G. Huerta-Angeles et al. / Carbohydrate Polymers 111 (2014) 883–891
887
is more expensive when compared to a direct use of the commer-
are more interesting due to their potential antineoplastic proper-
ties (Ali, Alazani, Mumtaz, Arun, & Sabir, 2013; Hassani, Hendra, &
Bouchemal, 2012; Khan, Husain, Jabeen, Mustafa, & Owais, 2012;
Wang, Luo, Li, & Zhao, 2012). The reaction was aimed to be scale
up and attain similar products even though the scale factors were
up to 50. In addition, the proposed reaction pathway overcomes
the problem of instability of all commercially available anhydrides
and thus represents an elegant alternative for scale up production
of esterified polysaccharides.
Table 1 indicates that the reaction can be applied for activa-
tion of short (C6–C8), medium (C-10) and long (C-18) fatty acids.
Table 1 shows that higher DS is obtained for fatty acids with shorter
alkyl chain length (entries 1, 5, 8, 12, 17 and 20 in Table 1). To
ensure reproducibility of the reaction, each synthesis was repeated
at least five times. In general, the procedure resulted in at least 5 g
scale. In all the cases, products with identical structural character-
istics, consistent degree of substitution (DS), yield and purity were
obtained.
The increased reactivity for shorter aliphatic acids may be
incorporation of these components in the reaction feed. Addition-
ally, there are also steric factors and increasing hydrophobicity
which cause a decrease of reaction efficiency with increasing alkyl
chain length. Table 1 also shows that the highest DS was reached
anhydride were used in the reaction feed. In this case maximum DS
for hexanoic, octanoic, and decanoic acid was about 35%, while 18%
and 13% of modified HA dimers were obtained for oleic and linoleic
acid, respectively (entries 2, 4, 6, 13 and 18 in Table 1). With higher
molar amounts of the mixed anhydride than two HA equivalents,
the reaction mixture was becoming turbid and an increase in DS
was not observed. In this latter case, large amounts of unbound fatty
acid (more than 15 wt.%) (Table 1) remained in the reaction prod-
uct, artificially increasing the reaction yields above 100% (entries 14
and 15). The presence of high amount of unbound fatty acid could
overlaps in 1H NMR spectrum with the integrated signal of bound
fatty acid (terminal CH3 at 0.8 ppm). Such overestimation did not
occur in other cases (entries 1–13 and 16–20), because there the
amount of free fatty acid was negligible (between 0.01 and 0.5 wt.%,
Table 1).
The molecular weight of starting HA was found to be an impor-
tant parameter affecting the success of the presented modification
pathway. The effect of Mw of starting HA material on DS of the
product is summarized in Table 2. It was observed that low molec-
ular weight HA (4800 g/mol) is more reactive than HA with larger
Mw. For low molecular weight HA high degree of substitution
was obtained even when low molar equivalents of mixed aliphatic
aromatic anhydride were used (Table 2, entry 23). However, the
recovery of the reaction was only about 80%. This is unlikely to be
explained by degradation of the polysaccharide during the reaction.
More likely, the low molecular weight esters of HA were washed
out during the purification process. An improved reaction yield was
reached in case of 15 000 and 38 000 g/mol with DS 9 and 8%, respec-
tively. Otherwise decreasing tendency of DS and reaction yields
were observed with increasing Mw of HA (Table 2). This was prob-
ably caused by an inefficient reaction progress due an increased
solution viscosity driven by HA molecular weight. There was almost
no apparent substitution in the case of high molecular weight HA
(1 800 000 g/mol).
unsaturated (C18:1, C18:2 and C-18:3) fatty acids. However, in the
case of unsaturated long fatty acids, a significant decrease of the
yield was observed with increasing number of unsaturated bonds in
the chain (entries 10–20, Table 1). The lowest yield (about 31%) was
observed for ␣-linolenic acid (entry 20). This observation can be
explained by possible liability of the double bond toward oxidation.
It was our intention to extend the presented methodology
for the covalent attachment of very long fatty acids such as
arachidic (C-20:0) or behenic acids (C-22:0) (entries 21 and 22).
Unfortunately, the results were unfruitful for all the tested reac-
tion conditions. Analyzing the reaction product by FTIR, we have
symmetric anhydride did not react with the polysaccharide. In
agreement with literature data, similar reaction intermediate was
detected during the reaction mechanism of Yamaguchi esterifica-
tion (Dhimitruka & SantaLucia, 2005).
drides on DS, these anhydrides were applied under similar reaction
conditions and the final products were compared. There was no
effect on DS in the case of short and medium fatty acids (data not
shown). However, significant differences were noted for fatty acids
with long chain. As it is shown in Table 1, entries 8 and 9, the DS
was much higher when asymmetric aromatic carboxylic anhydride
was used for the activation.
3.4. Influence of DMAP on the reaction success
alyst increases the nucleophilic acyl transfer of the long aliphatic
fatty acids. This reagent presents a high catalytic activity derived
from the electron-donating capability of the dialkylamino group
(Baidya et al., 2007). Despite the high toxicity of the chemical, at
least a catalytic amount of DMAP is required in order to success-
fully modify HA in the proposed way. There was no substitution
observed in the absence of DMAP.
3.5. Estimation of residual reagents in reaction products
The purity of the derivatives was determined by quantifying
possible entrapped residues of DMAP and TCBC in the product. The
exact DMAP and TCBC contents in final products were estimated by
HPLC methods and were found to be below 0.01% (w/w) for all prod-
ucts. Additionally, non-covalent bonded fatty acids were analyzed
as well. Residual free fatty acids were determined by GC analy-
ses and were found to be between 0.01 and 0.5% (w/w), except
for the samples described in entries 14 and 15 in Table 1. There-
fore, the experimental data revealed that impurities were present
in negligible amounts.
3.6. Analytical characterization of amphiphilic derivatives
It should be noted here that in case of short aliphatic acids
such as hexanoic, heptanoic and octanoic acid the presented activa-
tion methodology involving mixed aliphatic aromatic anhydrides
SEC-MALLS analyses showed an insignificant change of the
weight average molecular weight of the polysaccharide before and
Huerta-Angeles, Gloria
