations for SW, SW-PSP, and SW-PMP were evaluated,
since PF4 has been reported to play a vital role in inflammation
and injury response.34 Fig. 4G shows the PF4 concentrations in
plasma after the samples contacted with blood. It could be
observed that the PF4 concentration for the SW-PSP was lower
than that of blood and SW; thus, the platelet adhesion on the
SW-PSP surface was not activated. While, for the SW-PMP, the
PF4 concentration was comparative with that for the blood and
the platelets were not heavily activated by the SW-PMP.
Surface wettability
The surface hydrophilic/hydrophobic property is a vital character
for biomaterials. The pH-switchable wettability of the as-prepared
surfaces was evaluated via a static WCA test. As shown in Fig. 3,
the WCA for the SW-AD was about 54 ꢀ 1.51. After the anchoring
of Ald-PSP, the surface became more hydrophilic (47.4 ꢀ 1.51);
while, after the detaching of the Ald-PSP, the WCA returned to the
level of SW-AD again. Similar results have been obtained from
the attaching/detaching of Ald-PMT and Ald-PMP. Thus, the
wettability transitions were fully reversible and the surface
wettability could be regulated via this reversible immobilization
of the Ald-polymers.
Platelet adhesion and activation
Platelet adhesion is a vital event that induces thrombus for-
mation and blood clotting at biointerface, which is related to
the acceleration/prevention of blood clotting.33 To evaluate the
Protein adsorption
For blood contacting materials, protein adsorption is another
key event with respect to blood compatibility.35 It has been
known that the hydrophilic surface exhibits better antifouling
capability.36 In the present study, both the Bovine serum
albumin (BSA) and bovine serum fibrinogen (FBG) adsorptions
were detected when switching the surface polymer brushes, and
the results are shown in Fig. 5A. The pristine SW substrate
exhibited the highest level of protein adsorption. After the
immobilization of Ald-PSP, the protein adsorption for SW-PSP
dramatically decreased compared to those for SW-AD and SW,
which should be attributed to the PEGMA segments in the polymer
chains. After detaching the PSP and subsequently attaching the
PMT, the protein adsorption on the SW-PMT surface dramatically
increased. Although the surface of the SW-PMT was the most
hydrophilic among these substrates, the protein adsorption was
comparable with that for the SW-AD and higher than that for
the SW-PSP, which could be due to the electrostatic interaction
between the positively charged surface of SW-PMT and the
negatively charged protein. However, when switched to PMP
polymer brushes (with copolymerization of antifouling PEGMA
Fig. 3 Static water contact angles for SW, SW-OH, SW-GPS, SW-AD,
SW-PSP, SW-DPSP, SW-PMT, SW-DPMT, SW-PMP, and SW-DPMP.
J. Mater. Chem. B
This journal is ©The Royal Society of Chemistry 2015

View Article Online
Journal of Materials Chemistry B
Paper
Fig. 4 SEM images of platelets adhered on the SW (A, a), SW-AD (B, b), SW-PSP (C, c), SW-PMT (D, d), and SW-PMP (E, e); average numbers of the
platelets adhered on the surfaces at different modification stages from PRP calculated form six SEM images (F), values are expressed as means ꢀ SD,
n = 3, *P o 0.05 vs. SW; generated concentrations of PF4 of these samples with whole blood, which can act as an index for the platelet activation (G),
values are expressed as means ꢀ SD, n = 3, #P o 0.05 vs. blood.
Fig. 5 BSA and FBG adsorption amounts on the substrates (A); values are expressed as means ꢀ SD, n = 3, *P, #P o 0.05 vs. SW. APTT, PT, and TT assays
of the pristine and functionalized surfaces (B); values are expressed as means ꢀ SD, n = 3, **P, ##P o 0.05 vs. PPP. TAT generation of all the samples with
whole blood flowing for 3 h (C); values are expressed as means ꢀ SD, n = 3, &P o 0.05 vs. blood. Generated concentrations of C3a after the samples were
contacted with whole blood (D); values are expressed as means ꢀ SD, n = 3, ###P o 0.05 vs. blood. Generated concentrations of C5a after the samples
were contacted with whole blood (E); values are expressed as means ꢀ SD, n = 3, ***P o 0.05 vs. blood.
This journal is ©The Royal Society of Chemistry 2015
J. Mater. Chem. B

View Article Online
Paper
Journal of Materials Chemistry B
segments in Ald-PMT), the antifouling capability of the sub-
strate, SW-PMP, was retained again, which further verified that
the protein repelling capability was ascribed to the antifouling
PEG segments in the polymer chain.
Clotting time
Activated partial thromboplastin time (APTT) and prothrombin
time (PT) have been widely used to detect the coagulation
abnormalities in the intrinsic and extrinsic pathway, respectively.37,38
While thrombin time (TT) is a commonly adopted method to
evaluate the anticoagulation ability of blood contacting materials;
TT is mainly affected by the content and coagulation activity of
fibrinogen in plasma.39 The TT value reflects the activity of the
fibrinolytic system in the common pathway of the coagulation
cascade. Fig. 5B shows the results of APTT, TT, and PT measure-
ments for the samples immobilized with different Ald-polymers.
It could be observed that the clotting times for the SW and
SW-AD showed no significant variations compared to those for
platelet poor plasma (PPP).27 However, with the immobilization
of Ald-PSP, the APTT and PT dramatically increased to 136 and
33 s, respectively, which indicated that the PSP brushes showed
excellent heparin-mimicking anticoagulant activity. Meanwhile,
when the surface was switched to SW-PMT or SW-PMP, no
anticoagulant ability is detected. Thus, the clotting times of
the functionalized surfaces could be regulated via the reversible
and alterable immobilization of these three types of Ald-polymers.
Moreover, Thrombin generation was evaluated via the detection of
the thrombin–antithrombin III complex (TAT) concentration, since
the thrombin would be immediately neutralized by antithrombin
III, resulting in TAT complex formation. As shown in Fig. 5C, the
TAT concentration for the SW-PSP was lower than that for blood
and SW, which indicated that the TAT generation by the SW-PSP
was successfully suppressed by the heparin-mimicking PSP
brushes.
Fig. 6 Typical fluorescence microscopy images of the attached E. coli
and S. aureus on the switchable biointerfaces before and after immobiliza-
tion of functional polymer brushes. Green staining indicates live bacteria,
and red staining indicates dead bacteria.
were stained with red fluorescent nucleic acid stain propidium
iodide and green fluorescent nucleic acid stain SYTO9, respectively.
It could be observed that a large number of live bacteria were
adhered on the pristine and AD anchored silicon surfaces.
With the immobilization of Ald-PSP, the number of adhered
Complement activation in the human blood system
Generally, complement activation could be measured via C3a,
C4a, and C5a.40 In the present study, C3a and C5a measure-
ments were carried out to determine the complement activation
via an ELISA assay. Fig. 5D and E show the results of C3a and
C5a concentrations, respectively. It was found that the C3a
concentrations for the SW, SW-PSP, and SW-PMP showed no
significant variations, which were comparative with that for
blood. While the C5a concentrations for the SW, SW-PSP, and
SW-PMP were lower than that for blood. Thus, the functional
surfaces of SW-PSP and SW-PMP might show low blood-related
complement activation, which might suppress the inflammation
responses when contacting with blood.
Bactericidal activity
Biointerfaces with controllable bacterial adhesion and killing
are of vital importance in the design of antibacterial biomaterials.
In this study, we can facilely control the antibacterial adhesion and
bactericidal performance by switching the grafted polymer brushes
via DCBs. The results of bactericidal activity evaluation of these
Fig. 7 Calculated numbers of bacteria attachment on the substrate surfaces
based on the live/dead bacteria tests; E. coli (A), S. aureus (B). Error bars
represent the standard deviation of the mean (n = 3). *P, **P, ***P, #P, ##P,
functional surfaces are shown in Fig. 6. Dead and live bacteria ###P o 0.05 vs. SW.
J. Mater. Chem. B
This journal is ©The Royal Society of Chemistry 2015

View Article Online
Journal of Materials Chemistry B
Paper
bacteria dramatically decreased (Fig. 6 and 7), which was consistent sulfonate (SSNa, 90%), poly(ethylene glycol) methyl ether
with the tendency of platelet adhesion and protein adsorption methacrylate (PEGMA, Mw = 475), [2-(methacryloyloxy)ethyl]tri-
evaluations. To achieve a more bactericidal property, the surface methylammonium chloride solution (MT, 75 wt% in water), and
was switched from SW-PSP to SW-PMT via the substitution of N,N,N0,N00,N00-pentamethyldiethylenetriamine (PMDETA, 99%)
the PSP brushes with PMT brushes. As shown in Fig. 6, the were purchased from Aladdin (Shanghai, China). Toluene,
number of dead bacteria adhering on the SW-PMT surface tetrahydrofuran (THF, AR) and triethylamine (TEA, AR) were
dramatically increases compared to that for SW-PSP. However, purchased from Chengdu Kelong Chemical Reagent Co. Ltd
after the substitution of the Ald-PMT with Ald-PMP, both the live (China), and were distilled before use. The cyclic ligand
and dead bacteria adhering on the surface of SW-PMP obviously 5,7,7,12,14,14-hexamethyl-1,4,8,11-tetraazacyclotetradecane
decreased compared to that for the SW-PMT, which should be (Me6TATD) was synthesized according to a previous paper.41
ascribed to the antifouling capability of the PEGMA segments; thus CuBr (AR, Aladdin) was purified by stirring in glacial acetic acid
the combined antibacterial and antifouling capabilities were overnight, filtered and washed with ethanol, which was then
achieved via the anchoring of PMP brushes on the surface. In dried at 30 1C under vacuum for 24 h. Bovine serum albumin
general, the bacterial adhesion and bactericidal capabilities could (BSA, fraction V, 95%) and bovine serum fibrinogen (FBG) were
be facilely regulated via the switching immobilization of Ald-PSP, purchased from Sigma Aldrich, USA. Micro BCAt Protein Assay
Ald-PMT, and Ald-PMP.
Reagent kits were obtained from PIERCE. Dialysis membranes
(MWCO = 3500 Da) were purchased from Solarbio (Canada).
Water used throughout this study was ultrapure water, which
was obtained from a pure water production system. All the other
reagents (analytical grade) were purchased from Kelong Chemical
Reagent Co. Ltd, and were used without further purification.
Conclusion
In conclusion, we have demonstrated a new reversible surface
functionalization strategy to achieve switchable biointerfaces
via acylhydrazone bonds. Different Ald-polymers could be
reversibly and alterably immobilized onto the acylhydrazide
functionalized surface; thus, the surface performances could
be dramatically altered with the immobilization of different
Ald-polymers. XPS data, AFM images, and WCA measurements
confirmed the successful, reversible, and alterable immobilization of
the Ald-polymers. Moreover, platelet adhesion, protein adsorption,
clotting times, complement activation, and bactericidal activity were
carried out to evaluate the alterable biofunctions when switching the
grafted functional polymer brushes. The results indicated that the
blood compatibility of the surface was dramatically enhanced by the
immobilization of Ald-PSP brushes. Afterward, by the substitution of
Ald-PSP with Ald-PMT, the surface performance switched from blood
compatible to antibacterial. Moreover, to achieve a surface with
combined antibacterial and enhanced antifouling capabilities,
Ald-PMP was subsequently immobilized onto the surface to replace
the Ald-PMT. Thus, the surface performance dramatically altered
with the reversible and alterable immobilization of different types
of functional Ald-polymers. This advanced functionalization
methodology provides a dynamically programmable design
criterion for biomaterials, such as blood compatibility, antifouling,
and antibacterial capabilities. Moreover, other properties with
specific functions of interests can also be included via the
design of aldehyde-terminated functional molecules.
Silicon wafer substrates
The pretreatments of silicon wafers were according to our
previous report.9 Briefly, silicon wafers were sequentially washed
with acetone, ethanol, and ultrapure water via sonication, 30 min
for each. The silicon wafers were then immersed into a freshly
prepared piranha solution ((H2SO4 (98%)/H2O2 (30%) 7 : 3 v/v)) at
90 1C for 150 min. Caution! Piranha solution is an extremely strong
oxidant and should be handled only with the appropriate equip-
ment. Afterward, the substrates were washed with a copious
amount of ultrapure water and dried under a flow of ultrapure
N2 (99.99%). The substrates were then immediately used for
further modification.
Preparation of SW-AD
The piranha solution treated silicon wafers were immediately
used to couple with GPS. Briefly, ten pieces of silicon wafers
were immersed in 40 mL of 10% (v/v) GPS in anhydrous
toluene, and the reaction was performed at 80 1C for 12 h with
stirring under a N2 atmosphere. Then, the GPS coupled silicon
wafers were thoroughly rinsed and sonicated in pure anhydrous
toluene for three times to remove any adhered GPS and then
dried under nitrogen. Afterward, ten pieces of SW-GPS were
immersed in 50 mL of dimethyl sulfoxide containing 0.5 g of
AD. The reaction was carried at 60 1C for 24 h with stirring
under a N2 atmosphere to obtain SW-AD. Then, the substrates
were washed with a copious amount of water to remove any
physically attached molecules.
Experimental
Materials and reagents
Synthesis of Ald-Br
Non-doped silicon wafers (h1 0 0i, n-type, size: 10 ꢁ 10 ꢁ
0.5 mm3, surface: both sides polished) were purchased from The procedure for the synthesis of Ald-Br was according to the
Hefei Crystal Technical Material Co. Ltd (China). 3-Glycidoxypropyl literature.30,42 Briefly, 4-hydroxybenzaldehyde (5 g, 41 mmol)
trimethoxysilane (GPS, 499%), adipic acid dihydrazide (AD, was dissolved in 180 mL of dry THF containing 10 mL of TEA.
HPLC, 499%), 4-hydroxy benzaldehyde (AR, 98%), a-bromo- Afterward, BIBB (6 g, 48.7 mmol) in 20 mL of dry THF was
isobutyryl bromide (BIBB, 98%), sodium 4-vinylbenzene- added slowly to the above solution at 0 1C. The reaction was
This journal is ©The Royal Society of Chemistry 2015
J. Mater. Chem. B

View Article Online
Paper
Journal of Materials Chemistry B
maintained at 0 1C for 2 h and at room temperature for another (FE-SEM) images were acquired by using a scanning electron
24 h. Then the white precipitate was filtered off, and the microscope (JSM-7500F, JEOL, Japan).
remaining solution was concentrated under vacuum. The oil
The surface contact angles of double-distilled (DI) water
like product was dissolved in 400 mL of CH2Cl2, and then on the substrates surfaces were measured via a contact angle
washed with saturated sodium bicarbonate solution (3 ꢁ 100 mL). goniometer (OCA20, Dataphysics, Germany).
The organic layer was collected and dried with anhydrous
Clotting time
magnesium sulfate overnight, filtered and then concentrated
under vacuum to obtain a yellowish oil product. For further APTT, TT, and PT tests were performed to evaluate the anticoagulant
purification, the product was purified using silica gel chromato- capability of the functionalized surfaces. The as-prepared samples
graphy with petroleum and ethyl acetate (20 : 1) mixture as the were first immersed in a phosphate buffer (pH 7.4) overnight at 4 1C,
eluent and then vacuum dried to obtain a white solid.
followed by incubation at 37 1C for 1 h. One piece of each sample
was then placed in a 24-well cell culture plate. Afterward, PPP
(220 mL) was added and incubated in the well at 37 1C for 0.5 h.
The incubated PPP solutions were then separately collected and
analyzed through an automatic coagulation analyzer (sysmex
CA-530, Japan). The clotting times for APTT, PT and TT were
automatically collected and printed by the equipment. The
assays were performed in triplicate for each sample.
Synthesis of Ald-polymers via ATRP
For Ald-PSP, briefly, Ald-Br (0.1356 g, 0.5 mmol) was dissolved
in 25 mL of DMSO/H2O (4 : 1) (for the synthesis of Ald-PMT and
Ald-PMP, 25 mL of DMSO was added). Then, 2.516 g of SSNa
and 2.484 g of PEGMA were added to the above solution (for
Ald-PMT, 6.67 g of MT was added; for Ald-PMT, 3.35 g of MT
and 2.484 g of PEGMA was added). Afterward, Me6TATD (0.17 g,
0.6 mmol) (for Ald-PMT and Ald-PMP, PMDETA, 0.152 g,
0.88 mmol was added) and CuBr (0.0715 g, 0.5 mmol) were
added under a N2 atmosphere. The Schlenk tubes were closed
by a three-way stopcock. The tubes were sealed off in a N2
atmosphere by three freeze–pump–thaw cycles. The reactions
were carried out at 70 1C for 24 h and then terminated by
exposing to air. The obtained polymer solutions were purified
via dialysis (MWCO = 3500 Da) against water for three days. The
resultant solutions were freeze-dried to obtain the products.
Platelet adhesion
Platelet adhesion experiments were carried out by using
platelet-rich plasma (PRP), which was obtained via centrifugation
of fresh human blood (male, 26 years old, collected using a clinical
anticoagulant vacuum tube) at 1000 rpm for 15 min. Afterward,
1 mL of the fresh PRP was added to each well containing one piece
of normal saline pre-equilibrated sample and then incubated at
37 1C for 2 h. Subsequent procedures for the treatment of the
substrates were described in detail in our previous report.27
FE-SEM was applied to observe the platelets adhered on these
substrates. The average number of platelets adhered on each
sample was calculated via six SEM images at 500ꢁ magnification
at different places.
Surface reversible and alterable functionalization
First, 0.2 g of the Ald-polymer was dissolved in 20 mL of 0.1 M
NH4OAc/AcOH (pH 4.5) to obtain its aqueous solution.43 The
SW-AD was immersed into the solution of Ald-PSP for 12 h at Protein adsorption
room temperature. After the immobilization, the substrate was
BSA and FBG were employed as two model proteins to measure
rinsed with a copious amount of water and dried under N2 to
obtain SW-PSP. Afterward, the SW-PSP was immersed into 1 M
HCl for 2 h at room temperature to obtain SW-DPSP. To
fabricate SW-PMT, the SW-DPSP substrate was immersed into
the solution of Ald-PMT for 12 h at room temperature, then
washed with water and dried under N2. Afterward, the SW-PMT
was immersed into 1 M HCl for 2 h at room temperature to
obtain SW-DPMT. Subsequently, the SW-DPMT substrate was
immersed into the solution of Ald-PMP for 12 h at room
temperature, followed by washing with a copious amount of
water and then drying under N2 to obtain SW-PMP. Then, the
SW-PMP was immersed into 1 M HCl for 2 h at room temperature
to obtain SW-DPMP.
the protein adsorption properties of the pristine and functionalized
silicon substrates. Briefly, every 2 mL of BSA (or FBG) in phosphate
buffer solution with a concentration of 1 mg mLꢂ1 was added to a
24-well cell culture plate containing one piece of normal saline
pre-equilibrated sample each well. The plate was kept static
for 1 h to reach the adsorption equilibrium. Afterward, the
substrates were scrupulously rinsed with PBS solution and DI
water, and then put in a new 24-well cell culture plate followed
with the addition of 2 mL of 2% sodium dodecyl sulfate solution
to each well. The plate was shaken at 37 1C for 2 h to wash down
the adsorbed protein on the surface. The protein concentration
of the washing solution was determined via the Micro BCATA
Protein Assay Reagent Kit.
Characterization
1H NMR spectra were acquired via a Bruker spectrometer
Contact activation of the coagulation system and complement
activation in the human blood system
(400 MHz). An X-ray photoelectron spectrometer (XSAM800, Platelet activation and coagulation activation were evaluated by
Kratos Analytical, UK) was applied to confirm the surface PF4 and TAT, respectively, through commercial enzyme-linked
compositions of the substrates. AFM images were obtained via immunosorbent assays (ELISA, CUSABIO BIOTECH CO., Ltd,
a Multimode Nanoscope V scanning probe microscopy system China). The same as the PF-4 and TAT detections, the complement
(Bruker, USA). Field emission scanning electron microscope activation (C3a and C5a) evaluation for these samples was also
J. Mater. Chem. B
This journal is ©The Royal Society of Chemistry 2015

View Article Online
Journal of Materials Chemistry B
Paper
carried out using the ELISA method.40 Briefly, whole blood was
incubated with the as-prepared samples for 2 h and then centrifuged
at 1000g (4 1C) centrifugal force for 15 min to obtain plasma.
Afterward, evaluations were performed according to the respective
instructions from the manufacturer.
6 G. Pan, B. Guo, Y. Ma, W. Cui, F. He, B. Li, H. Yang and
K. J. Shea, J. Am. Chem. Soc., 2014, 136, 6203–6206.
7 X. Du, J. Li, A. Welle, L. Li, W. Feng and P. A. Levkin, Adv.
Mater., 2015, 27, 4997–5001.
8 C. Hu, F. Tian, Y. Zheng, C. S. Y. Tan, K. R. West and
O. A. Scherman, Chem. Sci., 2015, 6, 5303–5310.
9 J. Deng, X. Liu, W. Shi, C. Cheng, C. He and C. Zhao, ACS
Macro Lett., 2014, 3, 1130–1133.
Bactericidal activity
The method for determining the bactericidal capability was
according to the previous report.44,45 Briefly, E. coli (or S. aureus)
was first cultured in a pure culture at 37 1C on a Luria–Bertani
(LB) agar plate. One colony was utilized to cultivate with 50 mL
of LB medium at 37 1C for 22 h. Subsequently, the bacterial
suspension was centrifuged at 8000 rpm for 10 min to obtain
the cell pellet, which was then washed with normal saline three
times, subsequently suspended in normal saline to obtain a
final suspension with a concentration of 108 cells per mL. Then,
the sterilized samples were put in a 24-well cell-culture plate,
test surface up. Every 2 mL of the above bacterial suspension
was added to each well, and then incubated at 37 1C for 2 h
without shaking. Afterward, the samples were scrupulously
rinsed with normal saline to remove loosely attached cells. A
live/dead staining assay was actualized by using the BacLight
viability kit to determine the bactericidal activity of the sample
surface. Briefly, each sample was immersed in 1 mL of staining
solution containing 3 mL of a 1 : 1 mixture of propidium iodide
and SYTO9 in normal saline, incubated in the dark for 15 min,
and then rinsed with normal saline, followed by examination
using fluorescence microscopy. Three separate samples were
analyzed for each type of substrate to obtain reliable data.
¨
10 L. Tauk, A. P. Schroder, G. Decher and N. Giuseppone, Nat.
Chem., 2009, 1, 649–656.
11 W. Laure, P. Woisel and J. l. Lyskawa, Chem. Mater., 2014,
26, 3771–3780.
12 Y. Imai, H. Itoh, K. Naka and Y. Chujo, Macromolecules,
2000, 33, 4343–4346.
13 C. Godoy-Alcantar, A. K. Yatsimirsky and J. M. Lehn, J. Phys.
Org. Chem., 2005, 18, 979–985.
14 A. Dirksen, T. M. Hackeng and P. E. Dawson, Angew. Chem.,
Int. Ed., 2006, 45, 7581–7584.
15 V. T. Bhat, A. M. Caniard, T. Luksch, R. Brenk, D. J. Campopiano
and M. F. Greaney, Nat. Chem., 2010, 2, 490–497.
16 W. G. Skene and J.-M. P. Lehn, Proc. Natl. Acad. Sci. U. S. A.,
2004, 101, 8270–8275.
17 L. Li, J. Gu, J. Zhang, Z. Xie, Y. Lu, L. Shen, Q. Dong and
Y. Wang, ACS Appl. Mater. Interfaces, 2015, 7, 8033–8040.
18 A. G. Olive, N. H. Abdullah, I. Ziemecka, E. Mendes,
R. Eelkema and J. H. Van Esch, Angew. Chem., Int. Ed.,
2014, 53, 4132–4136.
19 J. Li, S. Yang, L. Wang, X. Wang and L. Liu, Macromolecules,
2013, 46, 6832–6842.
20 Q.-Q. Jiang, W. Sicking, M. Ehlers and C. Schmuck, Chem.
Sci., 2015, 6, 1792–1800.
´
21 C. S. Mahon, A. W. Jackson, B. S. Murray and D. A. Fulton,
Polym. Chem., 2013, 4, 368–377.
Acknowledgements
22 G. Deng, F. Li, H. Yu, F. Liu, C. Liu, W. Sun, H. Jiang and
Y. Chen, ACS Macro Lett., 2012, 1, 275–279.
23 N. Kuhl, S. Bode, R. K. Bose, J. Vitz, A. Seifert, S. Hoeppener,
S. J. Garcia, S. Spange, S. van der Zwaag and M. D. Hager,
Adv. Funct. Mater., 2015, 25, 3295–3301.
24 W.-S. Yeo and M. Mrksich, Adv. Mater., 2004, 16, 1352–1356.
25 F. M. Brunel, J. D. Lewis, G. Destito, N. F. Steinmetz,
M. Manchester, H. Stuhlmann and P. E. Dawson, Nano
Lett., 2010, 10, 1093–1097.
This work was financially sponsored by the National Natural
Science Foundation of China (No. 51225303 and 51433007),
and the Sichuan Province Youth Science and Technology
Innovation Team (No. 2015TD0001). We would also thank our
laboratory members for their generous help, and gratefully
acknowledge the help of Ms Hui Wang, of the Analytical and
Testing Center at Sichuan University, for SEM observation.
26 H. Nandivada, H. Y. Chen and J. Lahann, Macromol. Rapid
Commun., 2005, 26, 1794–1799.
Notes and references
27 J. Deng, X. Liu, S. Zhang, C. Cheng, C. Nie and C. Zhao,
Langmuir, 2015, 31, 9665–9674.
1 M. A. C. Stuart, W. T. Huck, J. Genzer, M. Mu¨ller, C. Ober,
M. Stamm, G. B. Sukhorukov, I. Szleifer, V. V. Tsukruk and
M. Urban, Nat. Mater., 2010, 9, 101–113.
2 H. Yang, B. Yuan, X. Zhang and O. A. Scherman, Acc. Chem.
Res., 2014, 47, 2106–2115.
3 M. Delcroix, G. Huet, T. Conard, S. Demoustier-Champagne,
F. E. Du Prez, J. Landoulsi and C. C. Dupont-Gillain,
Biomacromolecules, 2012, 14, 215–225.
28 W. J. Yang, T. Cai, K.-G. Neoh, E.-T. Kang, S. L.-M. Teo and
D. Rittschof, Biomacromolecules, 2013, 14, 2041–2051.
29 Z. X. Voo, M. Khan, K. Narayanan, D. Seah, J. L. Hedrick and
Y. Y. Yang, Macromolecules, 2015, 48, 1055–1064.
30 X. Wang, B. e. Lv, G. Cai, L. Fu, Y. Wu, X. Wang, B. Ren and
H. Ma, Sci. Rep., 2012, 2, 892.
31 C. S. Mahon, A. W. Jackson, B. S. Murray and D. A. Fulton,
Chem. Commun., 2011, 47, 7209–7211.
4 X. Huang, Y. Sun and S. Soh, Adv. Mater., 2015, 27,
4062–4068.
32 X. Liu, J. Deng, L. Ma, C. Cheng, C. Nie, C. He and C. Zhao,
Langmuir, 2014, 30, 14905–14915.
5 S. Kumar, X. Tong, Y. L. Dory, M. Lepage and Y. Zhao, Chem.
Commun., 2013, 49, 90–92.
This journal is ©The Royal Society of Chemistry 2015
J. Mater. Chem. B

View Article Online
Paper
Journal of Materials Chemistry B
´
´
´
33 M. Tanaka, T. Motomura, M. Kawada, T. Anzai, Y. Kasori, 39 M. Popovic, K. Smiljanic, B. Dobutovic, T. Syrovets, T. Simmet
´
T. Shiroya, K. Shimura, M. Onishi and A. Mochizuki, Bio-
materials, 2000, 21, 1471–1481.
34 S. Niewiarowski and D. P. Thomas, Nature, 1969, 222,
1269–1270.
35 K. Ishihara, K. Fukumoto, Y. Iwasaki and N. Nakabayashi,
Biomaterials, 1999, 20, 1553–1559.
36 S. Herrwerth, W. Eck, S. Reinhardt and M. Grunze, J. Am.
Chem. Soc., 2003, 125, 9359–9366.
and E. R. Isenovic, J. Thromb. Thrombolysis, 2012, 33, 160–172.
40 S. Nie, H. Qin, C. Cheng, W. Zhao, S. Sun, B. Su, C. Zhao and
Z. Gu, J. Mater. Chem. B, 2014, 2, 4911–4921.
41 R. W. Hay, G. A. Lawrance and N. F. Curtis, J. Chem. Soc.,
Perkin Trans. 1, 1975, 591–593.
42 R. Narain and S. P. Armes, Biomacromolecules, 2003, 4,
1746–1758.
43 C. S. Mahon and D. A. Fulton, Chem. Sci., 2013, 4, 3661–3666.
´
37 A. H. Kamal, A. Tefferi and R. K. Pruthi, Mayo Clinic 44 Q. Yu, J. Cho, P. Shivapooja, L. K. Ista and G. P. Lopez,
Proceedings, Elsevier, 2007, pp. 864–873. ACS Appl. Mater. Interfaces, 2013, 5, 9295–9304.
38 L. L. Leung, ASH Education Program Book, 2006, vol. 2006, 45 G. Cheng, H. Xue, Z. Zhang, S. Chen and S. Jiang, Angew.
pp. 457–461.
Chem., Int. Ed., 2008, 47, 8831–8834.
J. Mater. Chem. B
This journal is ©The Royal Society of Chemistry 2015