Small gold nanoparticles for interfacial Staudinger-Bertozzi ligation

Article abstract of DOI:10.1039/c5ob00372e

Small gold nanoparticles (AuNPs) that possess interfacial methyl-2-(diphenylphosphino)benzoate moieties have been successfully synthesized (Staudinger-AuNPs) and characterized by multi-nuclear MR spectroscopy, transmission electron microscopy (TEM), UV-Vi

Full text of DOI:10.1039/c5ob00372e

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Small gold nanoparticles for interfacial
Staudinger–Bertozzi ligation†
Cite this: Org. Biomol. Chem., 2015,
13, 4605
Pierangelo Gobbo,*a Wilson Luo,a Sung Ju Cho,a Xiaoxiao Wang,a
Mark C. Biesinger,b Robert H. E. Hudsona and Mark S. Workentin*a
Small gold nanoparticles (AuNPs) that possess interfacial methyl-2-(diphenylphosphino)benzoate moi-
eties have been successfully synthesized (Staudinger-AuNPs) and characterized by multi-nuclear MR
spectroscopy, transmission electron microscopy (TEM), UV-Vis spectroscopy, thermogravimetric analysis,
and X-ray photoelectron spectroscopy (XPS). In particular, XPS was remarkably sensitive for characteriz-
ation of the novel nanomaterial, and in furnishing proof of its interfacial reactivity. These Staudinger-
AuNPs were found to be stable to the oxidation of the phosphine center. The reaction with benzyl azide
in a Staudinger–Bertozzi ligation, as a model system, was investigated using 31P NMR spectroscopy. This
demonstrated that the interfacial reaction was clean and quantitative. To showcase the potential utility of
these Staudinger-AuNPs in bioorganic chemistry, a AuNP bioconjugate was prepared by reacting the
Staudinger-AuNPs with a novel azide-labeled CRGDK peptide. The CRGDK peptide could be covalently
attached to the AuNP efficiently, chemoselectively, and with a high loading.
Received 24th February 2015,
Accepted 11th March 2015
DOI: 10.1039/c5ob00372e
www.rsc.org/obc
The conjugation of biomolecules to nanomaterials is of high uncontrolled “shotgun” type bioconjugation,9 whereas for
interest because of the potential applications of the resulting many applications it is desirable to have more precise control
bioconjugate in biological and medical diagnostics, thera- over the amount of biomolecule that is introduced. This can
peutics and drug vectors.1–5 Among all nanomaterials, gold be achieved by designing a single stable AuNP that exhibits a
nanoparticles are often regarded as the most promising tem- chemoselective reactive moiety at its interface that serves as an
plate because of their low toxicity and chemical stability. In anchoring group to incorporate a discrete number of bio-
addition, their optical properties can be easily tuned by molecules. This could result in the precise tuning of the final
varying the gold core diameter6 and the chemical properties biophysical properties of the AuNP-bioconjugate. It is also
can be tuned by changing the ligand corona.7 In the past desirable that the interfacial reactive group on the nanoparti-
decade, many different approaches have been developed to cle leads to a robust conjugate that relies on a strong bond
synthesize AuNP bioconjugates for different purposes. Gener- between nanoparticle and biomolecule. In order for the inter-
ally these strategies rely on electrostatic interactions between facial reaction to be chemoselective, this nanoparticle template
the gold surface and the side chains of the biomolecule’s resi- must undergo a reaction with a functional group that is not
dues, on hydrophobic–hydrophilic type of interactions, on present in nature, atoxic, and small so that it does not affect
direct place exchange exploiting the affinity of cysteine for the activity of the conjugated biomolecule.
gold, or on interfacial reactions (e.g. interfacial EDC coupling,
The functionality that addresses all these requirements is
imine formation etc.).8 These methods generally result in an the azide group. The azide can undergo four different reac-
tions. The [3 + 2] Huisgen cycloaddition and the copper
catalyzed [3 + 2] Huisgen cycloaddition with dipolarophiles,
aThe University of Western Ontario and the Centre for Materials and Biomaterials
the strained-promoted alkyne–azide cycloaddition (SPAAC)
with strained alkynes and the Staudinger–Bertozzi ligation.10,11
Research, Richmond Street, London, Ontario, Canada. E-mail: mworkent@uwo.ca,
pgobbo2@uwo.ca; Tel: +1 519-661-2111 extn 86319
bSurface Science Western, The University of Western Ontario, 999 Collip Circle,
While the [3 + 2] Huisgen cycloaddition and its catalyzed
version are not completely compatible with AuNPs chemistry
London (ON), N6G 0J3, Canada
because the high temperature or the Cu(I) required to push the
reaction to completion cause severe nanoparticle aggrega-
tion,12 the SPAAC reaction presents limited chemoselectivity
due to the possibility of nucleophilic attacks (especially from
thiols and amines largely present in biomolecules) to the
highly reactive strained triple bond.13–15 A recent work of ours
†Electronic supplementary information (ESI) available: Detailed synthesis and
characterization of the methyl-2-(diphenylphosphino)benzoate-thiol ligand;
detailed procedure for the synthesis of the Staudinger-AuNP; TGA data; complete
XPS characterization for the Staudinger-AuNP, the Staudinger-AuNP reacted with
benzyl azide, and Staudinger-AuNP reacted with peptide; details of the calcu-
lations for the nanoparticles raw formula; detailed synthesis and characteriz-
ation of the azide-modified peptide. See DOI: 10.1039/c5ob00372e
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 4605–4612 | 4605

View Article Online
Paper
Organic & Biomolecular Chemistry
highlighted this issue showing how post assembly deprotec- air bubbling failed. However, small amounts of phosphine
tion of peptides once “clicked” on the AuNP surface was oxide can form during the different stages of the ligand syn-
necessary to efficiently synthesize a nanoparticle bioconjugate thesis. It was possible to reduce the phosphine oxide back to
through the SPAAC reaction.16 The Bertozzi–Staudinger the active phosphine using the synthetic procedure reported
ligation is a reaction that was specifically developed for investi- by H.-C. Wu et al.26 Before deprotecting the thiol end, the
gating the metabolic engineering of cell surfaces17 and takes mixture of phosphine oxide and phosphine can be dissolved
place between the azide and a substituted triphenylphosphine. in a solution of triphenylphosphine and HSiCl3 in dry toluene.
Whereas the cycloaddition reactions lead to a triazole ring, the Heating of this reaction mixture to 100 °C leads to the quanti-
Staudinger–Bertozzi ligation forms a stable amide bond. From tative reduction of the phosphine oxide. It is noteworthy that
its very first development the Staudinger–Bertozzi ligation has this reaction fails in presence of free thiols and, if the phos-
been successfully employed in biochemistry to immobilize phine-oxide reduction is carried out at the nanoparticles inter-
peptides and proteins onto different materials,18–22 for in vivo face, HSiCl3 causes a fast and total decomposition of the
and in vitro imaging, and for protein detection.23–25
AuNP.
The Staudinger-AuNPs have been synthesized through a
This reaction seems to be an ideal candidate for designing
a new AuNP template that can react in a highly chemoselective place-exchange reaction using triethylene glycol monomethyl
and efficient way with any azide-modified (bio)molecule. ether AuNP (MeO-EG3-AuNP) as the starting material,
For these reasons we report the synthesis of a thiol ligand Scheme 1. Details of their synthesis of the latter can be found
that contains a methyl-2-(diphenylphosphino)benzoate moiety in the ESI.† These nanoparticles have a gold core diameter of
designed to undergo a Staudinger–Bertozzi ligation with azide
3
1 nm and are soluble in water and polar organic solvents.27
functionalities. This ligand was introduced onto the surface of MeO-EG3-AuNPs represent a very resilient substrate for further
small water-soluble AuNPs through a place-exchange reaction modification and interfacial organic chemistry because they
and the reactivity of the resulting nanomaterial was investi- are resistant to acidic16 and basic conditions,12 they can be
gated using benzyl azide as model clickable partner. Finally, to heated to over 100 °C,27 and they can be repeatedly dried and
showcase the high chemoselectivity of these novel reactive re-dissolved in a solvent with little to no aggregation. The
AuNPs for bioconjugation, we react them with an azide-modi- place-exchange reaction to introduce the phosphine-thiol (4)
fied CRGDK peptide. CRGDK peptides are promising targeting onto the nanoparticle’s corona was carried out by dissolving
biomolecules for diverse cancer cells lines and therefore the 200 mg of MeO-EG3-AuNP and 100 mg of phosphine-thiol (4)
resulting AuNP-CRGDK bioconjugate would be a potential in dry and argon purged dichloromethane. The reaction was
broad-spectrum targeting agent.1 In addition, the presence of carried out for 15 minutes at room temperature, under inert
the cysteine in the CRGDK peptide represents a challenge for atmosphere (to minimize potential oxidation of the reactive
the chemoselective conjugation on nanoparticles. Herein, we phosphine) and under vigorous stirring. The particles were
present a facile, versatile, and straightforward protocol for the then cleaned from the excess thiol by re-dissolving them in
conjugation of biomolecules onto the AuNP surface through 6 ml of dichloromethane and by precipitating them by adding
the Staudinger–Bertozzi ligation. Importantly, the study show- 33 ml of hexanes in which the nanoparticles are not soluble.
cases the use of XPS as a method to easily characterize the This washing operation was repeated five times overall. The
interfacial reactivity of this new nanomaterial, and to quantify Staudinger-AuNPs were characterized by 1H and 31P NMR
the amount of conjugated biomaterial, information that is very spectroscopy, transmission electron microscopy (TEM), UV-Vis
important for practical applications in biochemistry, medical spectroscopy, thermogravimetric analysis, and X-ray photo-
diagnosis, therapies and drug delivery.
electron spectroscopy (XPS). The 1H and 31P NMR spectra
showed presence of broad peaks typical of a AuNP sample indi-
cating that the washing procedure was efficient and that the
most of the free thiol ligands were successfully removed. Cur-
iously the spectra did not show a perfect correspondence of
Results and discussion
The approach used to introduce the reactive triphenylphos- the peaks with the free ligand (4). Usually the successful in-
phine moiety for the Staudinger–Bertozzi ligation onto the corporation of a second thiol ligand through the place-exchange
water-soluble AuNPs required the synthesis of the appropriate reaction is easily determined by comparing the 1H NMR
thiol containing ligand for a place-exchange reaction; see spectra of the free ligand with that of the place-exchanged
Scheme 1. The first step involved coupling 3-(diphenylphos- AuNPs sample. The chemical shifts typically match even
phino)-4-(methoxycarbonyl)benzoic acid (1) to an amino-tetra- though the peaks on the AuNPs spectrum are broader.28,29 In
1
ethylene glycol-trityl-protected thiol (2). The deprotection of this case H NMR spectrum shows a substantial difference in
the thiol functionality in 5% trifluoroacetic acid–dichloro- the aromatic region as highlighted in Fig. 1. An even more
methane led to the desired phosphine-thiol ligand (4). Details drastic difference is observed in the 31P NMR spectrum of the
of the ligand synthesis can be found in the ESI.† Compound Staudinger-AuNPs with respect to that of compound (4). The
(4) was found to be remarkably stable to the oxidation of both 31P NMR spectrum of the Staudinger-AuNPs did not show a
phosphine and thiol ends. Attempts to oxidize the phosphine peak for the phosphine at −3.8 ppm, but two signals one
and the thiol by dissolving the molecule in wet solvents and by broad and intense at 40.3 ppm and one at 31.3 ppm. By
4606 | Org. Biomol. Chem., 2015, 13, 4605–4612
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 1 Synthetic strategy for the synthesis of the Staudinger-AuNP.
comparison with the data of the phosphine oxide formation starting material due to the very short distance between the
gathered during the synthesis of compound (4), the signal at gold cores involved in the AuNPs self-assemblies. The reason
31.3 ppm can be attributed to the oxidized methyl-2-(diphenyl- for this collective behavior of the Staudinger-AuNPs is probably
phosphino)benzoate. Integration of the 31P NMR peaks shows due to hydrophobic–hydrophilic-type of interactions between
that the AuNP sample has 9% of oxidized phosphine that is the hydrophilic ethylene glycol linker and the strongly hydro-
most likely generated during the washing procedure where it is phobic character of the triphenylphosphine. As reported by
exposed to air. The shift of the phosphine peak in the 31P K. J. M. Bishop and coworkers, ligands with different solubility
NMR and the difference in the aromatic region shown in the properties introduced through a place-exchange reactions do
1H NMR seem to suggest that there is an interaction at the not distribute homogenously on the particle’s corona but
phosphorous center that changes the electronic properties of rather tend to form patches or groups. These patches are
the phosphine.
responsible for the formation of inter-nanoparticles hydro-
Transmission electron microscopy provided relevant phobic “bonds” resulting in the self-assembly of nanoparticle
insights on this issue. The TEM images reported in Fig. 3A islands (like in the present work) or chain-like structures.31
clearly show that 3.63 1.4 nm Staudinger-AuNPs are interact-
High-resolution XPS analysis, see Fig. SI3,† further con-
ing with each other forming AuNPs self-assemblies with an firms the presence of inter-nanoparticle interactions showing
inter-particle distance of ∼1 nm.30,31 TEM images of similar a peak for the Au 4f core line that displays two different com-
ethylene glycol small AuNPs usually show well isolated and ponents: a small component at 84.62 eV typical of isolated
evenly distributed dark spots on the TEM grid, like after the AuNPs and a larger component at 83.75 eV that is character-
interfacial Staudinger–Bertozzi ligation, see Fig. 3B. This istic of bulk gold and is most likely related to the AuNP net-
AuNPs self-assembly was also confirmed by UV-Vis spectro- works shown in the TEM images.32 XPS analysis also shows a
scopy. Typically AuNPs of such a small size do not show any peak for the amide nitrogen at 399.75 eV. The high resolution
plasmon resonance band in the UV-Vis spectrum. However, scans of the P 2p peak, see Fig. 4, shows a major component at
the UV-Vis derivative reported in Fig. 3C highlights the appear- 131.52 eV (P 2p3/2) related to the triphenylphosphine and a
ance of a weak plasmon band in the region 525–650 nm for minor component (13%) at 132.60 eV (P 2p3/2) related to
the Staudinger-AuNP sample compared to the MeO-EG3-AuNP triphenylphosphine oxide, confirming independently and with
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 4605–4612 | 4607

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 1 1H (left) and 31P (right) NMR spectra of phosphine-thiol (4) (top) versus Staudinger-AuNP (bottom). *Denotes residual solvent peaks. Spectra
calibrated against residual chloroform and H3PO4.
higher precision the result obtained through 31P NMR specifically these Staudinger-AuNPs contain methyl-2-(diphe-
spectroscopy. The high resolution scans of the C 1s, see Fig. 4, nylphosphino)benzoate in concentration 0.432 µmol mg−1
.
shows a distinct component at 289.08 eV that is related to the Details of these calculations are reported in the ESI.†
carbonyl of the ester functional group, a component at
The reactivity of the Staudinger-AuNP through the Staudin-
288.00 eV that is related to the amide’s carbonyl, and the ger–Bertozzi ligation was then investigated using benzyl azide
typical components for the ethylene glycol ligands at 286.36 eV and the reactions were monitored through 31P NMR spectro-
and 285.00 eV. Finally, the high-resolution scan of the S 2p scopy. In the 31P NMR spectrum it was possible to follow the
peak, see Fig. SI3,† shows the typical components of the disappearance of the broad peak at 40.3 ppm corresponding to
sulfur–gold bond and presence of a negligible amount of free the active phosphine and the appearance of the peak of the
thiol and sulfates.33 The presence of free thiols and sulfates ligation product at 34.9 ppm, refer to Fig. 1 and 2. Kinetic
(generated by the slow oxidation of thiols over time) is due to experiments were carried out under pseudo-first-order reaction
the fact that they have been trapped in between the AuNPs conditions at 25 °C and the reaction kinetics at the AuNP inter-
self-assemblies and are difficult to wash away.
face was compared with that in solution using ligand (4) as a
Despite the peculiar behavior of these AuNPs, exposure to model compound. For this investigation Staudinger-AuNP
benzyl azide provided clear evidence of the Staudinger– (0.0259 M in phosphine, 30.0 mg of Staudinger AuNP) were
Bertozzi product. In fact the peaks of the 1H and 31P NMR reacted with benzyl azide (0.518 M) in wet CDCl3. Chloroform
spectra returned to match those of the model Staudinger– was chosen due to the high solubility of the nanoparticles in
Bertozzi product obtained using the phosphine-thiol (4) with this solvent that permitted the high AuNPs concentration
benzyl-azide (vide infra).
required for the experiments. A reaction using the same con-
Thermogravimetric analysis of the Staudinger-AuNP ditions was carried out between model compound (4) and
showed that the 43% of the total weight is due to the organic benzyl azide. This same reaction was then carried out at
corona (see Fig. SI5†), of this the 60% is constituted by ligand higher concentration using CDCl3 and DMSO-d6 as the sol-
(4), while the 40% is constituted by MeO-EG3-S−. Finally, from vents to compare the results with those previously reported in
the deconvolution of the TGA curve, the 31P NMR spectrum the literature.34 Table 1 summarizes the results of these
and the gold core diameter obtained from TEM images it was kinetic experiments. During the course of all these reactions
possible to calculate a raw formula for the Staudinger-AuNP: no intermediates were observed through 31P NMR spectro-
Au1500 (MeO-EG3-S)500 (Ph3P-EG4-S)200 (Ph3PvO-EG4-S)20. More scopy. This indicates that the reaction mechanism remains
4608 | Org. Biomol. Chem., 2015, 13, 4605–4612
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 2 1H (left) and 31P (right) NMR spectra of phosphine-thiol (4) after reaction with benzyl azide (top) versus Staudinger-AuNP after reaction with
benzyl azide (bottom). * Denotes residual solvent peaks. Spectra calibrated against residual chloroform and H3PO4.
Table 1 Second order rate constants of the Staudinger–Bertozzi ligation with benzyl azide obtained at 25 °C under different experimental
conditions
[Phosphine]
(M)
[Benzyl azide]
(M)
[H2O]
(M)
Second order rate
Solvent
constant (10−3 M−1 s−1
)
Staudinger-AuNP
Ligand (4)
Ligand (4)
0.0259
0.0259
0.0259
0.0440
0.518
0.518
0.518
0.830
5.56
5.56
5.56
5.56
CDCl3
CDCl3
DMSO-d6
DMSO-d6
0.027 0.002
1.6 0.1
3.5 0.1
Ligand (4)
4.7 0.1
that proposed by F. L. Lin and coworkers with the rate-limiting the reaction yields the Staudinger–Bertozzi product
step being the formation of the phosphazide.34 While this is quantitatively.
not surprising for the model reactions with ligand (4), it is
The product of the interfacial Staudinger–Bertozzi ligation
remarkable that the Staudinger–Bertozzi ligation takes place at was then separated from the free benzyl azide by washing a
the AuNPs interface in a comparable manner despite the film of nanoparticles made in a round-bottom flask and
markedly different chemical environment represented by the rinsing it with hexanes where the nanoparticles are insoluble.
nanoparticle structure and the networks that they form in solu- The purified sample was then characterized by 1H and 31P
tion. Our results show that the reaction of ligand (4) has NMR spectroscopy, XPS, UV-Vis spectroscopy and TEM. As
second-order rate constants comparable to those reported by anticipated, TEM images show the absence of AuNPs self-
F. L. Lin and coworkers and that they increase with an increase assemblies and only well isolated nanoparticles, see Fig. 3B.
in solvent dielectric constant. The reaction at the AuNP inter- The 1H NMR spectrum of the reacted Staudinger-AuNPs
face instead results to be a hundred times slower. This is sample now shows perfect correspondence with the chemical
probably due to additional reaction parameters introduced shift of the model reaction with compound (4), see Fig. 2. The
with an interfacial reaction and because of the inter- UV-Vis spectrum of the reacted nanoparticles sample now
nanoparticle interactions observed that need to be overcome shows a spectrum similar to that of MeO-EG3-AuNP starting
before the ligation occurs. However, the interfacial phosphines material, characterized by the absence of a plasmon resonance
are still reactive towards the Staudinger–Bertozzi ligation and band in the region 525–650 nm (see Fig. 3C).
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 4605–4612 | 4609

View Article Online
Paper
Organic & Biomolecular Chemistry
ponent at higher binding energy is now that related to the newly
formed amide at 288.00 eV. This is expected because of the
generation of methanol that occurs during the reaction mechan-
ism. In addition, the high-resolution scan of the P 2p core line
now appears narrower and shows only the component for the
triphenylphosphine-oxide. These results together confirm that
the interfacial reaction went to completion. The high-resolution
scan of the S 2p core line now shows the absence of free thiols
and disulfides as a result of efficient trituration once the nano-
particles self-assemblies has been abrogated.
In summary, despite the peculiar inter-Staudinger-AuNPs
interaction that leads to AuNPs self-assemblies and changes
the electronic properties of the phosphine, the Staudinger-
AuNP are still readily reactive towards the azide bioorthogonal
partner as highlighted by the kinetic study using 31P NMR
spectroscopy and by the XPS analysis of the interfacial ligation
product. More importantly, the reaction is quantitative as inde-
pendently confirmed by 31P NMR spectroscopy and XPS ana-
lysis, and the reacted Staudinger-AuNP could be recovered
easily and quantitatively.
To showcase the high chemoselectivity of these Staudinger-
AuNPs and their potential for bioconjugation, we reacted them
with an azide functionalized CRGDK peptide. The CRGDK
peptide was synthesized via standard Fmoc SPPS method and
the azide functionality was coupled to the N-terminus. The
Fig. 3 TEM images of (A) Staudinger-AuNP, and (B) Staudinger-AuNP
after reaction with benzyl azide. (C) Derivative of UV-Vis spectrum of
MeO-EG3-AuNP (blue), Staudinger-AuNP (red), and Staudinger-AuNP
after reaction with benzyl azide. The derivative highlights the presence
of the plasmon resonance band for the Staudinger-AuNP sample due to
the AuNPs self-assemblies.
XPS proves to be a powerful technique to furnish proof of AuNP bioconjugate was synthesized simply by mixing 20 mg of
interfacial reactivity. Fig. 4 shows that after the interfacial reac- Staudinger AuNP (8.64 µmol of phosphine) with 30 µmol of
tion with the azide, the C 1s core line no longer shows the azide-CRGDK in DMSO-d6 at 37 °C. The reaction was moni-
component for the ester bond at 289.08 eV, and the com- tored through 31P NMR spectroscopy following the progressive
Fig. 4 High-resolution XPS of C 1s and P 2p core lines of the Staudinger-AuNP (left) and Staudinger-AuNP after reaction with benzyl azide (middle)
and azide-CRGDK peptide (right).
4610 | Org. Biomol. Chem., 2015, 13, 4605–4612
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
disappearance of the broad peak at 40.3 ppm and the appear- NMR spectra and TEM images. This represents important
ance of the product peak at 33.2 ppm (see Fig. SI11†) indicat- information in the bioconjugation field because of the
ing proper interfacial reactivity with formation of the expense of both starting materials and consequent difficulty to
Staudinger–Bertozzi product. During the course of the reactiv- work in excess of one or the other clickable partner.
ity we did not observe displacement of any of the AuNP
The reactivity of the Staudinger-AuNP was then investigated
ligands due to a place exchange reaction by the peptide’s through 31P NMR spectroscopy using benzyl azide as the click-
cysteine. This would have resulted in the appearance of a peak able partner. The possibility to easily monitor the reaction by
at −3.80 ppm for the free phosphine ligand and easily identifi- 31P NMR spectroscopy by following the disappearance of the
1
able disulphides peaks in the H NMR spectrum. When reac- phosphine peak at 40.3 ppm and the appearance of the phos-
tion was completed the excess peptide was washed away with phine oxide peak of the product at 34.9 ppm represents a great
nanopure water using centrifugal filtration (Millipore centrifu- advantage of this novel nanomaterial, and makes their use
gal filter units MWCO 10 kDa). Interestingly the reaction easy and straightforward. The data obtained from the kinetic
stopped at ∼30% (calculated from the 31P NMR spectra), indi- study showed that the reaction mechanism with which these
cating that most likely the AuNPs reached saturation due to AuNPs react appears to be identical to that of small methyl-2-
the larger size of the peptide compared to the smaller model (diphenylphosphino)benzoate molecules in solution. The only
molecule benzyl azide. The AuNP-bioconjugate was character- difference appears to be in the kinetic constant that resulted
ized by TEM and XPS. TEM images (see Fig. SI10†) show again to be a hundred times slower, due to the interfacial character
isolated AuNPs of ∼3 nm in diameter and no nanoparticles of the reaction. More importantly, these studies showed
networks were observed. The high-resolution scans of the C 1s quantitative reactivity of the triphenylphosphine moieties with
peak (see Fig. 3) show still the presence of the ester component the azide partner small molecule and the reacted Staudinger-
at 289.00 eV and a marked increase in the component related AuNPs could be quantitatively recovered.
to amide’s carbonyl at 288 eV due to the presence of the
XPS confirms to be a reliable and highly sensitive surface
numerous peptide bonds. The high-resolution scans of the analysis technique for the study of interfacial organic chemi-
S 2p peak (see Fig. SI12†) now show a marked increase of the stry. In fact, the Staudinger-AuNP and the successful formation
S–H component at 163.21 eV (S 2p3/2) and of the disulfide com- of the interfacial ligation product could be effectively charac-
33
ponent at 168.45 eV (S 2p3/2
)
due to the presence of the pep- terized through the high-resolution scans especially of the
tide’s cysteines that do not interfere with the interfacial C 1s and the P 2p core lines. The former core line after the
reactivity thanks to the high chemoselectivity of the Staudin- interfacial Staudinger–Bertozzi ligation clearly shows the dis-
ger–Bertozzi ligation. Finally, the high-resolution scans of the appearance of the component at 289.08 eV that corresponds to
P 2p peak confirms that the 30% of the phosphine reacted the methyl-ester functionality of the starting material; the
with the peptide to form the triphenylphosphine-oxide. Inter- latter distinctively show two components one for the triphenyl-
estingly this 30% of interfacial reaction yield seems to be phosphine at 131.52 eV and the other for the triphenylphos-
recurring when employing 3 nm AuNPs. The same result was phine oxide at 132.43 eV, allowing for quantifying the amount
indeed obtained when
peptide was clicked onto azide-functionalized AuNPs of the
a
DBCO-functionalized protected- of azide partner that has been clicked onto the nanoparticles.
Another great advantage of our Staudinger-AuNPs is the
same size using the SPAAC-PAD strategy.16 Because the smaller insensitivity of the Staudinger–Bertozzi ligation to nucleo-
reactive partner (benzyl azide) reacted quantitatively, the lower philes such as thiols or amines that are largely present in bio-
extent of reaction with the peptide may simply be an issue of systems. The high chemoselectivity of the Staudinger-AuNPs
sterics with the peptide blocking additional reactive sites. makes them an attractive alternative to faster but less chemo-
These results together are consistent with successful conju- selective bioconjugation techniques that for example involve
gation of the peptide onto the Staudinger-AuNPs and the the Michael addition to maleimide functionalities or the
efficiency of the rinsing/trituration. Finally, the nanoparticle SPAAC reaction. In fact, the reactivity of the maleimide or the
bioconjugate was found to be soluble in water and in polar side-reactivity of cyclooctyne towards thiols and amines can
organic solvents and could be repeatedly dried and redissolved lower the chemoselectivity of the corresponding nanomaterial
in solvent with little to no aggregation.
representing a threat to the biomolecule’s conformation and
biochemistry. In contrast our Staudinger-AuNPs react exclu-
sively with the azide, as highlighted by the successful conju-
gation of an azide-functionalized CRGDK peptide. This direct
conjugation would have required additional precautions if the
Conclusions
In this paper we reported the first synthesis of small gold Michael addition or the strain-promoted cycloaddition tech-
nanoparticles that present interfacial methyl-2-(diphenylphos- niques were employed. Instead, using the Staudinger–Bertozzi
phino)benzoate functionalities able to undergo Staudinger– ligation the AuNP-CRGDK bioconjugate could be synthesized
Bertozzi ligation for the creation of nanoparticle bioconju- efficiently and with a high loading onto the nanomaterial.
gates. These small Staudinger-AuNPs were completely charac-
Finally the Staudinger-AuNPs can be used also in material
terized and a nanoparticle raw formula could be estimated chemistry. Thanks to their clean way of reacting they can react
with good precision from the combination of the TGA data, easily with any azide modified surface or material for the
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 4605–4612 | 4611

View Article Online
Paper
Organic & Biomolecular Chemistry
creation of nanohybrid materials and sensors. We are currently 15 M. Chigrinova, C. S. McKay, L. P. B. Beaulieu,
exploring these possibilities.
K. A. Udachin, A. M. Beauchemin and J. P. Pezacki, Org.
Biomol. Chem., 2013, 11, 3436–3441.
16 X. Wang, P. Gobbo, M. Suchy, M. S. Workentin and
R. H. E. Hudson, RSC Adv., 2014, 4, 43087–43091.
17 E. Saxon and C. R. Bertozzi, Science, 2000, 287, 2007–
2010.
18 S. H. Yoshimura, S. Khan, S. Ohno, T. Yokogawa,
K. Nishikawa, T. Hosoya, H. Maruyama, Y. Nakayama and
K. Takeyasu, Bioconjugate Chem., 2012, 23, 1488–1493.
19 A. Watzke, M. Kohn, M. Gutierrez-Rodriguez, R. Wacker,
H. Schroder, R. Breinbauer, J. Kuhlmann, K. Alexandrov,
C. M. Niemeyer, R. S. Goody and H. Waldmann, Angew.
Chem., Int. Ed., 2006, 45, 1408–1412.
Acknowledgements
We (RHEH, MSW) acknowledge the Natural Sciences and
Engineering Research Council of Canada and The University
of Western Ontario financial support. PG thanks the
Government of Canada for a Vanier Scholarship and Research
Western.
Notes and references
1 A. Kumar, H. L. Ma, X. Zhang, K. Y. Huang, S. B. Jin, J. Liu, 20 M. B. Soellner, K. A. Dickson, B. L. Nilsson and
T. Wei, W. P. Cao, G. Z. Zou and X. J. Liang, Biomaterials,
2012, 33, 1180–1189.
2 A. J. Mieszawska, W. J. M. Mulder, Z. A. Fayad and
D. P. Cormode, Mol. Pharmaceutics, 2013, 10, 831–847.
3 D. Xi, S. Dong, X. X. Meng, Q. H. Lu, L. J. Meng and J. Ye,
RSC Adv., 2012, 2, 12515–12524.
4 Y. C. Yeh, B. Creran and V. M. Rotello, Nanoscale, 2012, 4,
1871–1880.
5 C. J. Murphy, A. M. Gole, J. W. Stone, P. N. Sisco,
R. T. Raines, J. Am. Chem. Soc., 2003, 125, 11790–11791.
21 D. E. Borchmann, N. T. Brummelhuis and M. Weck, Macro-
molecules, 2013, 46, 4426–4431.
22 C. I. Schilling, N. Jung, M. Biskup, U. Schepers and
S. Brase, Chem. Soc. Rev., 2011, 40, 4840–4871.
23 M. King and A. Wagner, Bioconjugate Chem., 2014, 25, 825–
839.
24 B. E. Ramakers, J. C. van Hest and D. W. Lowik, Chem. Soc.
Rev., 2014, 43, 2743–2756.
A. M. Alkilany, E. C. Goldsmith and S. C. Baxter, Acc. Chem. 25 D. Grohmann, F. Werner and P. Tinnefeld, Curr. Opin.
Res., 2008, 41, 1721–1730.
Chem. Biol., 2013, 17, 691–698.
6 S. Eustis and M. A. El-Sayed, Chem. Soc. Rev., 2006, 35, 209– 26 H. C. Wu, J. Q. Yu and J. B. Spencer, Org. Lett., 2004, 6,
217.
4675–4678.
7 D. Pissuwan, T. Niidome and M. B. Cortie, J. Controlled 27 P. Gobbo and M. S. Workentin, Langmuir, 2012, 28, 12357–
Release, 2011, 149, 65–71.
12363.
8 J. B. Delehanty, K. Boeneman, C. E. Bradburne, R. Kelly, 28 K. D. Hartlen, H. Ismaili, J. Zhu and M. S. Workentin, Lang-
J. E. Bongard and I. L. Medintz, Ther. Delivery, 2010, 1,
411–433.
9 W. R. Algar, D. E. Prasuhn, M. H. Stewart, T. L. Jennings,
J. B. Blanco-Canosa, P. E. Dawson and I. L. Medintz, Bio-
conjugate Chem., 2011, 22, 825–858.
muir, 2012, 28, 864–871.
29 J. Zhu, B. M. Lines, M. D. Ganton, M. A. Kerr and
M. S. Workentin, J. Org. Chem., 2008, 73, 1099–
1105.
30 J. He, X. L. Huang, Y. C. Li, Y. J. Liu, T. Babu,
M. A. Aronova, S. J. Wang, Z. Y. Lu, X. Y. Chen and
Z. H. Nie, J. Am. Chem. Soc., 2013, 135, 7974–7984.
10 N. J. Agard, J. M. Baskin, J. A. Prescher, A. Lo and
C. R. Bertozzi, ACS Chem. Biol., 2006, 1, 644–648.
11 S. S. van Berkel, M. B. van Eldijk and J. C. M. van Hest, 31 H. Y. Lee, S. H. R. Shin, A. M. Drews, A. M. Chirsan,
Angew. Chem., Int. Ed., 2011, 50, 8806–8827.
S. A. Lewis and K. J. M. Bishop, ACS Nano, 2014, 8, 9979–
12 P. Gobbo, Z. Mossman, A. Nazemi, A. Niaux,
9987.
M. C. Biesinger, E. Gillies and M. S. Workentin, J. Mater. 32 P. Gobbo, S. Novoa, M. C. Biesinger and M. S. Workentin,
Chem. B, 2013, 2, 1764–1769.
Chem. Commun., 2013, 49, 3982–3984.
13 K. E. Beatty, J. D. Fisk, B. P. Smart, Y. Y. Lu, J. Szychowski, 33 J. C. Azcarate, M. A. F. Addato, A. Rubert, G. Corthey,
M. J. Hangauer, J. M. Baskin, C. R. Bertozzi and
G. S. K. Moreno, G. Benitez, E. Zelaya, R. C. Salvarezza and
D. A. Tirrell, ChemBioChem, 2010, 11, 2092–2095.
M. H. Fonticelli, Langmuir, 2014, 30, 1820–1826.
14 P. V. Chang, J. A. Prescher, E. M. Sletten, J. M. Baskin, 34 F. L. Lin, H. M. Hoyt, H. van Halbeek, R. G. Bergman and
I. A. Miller, N. J. Agard, A. Lo and C. R. Bertozzi, Proc. Natl.
Acad. Sci. U. S. A., 2010, 107, 1821–1826.
C. R. Bertozzi, J. Am. Chem. Soc., 2005, 127, 2686–
2695.
4612 | Org. Biomol. Chem., 2015, 13, 4605–4612
This journal is © The Royal Society of Chemistry 2015