Preparation and characterisation of fluorescent chitosans using 9-anthraldehyde as fluorophore

Article abstract of DOI:10.1016/S0008-6215(01)00275-0

Chitosans with chemical composition ranging from a fraction of N-acetylated units (F_A) of 0.01 to 0.61 were used to prepare fluorescence labelled chitosans by reductive amination with 9-anthraldehyde. Fluorescent chitosans with a low theoretical degree of substitution (DS, 0.001-1%) were prepared, and the actual DS of the products were determined by UV and ¹H NMR spectroscopy. The fluorescence excitation and emission spectra of the chitosan with F_A of 0.09 and DS 1% showed an excitation maximum at 254 nm and an emission maximum at 413 nm. The intrinsic viscosities ([η]) of the fluorescent chitosans were compared to those of the original chitosans, showing that the derivatisation procedure lead only to a negligible decrease in [η]. The conformation of these fluorescent chitosans with very low DS-values is not altered and they can conveniently be directly quantified by UV or fluorescence spectroscopy.

Full text of DOI:10.1016/S0008-6215(01)00275-0

Carbohydrate Research 336 (2001) 291–296
www.elsevier.com/locate/carres
Preparation and characterisation of fluorescent chitosans
using 9-anthraldehyde as fluorophore
Kristoffer Tømmeraas,a Sabina P. Strand,a Wei Tian,b Lennart Kenne,b
Kjell M. Va˚ruma,*
aNorwegian Biopolymer Laboratory (NOBIPOL), Department of Biotechnology,
Norwegian Uni6ersity of Science and Technology, N-7491 Trondheim, Norway
bDepartment of Chemistry, Swedish Uni6ersity of Agricultural Sciences, Box 7015, SE-750 07 Uppsala, Sweden
Received 28 May 2001; accepted 18 October 2001
Abstract
Chitosans with chemical composition ranging from a fraction of N-acetylated units (FA) of 0.01 to 0.61 were used
to prepare fluorescence labelled chitosans by reductive amination with 9-anthraldehyde. Fluorescent chitosans with
a low theoretical degree of substitution (DS, 0.001–1%) were prepared, and the actual DS of the products were
1
determined by UV and H NMR spectroscopy. The fluorescence excitation and emission spectra of the chitosan with
FA of 0.09 and DS 1% showed an excitation maximum at 254 nm and an emission maximum at 413 nm. The intrinsic
viscosities ([p]) of the fluorescent chitosans were compared to those of the original chitosans, showing that the
derivatisation procedure lead only to a negligible decrease in [p]. The conformation of these fluorescent chitosans
with very low DS-values is not altered and they can conveniently be directly quantified by UV or fluorescence
spectroscopy. © 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Fluorescent chitosans; Degree of substitution; Intrinsic viscosities
1. Introduction
teries,10 antimicrobial effect,11 and metal ion
removal from aqueous solutions.12
Chitosan is a copolymer of 2-acetamido-2-
Quantitative analysis of polysaccharides im-
plies total hydrolysis of the monosaccha-
ride(s). However, highly de-N-acetylated
chitosans are relatively resistant to acid hy-
drolysis due to their positively charged 2-
amino groups.13 Enzymes such as chitosanases
may be used for depolymerisation of chi-
tosans,14 however both availability and cost
limit the use of enzymes. Chitosans can be
quantified by the use of anionic dyes,15 or a
radioactive label.
deoxy-b- -glucopyranose (GlcNAc, A-unit)
D
and
2-amino-2-deoxy-b- -glucopyranose
D
(GlcN, D-unit) obtained by de-N-acetylation
of chitin, a structural polysaccharide in the
exoskeleton of arthrophods and the cell wall
of certain fungi.1 Chitosans and its derivatives
are currently being studied for potential uses
in such diverse fields as drug and gene deliv-
ery,2–8 enzyme immobilisation,9 polymer bat-
We have previously reported on the use of
ninhydrin for quantitative analysis of chi-
tosan, finding that both the chemical composi-
tion and the chain length of the chitosans were
* Corresponding author. Tel.: +47-73-593324; fax: +47-
73-591283.
E-mail address: kjell.morten.vaarum@chembio.ntnu.no
(K.M. Va˚rum).
0008-6215/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0008-6215(01)00275-0

292
K. Tømmeraas et al. / Carbohydrate Research 336 (2001) 291–296
important in order to obtain reliable calibra-
tion curves16. However, other primary amines
(e.g. proteins) can also react with ninhydrin,
excluding the use of the ninhydrin method to
quantify chitosans in e.g. food systems.
and emission scan spectra of fluorescent chi-
tosans (3) with different degrees of substitu-
tion (DS) were acquired from 2.0×10−6
M
(GlcN/GlcNAc units) water solutions using a
10-mm cuvette. The fluorescent emission scan
spectra of 2 and ANM were acquired from
In this paper we describe a simple method
to introduce a certain amount of fluorescent
groups into chitosans to make them fluores-
cent without essentially affecting their poly-
mer properties. These labelled fluorescent
chitosans may be used as standard samples for
the quantitative analysis of chitosans by either
fluorescence or UV spectroscopy. Fluores-
cence labelling of chitosan can be used as an
alternative to radioactive assays with the addi-
tional advantage that the chitosans can be
observed directly with e.g. fluorescence mi-
croscopy.17 Another example is the use in
ultracentrifugation studies of chitosans.18
Fluorescence labelled chitosans may also be
useful to detect chitosans in biological
systems.
1
2.0×10−8 M MeOH solutions. The H NMR
spectrum of 3 was acquired by a Bruker DPX
400 spectrometer at 400.13 MHz and 90 °C
(10 mg in 0.7 mL D2O). All samples were
supplied with 5 mL of a 1% solution of sodium
3-(trimethylsilyl)propionate-d4 in D2O as an
internal standard.19 In order to ensure that the
relative areas of the observed resonances rep-
resent the relative amounts of protons in-
volved, especially for the protons in the
9-anthryl group relative to the chitosan pro-
tons, we accumulated free induction decays
with a 30° pulse and a repetition time of 13 s.
Spectra with high signal-to-noise ratios were
recorded by obtaining 256 scans of each sam-
ple. UV spectra were recorded on a Perkin–
Elmer
UV–Vis
spectrophotometer
on
solutions of 2, ANM (2.0×10−6 M in
MeOH) and fluorescent chitosan (2.0×10−4
M GlcN/GlcNAc units in water). The calibra-
tion curve was obtained by measuring the
absorbance of increasing concentrations of
ANM (0.50–11×10−6 M) at 250 nm. Deter-
mination of DS for the different fluorescent
chitosans was also performed at this wave-
length. Intrinsic viscosities were measured at
20 °C by the use of an Ubbelohde Viscosime-
ter (Type Schott 53101/0a) as previously
described.20
The coupling reaction.—To a solution of
chitosan 1 (1.0 g, 2–5 mmol GlcN) in 2%
AcOH (120 mL), a solution of 2 (in the cho-
sen amount) in MeOH (120 mL) was added.
The mixture was stirred for 24 h at rt, sodium
cyanoborohydride (4.5 mg, 0.07 mmol) added
and the stirring was continued for 12 h. The
reaction mixture was evaporated under vac-
uum at a temperature below 30 °C until the
remaining volume was approximately 100 mL,
and then extracted with EtOAc (3×100 mL)
to remove excess reagents. The water phase
was dialysed first against 0.2 M NaCl and
then against deionised water. The pH of the
dialysed solution was adjusted to 4.5 with HCl
before lyophilisation. White fluorescent chi-
2. Experimental
Materials.—The following chitosans with
different intrinsic viscosities ([p]) and fractions
of N-acetylated units (FA), were provided by
Pronova Biopolymer (Drammen, Norway)
and used as starting material (1): No. 1 ([p]=
1100 mL/g, FA=0.15), No. 2 ([p]=230 mL/g,
FA=0.09), No. 3 ([p]=900 mL/g, FA=0.61)
and No. 4 ([p]=830 mL/g, FA =0.01). All
final products were purified by dialysis using
Spectra/por 3 (molecular weight cut-off 3500
Da). The fluorophore 9-anthraldehyde (2) was
purchased from Aldrich. A mixture of 2 (vary-
ing amounts) and sodium cyanoborohydride
(157 mg, 2.5 mmol) in MeOH (20 mL) was
stirred at rt over night. The mixture was evap-
orated to one third of the original volume,
and then water (10 mL) was added. The pre-
cipitated product was collected and recrys-
tallised (1:1 MeOH–water) as light yellow
1
needles, confirmed by H NMR and melting
point to be 9-anthracenemethanol (ANM),
also available commercially.
Analysis.—The fluorescence analysis was
performed on a Perkin–Elmer LS-5B lumines-
cence spectrometer. The fluorescent excitation

K. Tømmeraas et al. / Carbohydrate Research 336 (2001) 291–296
293
tosan-chlorides with DS of 0.001–1% were
obtained in yields generally exceeding 90%.
The data of the starting chitosans and the
fluorescent chitosans are listed in Table 1.
shown in Scheme 1. The condensation reac-
tion of a primary amino group (nucleophile)
with a carbonyl group (an aldehyde or a ke-
tone) results in an unstable imino bond by the
elimination of a water molecule. This Schiff
base type of equilibrium reaction is strongly
dependent on the pH of the solution and other
factors stabilising or destabilising the product,
such as water concentration, degree of conju-
gation of the resulting product,21 and the
3. Results and discussion
9-Anthraldehyde (2) was introduced on the
different chitosans by reductive amination as
Table 1
Theoretical and actual degree of substitution (DS), yield, intrinsic viscosity ([p]) and measured log o for the fluorescent chitosans
Starting chitosan
Fluorescent chitosan
DS a (mol%)
DS b (mol%)
Yield (%)
[p] (mL/g)
Log m (254 nm)
Chitosan No. 1
FA=0.15, [p]=1100 mL/g
0.014
0.14
1.4
0.02
0.14
1.2 [1.0]
98
97
97
950
810
1030
1.34
2.18
3.11
Chitosan No. 2
FA=0.09, [p]=230 mL/g
0.0014
0.014
0.14
0.008
0.06
0.26
98
96
95
96
230
240
170
210
0.92
1.79
2.45
3.47
1.4
2.7 [1.0]
Chitosan No. 3
FA=0.61, [p]=900 mL/g
0.014
0.14
1.4
0.02
0.12
1.6 [0.9]
97
98
96
850
810
700
1.26
2.12
3.24
Chitosan No. 4
FA=0.01, [p]=830 mL/g
0.014
0.14
1.4
0.04
0.25
1.6
97
96
96
850
850
840
1.67
2.44
3.24
m: the extinction coefficient from the Lambert–Beer law: A=mcl. DS: degree of substitution (mol% of sugar units (GlcNAc+GlcN)
substituted with 2).
a Theoretical value.
b Value calculated from UV measurements using calibration curve (Fig. 4). Values in brackets are determined by 1H NMR
spectroscopy.
Scheme 1. Reductive amination of chitosan (1) with 9-anthraldehyde (2) in the presence of NaCNBH3 to obtain the fluorescent
chitosan (3).

294
K. Tømmeraas et al. / Carbohydrate Research 336 (2001) 291–296
groups of chitosan (1) is the nucleophile con-
densing with the aldehyde group of 9-an-
thraldehyde (2), resulting in the secondary
amine of the fluorescent chitosan (3).
The molar ratios of monomers of 1:2 in the
coupling reaction were ꢀ100:1, 1000:1,
10,000:1 and 100,000:1, in order to synthesise
chitosans with a chromophore which can con-
veniently be visualised and quantified by UV or
fluorescence spectroscopy, without altering the
conformation of the polysaccharide. The
fluorescent chitosans were characterised, and
the data are shown in Table 1. Although a
similar reaction has been reported previously,22
the properties and spectroscopic data of the
product were not given. In the following dis-
cussion, we have chosen the modified chitosan
No. 2 (see Table 1) with DS 1% as an arbitrary
example. The fluorescent properties of 3 were
compared with 9-anthracenemethanol (ANM).
The emission scan spectrum of 3 in water
(excitation wavelength 254 nm) is shown in
Fig. 1. (A) Fluorescent emission scan spectra obtained by
excitation at 254 nm. The reductive amination product 3 with
DS 1% (FA=0.09) in water (—) and 9-anthracenemethanol
(ANM) in methanol (---). (B) Fluorescent excitation scan
spectrum of product 3 with DS 1% (FA=0.09) in water.
choice
of
sodium
cyanoborohydride
(NaCNBH3) as the reducing compound has the
advantage that it selectively reduces the imino
bonds at pH\4.21 In this case, the amino
Fig. 2. 1H NMR spectrum of the fluorescent chitosan 3, where the aromatic region is enlarged (DS 1%, FA=0.15, 400 MHz, D2O,
pD 4).

K. Tømmeraas et al. / Carbohydrate Research 336 (2001) 291–296
295
ppm) and GlcN (4.85 ppm) in the proton
NMR spectrum (Fig. 2) were used to deter-
mine the DS. The aromatic resonances were
too weak for those chitosans with DS lower
Fig. 1(A, solid line) and this is similar to that
of ANM in methanol (dashed line), but has
shifted 5 nm towards longer wavelengths. The
two starting materials 1 and 2 have no emis-
sion in this region. The maximum emission
wavelength of 254 nm was found in the excita-
tion spectrum of 3 (Fig. 1(B)). Fluorescent
chitosans with different DS show the same
emission and excitation scan spectra (data not
shown). No significant difference was ob-
served between the maximum emission wave-
lengths (Emmax) of different chitosan samples.
1
In the H NMR spectrum of 3 (Fig. 2), the
signals at l 2.05 (MeCOꢀ), 3.15 (H-2, D-unit),
3.5–4.0 (H-3, H-4, H-5 and H-6) and 4.85
(H-1, D-unit) and 4.60 (H-1, A-unit) showed
the characteristic protons of chitosan. The
assignment was made by direct comparison to
reported data.23,24 The five small resonances in
the aromatic region (7.65–8.85 ppm) were as-
signed to the protons in the 9-anthryl group as
indicated in Fig. 2. The signals for the two-
methylene protons (ꢀCH2Nꢀ) were not as-
signed, and probably overlap with the
chitosan resonances.
The intrinsic viscosities ([p]) of the fluores-
cent chitosans were only marginally lower
than those of the starting chitosans (Table 1).
It is therefore concluded that the chitosans
can be derivatised by this procedure without
significant degradation. The fluorescent chi-
tosans were prepared with affirmed low substi-
tutions, making it unlikely that the fluorescent
chromophore should affect the conformation
and other properties of the polysaccharide, as
also concluded from conformation studies us-
ing analytical ultracentrifugation.18 The UV
spectrum of the modified chitosans 3 (Fig.
3(a)) is similar to that of the reference ANM
(Fig. 3(b)), but different from that of the
9-anthraldehyde (2) (Fig. 3(c)). Underivatised
chitosan has no UV adsorption in this region
(data not shown). The umax of 3 (254 nm) is
very close to that of ANM (250 nm). This
makes it possible to determine the DS of 3 by
using ANM as a standard reference, assuming
that the extinction coefficients of 3 and ANM
are similar. Two methods have been used for
the determination of the DS of 3. Firstly, the
integrals of the aromatic proton signals and
that of the signals from H-1 of GlcNAc (4.60
Fig. 3. UV spectra of (a) fluorescent chitosan 3 (DS 1%,
FA=0.09) in water; (b) 9-anthracenemethanol (ANM) in
methanol; (c) 9-anthraldehyde (2) in methanol.
Fig. 4. Calibration curve for quantitative determination of DS
values for fluorescent chitosans by UV spectroscopy, prepared
by measuring the absorbance of increasing concentrations of
9-anthracenemethanol (ANM) in methanol.

296
K. Tømmeraas et al. / Carbohydrate Research 336 (2001) 291–296
than 1% to provide a reliable estimation of the
DS. Secondly, UV spectroscopy was used. The
UV method is based on the assumption that
the 9-anthryl group is the main chromophore
in the modified chitosans and the reference,
and that this chromophore determines the UV
absorption in both 3 and ANM. These two
compounds have similar UV curves and the
correlation of the absorbance to the concen-
tration of the fluorescent chitosan 3 is also
linear. This can be used to obtain a calibration
curve to determine the concentration of the
9-anthrylmethylated unit in the modified chi-
tosan 3, as shown in Fig. 4.
Østgaard is thanked for critical reading of the
manuscript.
References
1. Roberts, G. A. F. Chitin Chemistry; Macmillian: Hong
Kong, 1992.
2. Schipper, N. G. M.; Va˚rum, K. M.; Artursson, P. Pharm.
Res. 1996, 13, 1686–1692.
3. Schipper, N. G. M.; Olsson, S.; Hoogstraate, J. A.;
DeBoer, A. G.; Va˚rum, K. M.; Artursson, P. Pharm. Res.
1997, 14, 923–929.
4. Schipper, N. G. M.; Va˚rum, K. M.; Stenberg, P.; Ock-
lind, G.; Lennerna¨s, H.; Artursson, P. Eur. J. Pharm. Sci.
1999, 8, 335–343.
5. Ko¨ping-Ho¨ggard, M.; Tubulekas, K.; Guan, H.; Ed-
wards, K.; Nilsson, M.; Va˚rum, K. M.; Artursson, P.
Gene Ther. 2001, 8, 1108–1121.
6. van der Lubben, M.; Verhoef, J. C.; van Aelst, A. C.;
Borchard, G.; Junginger, H. E. Biomaterials 2001, 22,
687–694.
4. Conclusions
7. Mao, H.-Q.; Roy, K.; Troung-Le, V. L.; Janes, K. A.;
Lin, K. Y.; Wang, Y.; August, J. T.; Leong, K. W. J.
Controlled Release 2001, 70, 399–421.
8. Kato, Y.; Onishi, H.; Machida, Y. J. Controlled Release
2001, 70, 295–307.
9. Noda, T.; Furuta, S.; Suda, I. Carbohydr. Polym. 2001,
44, 189–195.
10. Osman, Z.; Ibrahim, Z. A.; Arof, A. K. Carbohydr.
Polym. 2001, 44, 167–173.
11. Rhoades, J.; Roller, S. Appl. En6iron. Microbiol. 2000, 66,
80–86.
12. Juang, R.-S.; Shiau, R.-C. J. Membr. Sci. 2000, 165,
159–167.
13. Va˚rum, K. M.; Ottøy, M. H.; Smidsrød, O. Carbohydr.
Polym. 2001, 46, 89–98.
14. Izume, M.; Ohtakara, A. Agric. Biol. Chem. 1987, 51,
1189–1191.
15. Roberts, G. A. F.; Taylor, K. E. Macromol. Chem. Rapid
Commun. 1989, 10, 339–343.
16. Prochazkova, S.; Va˚rum, K. M.; Østgaard, K. Carbo-
hydr. Polym. 1999, 38, 115–122.
17. Ga˚serød, O.; Smidsrød, O.; Skja˚k-Bræk, G. Biomaterials
1998, 19, 1815–1825.
18. Co¨lfen, H.; Harding, S. E.; Va˚rum, K. M.; Winzor, D. J.
Carbohydr. Polym. 1996, 30, 45–53.
The condensation of chitosans with 9-an-
thraldehyde, i.e., Schiff base formation, fol-
lowed by reduction with sodium cyanoboro-
hydride renders an easy way to obtain fluores-
cent chitosans that emit at Emmax 413 nm
when excited at Exmax 254 nm. The degree of
substitution of the obtained fluorescent chi-
tosan 3 will depend on the molar ratio of the
two starting materials, and can be quantita-
tively determined by UV or NMR spec-
troscopy. The fluorescent chitosans can be
prepared without significant depolymerisation
of the polysaccharide. Like chitosan, polymer
3 was water soluble at acidic pH values, and
insoluble in methanol and ethanol. Incorpora-
tion of the fluorophore label will allow both
detection and quantification of chitosan in
different applications.
19. Wishart, D. S.; Bigam, C. G.; Yao, J.; Abildgaard, F.;
Dyson, H. J.; Oldfield, E.; Markley, J. L.; Sykes, B. D. J.
Biomol. NMR 1995, 6, 135–140.
20. Draget, K. I.; Va˚rum, K. M.; Moen, E.; Gynnild, H.;
Smidsrød, O. Biomaterials 1992, 13, 635–638.
21. Borch, R. F. J. Am. Chem. Soc. 1971, 93, 2897.
22. Yalpani, M.; Hall, L. D. Can. J. Chem. 1981, 59, 2934–
2939.
23. Domard, A.; Gey, C.; Rinaudo, M.; Terrassin, C. Int. J.
Biol. Macromol. 1987, 9, 233–237.
24. Va˚rum, K. M.; Anthonsen, M. W.; Grasdalen, H.;
Smidsrød, O. Carbohydr. Res. 1991, 211, 17–23.
Acknowledgements
This work was supported by grants from
the Nordic Fund for Technology and Indus-
trial Development (NF), the Swedish Natural
Science Research Council and the Research
Council of Norway (NFR). Professor Kjetill