Solute Energetics in Aqueous Xanthan Gum Solutions: What Can Be Learned from a Fluorescent Probe?

Abstract

Xanthan gum (XG) is a well-known anionic polysaccharide that finds broad application in the food and petroleum industries because of its ability to enhance solution viscosity at low concentrations and moderate temperatures. The aim of this work was to use the solvation probe coumarin 153 (C153) to characterize changes in the xanthan gum (XG) solution microstructure as a function of XG concentration and temperature from the perspective of a dissolved solute molecule. We established the utility of C153 fluorescence to track solution changes for XG concentrations that span the transition region from a dilute to a semi-dilute solution, defined by the xanthan gum overlap concentration, C*~0.02 g/dL. The temperature was varied from 293 to 353 K to probe solution conditions wherein XG has been reported to undergo a structural change from helix to random coil conformation, the details of which are still under debate. While C153 fluorescence does not elucidate direct structural information, the emission response is a simple means by which changes in aqueous XG solution can be identified. C153 spectroscopy is observed to correlate with XG conformational changes, as reported in the literature.

1. Introduction

Biopolymers, including polysaccharides such as xanthan gum, chitin, pullulan, hyaluronic acid, carrageenan, glucans, guar gum, cellulose, and dextran, are renewable resources derived from plants, seaweeds, animals, and bacterial fermentation [1]. These polymers have a diversity of chemical structures and thus properties that make them good candidate materials for sustainable applications because they demonstrate biodegradability, biocompatibility, and bio-adhesion [2]. These materials are important to understand as they provide guidelines for the selection of complex fluids or rheology modifiers [3]. In the class of biopolymers, the anionic polysaccharide xanthan gum (XG, Scheme 1) is an environmentally friendly and water-soluble biopolymer. Excellent reviews that highlight the extensive utility of XG are available [2,4,5,6,7,8]. It has been approved by the United States Food and Drug Administration (FDA) since 1969 as a food additive. XG is among the most important commercially available and low-cost microbial polysaccharides. Because of its rheological and pseudoplastic properties, it is widely used as a thickener or emulsifier agent that finds many applications in food products [4], including salad dressing, ice cream, and yogurt [9]. Many diverse applications of XG are seen across various fields, including oil-field development [10], drug delivery [8], drag reduction [1], biomedical [11], pharmaceutical [12], agriculture [13], mining [6], battery [14], and environmental fields [5,7]. XG is attractive for a considerable range of applications as it is a well-known, nontoxic, biodegradable, biocompatible, and bio-adhesive polymer [15]. XG is produced by fermentation of Xanthomonas campestris, with a primary structure consisting of the trisaccharide β-D-mannose-(1-4)-α-D-glucuronic acid-(1-2)-α-D-mannose linked to the O(3) position of every other glucose residue [16]. Pyruvic acid residues are linked through ketal bonds to roughly 50% of the terminal mannose residues [17]. These mannose residues also include acetate groups. Structural variability—the extent of pyruvate and acetate group substitution—is controlled by the culture and fermentation process conditions [18]. The degree of acetate and pyruvate inclusion influences the XG threshold temperature, where XG is reported to undergo an order/disorder transition from a helical conformation to a random coil conformation [19,20]. The XG backbone consists of both hydroxyls and carboxyls along with glycosidic oxygen atoms, all of which govern intermolecular hydrogen bonding [21]. XG molecular weights are reported to range between 1 and 20 × 10⁶ g/mol [16] with an estimated average molecular weight of 2 × 10⁶ g/mol [22]. The large molecular weight, coupled with intermolecular hydrogen bonding interactions, results in XG behaving as a pseudoplastic, and its viscosity increases with polymer concentration in water [23].

The secondary structure of XG exists in three forms: (i) native, single-stranded helix; (ii) double or triple helix, rigid structure; and (iii) disordered, flexible chains [24,25]. The structural details of the helical conformation remain controversial. The conformation transitions are influenced by temperature, pH, salt content, solvent quality, shear rate, and other environmental conditions [26]. The disordered conformation, or random coil, can be described as a broken or imperfect helix [9]. XG chains extend in non-saline aqueous media because of intra-chain electrostatic repulsion between charged groups on the side chains. In salt-free aqueous solutions, xanthan gum chains retain their high degree of rigidity with some degree of flexible conformation [16,27]. In the absence of ionic content, XG is surrounded by a cloud of counterions that exactly balances the charge on the chains, and its aqueous solution maintains charge neutrality [27]. Due to the charge balance in non-saline environments, the XG molecular configuration is sensitive to added salt ions in solution [28]. In the presence of salts (brine solution), XG assumes an ordered, helical conformation that is more rigid and rod-like than the disordered conformation as a function of salt concentration and solution temperature [29,30]. Specifically, XG chains in the ordered conformation are much more rigid, with an estimated Kuhn segment length [31] (L_k) of 240 ± 40 nm in terms of the wormlike chain model compared to the disordered structure (L_k = 42 nm). The XG intrinsic persistence length is similar to that of the single helical structure of DNA (~50 nm) and the double helical structure (~120–148 nm) in salt solution [32]. The presence of salt in solution causes shielding of electrostatic repulsion between charged groups, making the XG macromolecules assume a more compact structure [33]. This decreases the polymer chain size and the number of interactions with the neighboring chains [16]. The degree to which a polymer molecule’s conformation shrinks may be dependent on the stiffness of the polymer chain in solution [27].

Results from Rochefort et al. show that the ordered conformation of keltrol XG (~3.5 × 10⁶ g/mol, 0.5 g/L) is attained at a salt concentration above 10⁻² M NaCl, and no further salt effects should be noted [9]. Comparing the molecular weight of XG (3 to 5 × 10⁵ g/mol, 3.1–6.4 mg/mL) in 5 mM and 100 mM NaCl, the molecular weight values are smaller at 5 mM, which indicates a partly unfolded structure in the low-ionic-strength solvent [31]. Brunchi and colleagues found that the intrinsic viscosity of XG (1.165 × 10⁶ g/mol) increases with temperatures rising from 25 to 75 °C in salt-free solutions and in 10⁻⁵ and 10⁻² M NaCl [26]. A low intrinsic viscosity indicates ordered conformation chains, while an increase in intrinsic viscosity suggests a disordered conformation. Rinaudo et al. estimated a persistence length for XG (~10⁶ g/mol, 0.01 M NaCl) of 30 nm and 125 nm for the double helix structure after renaturation or cooling below 25 °C [24,34]. Philippova and colleagues reported a side-by-side association of xanthan double helices at 0.15 and 0.5 wt % in 0.64 M KCl (via FF-TEM, 1 × 10⁶ g/mol, and 0.5–2 wt %) [35]. At a fixed ionic strength of 10 mM NaCl, the observed helix–coil transition occurs at high temperatures (near 50 °C) with reduced XG pyruvate levels, while reducing acetate content reduces the transition temperature [36]. Acetate content stabilizes the ordered conformation because of intramolecular attractive interactions, while pyruvate inhibits the formation of the ordered structure due to intermolecular electrostatic interactions [26].

While much attention has been given to XG solutions under a variety of experimental conditions such as pH, ionic strength, temperature, and shear rate, the XG conformational transition from helix to random coil in response to experimental stimuli is still a subject of interest as it is not fully understood. Moreover, most work has been conducted using direct solution probing by means of bulk physicochemical measures such as rheology, density, viscosity, and conductivity experiments. Various scattering techniques, including small- and wide-angle neutron scattering and x-ray scattering, have also been applied to examine XG structural organization. To the best of our knowledge, there is no report using optical emission spectroscopy, e.g., fluorescence emission spectroscopy, to probe and characterize XG solution characteristics at a microscopic level from the perspective of an extrinsic molecular probe. Fluorescent probes like BODIPY (4,4′-difluoro-4bora-3a,4a-diaza-s-indacene) and DASPMI (4-(4-(dimethylamino)styryl)-N-methyl-pyridinium iodide) have been used to monitor changes in fluorescence lifetime, which were related to changes in gellan gum solution viscosity [37]. Fluorescence spectroscopy provides a convenient, direct method for probing the microenvironment and has been used to track organizational changes that occur at the micro- and nanoscopic level in diverse media such as sol-to-gel transitions [37], aggregation in surfactant solutions [38], protein–ionic liquid interactions [39], and strand architecture of amyloid fiber formation [40]. For XG, the hydrophobic association with grafted C16 carbon chains was studied using pyrene and anilinonaphthalene sulfonate (ANS) to determine the modified XG change in hydrophobicity [41]. While it is important to understand the fundamental nature of XG-XG interactions under various solution compositions and conditions, the numerous chemical and engineering applications that use XG as an enhancement medium suggest that there is also value in examining solution behavior using indirect probing via a molecular solute [15,21,42,43,44,45,46,47,48]. Thus, here we examine the steady-state and time-resolved fluorescence emission of coumarin 153 (C153) as an independent means of probing XG interactions, with the goal of assessing the solution microenvironment effects on a solute molecule. Simultaneously, the observed effects provide direct insight into XG solution changes. The data presented here augment our understanding of XG concentration and temperature effects from the unique perspective that an extrinsic fluorophore offers, helping to elucidate the medium modifier behavior in the XG/water system.

2. Materials and Methods

The food-grade xanthan gum used in this study is commercially available in a wide range of particle sizes, (CPKelco, Atlanta, GA, USA; material number: 10040281), with an estimated molecular weight of 2 × 10⁶ g/mol and a polydispersity index ~2 [28,49]. The polymer was received in powder form and used without further purification or modification. The deionized water used was purified with a MilliQ system with a measured resistance of 18.2 MΩ cm. A 0.16 wt % xanthan gum stock solution was prepared by dispersing dried xanthan gum in deionized water. The stock solution was gently stirred for 24 h at 20 °C to obtain a fully hydrated solution that appeared translucent and was kept refrigerated at 4 °C to minimize bacterial growth. When solutions were diluted for measurement, they were equilibrated at room temperature for at least 1 h and were observed to be optically transparent. Appropriate volumes of the xanthan gum stock solution were pipetted into deionized water to prepare the desired xanthan gum solution concentrations.

Coumarin 153 (C153, Scheme 1) was purchased from Exciton (Lockbourne, OH, USA), stored under desiccation, and used as received. To prepare a sample for measurement, C153 was pipetted from a methanol stock solution, and the solvent was thoroughly evaporated using a gentle stream of dry ultrapure nitrogen prior to the addition of XG solution. The probe concentration was ~10 μM, and the resulting solution optical density (absorbance) was not more than 0.1 at the laser excitation wavelengths, 405 and 425 nm, as is typical to minimize inner filter effects [50,51]. Solutions were mixed by repeatedly inverting the cuvette to form a single-phase, transparent solution.

Steady-state absorbance was measured using a Perkin-Elmer Lambda 800 UV–vis spectrometer, with slit settings that provided 2 nm spectral resolution. Fluorescence excitation and emission spectra were recorded on a Horiba Scientific Fluorolog-3 fluorescence spectrometer (Horiba Scientific, Piscataway, NJ, USA) with a 2 nm resolution. A 450 W Xe arc lamp and a single-grating monochromator were used for excitation, while emission was measured through a double-subtractive grating monochromator for enhanced stray light rejection. Detection was performed using a Hamamatsu R928P PMT (Hamamatsu Photonics, Shizuoka, Japan). Calibration was performed daily using the Raman signal from deionized water with 350 nm excitation and observing Raman scatter at 397 nm, and the signal-to-noise ratio was always greater than 5000:1. A National Institute of Standards and Technology (NIST) calibration lamp was used to generate excitation and emission correction files for both intensity and wavelength instrument correction procedures. All spectra were further corrected by applying blank solvent subtractions to correct for sample response. Replicate data were combined to yield spectral uncertainties that were not larger than 100 cm⁻¹ in all cases, and the intensity percent variation at the peak of emission was not more than 5%. To compute the normalized intensity for each spectrum, the entirety of each spectrum was ratioed to its absolute maximum intensity value.

Time-correlated single photon counting (TCSPC) fluorescence instrumentation (Tau-3 system, Horiba Scientific, Piscataway, NJ, USA) was used as previously described [52]. Intensity decays were measured with excitation by a 405 nm NanoLED (405-L) diode laser operated at a 1 MHz repetition rate. Excitation photons were vertically polarized prior to entering the sample chamber, and the subsequent emission was passed through a Glan-Thompson polarizer set at the “magic” angle (54.7°) for excited-state decay kinetics, spectrally resolved through the Fluorolog-3 emission monochromator, and detected with an IBH TBX 850 detector (Horiba Ltd., Kyoto, Japan). Data were collected to 10,000 counts in the peak channel in accord with Poisson statistics for photon counting measurements. Instrument response was determined using a glycogen scattering solution, and typical values were ~180 ps. The instrument time calibration was 7.1 ps per channel. Decay data were measured at the peak of the steady-state emission, with a 5 nm bandpass unless otherwise stated. For TCSPC analysis, the intensity decays were evaluated using DAS6 (version 6.6) decay analysis software. We estimated that the effective time resolution was ~50 ps following a least-squares reconvolution fitting of the intensity decay. The quality of fits to a sum-of-exponentials model was assessed using reduced chi-squared values, and a fit was judged acceptable if χ_r² < 1.2. When the data appeared to need multiple time constants, we allowed the inclusion of an additional time constant only if there was at least a 10% improvement (decrease) in χ_r², along with a discernable improvement in fitting residuals with a concomitant improvement in the autocorrelation of the residuals. In this way, the estimated uncertainty in fitted decay time constants was not more than ±0.08 ns.

An independent set of time-resolved 2D fluorescence measurements was performed using a Ti:sapphire streak camera spectrometer (HAMAMATSU PHOTONICS, Hamamatsu City, Japan) to check for solute dynamics that exceed the resolution lower limit of the TCSPC instrument. Fluorophore excitation was achieved using a Spectra-Physics Mai Tai femtosecond oscillator (Spectra-Physics, Milpitas, CA, USA) producing 2.95 W average power at 850 nm with a repetition rate of 80 MHz and pulse width of ~80 fs FWHM. The beam diameter was approximately 100 μm. The fundamental beam fed a GWU (Erftstadt, Germany) Ultrafast Harmonic Generator (UHG) unit that housed a Bragg cell pulse selector and BBO crystal for frequency doubling. The 80 MHz pulse train was first reduced to 80 kHz with the Bragg cell and then the frequency was doubled to 425 nm for sample excitation, with an average doubled power of ~200 μW. The various table optics used here were purchased from Newport Corporation (North Logan, UT, USA). Prior to entering the sample, the beam was passed through a half-wave plate followed by a Glan-Laser polarizer and was then focused on the sample using a 100 mm focal length 25.4 mm diameter plano-convex lens. Emission from the sample was collected, collimated, and passed through a matching Glan-Laser polarizer to set emission polarization at the magic angle (54.7°) with respect to excitation. A polarization scrambler and focusing lens steered the emission into a SpectroPro 150 mm spectrograph (Teledyne Princeton Instruments, Trenton, NJ, USA). The output of the spectrograph was finally focused on the universal streak camera, model C10910 with model M10912 (Hamamatsu, Bridgewater, NJ, USA). Spectrograph and streak slits were set at 50 μm and 30 μm to maximize spectral and time resolution, respectively. Analysis code was written in-house using Igor Pro 9 (WaveMetrics, Portland, OR, USA).

3. Results and Discussion

3.1. Steady-State Spectroscopy

C153 is known for its sensitive spectral response to the surrounding microenvironment, resulting from the ~8 D change in the dipole moment upon excitation to the electronic excited state [53,54]. The polar nature of the C153 ground state, 6.55 D [55], and excited state, 14.5 D [56], makes C153 a reasonable choice to study the XG/water interactions. The fluorinated moiety of C153 enhances the ability of this polar molecule to respond to ensemble microenvironments that are less polar while simultaneously solubilizing in polar microdomains. Its solvatochromatic response is well documented [57,58]. Moreover, the presence of the fluorine and oxygen atoms within the molecular structure also favors hydrogen bonding interactions, further enhancing observed spectral shifts in response to solution properties [59]. In typical polar solvents such as acetonitrile or methanol, C153 is strongly fluorescent and exhibits larger quantum yields [58,60,61] compared to that observed in water, where the C153 emission is redshifted and significantly quenched. Temperature also influences the C153 spectral response. In chloropropane, a decrease in molar absorptivity and a blueshift in the absorbance maximum is observed with increasing temperature, with similar behavior noted for the emission [62,63]. Through these spectral response features, C153 is expected to differentiate between XG domains and water domains insofar as XG forms a solvated structure. Steady-state spectroscopy alone is unlikely to clearly distinguish between specific structural motifs such as conformational change or XG-XG aggregation because either situation may modulate the micro- or nanoheterogenous solution structure.

3.1.1. XG Concentration Effect

The concentration dependence of XG ordering in aqueous solution has been shown through scattering studies [64] and physical property measurements [28,65,66,67]. For example, increasing the XG concentration from 0.4 to 2.0% increased viscoelastic stability, viscoelastic moduli, and widened linear viscoelastic regions [68]. XG reduced viscosity increases with dilution, indicating the characteristics of polyelectrolyte behavior due to electrostatic repulsion interactions [69]. Milas and coworkers used small-angle neutron scattering to suggest an XG conformation transition that favors a helical (ordered) conformation at decreased temperature (298 K) and with decreasing XG concentration, interpreting the ordered-to-disordered conformation [64]. It is also plausible to consider that interaction mechanisms may include concentration effects due to molecular aggregation and entwined interactions but at high concentrations, even up to 1 g/dL, as reported by Papagiannopoulos and coworkers [32,70]. We note that this effect was only observed at concentrations in semi-dilute or more concentrated solutions, well beyond our highest solution concentrations. Thus, XG aggregation is unlikely when working in a dilute solution and at best may only begin to manifest in a semi-dilute solution, relative to C*. However, the detailed molecular underpinnings of the XG concentration dependence are not yet fully explained, and spectroscopy offers an additional perspective. Examples of the C153 spectral response to XG concentration are presented in Figure 1. The upper panels of Figure 1 show the absolute emission intensity changes for XG concentrations at 303 K and 343 K. We expect the C153 quantum yield to be sensitive to XG solution microstructure, depending on the water environment sensed, as water quenches C153 emission, and the observed intensity variation confirms this response. The spectral response further showed that at higher temperatures (Figure 1, upper right panel) the intensity maximum at each concentration consistently increased by at least threefold at 343 K compared to 303 K for all XG concentrations measured. We note that some care must be used with the interpretation of absolute intensity changes since small variations in sample preparation may occur, such as pipetting volumes of stock solution, which could result in small but potentially detectable variations in the measured absolute intensity. Nonetheless, the magnitude changes are very much larger than what might be caused by less accurate pipetting, and although we cannot be completely quantitative in the absence of measured quantum yields, the gross trends observed here are consistent and demonstrate the C153 fluorescence sensitivity to concentration and temperature effects. Given this, we prefer to use a more reliable evaluation metric and, therefore, view the spectral response using a normalized intensity metric, as shown in the lower panels of Figure 1. Several features of these data emerge. First, at 303 K the spectra reveal only a slight variation in the overall emission as [XG] increases, though from the spectral plots, the magnitude is difficult to discern. While peak position suggests only a slight variation, the half-maximum intensities show an XG-concentration-dependent spectral shift and shape change, where the longer wavelength “red” edge emission remains constant to within uncertainty, and the “blue” edge emission shifts to higher energy, as indicated by the arrow. In contrast, at higher temperatures, we observe a negligible effect on wavelength and spectral shape. Thus, at higher temperatures, the intensity variation is the key parameter that shows the C153 response to concentration changes, and the lack of spectral shift indicates a more homogeneous chemical environment about C153 at 343 K compared to 303 K.

To fully characterize the spectra, the concentration dependence is quantified in Figure 2 for absorption (data above the axis break, ~24,000 cm⁻¹) and emission (data below the axis break, ~17,500 cm⁻¹) at each of the temperatures measured here.

The absorption response shows that at 293 K, XG concentration has a negligible effect to within uncertainty (~100 cm⁻¹), until the [XG] = 0.032 g/dL. Upon reaching a sufficiently high XG concentration, there is a substantial blueshift in the C153 absorption that is consistent with C153 more directly interacting with hydrophobic domains within the XG solution structure. The solid blue line is a linear regression applied to the entire data set and is included primarily to emphasize the nonlinear shift in absorption peak frequencies. In their descriptions of linear solvation theory, Lippert [71] and Mataga [72] both predict that the shift in spectral frequency should change linearly as a function of the solvent reaction field, Δf

$$\Delta f=\left({\displaystyle \frac{\epsilon -1}{2\epsilon +1}}-{\displaystyle \frac{n_D^2-1}{2n_D^2+1}}\right),$$

(1)

which functionally depends on the medium dielectric constant (ε) and index of refraction (n_D), provided that these parameters also change in a linear manner. For frequency shifts that deviate nonlinearly, there must be other solution mechanisms that account for the observed shift. Consistent with other XG literature reports that used spectroscopy, our observed absorption responses suggest changes in the XG conformation [73]. Fluorescence from pyrene showed evidence of concentration-dependent hydrophobicity changes resulting from the acetyl group interactions, which help maintain the XG helical conformation [73]. NMR results also confirm the same observation [26]. We note the same general pattern in spectral shift at each of the temperatures tested here, from 293 K to 353 K. As a general comment, increasing the temperature at a given XG concentration resulted in a lower C153 energy, discussed in more detail below. The absorption full width at half maximum (FWHM, Figure 2 lower panel) mirrors the concentration-dependent frequency changes, where [XG] = 0.0016 g/dL produces a broad spectrum that narrows with increasing XG concentration, up to [XG] = 0.0160 g/dL. Increasing temperature also produced a narrower absorption spectrum, and C153 reports the formation of a more homogeneous solution. From these absorption results, we see that, upon transition to the excited state, the enhanced C153 dipole moment is sensitive to the local changes driven by XG concentration variation.

Briefly, the emission response seen in Figure 2 is generally similar to absorption with a few noted differences. First, the emission frequencies consistently redshift at all concentrations and are better described by a linear response. Second, while temperature lowers the C153 energy at each concentration, the concentration-dependent peak position becomes negligible at higher temperatures. The two dashed lines in Figure 2 indicate the different concentration-dependent slopes as the temperature increases. However, the redshift between [XG] = 0.0016 g/dL and [XG] = 0.0080 g/dL shows a steeper change compared to [XG] ≥ 0.0080 g/dL. If we assume that a linear response should follow from the three most dilute XG concentrations, then the higher concentration emission frequencies deviate negatively from expected values. This suggests that the XG microenvironment at greater XG concentrations either stabilizes the C153 excited state or raises the ground state energy, either of which lowers the transition energy. We have not varied excitation energy here, so at this juncture we cannot differentiate these effects. Nonetheless, the observed emission response shows that the most significant solution changes occur at [XG]~0.0080 g/dL and higher. Emission FWHM data indicate only a small overall change as indicated by the smaller slope of the red regression line compared to the blue absorbance regression line (Figure 2, lower panel).

There are at least two possible scenarios for this observation that may be considered. First, as [XG] increases C153/XG interactions diminish because of stronger and/or more extensive XG-XG interactions that drive intra- and intermolecular associations. A measure of the XG chain associations is given by the overlap concentration, C*, the polymer concentration at which chain entanglements begin to alter the solution behavior [26,74,75]. In this concentration regime, the solution is considered to transition from a dilute to a semi-dilute system. For XG in water, C* has been reported to be ~0.02 g/dL [76]. Below this value, XG is in a dilute solution where entanglements between separate chains should be minimized. Takamasa and coworkers cited conformational change between coil and helix along with motional restriction in deacylated XG as their explanation for the NMR line broadening at 288 K [77]. In the same report, they identified the coil-to-helix transition using differential scanning calorimetry. Brunchi and coworkers also used high-resolution NMR to study the interactions of acetate and pyruvate groups, which are known to influence XG chain conformation, and reported that, in 0.0021 g/dL XG solution at 298 K, XG is in the ordered conformation [26]. Consistent with these reports, in dilute XG solution, the C153 spectral response seems to indicate it is better shielded from water interactions compared to more concentrated XG solutions. If one assumes a helical XG conformation at low [XG] and lower temperature [31,78], and if C153 is strongly associated with this motif, then on average it apparently senses a less water-like environment, or at least an environment that is on average less polar. A second possibility might be gleaned by considering an XG organization that may allow for water pool formation within XG nanocages. In such an example, C153 intensity is quenched because of C153/water interactions. As [XG] increases, the decreased emission intensity combined with redshifted emission energy argues that C153 is becoming water-exposed. If local water domains are forming or C153 is expelled away from XG, either possibility would show the observed behavior. The transformation from helix to coil conformation may be complicit in these spectral changes. The most probable interpretation here is a combination of these, and likely other, interaction effects assuming a randomized distribution of C153 in the bulk solution. What then of temperature effects?

3.1.2. Temperature Effect

Temperature is an important parameter in investigating the XG aqueous solution. The intrinsic viscosity of XG in a salt-free solution increases with temperature from 298 K to 348 K, indicating a transition from an ordered to a disordered conformation [26]. In water, the intrinsic viscosity of XG increased by 30% at 323 K compared to 293 K [76]. XG (1.0 wt % at 288 K) in D₂O shows broadening peaks in ¹H NMR spectra, suggesting helix conformation and coil state. Based on the DSC results, the transition temperature from coil to helix is estimated to occur at 305.5 K [77]. The positive peak of the circular dichroism spectrum of XG in pure water decreased at 353 K and recovered when cooled at 313 K, indicating that the double helical conformation of XG was locally reconstructed by renaturation [79]. How then, does the response from an extrinsic probe molecule compare?

Temperature influence on the C153 spectra is readily observed in Figure 3 and the spectral analysis is given in Figure 4. On heating, the fluorescence intensity increases by approximately 4× from 293 K to 343 K at the limits of XG concentration studied here (upper panels). A more extensive spectral shape change was observed at increased [XG] (Figure 3, lower right panel). Figure 4 shows that the absorption peak frequencies vary smoothly, with a small redshift on the order of ~350 cm⁻¹. The notable exception is [XG] = 0.0160 g/dL (yellow diamonds), where the redshift shows a marked minimum at 313 K. Similarly, C153 emission also shows an overall modest redshift of ~300 cm⁻¹, but interestingly there is first a small rise between 293 K and 303 K that is at the edge of our spectral resolution for these experiments. We also note that [XG] = 0.0320 g/dL shows no such rise and redshifts across the entire temperature range. Spectral FWHM data evidence the pronounced temperature effect. For the [XG] = 0.0016 g/dL, the most dilute solution, the absorption spectra at 293 K are collectively the broadest, with a width of ~6300 cm⁻¹ that narrows by ~700 cm⁻¹ up to [XG] = 0.0160 g/dL. While increasing the temperature to 303 K reduces the spectral width by >1000 cm⁻¹, there is a secondary effect that indicates the formation of a minimum width for each concentration. The observed width minimum is rather weak up to [XG] = 0.0080 g/dL but significant for [XG] = 0.0160 g/dL at 313 K. For T > 313 K, the three lowest XG concentrations, the widths are essentially constant and the [XG] = 0.0160 g/dL width moderates. The temperature dependence provides further evidence that the most significant solution changes occur at [XG] = 0.0160 g/dL and 313 K. Presuming XG assumes a random coil conformation at increased temperatures, the solution should be generally more homogeneous and C153 appears to sense the solution change. This is also supported by the consistent temperature-dependent emission redshift. Our data compare favorably with the literature that suggests conformational changes occur at roughly 313 K, where we observe a significant shift in spectral response [26,80].

3.2. Time-Resolved Fluorescence

Steady-state absorption and emission give a general sense of solution response to concentration and temperature changes, where the measured responses are the ensemble-averaged data. Through the time-resolved intensity decay, we can extract more detailed information from the excited state radiative relaxation. For example, a probe that experiences multiple chemical environments may show multiple excited-state emission lifetimes that are characteristic of how the microenvironment influences the intensity decay kinetics. In this way, the decay kinetics will be more complicated than is modeled by simple monoexponential decay, i.e., a single lifetime. As mentioned above, C153 emission intensity is quenched in the presence of water, which results from the decrease in fluorescence quantum yield and, therefore, a decrease in the excited-state lifetime. Thus, the determination of lifetime provides us with an additional and direct means of assessing C153’s relative exposure to water.

3.2.1. XG Concentration Effect

The intensity decay data were fitted using a simple sum-of-exponentials model and the decay data were best described using two time constants. Reduced chi-squared values were between 1.0 and 1.2 for all fits. Analysis results are presented in Figure 5, showing the lifetime as a function of XG concentration. Whatever the specific XG conformational organization is in water, it should be represented at these most dilute XG concentrations. With that established, we observe C153 lifetimes that indicate no fewer than two distinct solution domains, characterized by a constant lifetime component of 1.7 ns and a second lifetime component of 2.5–4.5 ns that gradually increased with added XG at temperatures less than 333 K. We determined the C153 lifetime in neat water at its emission maximum and found that τ_flr = 1.75 ± 0.07 ns. Given this, we can clearly assign the 1.7 ns lifetime value to C153 in a predominantly water environment and, therefore, the longer lifetime must be associated with the C153/XG interactions. Fractional contributions were computed from preexponential decay factors as a simple normalized ratio, a_i/Σa_i. We note here as well that a biexponential decay model is the best for all XG concentrations presented here.

Lifetime contributions at all XG concentrations show that the water component lifetime is the dominant contribution. Importantly, there is a clear break in the linearity for the longer lifetime values (and associated fractional contributions) occurring between [XG] = 0.0032 and 0.0080 g/dL, where the regression lines intersect in Figure 5, mirroring the observations in the steady-state analysis. Below the intersection point, the slope is 0.035 ns dL g⁻¹, and this lifetime variation is less than our measurement uncertainty of ~0.05 ns. The “XG concentrated slope” is 4-fold larger at 1.22 ns dL g⁻¹. It is also important to note that for temperatures above 313 K, the concentration dependence of the lifetime data yields a relatively constant value for the longer time constant that we assigned to C153/XG interactions. The time-resolved data show two distinctly different C153 associations and therefore augment the conclusions reached from the steady-state analysis. Thus, a solidified picture of solution transition emerges. There is a clear and substantiated transition sensed by C153 in the concentration-dependent changes induced by added XG, the onset of which occurs at [XG]~0.0080 g/dL. Although C* is larger than this concentration, the C153 molecularity seems to report on what might be considered as a pre-association of XG chain interactions as concentration builds toward the value predicted from bulk viscosity measurements [76].

To thoroughly explore the XG concentration range, we included two additional XG concentrations that are 10-fold and 100-fold diluted from [XG] = 0.0016 g/dL, as reported for the steady-state data, shown in Figure S1. These added concentrations were measured to ensure that we established the decay kinetics approaching XG infinite dilution so that the effects from XG intramolecular interactions were minimized. The constancy of C153 response over at least two orders of magnitude of dilute XG solution maps the concentration range over which there is not likely to be conformational transition(s) or XG-XG associations because the probability of aggregation in dilute solution is small in the concentration range studied here. It appears that what C153 senses is akin to a typical dilute solute/solvent (XG/water) solution. Near to [XG] = 0.01 g/dL, the marked change in C153 lifetime response clearly shows the concentration-dependent onset of XG change. One further point of note is that, along with the two additional diluted XG solutions, we also measured one solution beyond 0.0320 g/dL, at [XG] = 0.0800 g/dL, which was not measured by steady-state emission. Examination of the lifetimes and contributions show that these additional data sets are well aligned with the data patterns of the “main” XG concentrations, which bolsters our confidence in determining a transition point from all of the combined spectroscopic data presented. As one final assessment of the lifetime behavior, we also computed the average lifetime response,

$$〈{\tau }_{flr}〉={\displaystyle \sum _i{a}_i{\tau }_i},$$

(2)

where a_i is the preexponential factor for lifetime τ_i, for what it might reveal about the lifetime landscape. Those data are presented in Figure S2, and we found that they track nearly identically with the shape of the longer lifetime in Figure 5, although shifted to smaller lifetime values because of the constant 1.7 ns contribution.

3.2.2. Temperature Effect

The temperature-dependent lifetimes present an additional perspective and are summarized in Figure 6.

The faster lifetimes show a slight decrease in value, but this is about the same magnitude as the experimental uncertainty, coupled with a corresponding increase in its associated fractional contribution as temperature increases. The contribution approaches ~90% at the highest temperatures for all XG concentrations considered. This is further evidence that C153 becomes more water-exposed as the XG solution structure undergoes transformation by thermal effects. With increased temperature, XG chains become more flexible, and, as mentioned above, XG heating induces a transition to a random coil conformation, as evidenced by the literature reports using SANS, NMR, viscosity, rheometry, and DSC [26,32,64,70,77]. While the shorter lifetime remains essentially constant, the longer lifetime presents a more significant decrease along with a decrease in fractional contribution. Referencing the concentration-dependent data for [XG] < 0.0032 g/dL, we see that the ~4 ns lifetime shortens by ~1 ns. We interpret this to mean that, although there is a significant change to the XG structure, there is likely to be an ensemble effect where C153 still senses XG. It is likely that the solvation environment of the XG structure, both backbone and side chain components, contributes to decreasing the C153 lifetime from 4 ns toward that observed in neat water. Nonetheless, we see a remaining 10% contribution even at the highest temperature studied here.

4. Summary and Conclusions

We have demonstrated the utility of fluorescence spectroscopy to provide information on XG solution behavior through the perspective of the sensitive solute molecule, C153. Few reports in the XG literature leverage the advantages and sensitivities offered by fluorescence spectroscopy. The spectroscopic response of a notably sensitive fluorophore, C153, has been used to delineate and quantify the XG concentration and temperature dependence in plain water solutions.

Spectral shifts and widths derived from the C153 ground- and excited-state transition energies support the idea that XG solution heterogeneity becomes significant at [XG]~0.0080 g/dL. Upon increasing temperature, the local polarity enhancement of the C153 microenvironment by water is evidenced by a spectral redshift. As XG adopts an open, random coil configuration, water more uniformly solvates the XG backbone and side chains, lowering the C153 energy. Spectral widths narrowed and also suggested significant solution changes occurring around [XG] = 0.0080 g/dL. These spectroscopic observations align well with the literature that reports the XG motif transformations between helix and random coil.

The inclusion of intensity decay data can help, to some extent, clarify the emission spectroscopy. Fluorescence lifetimes have resolved two C153 populations of varying percentage contributions. The faster 1.7 ns time constant we assigned to a C153-water microenvironment and a slower time constant at ~3.0 ns associated with C153 interactions within an XG microenvironment. Excited-state fluorescence lifetimes corroborated steady-state observations and clearly showed a distinct transition in the longer of the two time constants at [XG]~0.006 g/dL.

At least three plausible molecular explanations can account for our overall observations. First, C153 is responding to the helix-to-coil transition. If buried in and among XG helices, C153 specifically senses the XG molecular structure. Second, increasing the XG concentration increases the probability of side-chain interactions, thus creating a localized region wherein C153 resides. But, at elevated temperatures the drive by an entropic effect seems to counteract local organization, thus creating a mixed spectroscopic response. A third consideration is the formation of localized water domains by a caging type of XG chain organization. From spectral data alone it is challenging to completely harmonize all of these effects. However, we favor the first possibility given the literature comparisons that apply other instrumental methods such as NMR, SANS, DSC, and viscosity.

Given the complexity of the interactions in this system and the responses we have highlighted here, continued investigations are required to better describe the underlying behavior exhibited by XG in plain water solutions and how it influences spectroscopy. Toward this end, future research efforts aim to use fluorescence polarization to elucidate the dynamic response to varying XG concentration and temperature. The expectation is that the time course of molecular diffusion will help to more clearly differentiate between competing mechanisms such as aggregation vs. conformational change.