Luminescent Nanocellulose Platform: From Controlled Graft Block Copolymerization to Biomarker Sensing

Article abstract of DOI:10.1021/acs.biomac.5b01716

A strategy is devised for the conversion of cellulose nanofibrils (CNF) into fluorescently labeled probes involving the synthesis of CNF-based macroinitiators that initiate radical polymerization of methyl acrylate and acrylic acid N-hydroxysuccinimide ester producing a graft block copolymer modified CNF. Finally, a luminescent probe (Lucifer yellow derivative) was labeled onto the modified CNF through an amidation reaction. The surface modification steps were verified with solid-state ¹³C nuclear magnetic resonance (NMR) and Fourier transform infrared spectroscopy. Fluorescence correlation spectroscopy (FCS) confirmed the successful labeling of the CNF; the CNF have a hydrodynamic radius of about 700 nm with an average number of dye molecules per fibril of at least 6600. The modified CNF was also imaged with confocal laser scanning microscopy. Luminescent CNF proved to be viable biomarkers and allow for fluorescence-based optical detection of CNF uptake and distribution in organisms such as crustaceans. The luminescent CNF were exposed to live juvenile daphnids and microscopy analysis revealed the presence of the luminescent CNF all over D. magna's alimentary canal tissues without any toxicity effect leading to the death of the specimen. (Graph Presented).

Full text of DOI:10.1021/acs.biomac.5b01716

Subscriber access provided by La Trobe University Library
Article
A Luminescent Nanocellulose Platform: from Controlled
Graft Block Copolymerization to Bio-marker Sensing
Julien R.G. Navarro, Stefan Wennmalm, Jamie Godfrey, Magnus Breiholtz, and Ulrica M Edlund
Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.5b01716 • Publication Date (Web): 20 Jan 2016
Downloaded from http://pubs.acs.org on January 25, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Biomacromolecules is published by the American Chemical Society. 1155 Sixteenth
Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 41
Biomacromolecules
1
2
3
4
5
6
7
8
9
A Luminescent Nanocellulose Platform: from
Controlled Graft Block Copolymerization to Bioꢀ
marker Sensing
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
†
‡
†
§
Julien R.G. Navarro , Stefan Wennmalm , Jamie Godfrey , Magnus Breitholtz
,†
and Ulrica Edlund*
†
Fiber and Polymer Technology, Royal Institute of Technology (KTH), Teknikringen 56, SEꢀ100
4, Stockholm, Sweden
4
‡
Science for Life Laboratory, Department of Applied Physics, KTHꢀRoyal Institute of
Technology, SEꢀ171 65 Solna, Sweden
§
Department of Environmental Science and Analytical Chemistry, Stockholm University, SEꢀ
14 18 Stockholm, Sweden
1
ACS Paragon Plus Environment
1

Biomacromolecules
Page 2 of 41
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT
A strategy is devised for the conversion of cellulose nanofibrils (CNF) into fluorescently labelled
probes involving the synthesis of CNFꢀbased macroinitiators that initiate radical polymerization
of methyl acrylate and acrylic acid Nꢀhydroxysuccinimide ester producing a graft block
copolymer modified CNF. Finally, a luminescent probe (Lucifer yellow derivative) was labelled
onto the modified CNF through an amidation reaction. The surface modification steps were
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
3
verified with solidꢀstate C nuclear magnetic resonance (NMR), and Fourier transform infrared
spectroscopy. Fluorescence Correlation Spectroscopy (FCS) confirmed the successful labelling
of the CNF, the CNF have a hydrodynamic radius of about 700 nm with an average number of
dye molecules per fibril of at least 6600. The modified CNF was also imaged with confocal laser
scanning microscopy. Luminescent CNF proved to be viable bioꢀmarkers and allow for
fluorescenceꢀbased optical detection of CNF uptake and distribution in organisms such as
crustaceans. The luminescent CNF were exposed to live juvenile daphnids and microscopy
analysis revealed the presence of the luminescent CNF all over D. magna’s alimentary canal
tissues without any toxicity effect leading to the death of the specimen.
KEYWORDS: Cellulose nanofibrils, controlled radical polymerization, grafting, luminescent
cellulose, fluorescent correlation spectroscopy, bioꢀmarker
ACS Paragon Plus Environment
2

Page 3 of 41
Biomacromolecules
1
2
3
4
5
6
7
8
9
INTRODUCTION
Nanocellulose, fibril nanoparticles of subꢀmicrometer thickness from cellulose, have attracted
considerable attention over the last decades as they are biodegradable, renewable and recyclable
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
with strong potential for a range of applications, from composites to drug delivery.
Nanocellulose can be extracted from various sources (cotton, woods or algae) and under different
forms: Cellulose nanofibrils (CNF) assimilated to long wires are composed of amorphous and
crystalline domains, while cellulose nanocrystals (CNC) are described as small rods, where the
2
–4
amorphous regions are removed through acid hydrolysis. The nanocellulose structure provides
good mechanical and thermal properties, and thereby could be used as a powerful reinforcing
1
,5
filler for composite films . However, surface modification remained necessary for their further
use in specific applications in order to obtain a sufficient interaction of the nanocellulose with
6
the polymer matrix. Thanks to the abundance of hydroxyl groups on the nanocellulose surface,
7
,8
9,10
several synthetic approaches have been exploited such as clickꢀchemistry, esterification,
1
1
12
13
etherification, silylation, and oxidation. In most cases, the substitution degree, and hence
polymer matrix compatibility was relatively low, and nanocellulose dispersion in various organic
solvents remained difficult. Therefore, there is a real need of a surface modification approach
that produce chemically modified CNF which yield stable CNF suspensions in numerous organic
solvents and allow for facile handling and subsequent chemical functionalization. Controlled ring
opening polymerization (ROP) has been used to graft and polymerize εꢀcaprolactone onto
14–16
nanocellulose, yielding a stable suspension in toluene.
Moreover, ROP modification allowed
for processing of a nanocelluloseꢀbased nanocomposite with a significant improvement of the
1
7
initial, undoped, mechanical properties. Majoinen et al. demonstrated that it was possible to
obtain stable nanocellulose suspension, either in aqueous or organic solvents, through surface
ACS Paragon Plus Environment
3

Biomacromolecules
Page 4 of 41
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
18
initiated controlled radical polymerization. Yi et al. modified cellulose nanocrystals through
the controlled polymerization of N,NꢀDimethylaminoethyl methacrylate and obtained a
thermally sensitive chiral nematic phase change in suspension.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Recently, it has been shown that polysaccharides can be efficiently converted to brominated or
chlorinated macroꢀinitiators for Cu(0)ꢀmediated controlled radical polymerization through partial
1
9–23
substitution of their hydroxyl groups.
Cu(0)ꢀmediated radical polymerization, introduced as
2
4–
Single Electron Transfer Living Radical Polymerization (SETꢀLRP) by Percec and coꢀworkers
2
7
, proved to be an efficient technique as it can be performed in water and under nonꢀinert
28–32
conditions with full control on the molecular weight.
SETꢀLRP is described to occur via an
electron transfer from the Cu(0) catalyst to an alkyl halide initiator and the dormant propagating
2
5
chain end. Cu(0)ꢀmediated controlled radical polymerization has been demonstrated to be
33
viable for a wide range of vinyl monomers. In view of the above, Cu(0)ꢀmediated controlled
radical polymerization offers a strong potential applied to polysaccharide functionalization and
offer new opportunities to customize nanocellulose chemistry for specific applications and
desired property profiles.
The labelling of nanocellulose with fluorescent probes is of great interest as bioꢀmarkers and in
3
4–39
sensor applications.
Within the standard fluorescence spectroscopy and microscopy,
4
0
Mahmoud et al. monitored the in situ cellular uptake of fluorescently labelled cellulose
41
nanocrystals and evaluated its cytotoxicity. Colombo et al. modified cellulose nanocrystals
with Alexa Fluor 633 and studied their toxicity through intravenous injection in living
organisms. In both studies, nanocellulose proved to be an efficient material platform for
biomedical applications.
ACS Paragon Plus Environment
4

Page 5 of 41
Biomacromolecules
1
2
3
4
5
6
7
8
9
Among the rich library of commercially available dyes, Lucifer Yellow has the advantage of
4
2–46
being water soluble, stable against photoꢀbleaching and useful for biological applications.
So
far, to study the photophysical properties of a dye at the single molecule level, and to verify the
correct labelling of the dye on an entity, fluorescence correlation spectroscopy (FCS) has proven
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
7,48
to be one of the most potent techniques.
FCS detects fluorescently labelled particles diffusing
through a femtoliter detection focus, and provides information about the diffusion coefficient and
4
9,50
the particle concentration (picomolar to micromolar range).
From the diffusion coefficient
the hydrodynamic radius can be derived which indicates the particle size. In single colour mode,
FCS can differentiate the detected fluorescence signal of a free dye from a signal of a dye
5
1
52
labelled larger objects such as proteins, and nanoparticles. During the last 20 years FCS and
related fluctuation techniques have developed into a widely used and commercially available tool
53–55
to analyse biomolecular interactions and fluctuations, in solution or in living cells.
In terms of water ecology and ecotoxicology studies, the cladoceran Daphnia magna (D. magna),
a small freshwater shellfish, has been used for several years now as a standardized test
56
organism. Daphnia magna are filter feeders, ingesting algae into their mouth through a selfꢀ
produced water current. These animals are stable even if exposed to different feeding sources
5
7,58
and various stresses.
By exposing them to fungi, viruses or bacteria, the stressꢀinduced
One remarkable feature of the D. magna resides in their transparent
5
9,60
response is detectable.
6
1
shell, which makes them an ideal candidate for microscopy studies. Teplova et al. used
different dyes to separate and image the intoxicationꢀdeath step in the D. magna body. By
feeding D. magna with luminescent objects, it remained possible to study the ecotoxicity of a
material.
ACS Paragon Plus Environment
5

Biomacromolecules
Page 6 of 41
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Our aim was to develop luminescent CNF by controlled radical polymerization of block
copolymer grafts from purposely synthesized NFCꢀbased macroinitiators. The first polymer
block comprises methyl acrylate to enhance the stability of the suspension, suppressing fibrilꢀ
fibril interactions, whilst the second block is composed of acrylic acid Nꢀhydroxysuccinimide
ester, in order to introduce functional anchoring sites that are modified with a luminescent dye.
FCS and confocal scanning laser microscopy enabled determination of the photophysical
properties in suspension and at the molecular scale of the resulting fluorescently tagged CNF.
We finally exposed the fluorescently modified CNF to living juvenile daphnids to explore the
bioꢀmarker potential.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
EXPERIMENTAL SECTION
Materials
1,1’ꢀCarbonyldiimidazole (CDI), 2ꢀBromoꢀ2ꢀmethylpropionic acid 98%, imidazole ≥ 99%,
methyl acrylate 99% and acrylic acid Nꢀhydroxysuccinimide ester ≥ 90% were purchased from
SigmaꢀAldrich. Dimethyl sulfoxide (DMSO, ≥ 99%) was purchased from Merck. Lucifer Yellow
cadaverine was purchased from Interchim. Copper wire (diameter 0.812 mm) was purchased
from Fisher. For the production of cellulose nanofibrils a bleached neverꢀdried softwood sulphite
dissolving pulp (Domsjö mill, Domsjö Fabriker AB, Sweden) was used. A monoꢀcomponent
endoglucanase (FiberCare R, Novozymes, Denmark) was used as received. MilliꢀQ water was
used for the solvent exchange procedure.
Extraction of cellulose nanofibrils (CNF) from wood pulp
ACS Paragon Plus Environment
6

Page 7 of 41
Biomacromolecules
1
2
3
4
5
6
7
8
9
The CNF was prepared by a combined refining and enzymatic preꢀtreatment procedure as
62
described by Pääkkö et al. prior to homogenization in a Microfluidizer Mꢀ110EH
Microfluidics Corp., USA). No biocide was added after the preꢀtreatment. The homogenization
(
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
was carried out at about 2% w/w concentration with repeated passes through fixed Zꢀshaped
chamber pairs connected in series. First, the fibre slurry was passed three times through a
chamber set with dimensions of 400 and 200 µm followed by five passes through a chamber set
with dimensions of 200 and 100 µm at 1700 bar. The total charge density of the CNF was
6
3
determined to be 40 µeq/g by conductometric titration of the unhomogenized pulp. Scanning
Elecron Microscopy micrographs of the lyophilized CNF are shown in Figure S1 (Supporting
information).
Solvent exchange procedure
7
g of CNF (2.06 % w/w) were suspended in 50 mL of water and subsequently stirred for 5 h.
The suspension was then sonicated for 5 min and again stirred for 3 h. The suspension was
finally distributed (10 mL) into several centrifugation tubes and DMSO (30 mL) was added. The
tubes was then centrifuged (4000 RPM, 20 min), the supernatants removed and replaced with
fresh DMSO. The centrifugation operation was repeated 4 times.
CNF-based macroinitiator synthesis
2ꢀBromoꢀ2ꢀmethylpropionic acid (2 g, 12 mmol) was dissolved in 30 mL DMSO and mixed with
CDI (1.95 g, 12 mmol) at room temperature for 1 h. Subsequently, the temperature was raised to
55°C and a suspension composed of CNF (5 g, 1 % w/w) and imidazole (1.5 g, 22 mmol) in 50
ACS Paragon Plus Environment
7

Biomacromolecules
Page 8 of 41
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
mL of DMSO was slowly added. The reaction proceeded for 16 h. Finally, the modified CNF
were purified by centrifugation (4 000 RPM / 20 min). The supernatants were discarded and
replaced with DMSO. The purification steps were repeated 8 times. The modified CNF was
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
13
characterized by FTIR spectroscopy and solidꢀstate MAS C NMR.
Tris[2-(dimethylamino)ethyl]amine (Me -TREN) synthesis
6
64
Me ꢀTREN was synthesized according to literature. Briefly, tris(2ꢀaminoethyl)amine (15 g, 0.1
6
mol) was added dropwise to a solution composed of formaldehyde (160 mL) and formic acid
(160 mL) previously cooled to 0°C. After 1 h stirring, the solution was allowed to warm to room
temperature and subsequently refluxed for 16 h. The solution was concentrated by rotary
evaporation, washed with a saturated solution of NaOH and the product was extracted with
dichloromethane. The solvent was finally removed and the slightly yellow liquid was dried
overnight under vacuum before purification by vacuum distillation.
Polymerization reactions
The CNFꢀbased macroinitiator was suspended in DMSO (10 mL), followed by the addition of 3
g of monomer, either methyl acrylate (MA) or acrylic acid Nꢀhydroxysuccinimide ester (NAS).
A piece of copper wire (diameter=0.812 mm, length=6.25 cm) was added and the suspension
was degassed via nitrogen sparging for 10 minutes and the temperature was raised to 40°C. The
Me TREN ligand was added and the reaction was allowed to proceed for 16h, under a nitrogen
6
atmosphere. The modified CNF were purified through several centrifugation steps (4 000 RPM /
ACS Paragon Plus Environment
8

Page 9 of 41
Biomacromolecules
1
2
3
4
5
6
7
8
9
2
0 min). The different reactants were used with the ratio [M] /[I] /[L] of 200/1/0.2 for the SETꢀ
0
0
0
LRP of MA and [M] /[I] /[L] = 30/1/0.2 for NAS. The modified CNF batches were
0
0
0
13
characterized by FTIR spectroscopy and solidꢀstate MAS C NMR.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fluorescent labelling with Lucifer Yellow cadaverine
The block copolymer modified CNF (2 g, 1 % w/w) was suspended in DMSO (8 mL) before
addition of 2 mL DMSO solution composed of Lucifer Yellow cadaverine (20 mg, 37 µmol) and
triethylamine (22 µL), and stirred at 40°C for 48 h. The luminescent CNF was purified through
several centrifugation steps (4 000 RPM / 20 min), and the purification step was repeated until
no fluorescent signal was detected in the supernatant. The luminescent CNF was characterized
by fluorescence spectroscopy, fluorescence correlation spectroscopy (FCS) and confocal
scanning laser microscopy.
CNF uptake in Daphnia magna.
Daphnia magna is a freshwater crustacean and an established model species in ecotoxicological,
ecological and evolutionary studies. The animals used in this study originate from the test strain
‘‘Klon 5’’, the State office for nature, environment, and customer protection NorthꢀRhine
Westfalia, Bonn, Germany; originally from the Federal Environment Agency, Berlin, Germany).
6
5,66
Daphnids were cultured in M7 medium
in groups of ~25 females in 2 L containers and fed a
ACS Paragon Plus Environment
9

Biomacromolecules
Page 10 of 41
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
mixture of the green algae Pseudokirchneriella subcapitata and Scenedesmus subspicatus three
times a week.
Characterization
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Infrared spectroscopy was performed using a PerkinElmer Spectrum 2000 spectrometer,
equipped with an attenuated total reflection (ATR) accessory. Measurements were normally
ꢀ1
performed by accumulating 64 scans in the spectral region of 4000–550 cm with a spectral
resolution of
ꢀ
1
4
cm .
Fluorescence spectra were obtained using a Varian Cary Eclipse Fluorescence
spectrophotometer. Single nanoparticle fluorescence visualization was performed using an
inverted Zeiss Axiovert Observer.Z1 microscope equipped with a LSM5 exciter. The sample was
visualized with a 63x/1.4 NA oilꢀimmersion objective lens, excited with a diode laser (405 nm),
together with a beam splitter (HFT 405/488/543/633) and a longꢀpass filter (LP 530).
Fluorescence microscopy was used to identify the uptake and presence of CNF in the guts of the
daphnids. Pictures were taken on live juvenile daphnids (age 24ꢀ48 old) sampled from the culture
and immediately exposed to 1 g/L CNF for 3 hours. The microscopeꢀcameraꢀsetup consisted of
a Canon EOS 5d Mark III camera (Canon Inc, Tokyo, Japan) mounted via a camera tube (model:
DD20DMT, Diagnostics Instruments Inc, Sterling Heights, USA) to a Leitz DMRBE microscope
(Leica microsystems GmbH, Wetzlar, Germany).
Fieldꢀemission Scanning electron microscopy (FEꢀSEM) imaging was performed with a Hitachi
Sꢀ4800 field emission scanning electron microscope, operating at 1 kV. Samples were mounted
on carbon tapeꢀcoated stubs and sputter coated with a 12 nm thick layer of Pt/Pd using a
Cressington 20HR sputter coater.
ACS Paragon Plus Environment
10

Page 11 of 41
Biomacromolecules
1
2
3
4
5
6
7
8
9
13
All C NMR experiments were performed with a Bruker 500 Avance III HD spectrometer at
1
3
1
Larmor frequencies of 125 MHz and 500 MHz for C and H, respectively. The samples were
13
packed in 4 mm zirconia rotors and spun at 8 kHz. Ramped crossꢀpolarization (CP) C MAS
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
NMR spectra were recorded with a C nutation frequency of 50 kHz and contact time 1.5
1
ms. Highꢀpower H decoupling was achieved by the TPPM technique using a nutation frequency
of 80 kHz. 4096 signal transients were accumulated with relaxation delays from 3 to 15 s,
depending on relaxation time estimated for each sample. Signal apodization by an 30 Hz
1
3
Lorentzian broadening was applied before Fourier transformation and C chemical shifts are
quoted relative to neat tetramethylsilane (TMS).
FCS measurements were performed on an FCSꢀequipped Zeiss 780 confocal microscope (Zeiss,
Jena, Germany). The sample was excited by a 405 nm laser line (300 µW) focused through a Cꢀ
Apochromat 40X/1.2 NA waterꢀimmersed objective via a dichroic mirror. The fluorescence was
detected by the same objective and was spectrally divided and detected by a 32 channel GaAsP
detector after passage through a pinhole in the image plane. The size of the confocal detection
volume was estimated by measurements on the dye Lucifer Yellow, which was assumed to have
2
a diffusion coefficient of about 400 ꢀm /s in in water and thereby a diffusion coefficient of about
2
2
00 ꢀm /s in DMSO. This yielded a detection volume of 0.2 fL. FCS analysis was performed
using the Zen 2012 software (Zeiss, Jena, Germany) as well as in house written functions using
Origin 9.1 (Originlab Corporation, USA). The autocorrelation function in an FCS experiment is
calculated as:
ꢄꢅꢁꢆꢃ∙ꢅꢁꢆꢇꢈꢃꢉ ꢄꢋꢅꢁꢆꢃ∙ꢋꢅꢁꢆꢇꢈꢃꢉ
ꢀ
ꢁꢂꢃ =
=
+1
ꢄꢅꢁꢆꢃꢉꢊ
ꢄꢅꢁꢆꢃꢉꢊ
Here F is the detected fluorescence intensity, δF is the deviation from the mean fluorescence
intensity at a certain time point, (δF(τ)=F(τ) ꢀ〈I〉), and brackets denote mean value.
ACS Paragon Plus Environment
11

Biomacromolecules
Page 12 of 41
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
For a sample containing particles of uniform size, and where translational diffusion is the only
process giving rise to fluorescence fluctuations, the autocorrelation function in an FCS
measurement is fitted to:
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢌ
ꢌ
ꢌ
ꢐ
ꢀ
ꢁτ)=
ꢎ
ꢒ/ꢊ+1
ꢊ
ꢎ
ꢍ ꢁꢌꢇꢎ
ꢃ
ꢁꢌꢇ
ꢃ
ꢏ
ꢑ
ꢊ
ꢎ
ꢏ
Where N is the mean number of particles in the detection volume, τD is the diffusion time of
particles through the detection volume, and ω and z denote the distances in the radial and axial
dimensions respectively, at which the average detected fluorescence intensity has dropped to eꢀ2
of its peak value.
RESULTS AND DISCUSSION
To produce a (multi)functional platform, bioꢀmarker or sensor, cellulose nanofibrils (CNF) were
chemically modified through an esterification reaction and two consecutive Cu(0)ꢀmediated
controlled radical polymerizations to further exploit the buildꢀup of block copolymer grafts and
1
3
the labelling of specific entities. In contrast to the soꢀcalled TEMPOꢀoxidized CNF, where the
native cellulose is converted/processed to individual fibrils through the introduction of surface
charges which induce an electrostatic repulsion between the fibrils, our CNF source was
composed of an abundant numbers of hydroxyl groups on the surface, enhancing the possibility
of an interꢀfibril interaction. The growth of the first polymer block aims to suppress this tendency
of fibrilꢀtoꢀfibril aggregation, yielding a stable CNF suspension over the time. The second
polymer block will allow introducing reactive functionality on to the CNF surface, producing a
highly reactive CNF platform toward amineꢀbased entities. Finally, the block copolymers
ACS Paragon Plus Environment
12

Page 13 of 41
Biomacromolecules
1
2
3
4
5
6
7
8
9
modified CNF was labelled with a waterꢀsoluble luminescent probe through an amidation
reaction.
Firstly, CNFꢀbased macroinitiators were successfully prepared via an esterification reaction
between 2ꢀBromoꢀ2ꢀmethylpropionic acid and the hydroxyl group of the nanocellulose
backbone, yielding–C(CH3)2ꢀBr pendant groups on the surface (Scheme 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Synthetic pathway for the preparation of CNFꢀbased macroinitiators.
Next, we explored the ability of the CNFꢀbased macroinitiator to initiate Cu(0)ꢀmediated
controlled radical polymerizations of methyl acrylate (MA) and acrylic acid Nꢀ
hydroxysuccinimide ester (NAS) monomers, respectively (Scheme 2). The polymerization of
MA, initiated by the modified CNF, was conducted run in the presence of Cu(0) and a
tetradendate tertiary amine ligand (Me TREN) in DMSO (Scheme 2a). After several purification
6
steps, the graft polymer modified CNF initiated the second SETꢀLRP of the NAS monomer,
yielding a reactive block copolymer modified CNF. In these two polymerization steps, the
monomer feeds were set to achieve a higher degree of polymerization of the first block (factor:
6.6). Finally, the Lucifer yellow cadaverine was labelled onto the modified CNF through an
amidation reaction (Scheme 2b).
ACS Paragon Plus Environment
13

Biomacromolecules
Page 14 of 41
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 2. Synthetic pathway for the labelling of fluorescent probes onto surface modified CNF:
(a) Synthesis of a block copolymer onto CNFꢀbased macroinitiators through SETꢀLRP with
methyl acrylate and acrylic acid Nꢀhydroxysuccinimide ester. (b) Fluorescent labelling of the
block copolymerꢀmodified CNF with the Lucifer yellow cadaverine by amidation.
The structures of the different chemically modified CNF were verified by ATRꢀFTIR and are
compared in Figure 1. Unmodified CNF (Figure 1a) shows the characteristic absorption bands
ꢀ1
ꢀ1
ꢀ1
ꢀ1
located at 3320 cm (OꢀH), 2950 and 2895 cm (CꢀH), 1430 cm (CꢀH) and 1161 cm (CꢀOꢀC).
In addition to these absorption bands, the CNFꢀbased macroinitiator (Figure 1b) shows an
ꢀ1
additional band located at 1736 cm , attributed to the carbonyl group. Unfortunately, this
additional absorption band is not intense, due to a relatively low substitution degree of the 2ꢀ
bromoꢀ2ꢀmethylpropionic acid onto the CNF. Nevertheless, the FTIR spectra of the polymerꢀ
ꢀ1
modified CNF (Figure 1c) shows strong additional absorption bands located at 2954 cm (CꢀH),
ꢀ1
ꢀ1
ꢀ1
1727 cm (C=O), 1437 cm (CꢀH), 1159 cm (CꢀOꢀC) indicating the expected grafting of PMA.
With the addition of the second monomer (NAS), the spectra of the block copolymer modified
CNF did not show any new absorption bands. The (initial) absorption bands of the NAS
ACS Paragon Plus Environment
14

Page 15 of 41
Biomacromolecules
1
2
3
4
5
6
7
8
9
ꢀ1
ꢀ1
ꢀ1
monomer were located at 1821 cm , 1775 cm and 1711 cm , attributed to the succinimide
ꢀ1
group, and 1200 cm (CꢀO, ester) overlapped with the absorption bands of the first polymer
block modifiedꢀCNF bands. Moreover, the carbonyl band intensity (or ratio if compared to the
OH band) did not increase since the precursor ratio was different for the two blocks (factor: 6.6).
Indeed, the second chain block will have a shorter degree of polymerization if compared to the
first block. The different FTIR spectra indicate that the surface modifications of the CNF through
an esterification reaction and subsequent graft polymerization were successful.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. FTIR spectra of (a) unmodified CNF, (b) CNFꢀbased macroinitiators, (c) Poly(MA)ꢀ
grafted CNF and (d) Poly(MAꢀblockꢀNAS)ꢀgrafted CNF.
ACS Paragon Plus Environment
15

Biomacromolecules
Page 16 of 41
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
13
CPMAS C spectra of unmodified cellulose nanofibrils, CNFꢀbased macroinitiator, and
poly(MAꢀblockꢀNAS)ꢀgrafted CNF, as well as the corresponding chemical structures are shown
in Figure 2, with the various chemical shifts provided at the top of each spectrum. The spectrum
of the initial, unmodified, CNF showed the characteristic bands of cellulose, with bands located
at 84 ppm (C4) and 62 ppm (C6) for the surface carbon sites, at 89 (C4) and 66 ppm (C6) for the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
7–69
carbon inside the crystalline region. These peak positions agreed with previous reports.
The
CNFꢀbased macroinitiator spectrum (Figure 2b) includes a new carbonyl bond signal (C7, 169
ppm), which has low intensity compared to the integration of the peaks from the initial cellulose,
from which a substitution degree of 2 % was estimated. This observation confirmed the
relatively low substitution degree of the 2ꢀbromoꢀ2ꢀmethylpropionate onto the CNF, in
agreement with the previous observations made with FTIR. The poly(MAꢀblockꢀNAS)ꢀgrafted
CNF spectrum (Figure 2c), on the other hand, shows the appearance of new intense peaks
1
3
characteristic of the MAꢀblockꢀNAS: the C NMR peak at 175 ppm is attributed to the sites
C12, C16 and C17 (carbonyl bond); the band located at 52 ppm corresponds to the methyl group
(
C13) and the methylene in βꢀposition of the carbonyl group (C10 and C14), whereas the bands
located between 41 and 36 ppm are attributed to the sites C11, C15 and C18. The integration of
the NMR peaks revealed a ratio of 0.9 of the poly(MAꢀblockꢀNAS)ꢀgrafted polymer to the signal
of the initial cellulose. Unfortunately, it was not possible to isolate the carbonyl bonds signals
from the MA block and the NAS block, and thus estimate the polymerization degree of each
block.
ACS Paragon Plus Environment
16

Page 17 of 41
Biomacromolecules
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
3
Figure 2. Solid state C NMR spectra at 500 mHz and chemical structures of a) CNF, b) CNFꢀ
based macroinitiator, and (c) Poly(MAꢀblockꢀNAS)ꢀgrafted CNF. The spinning side bands are
localized at 138.8, 136.2, 11.8 and 9.2 ppm.
UVꢀvisible and fluorescence spectra of Lucifer yellow and the Lucifer yellow labelled CNF
suspended in DMSO are shown in Figure 3. For a better comparison the absorption and emission
spectrum were rescaled with respect to the maximum intensity. The Lucifer yellow is
characterized by absorption and emission bands localized at 435 nm and 514 nm, respectively.
The spectra confirmed that tethering of the free Lucifer yellow to the block copolymerꢀmodified
CNF did not affect the position of the absorptionꢀemission bands.
ACS Paragon Plus Environment
17

Biomacromolecules
Page 18 of 41
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Normalized UVꢀVisible (solid line) and fluorescence spectra (dashed line) of Lucifer
yellow (black) and Lucifer yellow labelled onto CNF (red).
One of the most appropriate techniques for probing the photophysical properties of a dye at the
4
7,48
single molecule level is fluorescence correlation spectroscopy (FCS),
since it is possible to
detect the intensity fluctuation of a single chromophore passing through the focal point. Through
this approach, a small volume (< 1 µL) of highly diluted chromophore solution (pM) is excited
through a focused laser beam passing by a microscope objective. The subsequent fluorescence is
detected via a confocal pinhole and monitored through a singleꢀphoton detector. Each time a
chromophore diffuses through the observation volume a fluorescence signal is generated, and the
fluctuations thus generated are analyzed by the automatically calculated correlation function.
This technique will provide information on the photodynamics of the dye (blinking,
photobleaching), as well as the concentration and diffusion time of the luminescent species (i.e.
small molecules or large objects will have fast or slow diffusion times, respectively) in the focal
5
2,70
volume.
In our experiment, the unlabelled Lucifer Yellow (LY), i.e. the free dye, and the Lucifer yellow
labelled CNF (LYꢀCNF) were studied with FCS. The fluorescence correlation curves are shown
ACS Paragon Plus Environment
18

Page 19 of 41
Biomacromolecules
1
2
3
4
5
6
7
8
9
in Figure 4 and the results are summarized in Table 1. The free dye curve (Figure 4, top) is
composed only of fast diffusing molecules in the focal volume (τ =45 ꢀs). In the case of the
diff
dyeꢀlabelled CNF (Figure 4, bottom), the fluorescence signal is drastically different and is only
composed of slowly diffusing species, i.e. large objects. The diffusion time of the luminescent
fibrils was estimated to be 500 ms. This corresponds to a diffusion coefficient of fluorescent
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
2
CNF of DCNF≈0.018 ꢀm /s, approximately 11 000 times smaller than that of LY, D ≈200 ꢀm /s.
LY
In the most highly concentrated samples of CNF, FCS yielded a concentration of 0.35 nM on
average.
ACS Paragon Plus Environment
19

Biomacromolecules
Page 20 of 41
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Fluorescence correlation curves of Lucifer yellow (LY), unlabelled dye (top) and
Lucifer yellowꢀmodified CNF (bottom). Both samples were recorded in DMSO.
It is worth noting that the amplitude of the two autocorrelation curves is highly different. The
curve amplitude is inversely dependent of the average number of fluorescent molecules diffusing
in the focal volume. As such, as the autocorrelation amplitude increased, the number of observed
ACS Paragon Plus Environment
20

Page 21 of 41
Biomacromolecules
1
2
3
4
5
6
7
8
9
chromophores decreased. From the estimated concentration and the fluorescence countꢀrate the
brightness of the dye can be calculated. The brightness of LY was 18.7 kHz per molecule when
excited at 300 µW, whereas the brightness of the LYꢀ CNF was 247 kHz per molecule (i.e. per
fibril) when excited at 0.6 µW. Taking the measured difference in brightness as well as the
difference in laser power into account, the CNF fibrils were approximately 6600 times brighter
than individual LY dye molecules, and accordingly each CNF should contain at least 6600 LY
dye molecules.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As a control experiment, and to ensure that nonspecific binding between the free dyes with the
CNF was not possible, a separate measurement was performed by simply mixing the LY with the
chemically modified CNFꢀbased macroinitiators. That measurement yielded a diffusion time of
45 ꢀs, which corresponds to the diffusion time of the free dye. Thus, nonspecific binding
between the free dyes and the CNF was not observed, thus we conclude that the luminescent
signal stems from LY covalently bound to the CNF.
Table 1. Summary of the results obtained by FCS for the LY free dye and labelled onto the
modified CNF in DMSO.
Brightness of
CNF/ Brightness of
Samples
LY
τ
diff / ꢀs
Brightness / kHz
18.7
LY
45
ꢀꢀꢀꢀꢀꢀꢀ
6600
0
5
a
LY_CNF
Physical blend
5 x10
45
247
18.7
ACS Paragon Plus Environment
21

Biomacromolecules
Page 22 of 41
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
The brightness of the LYꢀlabelled CNF was 247 kHz per fibril.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In FCS the diffusion coefficient D is usually given by the measured diffusion time (τ ) from the
D
2
equation τ =ω /4D, and from D the hydrodynamic radius R of the particles can be obtained
D
h
6
2,71
from the StokesꢀEinstein equation. However in the current situation, the analyzed CNF
are
not spherical particle within the nanometer scale range but instead object with a size larger than
the detection focus. Thus, the measured τ must be interpreted differently in order not to
D
overestimate the R value.
h
72
Based on the results from Wu et al. , we extrapolate our result, and with changing the solvent
viscosity, a diffusion time of 500 ms corresponds to a hydrodynamic radius of 700 nm (Figure
5
).
ACS Paragon Plus Environment
22

Page 23 of 41
Biomacromolecules
1
2
3
4
5
6
7
8
9
Figure 5. Estimation of the hydrodynamic radius R of the CNF. Two particle sizes with their
h
7
2
diffusion time, plus the origo, were converted from Wu et al. , adjusted for the viscosity of
DMSO, and fitted with a second order polynomial for extrapolation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As concluded above, the average hydrodynamic radius of CNF was 700 nm and the average
fluorescence brightness per fibril corresponds to that of 6600 LY fluorophores. The
hydrodynamic radius corresponds to a spherical particle volume of about 1.5 fL, while the FCS
detection volume is about 0.19 fL. In other words, the part of the CNF which is inside the FCS
detection volume gives rise to a fluorescence signal corresponding to 6600 LY fluorophores, but
a spherically shaped CNF would be about eight times larger than the FCS detection.
Accordingly, if the fibrils are spherically shaped then each fibril is labelled about 8 times more
than the 6600 LY detected (50 000LY unit/CNF). If instead the fibrils are rodꢀshaped, they are
most likely aggregated, since a rodꢀshaped CNF of 5 nm diameter would not be able to
accommodate enough LY fluorophores inside the FCS detection volume without causing
significant fluorescence quenching.
Confocal laser scanning microscopy was used to image the LY labelled on the block copolymerꢀ
modified CNF. The fluorescence images were obtained with a 405 nm excitation wavelength,
with a beam splitter (HFT 405/488/543/633) and a longꢀpass filter (LP 530). The resulting
images (fluorescence and combined overlay fluorescenceꢀbrightꢀfield images) were obtained
with two different magnifications and are shown in Figure 6. For clearer visualization, the dyes
were spotted in blue. Luminescent spots were only observed along the CNF, which confirmed
the absence of unlabelled dye in the suspension. Moreover, the fluorescent spots were
homogeneously distributed onto the CNF.
ACS Paragon Plus Environment
23

Biomacromolecules
Page 24 of 41
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. Confocal scanning laser microscopy imaging of LY labelled on the block copolymerꢀ
modified CNF. (Left) Fluorescence image obtained with an excitation wavelength of 405 nm,
(middle) brightꢀfield image (right) combined overlay fluorescenceꢀbrightꢀfield images.
In the previous sections we have shown that the chromophoreꢀlabelled CNFs exhibit the desired
fluorescence, allowing precise imaging of the fibrils by confocal laser scanning microscopy. As
such it may serve as a potential bioꢀmarker that allow for the fluorescenceꢀbased optical
detection of CNF uptake and distribution in living organisms. For this purpose, we choose the
65,66
Daphnia magna. Daphnids were cultured in M7 medium
in groups of ~25 females and fed a
mixture of the green algae Pseudokirchneriella subcapitata and Scenedesmus subspicatus three
times a week. The live juvenile daphnids were then exposed to the LYꢀCNF (1g/L) for 3 h.
During this period, no dead species were observed. The experiment was repeated 5 times to
ensure a good reproducibility. Thanks to their transparent shell, fluorescence microscopy was
used to identify the uptake and presence of CNF in the guts of the daphnids (Figure 7). The
bright field images revealed the presence of green algae in the D. magna’s alimentary canal
ACS Paragon Plus Environment
24

Page 25 of 41
Biomacromolecules
1
2
3
4
5
6
7
8
9
tissues, while the fluorescence images revealed the presence of the luminescent CNF throughout
the alimentary canal (Figure 7, top). No fluorescence signal was detected on the control
specimen (Figure 7, bottom). Thus, the D. magna ingested the luminescent CNF, which then
dissipates within the algae in the guts of the daphnids, demonstrating the nonꢀtoxicity of our
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
1
material for the viability of the species. In case of toxic fluorescent material, the fluorescence
emission would have changed due to the lack of transmembrane potential. These results highlight
the nonꢀtoxicity of the modified CNF and as a new kind of nanomaterial based on wood, it
remains highly important to prove their biocompatibility for their future use in bioꢀapplications.
Figure 7. Bright field (left) and fluorescence (right) images of Daphnia magna in (bottom)
control experiment and (top) exposed to LYꢀCNF.
ACS Paragon Plus Environment
25

Biomacromolecules
Page 26 of 41
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
CONCLUSIONS
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
A strategy toward luminescent nanocellulose is devised based on the principle of controlled
radical graft copolymerization and photophysically probed on the molecular scale by FCS. CNF
were first converted to CNFꢀbased macroinitiators able to initiate Cu(0)ꢀmediated controlled
radical polymerization, producing block methyl acrylateꢀcoꢀacrylic acid Nꢀhydroxysuccinimide
ester grafted from the CNF backbones. A luminescent probe (Lucifer yellow derivative) was then
coupled to the modified CNF. FCS probing confirmed that CNF was successfully labelled by LY
fluorophores. For ecotoxicity purpose, the luminescent CNF were exposed to live juvenile
daphnids. After 3 hours, the specimens were subjected to microscopy analysis, revealing the
presence of the luminescent CNF throughout D. magna’s alimentary canal tissues with no
toxicity leading to the death of the specimen. Luminescent CNF were proved to be viable bioꢀ
markers and allow for fluorescenceꢀbased optical detection of CNF uptake and distribution in
organisms such as crustaceans.
ASSOCIATED CONTENT
Supporting Information. Figure S1. SEM picture of the unmodified CNF. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Email: edlund@polymer.kth.se
ACS Paragon Plus Environment
26

Page 27 of 41
Biomacromolecules
1
2
3
4
5
6
7
8
9
Author Contributions
The manuscript was written through contributions of all authors. All authors have given
approval to the final version of the manuscript.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Funding Sources
The Swedish research council Formas (project no 2014ꢀ151).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
The authors thank Formas (project number 2014ꢀ151) for financial support. We also thank E.
Ålander at Innventia AB for the preparation of nanocellulose.
REFERENCES
(
1)
2)
Kalia, S.; Dufresne, A.; Cherian, B. M.; Kaith, B. S.; Avérous, L.; Njuguna, J.;
Nassiopoulos, E. Int. J. Polym. Sci. 2011, 2011, 1–35.
(
Moon, R. J.; Martini, A.; Nairn, J.; Simonsen, J.; Youngblood, J. Chem. Soc. Rev. 2011,
40, 3941–3994.
(3)
Eichhorn, S. J.; Dufresne, A.; Aranguren, M.; Marcovich, N. E.; Capadona, J. R.; Rowan,
S. J.; Weder, C.; Thielemans, W.; Roman, M.; Renneckar, S.; Gindl, W.; Veigel, S.;
Keckes, J.; Yano, H.; Abe, K.; Nogi, M.; Nakagaito, A. N.; Mangalam, A.; Simonsen, J.;
Benight, A. S.; Bismarck, A.; Berglund, L. A.; Peijs, T. J. Mater. Sci. 2010, 45, 1–33.
(4)
(5)
Klemm, D.; Kramer, F.; Moritz, S.; Lindström, T.; Ankerfors, M.; Gray, D.; Dorris, A.
Angew. Chem. Int. Ed 2011, 50, 5438–5466.
Salas, C.; Nypelö, T.; RodriguezꢀAbreu, C.; Carrillo, C.; Rojas, O. J. Curr. Opin. Colloid
Interface Sci. 2014, 19, 383–396.
(
6)
7)
Habibi, Y. Chem. Soc. Rev. 2014, 43, 1519–1542.
(
Koschella, A.; Hartlieb, M.; Heinze, T. Carbohydr. Polym. 2011, 86, 154–161.
ACS Paragon Plus Environment
27

Biomacromolecules
Page 28 of 41
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
8)
Filpponen, I.; Argyropoulos, D. S. Biomacromolecules 2010, 11, 1060–1066.
(9)
Nielsen, L. J.; Eyley, S.; Thielemans, W.; Aylott, J. W. Chem. Comm. 2010, 46, 8929–
8931.
(10) Sassi, J.ꢀF.; Chanzy, H. Cellulose 1995, 2, 111–127.
(11) Eyholzer, C.; Bordeanu, N.; LopezꢀSuevos, F.; Rentsch, D.; Zimmermann, T.; Oksman,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
K. Cellulose 2010, 17, 19–30.
(12) Goussé, C.; Chanzy, H.; Cerrada, M. L.; Fleury, E. Polymer (Guildf). 2004, 45, 1569–
1575.
(13) Isogai, A.; Saito, T.; Fukuzumi, H. Nanoscale 2011, 3, 71–85.
(14) Habibi, Y.; Goffin, A.ꢀL.; Schiltz, N.; Duquesne, E.; Dubois, P.; Dufresne, A. J. Mater.
Chem. 2008, 18, 5002.
(15) Lönnberg, H.; Larsson, K.; Lindström, T.; Hult, A.; Malmström, E. ACS Appl. Mater.
Interfaces 2011, 3, 1426–1433.
(16) Carlmark, A.; Larsson, E.; Malmström, E. Eur. Polym. J. 2012, 48, 1646–1659.
(17) Majoinen, J.; Walther, A.; McKee, J. R.; Kontturi, E.; Aseyev, V.; Malho, J. M.;
Ruokolainen, J.; Ikkala, O. Biomacromolecules 2011, 12, 2997–3006.
(18) Yi, J.; Xu, Q.; Zhang, X.; Zhang, H. Cellulose 2009, 16, 989–997.
(19) Edlund, U.; RodriguezꢀEmmenegger, C.; Brynda, E.; Albersson, A.ꢀC. Polym. Chem.
2012, 3, 2920.
(20) Edlund, U.; Albertsson, A.ꢀC. J. Polym. Sci., Part A Polym. Chem. 2011, 49, 4139–4145.
(
(
(
21) Dax, D.; Xu, C.; Långvik, O.; Hemming, J.; Backman, P.; Willför, S. J. Polym. Sci. Part A
Polym. Chem. 2013, 51, 5100–5110.
22) Voepel, J.; Edlund, U.; Albertsson, A. C. J. Polym. Sci. Part A Polym. Chem. 2011, 49,
2366–2372.
23) Zoppe, J. O.; Habibi, Y.; Rojas, O. J.; Venditti, R. a.; Johansson, L.ꢀS.; Efimenko, K.;
Österberg, M.; Laine, J. Biomacromolecules 2010, 11, 2683–2691.
(24) Percec, V.; Guliashvili, T.; Ladislaw, J. S.; Wistrand, A.; Stjerndahl, A.; Sienkowska, M.
J.; Monteiro, M. J.; Sahoo, S. J. Am. Chem. Soc. 2006, 128, 14156–14165.
(25) Rosen, B. M.; Percec, V. Chem. Rev. 2009, 109, 5069–5119.
(26) Zhang, N.; Samanta, S. R.; Rosen, B. M.; Percec, V. Chem. Rev. 2014, 114, 5848–5958.
(27) Nguyen, N. H.; Levere, M. E.; Kulis, J.; Monteiro, M. J.; Percec, V. Macromolecules
012, 45, 4606–4622.
2
(28) Edlund, U.; Albertsson, A. C. J. Polym. Sci. Part A Polym. Chem. 2012, 50, 2650–2658.
(29) Zhang, Q.; Wilson, P.; Li, Z.; McHale, R.; Godfrey, J.; Anastasaki, A.; Waldron, C.;
Haddleton, D. M. J. Am. Chem. Soc. 2013, 135, 7355–7363.
(30) Nguyen, N. H.; Kulis, J.; Sun, H.ꢀJ.; Jia, Z.; van Beusekom, B.; Levere, M. E.; Wilson, D.
A.; Monteiro, M. J.; Percec, V. Polym. Chem. 2013, 4, 144–155.
(31) Samanta, S. R.; Nikolaou, V.; Keller, S.; Monteiro, M. J.; Wilson, D. A.; Haddleton, D.
ACS Paragon Plus Environment
28

Page 29 of 41
Biomacromolecules
1
2
3
4
5
6
7
8
9
M.; Percec, V. Polym. Chem. 2015, 2084–2097.
(32) Aksakal, R.; Resmini, M.; Becer, C. R. Polym. Chem. 2016, 7, 171–175.
(
(
(
33) Anastasaki, A.; Nikolaou, V.; Nurumbetov, G.; Wilson, P.; Kempe, K.; Quinn, J. F.;
Davis, T. P.; Whittaker, M. R.; Haddleton, D. M. Chem. Rev. 2015, 150730144649001.
34) Edwards, J. V.; Prevost, N.; Sethumadhavan, K.; Ullah, A.; Condon, B. Cellulose 2013,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
20, 1223–1235.
35) Schyrr, B.; Pasche, S.; Voirin, G.; Weder, C.; Simon, Y. C.; Foster, E. J. ACS Appl. Mater.
Interfaces 2014, 6, 12674–12683.
(36) Navarro, J. R. G.; Bergström, L. RSC Adv. 2014, 4, 60757–60761.
(
(
(
37) Navarro, J. R. G.; Conzatti, G.; Yu, Y.; Fall, A. B.; Mathew, R.; Edén, M.; Bergström, L.
Biomacromolecules 2015, 16, 1293–1300.
38) MoralesꢀNarváez, E.; Golmohammadi, H.; Naghdi, T.; Yousefi, H.; Kostiv, U.; Horák, D.;
Pourreza, N.; Merkoçi, A. ACS Nano 2015, 9, 7296–7305.
39) Zhou, D.; Zou, H.; Liu, M.; Zhang, K.; Sheng, Y.; Cui, J.; Zhang, H.; Yang, B. ACS Appl.
Mater. Interfaces 2015, 7, 15830–15839.
(40) Mahmoud, K. A.; Mena, J. A.; Male, K. B.; Hrapovic, S.; Kamen, A.; Luong, J. H. T. ACS
Appl. Mater. Interfaces 2010, 2, 2924–2932.
(
41) Colombo, L.; Zoia, L.; Violatto, M. B.; Previdi, S.; Talamini, L.; Sitia, L.; Nicotra, F.;
Orlandi, M.; Salmona, M.; Recordati, C.; Bigini, P.; La Ferla, B. Biomacromolecules
2015, 16, 2862–2871.
(
42) Furstenberg, A.; Vauthey, E. Photochem. Photobiol. Sci. 2005, 4, 260–267.
43) Stewart, W. W. Nature 1981, 292, 17–21.
(
(44) Navarro, J. R. G.; Lerouge, F.; Cepraga, C.; Micouin, G.; Favier, A.; Chateau, D.;
Charreyre, M.ꢀT.; Lanoë, P.ꢀH.; Monnereau, C.; Chaput, F.; Marotte, S.; Leverrier, Y.;
Marvel, J.; Kamada, K.; Andraud, C.; Baldeck, P. L.; Parola, S. Biomaterials 2013, 34,
8344–8351.
(
(
(
45) Pantoja, S.; Lee, C.; Marecek, J. F.; Palenik, B. P. Anal. Biochem. 1993, 211, 210–218.
46) Hanani, M. J. Cell. Mol. Med. 2012, 16, 22–31.
47) Elson, E. L. Biophys. J. 2011, 101, 2855–2870.
(48) Thompson, N. L.; Lieto, A. M.; Allen, N. W. Curr. Opin. Struct. Biol. 2002, 12, 634–641.
(
(
(
(
49) Magde, D.; Elson, E.; Webb, W. W. Phys. Rev. Lett. 1972, 29, 705–708.
50) Rigler, R.; Mets, U.; Widengren, J.; Kask, P. Eur. Biophys. J. 1993, 22, 169–175.
51) Langowski, J. Methods Cell Biol. 2008, 85, 471–484.
52) Navarro, J. R. G.; Plugge, M.; Loumaigne, M.; SanchezꢀGonzalez, A.; Mennucci, B.;
Débarre, A.; Brouwer, A. M.; Werts, M. H. V. Photochem. Photobiol. Sci. 2010, 9, 1042–
1054.
(
53) Schwille, P.; MeyerꢀAlmes, F. J.; Rigler, R. Biophys. J. 1997, 72, 1878–1886.
(54) Chen, Y.; Müller, J. D.; So, P. T. C.; Gratton, E. Biophys. J. 1999, 77, 553–567.
ACS Paragon Plus Environment
29