Accepted Manuscript

Coil-to-Helix Transition of Gellan in Dilute Solutions is a Two-Step Process

Alexander P. Safronov, Irina S. Tyukova, Galina V. Kurlyandskaya

PII: S0268-005X(17)30900-1

DOI: 10.1016/j.foodhyd.2017.07.019

Reference: FOOHYD 3988

To appear in: Food Hydrocolloids

Received Date: 22 May 2017

Revised Date: 05 July 2017

Accepted Date: 19 July 2017

Please cite this article as: Alexander P. Safronov, Irina S. Tyukova, Galina V. Kurlyandskaya, Coil-
to-Helix Transition of Gellan in Dilute Solutions is a Two-Step Process, Food Hydrocolloids (2017),
doi: 10.1016/j.foodhyd.2017.07.019

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form.
Please note that during the production process errors may be discovered which could affect the
content, and all legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

COIL-TO-HELIX TRANSITION OF GELLAN IN DILUTE SOLUTIONS IS A TWO-

STEP PROCESS
Alexander P. Safronov a,b*, Irina S. Tyukova a, Galina V. Kurlyandskaya a,c
aUral Federal University, 51 Lenin Av., 62000, Yekaterinburg, Russian Federation
bInstitute of Electrophysics UD RAS, 106 Amundsen St., 620016, Yekaterinburg, Russian
Federation
cUniversity of the Basque Country UPV-EHU, 48940, Bilbao, Spain
* corresponding author safronov@iep.uran.ru

ABSTRACT

Thermodynamics of coil-to-helix transition of natural polysaccharide - gellan was studied in

dilute solutions in 130 mM NaCL in concentration range 0.01 – 1.5% using SETARAM C80
3D Calvet calorimeter operated in a passive scanning mode at low cooling rate (0.04 K/min at
the transition temperature). It was shown that the negative values of the total enthalpy of the
coil-to-helix transition substantially increased in a step-wise fashion if the concentration of
the solution decreased below the threshold of coil overlapping in accordance with the results
of dynamic light scattering. The shape of thermograms depended on the concentration of
solutions. While in semi-dilute solutions with gellan concentration above 0.25% only one
exothermic peak was indicated in the cooling mode, in dilute solutions two distinct peaks of
the two-step transition: first - endothermic and second - exothermic were clearly observed.
The endothermic and exothermic peaks overlapped for semi-dilute solutions of gellan and
became progressively separated as the concentration diminished. The reasons, which underlie
the two-step nature of coil-to-helix transition of gellan are discussed based on the origin of
molecular interactions involved.

KEYWORDS: Gellan, coil-to-helix transition, differential scanning calorimetry, enthalpy

1
 ACCEPTED MANUSCRIPT

Introduction

Gellan is a commercially available industrial extracellular bacterial polysaccharide

synthesized by Sphingomonas elodea (formerly Pseudomonas elodea) (Kang, Veeder,
Mirrasoul, Kaneko, & Cottrell, 1982, Pollock, 1993). It is commercially used as a gelling
agent in food, pharmacy, medicine, cosmetics and related applications (Gibson, & Sanderson,
1997, Osmałek , Froelich, Tasarek, 2014, Prajapati, Jani, Zala, & Khutliwala, 2013,
Sanderson, 1990). Gellan has an anionic linear chain which comprises tetrasaccharide repeat
units of two residues of b-D-glucose, one of b-D-glucuronate, and one of a-L-rhamnose
(Jansson, Lindberg, 1983, Milas, Shi, Rinaudo, 1990). As biosynthesized, gellan repeat unit
has glyceryl and acetyl substituents in one of the glucose units (Kuo, Mort, & Dell, 1986).
However, in commercial products they are typically removed during the hot alkali treatment
of the fermentation broth. Like other industrial gelling agents such as xanthan, guar, agarose,
etc., gellan can be readily dissolved in hot water and undergoes gelation upon cooling of the
solution, which is in fact a target property for its application. Apart from its undisputable
value for the industrial applicability the process of gelation is a very interesting phenomenon
from the fundamental point of view in respect to the basic principles of behavior of native
polysaccharides in their water solutions. The basic features of gellan gelation were
extensively studied and are presented in several profound reviews (Morris, Nishinari,
Rinaudo, 2012, Nishinari, 1996, Rinaudo, & Milas, 2000, Valli, & Clark, 2010).
It was shown that the essence of the gelation phenomenon of gellan is the coil-to-helix
transition of its macromolecular chains (Morris, et al, 2012, Upstill, Atkins, & Atwool, 1986).
At elevated temperature the coil conformation of gellan macromolecule is thermodynamically
favorable. Upon cooling it, however, converts to the conformation of a double 3-fold left-
handed helix with a pitch of 5.64 nm (Chandrasekaran, Millane, Arnott, & Atkins, 1988,
Morris, et al, 2012). The temperature of this transition is around 40oC and it is affected by a
number of factors like salt concentration, pH, presence of other polysaccharides and organic
additives (Morris, et al, 2012). The coil-to-helix transition in gellan solution provided a
consequent effect on a variety of measurable properties, which were extensively studied
(Morris, et al, 2012). In the present study we focus on the thermodynamic aspects of this
transition, which are presented by the differential scanning calorimetry (DSC).
In several studies (Mazen, Milas, & Rinaudo, 1999, Miyoshi, Takaya, & Nishinari, 1994,
Miyoshi, Takaya, & Nishinari, 1995, Miyoshi, Takaya, & Nishinari, 1996, Miyoshi,
Nishinari, 1999, Ogawa, Takahashi, Yajima, & Nishinari, 2006) it was established that the

2
 ACCEPTED MANUSCRIPT

coil-to-helix transition upon cooling of gellan solution provides an exothermic peak at DSC
thermogram. The peak shifts to high temperatures if salt is added to the solution (Miyoshi, et
al 1995, Miyoshi, et al 1999). The enthalpy of the transition was found to be close to -10 J/g
in the 1% solution (Miyoshi, et al 1999). It was clearly shown that while there is a single
exothermic peak on DCS thermograms of the transition in cooling experiments there can be
one or two endothermic peaks on DSC plots in heating experiments depending on the
concentration of gellan solution (Morris, et al, 2012, Miyoshi, et al 1999). In the
concentration range 1 – 3% of gellan only one endothermic peak was indicated at the same
temperature as the coil-to-helix transition took place upon cooling. At higher concentration,
however, the second endothermic peak at higher temperature was also observed. The
temperature interval between the peaks increased as the concentration grew up. These two
peaks were attributed to the melting of the solitary helixes of gellan and to the melting of the
aggregated helixes, which formed the true gel (Morris, et al, 2012).
In general, it means that the helix-to-coil transition (melting of helixes) of gellan is affected
by the spatial restrictions. Meanwhile, it looks reasonable that the spatial restrictions due to
the overlapping of coils might as well influence the formation of double helixes of gellan in
the coil-to-helix transition upon cooling. The idea to clarify the limits of such restrictions
originated the objective of the present study. We intended to perform DSC measurements at
the conditions providing gradual diminishing of spatial restrictions for the coil-to-helix
transition of gellan. Apparently, it required the measurements in gellan solutions with
progressive dilution below the concentration around 1% which is up to date the lowest
concentration reported in DSC studies. It is shown below that the dilution of gellan solution
strongly affects the total value of the enthalpy and reveals two separate steps of coil-to-helix
transition.

Experimental Section

Materials
Gellan was purchased from Sigma-Aldrich under the brand name PhytagelTM (Sigma-
Aldrich, St. Louis, USA). Gellan solutions were prepared as described elsewhere (Tanaka,
Nishinari, 2007). Gellan powder was vigorously stirred in 130 mM NaCl in distilled water at
90oC until visual homogeneity was achieved. Then obtained solutions were hot filtered
through 10 micron Pyrex glass filter and thermostated for 4 h at 90oC before measurements.

3
 ACCEPTED MANUSCRIPT

Methods
Differential scanning calorimetry (DCS) measurements were performed in the passive
scanning mode, which was developed earlier (Vshivkov, Safronov 1997) for the observation
of a coil – globule phase transition in polystyrene solutions at low concentration and low
cooling rate. We elaborated SETARAM 3D C80 microcalorimeter with 15 cm3 cells. In this
case we gained both from the high sensitivity of the 3D calorimeter sensor and from the
extended capacity of the cell, which allowed to load up to 12 g of gellan solution compared to
maximum 1 g of it in the commercial SETARAM DSC3 evo. However, certain limitation
arose as well. Due to the comparatively large size of a thermostat and the consequently high
thermal inertia SETARAM C80 microcalorimeter is not well suited for the temperature
scanning in a linear mode regime. This mode implies low thermal inertia of the cell to
maintain the stability of a controlled linear baseline. To overcome this limitation we used the
regime which consisted of the four consequent steps: 1) linear heating from room temperature
up to 90oC at 1 K/min, 2) isothermal equilibration at 90oC for 1.5 h, 3) cooling down to 20oC
temperature at 1 K/min, 4) isothermal step for 20 h. In fact, in the third and in the fourth steps
the heater of the thermostat was effectively turned off and the calorimeter slowly cooled
down to the room temperature in a passive mode. The total duration of these steps took
approximately 24 h. The third and the fourth steps of the DSC regime were scheduled for the
nighttime to insure the stability of the external conditions of passive scanning mode. During
the passive scanning both the temperature and the baseline signal exponentially decreased.
Due to the exponential decrease of the temperature the scanning rate was not a constant but a
diminishing value. Meanwhile, it was very low: 0.1 K/min at the beginning of the scanning at
90 oC and 0.04 K/min at temperatures around the coil-to-helix transition. Such a low scanning
rate insured the equilibrium conditions for the transition.
Another limitation of the passive scanning was that the room temperature was the lowest
achievable temperature in the experiment. As the coil-to-helix transition of gellan in pure
water takes place near the room temperature (Morris, et al, 2012), its observation under such
a limitation becomes uncertain. Therefore the study was performed for the solutions of gellan
in 130 mM NaCl. It was reported in literature (Miyoshi, et al 1995, Miyoshi, et al 1999,
Morris, et al, 2012) that the temperature of the transition increases at the presence of simple
electrolytes. At 130 mM NaCl it was reported between 40 oC and 50oC (Morris, et al, 2012).
It made a necessary deviation from room temperature to perform reliable measurements.

4
 ACCEPTED MANUSCRIPT

During the experiments the weighted amount of gellan solution was placed in a 15 cm3
stainless steel working cell of a C80 calorimeter and was tightly closed. The reference cell
contained the equal amount of distilled water. The cells were weighted before and after the
experiment to control the absence of vapor leakage during temperature steps.
Dynamic light scattering (DLS) of gellan solutions at 60oC was performed using
Brookhaven ZetaPlus analyzer. Stock gellan solution (1%) for the measurements was
prepared as described in the Materials section. Then the hot solution was diluted to the
diminishing concentrations. The resulted solutions were not further filtered through fine (100
nm) Wattman filters but placed in optical cuvettes and stored for 20 h at 60oC for the
sedimentation of dust particles. Then they were placed in the analyzer and thermally
equilibrated at 60oC before the measurements.

Results and Discussion

According to the objective of the study DSC thermograms were taken for gellan solutions
with the diminishing concentration to see the effect of spatial limitations upon the enthalpy of
coil-to-helix transition. In all cases we used solutions of gellan in 130 mM NaCl.
Figure 1 presents the DSC thermogram (cooling) for 1% solution of gellan in 130 mM
NaCl obtained in the passive scanning mode. The temperature of the calorimeter decreased as
a smooth exponent. The exothermic peak of coil-to-helix transition is clearly indicated at the
DSC plot. The baseline for the transition was interpolated as a double exponential decay
function fitted to the parts of the thermogram before and after the transition. The baseline is
shown beneath the peak as a dashed line. In the same fashion DSC thermograms were
obtained for the series of diminishing concentrations of gellan solution starting from 1.5%
down to 0.01%. All these thermograms are presented in Figure 2. Each one was adjusted by
baseline subtraction. For the sake of clarity the thermograms are shifted upwards one against
the other. The results given in Figure 2 definitely show that the concentration of the gellan
solution affects the DSC plot for coil – helix transition.

5
 ACCEPTED MANUSCRIPT

Figure 1. DSC thermogram in passive scanning mode for the cooling of 1% solution of
gellan in 130 mM NaCl. 1 – Temperature plot. 2 – Heat flow plot. 3 – Baseline interpolated
as a double exponential decay function

Figure 2. DSC thermograms of coil-to-helix transition of gellan in solutions in 130 mM

NaCl. Gellan concentration 1.5% (1), 0.5% (2), 0.25% (3), 0.1% (4), 0.05% (5), 0.02% (6),

6
 ACCEPTED MANUSCRIPT

0.01% (7). Thermograms are corrected by baseline and shifted upward to compensate
overlapping.

At concentrations of gellan 1.5%, 0.5% (Figure 2), and 1% (Figure 1) a single exothermic
peak is observed on a thermogram. It is consistent with the results reported in the literature
(Mazen, et al 1999, Miyoshi, et al 1994, Miyoshi, et al 1995, Miyoshi, et al 1996, Miyoshi,
et al 1999, Ogawa, et al 2006) for the cooling of gellan solutions with concentration 1% and
higher. The position of this peak on a temperature scale around 45 – 50 oC correlates well
with reported reference values (Miyoshi, et al 1995, Miyoshi, et al 1999, Morris, et al, 2012).
Meanwhile, as the concentration of gellan solution was decreased down to 0.25% the shape
of the peak changed. A distinct and sharp endothermic sub-peak appeared before the front of
the exothermic peak: as the measurements were performed in a cooling mode the front of the
peak corresponded to its right side at temperatures higher than the summit. As the
concentration of gellan was further diminished down to 0.05% the endothermic peak became
more substantial as compared to the exothermic one. At lower concentrations both the
exothermic and the endothermic peaks gradually vanished until they fell below the sensitivity
of calorimeter at a concentration of gellan < 0.01%.
The DSC thermograms of the solutions show up that the concentration regions above and
below 0.25% of gellan are different in respect to the thermodynamics of coil-to-helix
transition. The concentration 0.25 % is close to the reported value for the coil overlap
concentration (1.7 g/L) in gellan solution (Morris, et al, 2012). Thus, it might be supposed
that the transition in the ensemble of isolated coils occurs differently compared to that above
the overlap concentration. To confirm that the solutions with concentrations below 0.25%
contained isolated gellan molecules the DLS study of gellan solutions was performed at 60
oC.

Figure 3 presents the concentration dependence of the weight average diameter of particles
in gellan solution at 60 oC together with the same dependence of the enthalpy of coil-to-helix
transition as calculated by the integration of DSC plots in Figure 2.

7
 ACCEPTED MANUSCRIPT

Figure 3. Concentration dependence of the weight average diameter of particles in gellan

solution (1) in 130 mM NaCl at 60 0C measured by DLS and of the enthalpy of coil-to-helix
transition (2) calculated by integration of DSC thermograms. Lines are for eye-guide only.

In the concentration range 0.01 –0.2 % the weight average diameter of particles in gellan
solutions was almost constant at a 130 nm level. At higher concentration the ten-fold increase
in the diameter was clearly observed. This result indicated that at 60 oC and at concentration
below 0.2% gellan molecules were separate coils and the solution might be considered as the
dilute one. At concentration approximately 0.25% an overlapping of these coils took place
(semi-dilute solutions), which resulted in the steep increase in the apparent particle diameter.
Based on this result we may conclude that endothermic peak appears at DSC thermograms
for dilute solutions of gellan, while in semi-dilute solutions only a single exothermic peak is
observed.
The concentration dependence of the enthalpy of coil-to-helix transition calculated by the
integration of DSC signal nicely matches the step-wise shape of the concentration
dependence of the particle diameter. At gellan concentration 0.25% and higher the enthalpy
of the transition is –16±3 J/g, which is close to the value of this transition reported by
Miyoshi, et al (1999) for the 1% gellan solutions. According to DLS data in this
concentration range the apparent particle diameter in solution strongly increases likely due to
the overlapping of coils. Meanwhile, at gellan concentration 0.1% and lower the negative
values of the enthalpy of transition increase approximately three-fold and reach the level
around –50 J/g. This concentration range corresponds to the dilute solutions of gellan, which

8
 ACCEPTED MANUSCRIPT

contain isolated coils. The values of the enthalpy of coil-to-helix transition in the
concentration region 0.01 – 0.1% were very close for all tested solutions. We have not tested
solutions with concentration below 0.01% because DCS signal approached the noise level.
The constancy of the enthalpy of transition seems reasonable if takes place in the system of
individual coils not influenced by other gellan macromolecules in the solution.
In the semi-dilute regime, as the concentration of gellan solution exceeds the threshold for
coil overlapping, the spatial restrictions for the folding of long chains in a double-helical
structure arose, and loose ends and loops not incorporated in double helixes are likely left. As
the fraction of monomer units incorporated into helical structure decreases, the negative
enthalpy of the transition decreases also.
As it was mentioned above, it is well documented that gellan forms double-helical structure
upon cooling. This result was obtained by X-ray diffraction (XRD) studies in dried stretched
fibers cast from gel-like solutions, which contained overlapping gellan coils (Upstill, et al
1986). As the enthalpy of the transition in the dilute solutions of separated coils is at least
twice larger than in semi-dilute solutions (Figure 3) it raises up the question of the specific
features of the transition in dilute solution of individual gellan coils. In principle, there are
three possibilities: 1) the transition takes place in an individual chain, which folds over itself
in a double helix; 2) the transition implies the aggregation of individual chains into a double-
helical structure; 3) the structure is not double-helical but completely different to that in semi-
dilute solutions. The last possibility seems unlikely as the double-helix is reported for the
variety of polysaccharides. The first possibility seems doubtful as it involves the polymer
chain folding back on itself in tight "hairpin bends", which would require the two strands of
the double helix to run antiparallel to one another, in contrast to the parallel arrangement
found by X-ray fibre diffraction. Parallel association of chain segments from the same
molecule would require long, disordered interconnecting sequences, and thus decrease
transition enthalpy. Thus, most likely is that the double helix formation in dilute solution is a
two-chain process. This conclusion is supported by the kinetics of the transition.
Let us turn to the analysis of the fine structure of the DSC plot of the transition. Figure 4
presents the concentration dependence of temperatures, which correspond to the endothermic
and to the exothermic peaks on thermograms. At concentration above 0.25% there is only one
exothermic peak. With the decrease in gellan concentration an endothermic peak appears,
both peaks shift to low temperatures, and the separation between them progressively
increases. Most likely it means that the endothermic and the exothermic processes take place

9
 ACCEPTED MANUSCRIPT

simultaneously in semi-dilute solutions, but in dilute solutions either the endothermic process
becomes more rapid or the exothermic becomes slower.

Figure 4. Concentration dependence of the temperature of the endothermic peak (1) and the
exothermic peak (2) at DSC thermograms of coil-to-helix transition in gellan solutions in 130
mM NaCl.

What is the origin of these two processes, which are distinctly separated in thermograms of
dilute solutions? There is no problem for the interpretation of the exothermic peak, which
finalize the transition and might be attributed to the negative enthalpy of the formation of H-
bonds between adjacent hydroxyl groups in the double-helical structure. As it is the first
order transition, its entropy is also negative as the helix conformation is more ordered than
the coil. The other possibility for the exothermal effect is the aggregation of water molecules
in “iceberg” structures due to the hydrophobic hydration of non-polar backbones of the
chains arranged in a double-helix. The enthalpy and the entropy of hydrophobic hydration in
polymer solutions are also negative (Tager, Safronov, Sharina, & Galaev, 1993, Tager,
Safronov, Berezyuk, & Galaev 1994). Both possibilities are certainly not the alternatives but
may take place simultaneously.
It is much more difficult to rationalize the existence of the endothermic peak, which starts
the coil-to-helix transition and corresponds to a completely different thermodynamic process.
As the enthalpy change is positive, this process is not driven by the formation of new bonds:
either H-bonds or Van der Waals ones. The only thermodynamic reason, which favors the
process with positive enthalpy, is the positive entropy change.

10
 ACCEPTED MANUSCRIPT

In our view, the most likely source for such a combination of thermodynamic functions is
the dehydration process, which necessarily precedes the folding of coils into a double-helix.
To provide the interchain H-bonds hydroxyl residues in monomeric units are likely to break
H-bonds with attached water molecules in the first hydration shell. In this process the
enthalpy is consumed but the entropy is gained as the free water molecules increase the
overall thermodynamic probability of the solution. Another dehydration process likely arises
from the ionic rearrangement during the transition. It is known (Morris, et al, 2012, Rinaudo,
2009) that the counterion condensation takes place if the polyanionic gellan macromolecule
changes the disordered coil conformation to the ordered helix with higher linear charge
density. Condensed counterions are obviously located at carboxyl residues of glucoronate
units in gellan. To establish electrostatic adhesion both a carboxyl group and a counterion
should lose their hydration shells. The dehydration provides positive enthalpy and positive
entropy. The scheme in Figure 5 illustrates the two steps of the transition.
Based on the given scheme it might be assumed that the process of endothermic
dehydration triggers the conformational transition of gellan macromolecules. Thus, it favors
the viewpoint that the conformational transitions in biomacromolecules are to a large extent
originated by the changes in the structure of water, which occur in the bulk and in the
hydration shells.

Figure 5. Schematic presentation of two steps of coil-to-helix transition in dilute gellan

solutions.

11
 ACCEPTED MANUSCRIPT

Of course, all these processes are not exceptionally limited to the dilute solutions but
take place at any gellan concentration. What is different in the dilute regime is that
endothermic and exothermic stages are somehow separated in time. Dehydration processes
took place at the scale of monomeric units and their rate obviously does not change if gellan
coils are separated or not. Meanwhile, the folding of the double-helix might depend on the
concentration if it involves movements of different gellan macromolecules. In such a case,
each gellan chain needs to find its counterpart for the formation of a double-helix. In dilute
solution they are separated and they had to diffuse to each other. Based on this, we may
assume that the coil-to-helix transition of gellan in dilute solution likely includes aggregation
of separate macromolecules rather than folding of individual chain over itself. This
conclusion correlates well with the temperature dependence of DLS for dilute gellan
solutions (Figure 6), which show that the substantial aggregation of molecules and the
increase in the apparent diameter of particles takes place upon the cooling of the solution of
gellan coils.

Figure 6. Temperature dependence (cooling) of the weight average diameter of particles in

0.02% gellan solution in 130 mM NaCl. Inset - lognormal distributions of particle diameter
determined by DLS at 50 oC (1) and 25 oC (2).

To consider the endothermic and the exothermic stages of the transition

quantitatively, let us analyze the shape of DSC thermograms, which correspond to the dilute
solutions. The splitting of the signal requires certain model assumptions. We have made such
separation based on the features of a signal in Calvet calorimeter, which is the result of the

12
 ACCEPTED MANUSCRIPT

superposing of the heat evolution in the cell and the heat dissipation from the cell (Calvet,
Prat, 1956). The simplest way of such superposing is the double-exponent, which stands for
the single peak of the heat evolution curve:
t t
W(t)  a( exp( )  exp( )) (1),
1 2
Here a is the scaling factor, 1 and 2 are the characteristic time of temperature ascent and
descent in a cell.
If assumed that both the endothermic peak and the exothermic peak might be written in the
double-exponent form (Equation 1), the DSC signal can be fitted by a combination of two
double exponents. The result of such a procedure applied to the thermogram of 0.25%
solution is given in Figure 7. The fitting parameters for all concentrations are given in Table
1.

Figure 7. Peak separation of DSC thermogram for 0.25% gellan solution in 130 mM NaCl. 1
– Experimental data on heat flow (only one point out of each sequence of 20 experimental
points is shown). 2 – Curve fitting of experimental data by the combination of two double
exponents (Equation 1). 3 – Separated endothermic peak. 4 – Separated exothermic peak.

Table 1. Parameters of quadruple-exponent fitting of DSC thermograms by Equation (1)

13
 ACCEPTED MANUSCRIPT

Gellan Parameters of peak-fit approximation

concentration
a1 (mW) 1 (min) 2 (min) a2 (mW) 3 (min) 4 (min) R2
(%)

0.25 0.64 31.8 3.1 -16.6 8.5 7.8 0.9979

0.1 0.20 73.4 1.0 -9.3 8.4 7.7 0.9525

0.05 0.10 103.5 0.2 -26.9 10.2 10.0 0.9062

0.02 0.22 82.9 6.5 -56.3 21.4 21.6 0.9200

0.01 0.18 113 21.2 -15.3 46.7 47.6 0.8318

Table 2. Enthalpy of endothermic and exothermic steps of coil-to-helix transition in dilute

gellan solutions in 130 mM NaCl

Gellan concentration Integral enthalpy (J/g of solution)

(%)
Exo Endo

0.25 -1.07 0.67

0.1 -0.85 0.40

0.05 -0.61 0.32

0.02 -0.99 0.72

0.01 -0.91 0.81

Using the fitting parameters the integration of separated exothermic and endothermic
peaks can be done analytically for the evaluation of the integral enthalpy for these two stages
of the transition. The obtained values are given in Table 2. As the splitting of the complex
DSC signal strongly depends on the applied model assumptions, and the variation of the
parameters of the double exponents (Equation 1) strongly affects the fitting procedure, these
data should be treated with caution. Nevertheless, we believe that some noticeable features
might be pointed out. One can see that the integral enthalpy of the exothermic steps and the

14
 ACCEPTED MANUSCRIPT

endothermic steps related to 1 g of solution is almost independent on the concentration of

gellan. Despite the 25-fold decrease in concentration from 0.25% down to 0.01%, the integral
value remains approximately the same – around 1 Joule per 1 g of the solution. The same can
be marked out for the values of the integral enthalpy of the endothermic step.
In our opinion, it is an indication that both steps of the coil-to-helix transition are
cooperative and involve not only the disorder-order transition of gellan chains but as well the
rearrangement of large ensembles of water molecules in such a fashion that the changes in
water structure trigger corresponding conformational changes in a macromolecule. At the
endothermic step it might be a cooperative dehydration which takes simultaneously over the
entire hydration shell. Considering the exothermic step it possibly means that the helixes of
gellan become hydrophobically hydrated due to the formation of ordered water structures
(“icebergs”) around them.

Conclusion
We have performed the DSC study of gellan solutions in 130 mM NaCl using the
experimental technique of passive scanning which enabled us to make precise measurements
of the enthalpy of coil-to-helix transition of gellan in dilute solutions at low cooling rates. It
was found that while in semi-dilute solutions the transition was characterized by a single
exothermic peak; in dilute solutions the distinct separate endothermic and exothermic peaks
associated with the transition were observed.
In our opinion the fine structure of the DSC signal observed for the dilute gellan
solutions indicates that the coil-to-helix transition proceeds in a step-wise manner. The first
step, which is endothermic, corresponds to more rapid process of the dehydration of polar
groups in monomeric units and of the dehydration of counterions, which condense on the
polyanionic chain then. The thermodynamic driving force for these processes is the entropy
gain due to the released water molecules. The second step, which is exothermic, relates to the
formation of the double-helical structure. It is a slower process, as it involves the diffusion of
gellan macromolecules to arrange in a double-helix. Therefore the coil-to-helix transition is
observed as a single exothermic peak in semi-dilute solutions with interpenetrating coils,
which are spatially close to each other, but it splits into separate endothermic and exothermic
peaks in dilute solutions of individual coils, which need time to diffuse.
The analysis of the values of enthalpy of endothermic and exothermic steps at the
diminishing gellan concentration shows that the concentration has minor influence on their
enthalpy. Most likely it indicates that both stages are cooperative transitions which involve

15
 ACCEPTED MANUSCRIPT

not only the conformational change of gellan macromolecule but the structural rearrangement
of water structure as well.

Aknowledgements

The study was performed with the financial support of the Russian foundation for basic
research (grant 16-08-00609). Special support of Ms. O.V. Merkur’eva is appreciated.

References

Calvet, E., Prat, H. (1956) Microcalorimetrie. Applications physico-chimiques et biologiques.

Paris. Masson and Cie.

Chandrasekaran, R., Millane, R. P., Arnott, S., & Atkins, E. D. T. (1988). The crystal
structure of gellan. Carbohydrate Research, 175, 1-15.

Jansson, P.-E., Lindberg, B., (1983). Structural studies of gellan gum, an extracellular
polysaccharide elaborated by Pseudomonas elodea. Carbohydrate Research 124, 135–139.

Gibson,W., & Sanderson, G. R. (1997). Gellan gum. In A. Imeson (Ed.), Thickening and
gelling agents for food (2nd ed.). (pp. 119-143) London: Blackie.

Kang, K.S., Veeder, G.T., Mirrasoul, P.J., Kaneko, T., Cottrell, I.W. (1982). Agar-like
polysaccharide produced by a Pseudomonas species: production and basic properties. Applied
and Environmental Microbiology, 43, 1086–1091.

Kuo, M. S., Mort, A. J., & Dell, A. (1986). Identi.cation and location of L-glycerate, an
unusual acyl substituent in gellan gum. Carbohydrate Research, 156, 173-187.

Mazen, F., Milas, M., & Rinaudo, M. (1999). Conformational transition of native and
modified gellan. International Journal of Biological Macromolecules, 26, 109-118.

Milas, M., Shi, X., Rinaudo, M., (1990). On the physicochemical properties of gellan gum.
Biopolymers, 30, 451–464.

Miyoshi, E., Takaya, T., & Nishinari, K. (1994). Gel-sol transition in gellan gum solutions. 2.
DSC studies on the effects of salts. Food Hydrocolloids, 8, 529-542.

16
 ACCEPTED MANUSCRIPT

Miyoshi, E., Takaya, T., & Nishinari, K. (1995). Effects of salts on the gel-sol transition of
gellan gum by differential scanning calorimetry and thermal scanning rheology.
Thermochimica Acta, 267, 269-287.

Miyoshi, E., Takaya, T., & Nishinari, K. (1996). Rheological and thermal studies of gel-sol
transition in gellan gum aqueous solutions. Carbohydrate Polymers, 30, 109-119.

Miyoshi, E., Nishinari, K. (1999). Rheological and thermal properties near the sol-gel
transition of gellan gum aqueous solutions. Progress in Colloid and Polymer Science, 114,
68-82.

Morris, E. R. Nishinari, K. Rinaudo, M. (2012). Gelation of gellan - A review. Food

Hydrocolloids, 28, 373-411.

Nishinari, K. (1996). Properties of gellan gum. In G. O. Phillips, D. J. Wedlock, & P. A.

Williams (Eds.), Gums and stabilisers for the food industry , (pp. 371-383).Oxford, UK: IRL
Press.

Ogawa, E., Takahashi, R., Yajima, H., & Nishinari, K. (2006). Effects of molar mass on the
coil to helix transition of sodium-type gellan gums in aqueous solutions. Food Hydrocolloids,
20, 378-385.

Osmałek , T., Froelich, A., Tasarek S. (2014). Application of gellan gum in pharmacy and
medicine. International Journal of Pharmaceutics, 466, 328–340

Pollock, T. J. (1993). Gellan-related polysaccharides and the genus Sphingomonas. Journal

of General Microbiology, 139, 1939-1945.

Prajapati, V. D., Jani, G. K., Zala, B. S., Khutliwala T. A. (2013). An insight into the
emerging exopolysaccharide gellan gum as a novel polymer. Carbohydrate Polymers, 93,
670– 678

Rinaudo, M., & Milas, M. (2000). Gellan gum, a bacterial gelling polymer. In G. Doxastaxis,
& V. Kiosseoglou (Eds.), Novel macromolecules in food systems (pp. 239-263). Amsterdam:
Elsevier.

Rinaudo, M. (2009). Polyelectrolyte properties of a plant and animal polysaccharide.

Structural Chemistry, 20, 277-298.

Sanderson, G. R. (1990). Gellan gum. In P. Harris (Ed.), Food gels (pp. 201-232). London:
Elsevier.

17
 ACCEPTED MANUSCRIPT

Tager, A.A., Safronov, A.P., Sharina, S.V., Galaev, I.Yu. (1993). Thermodynamic study of
poly(N-vinyl caprolactam) hydration at temperatures close to lower critical solution
temperature. Colloid and Polymer Science, 271, 868-872.

Tager, A.A., Safronov, A.P., Berezyuk, E.A., Galaev, I.Yu. (1994). Lower critical solution
temperature and hydrophobic hydration in aqueous polymer solutions. Colloid and Polymer
Science, 272, 1234-1239.

Tanaka, S., Nishinary, K. (2007). Unassociated molecular chains in physically cross linked
gellan gels. Polymer Journal, 39, 397-403.

Upstill, C., Atkins, E. D. T., & Atwool, P. T. (1986). Helical conformations of gellan gum.
International Journal of Biological Macromolecules, 8, 275-288.

Valli, R., & Clark, R. (2010). Gellan gum. In A. Imeson (Ed.), Food stabilizers, thickeners
and gelling agents (pp. 145-166). Chichester, UK: Wiley-Blackwell.

Vshivkov, S.A., Safronov, A.P. (1997). To the conformational coil-globule transition of

polystyrene in cyclohexane solutions. Die Makromolekulare Chemie, 198, 3015-3023.

18
 ACCEPTED MANUSCRIPT
Highlights

 Evidence of two-step mechanism of coil-to-helix transition of gellan is found by DSC

 Distinct step of heat absorption precedes heat evolution in the course of transition

 Dehydration of gellan is proposed for the mechanism of endothermic step