A method of water-soluble solid fraction saturation concentration evaluation in dry thalli of Antarctic lichenized fungi, in vivo

1Introduction

At initial steps of a very mild rehydration of some dry biological systems the liquid component of ¹H NMR signal does not increase proportionally with the increasing hydration level. Usually the observed dependency is well fitted using the rational function, e.g. in cellular organelles such as freeze-dried photosynthetic membranes [1], [2]; in plant tissues like horse chestnut bast [3]; at the initial phases of seed imbibition [4], as well as in whole living organisms, e.g. in thalli of several species of lichenized fungi [5], [6], [7], [8], [9].

In extremely dry biological systems the carbohydrates and polyols remain completely/partially solidified (actual water concentration is far below their saturation hydration concentration point) and gradually dissolve with the increasing hydration level.

Carbohydrates and/or polyols play a critical role either for freezing resistivity [10] or in acute desiccation shock resistivity for numerous living organisms. For example, before dehydration to cryptobiosis Polypedilum vanderplanki larvae [11], [12] accumulate large quantities of trehalose [13], [14], [15], [16]. The major carbohydrate constituents of seed are starch (up to 80% dry mass of seed) and cellulose. In developing seed enzyme action decomposes starch to simple sugars, which remain in aqueous medium [4], [17], [18], [19]. The accumulation of monosaccharides and polyols in lichen mycobiont is an adaptive feature for dealing with the complete dehydration [20], [21]. Ribitol level modifies the photosynthetic activity in some green algal foliose lichens [22].

¹H NMR relaxometry was applied to the analysis of extremely dry biological systems with a limited amount of water-soluble solid fraction. In horse chestnut bast it allowed to identify the water-soluble solid fraction as sucrose [3]. With the increased hydration level, as far as water-soluble solid fraction dissolves, the hydration dependency of total liquid ¹H NMR signal expressed in units of solid, L/S, is described by rational function, whereas for higher hydration levels at which the whole water-soluble solid fraction is completely dissolved the L/S hydration dependency becomes linear [3]. The presence of a threshold hydration level allows one to calculate the saturation concentration of water-soluble solid fraction, which may be used to identification of such a fraction.

In this report we propose a method for evaluation of the water-soluble solid fraction saturation concentration in case of a high amount of water-soluble solid fraction if the threshold hydration level is not detected, and if in the investigated biological system the signal coming from tightly bound water, L₁, and from loosely bound water signal, L₂, may be distinguished. A simple ¹H NMR experiment should be combined with sorption analysis for the investigated system.

As a model system we used a fruticose lichen Cetraria aculeata (Schreb.) Fr. thalli collected from sites in maritime Antarctica. Beside proton NMR spectra and relaxometry we recorded sorption kinetics, and sorption isotherms. The measurements were performed in vivo.

2Theory

Harańczyk, Węglarz and Sojka [3] proposed that if the sample is hydrated by the mass, Δm, of water and if the water soluble solid fraction is present in the system, the amplitude of FID for liquid/mobile signal component may be expressed as a sum of three terms:(1a)L=L0+αH2OρH2OΔm+αcdρcmcdwhere L₀ (if such a fraction is detected) is the signal of water'sealed’ in pores of dry solid matrix [1], [2], [23], the second term is proportional to the mass of water added, and the last one comes from water soluble solid fraction already dissolved. The amplitude of FID for solid signal component is then given by:(1b)S=S0−αcuρcmcd,and:(1c)S0=αsρsm0,where αH2O, αcd, αcu are the proportionality coefficients describing the effective contribution of a given proton pool to the total signal for water, and for water-soluble solid fraction in liquid and in solid phase, respectively (see Fig. 1). The values of coefficients α may be decreased e.g. by the presence of paramagnetic ions in the solution or on the surfaces of the solid matrix. If electron paramagnetism is absent in the investigated system αi=1[24]. However, in some biological systems solid NMR signal may be decreased by the presence of endogeneous electron paramagnetic ions, e.g. manganese [25]. The ρH2O and ρcare proton densities of water and of water-soluble solid fraction, respectively, whereas m_cd is the mass of dissolved water-soluble solid fraction.

Fig. 1

Proton subsystems distinguished at rehydration of C. aculeata in dry thalli (a), at low hydration level (b), and in wet thalli (c): solid thallus, S₀, (black space); air (white space); undissolved water soluble solid fraction (rectangular grid); immobilized liquid fraction consisted of water soluble solid fraction dissolved to saturation concentration, c_s (dots), mobile liquid fraction consisted of water soluble solid fraction dissolved to c_s, and mobile water sealed in pores of dry solid (L₀) (fine dots).

Fig. 1.

Knowing that the saturation concentration c_s of the water-soluble fraction is given by:(2)cs=mcdΔm+mcd.

The dissolved portion of water-soluble solid fraction, m_cd, may be obtained from:(3)mcd=cs1−csΔm.

In microheterogeneous systems bound water tend to orient depending on its mobility, and even tends to organize in some well defined layers [26]. If tightly bound water signal, L₁, may be distinguished from loosely bound water fraction L₂, the liquid signal component may be written as:(4)L=L0+L1+L2and:(5)Li=(αH2OiρH2O+αcdρccs1−cs)Δmi,where denumerator i=1, 2, and Δm₂=Δm−Δm₁.

If the loosely bound (mobile) water pool filling the pores of the solid matrix is absent in the dry system, i.e. L₀=0, and if the proportionality coefficients for mobile and for immobilized water are equal, αH2O1=αH2O2, and also for water soluble solid fraction is αcd=αcu, so, the L₁ to L ratio may be written as:(6)L1L=m1m1+(1+γcs1−cs)⋅Δm2,where m₁ is the mass of water saturating tightly bound water pool and the coefficient γ describes the water-soluble solid fraction proton density to water proton density ratio:(7)γ=ρcρH2O.

In the majority of living organisms which survive extreme dehydration the water-soluble solid fraction consists of sugars and/or polyols [5], [11], [12], [13], [14], [20]. Table 1 presents the saturation concentrations, and γ symbols for carbohydrates and polyols frequently occurring in biological systems. We used them to calculate the average values of γ. The coefficient γ does not vary much between biological carbohydrates with the averaged value equal γ=0.59±0.01. For polyols this value is quite similar and equals γ=0.67±0.06. If the type of water-soluble solid fraction is not known, the mean value averaged over sugars and polyols is equal γ=0.63±0.08.

The solid signal component, S₀, may be expressed as:(8)S0=αsρsms,where αs is the responsible proportionality coefficient for the solid matrix of the lichen thallus, ρs is the averaged proton density for the solid matrix of the thallus, and m_s is the mass of the solid matrix of the thallus which in the absence of ‘sealed’ water fraction equals the dry mass of the sample, m₀. If water-soluble solid fraction is present, we get for the solid signal, S:(9)S=αsρsm0−αcuρcmcd.

Thus, the total signal from liquid component may be expressed in units of solid component, L/S, [from (5), (9)] as:(10)L/S(Δm/m0)=(k+αcdαsγcs1−cs)Δm/m01−αcuαsρcρscs1−cs⋅Δm/m0.

Assuming that the decrease of the NMR signal for dissolved water-soluble solid fraction is comparable to this for water we get αcdαs≈k, and if the proportionality coefficient for the undissolved water-soluble the solid fraction is similar to the rest of the solid matrix of thallus, αcuαs≈1, we get:(11)L/S(Δm/m0)=k⋅(1+γcs1−cs)Δm/m01−γδcs1−cs⋅Δm/m0where:(12a)δ=ρsρH2O,and k is a slope of the L/S hydration dependence measured in case of the absence of water-soluble solid fraction expressed as:(12b)k=αLρLαSρS.

3Materials and methods

Cetraria aculeata (Schreb.) Fr thalli were collected on Penguin Island (South Shetlands Islands), Oceanic Antarctica (62°13′S, 58°25′W), in 2009 at the altitude 110 m above sea level. Air-dry thalli were stored in herbarium at relative humidity h≈40% and at room temperature with the hydration level, ∆m/m₀=0.088±0.011, where m₀ is the dry mass of the sample, and ∆m is mass of water taken up. Vitality tests performed using methylene blue staining showed that no less than 62% photobiont cells of thalli samples were alive before the measurements.

Polyol and sugar contents were measured using high resolution gas chromatography (HRGC) method. Dry lichen thalli were crushed in a mixed mill (MM 200, Retch, Germany) set at the vibrational frequency of 22 Hz for 2 min. Carbohydrates were extracted from 25 to 30 mg of pulverized thalli (in 1.5-ml tubes) with 800 μl of extraction mixture (methanol: chloroform: water, 12:5:3, v/v/v) containing 100 μg of d-chiro-inositol as an internal standard. The homogenate was shaken on Genie 2 vortex (Scientific Industries, USA) for 2 min, heated at 60 °C for 30 min and centrifuged at 20,000g for 20 min at 4 °C. Aliquots (400 μl) of clear methanol-water supernatant were desalted, evaporated to dryness. The dry residue was derivatized with a mixture of trimethylsilylimidazole: pyridine (1:1, v/v) as described previously [31]. The TMS-derivatives of carbohydrates were separated on a Zebron ZB-1 capillary column (Phenomenex, USA) in GC-2010 gas chromatograph (Shimadzu, Japan). The column was operated with an initial temperature 140 °C adjusted to 160 °C at 5 °C min⁻¹, then to 200 °C at 10 °C min⁻¹, and finally to 290 °C at 30 °C min⁻¹ (with total resolution time 11 min). The injection port operated in the split mode (10:1) at 325 °C, and the flame ionization detector was maintained at 350 °C. Helium was used as a carrier gas (with linear velocity 40 cm s⁻¹).

Soluble carbohydrates were quantified as described previously [33]. Standard of d-chiro-inositol (internal standard) was obtained from Industrial Research Ltd (New Zealand) and other carbohydrates were purchased from Sigma-Aldrich. The results of analyses are the means of three independent replicates±SE. Their values averaged for mycobiont and for photobiont are presented in Table 1.

Before the hydration courses thalli were incubated for 130 h over silica gel (h=0%), dehydrating to the hydration level ∆m/m₀=0.031±0.003. The dehydration process was well fitted by a single exponential function with the dehydration time t_D=(14.11±0.65) h (See Fig. 2).

Fig. 2

(a) The dehydration of air dry Cetraria aculeata (Schreb.) Fr. thalli to gaseous phase at the relative humidity h=0%, recorded as relative mass increase expressed in units of dry mass Δm/m₀. (b) The residual function calculated as the difference between the fitted and recorded values of the relative mass increase; the reduced value of χ²=2.48.

Fig. 2.

The hydration time-courses were performed from the gaseous phase with controlled humidity, at room temperature (t=22 °C), over the surface of H₃PO₄ (h=9%), over the surface of saturated solutions of CH₃COOK (h=23%), CaCl₂ (h=32%), K₂CO₃ (h=44%), Na₂Cr₂O₇ (h=52%), NH₄NO₃ (h=63%), Na₂S₂O₃ (h=76%), K₂CrO₃ (h=88%), Na₂ SO₄ (h=93%), K₂SO₄ (h=97%), and over the water surface (h=100%). Hydration kinetics courses were performed on one given sample per each value of relative humidity. After completing of the hydration courses, the dry mass of the thallus was determined after heating at 70 °C for 72 h. Higher temperatures were not used as they may cause decomposition of some organic constituents of the thallus [34].

The air-dry thalli designed for the NMR measurements were chopped, placed in NMR tubes, and incubated over silica gel, then the hydration procedure was performed. The NMR measurements were performed on fourteen samples. The measurements of a drier sample covered a part of hydration range for the next more wet sample, whereas the hydration level was calculated as ∆m/m₀.

¹H NMR spectra were collected on Bruker Avance III spectrometer (Bruker Biospin), operating at the resonance frequency 300 MHz (at B₀=7 T), with the transmitter power equal to 400 W. The pulse length was π/2=1.5 μs, bandwidths 300 kHz, and repetition time was 2 s.

¹H NMR free induction decays (FIDs) were recorded using WNS HB-65 high power relaxometer (Waterloo NMR Spectrometers, St. Agatha, Ontario, Canada). The resonance frequency was 30 MHz (at B₀=0.7 T); the transmitter power was 400 W; and the pulse length π/2=1.5 μs. In every FID 120 data points were collected, which were ordered in three equal groups with the different time-spacings equal to 0.4 μs, 2.5 μs, and 10 μs. The quadrature detection was applied, and the value of FID was a square root from the sum of squares of the phased and the dephased channel signals, Re²+Im². FIDs were acquired using Compuscope 2000 card of an IBM clone controlling the spectrometer and averaged over 2000 accumulations.

The data obtained were analyzed using the FID analyzing procedure of a two-dimensional (in time domain) NMR signal-analyzing software CracSpin written at the Jagiellonian University, Cracow [35], or by commercially available software Origin 7.0.

4Results

4.1Hydration kinetics

The hydration courses for Cetraria aculeata were performed from the gaseous phase, and were fitted well by single exponential function for the relative humilities, h_, controlled between 9% and 52% (see Fig. 3):(13a)Δm/m0=A0h+A1h⋅[1−exp(−t/t1h)],where ∆m/m₀ is the relative mass increase expressed in units of dry mass, m₀, A1h is the saturation level for the fast component solely observed at a given relative humidity range, and A0h is the hydration level at h=0% (very tightly bound water fraction).

Fig. 3

The rehydration of the lichen Cetraria aculeata (Schreb.) Fr. from gaseous phase at different values of relative humidity h, recorded as relative mass increase expressed in units of dry mass Δm/m₀. Targets humidity: h=9% – closed squares, h=23% – open squares, h=32% – closed circles, h=44% – open circles, h=52% – closed triangles, h=63% – open triangles, h=76% – closed reversed triangles, h=88% – open reversed triangles, h=97% – close starlets, h=100% – asterisks. The error bars are within the plot symbols.

Fig. 3.

As relative humidity exceeded 63% the slow hydration component appeared, and the hydration courses were fitted well by the two exponential function (Fig. 3):(13b)Δm/m0=A0h+A1h⋅[1−exp(−t/t1h)]+A2h⋅[1−exp(−t/t2h)],where A0h is the saturation level of very tightly bound water fraction (at h=0%), A1h and A2h are the saturation hydration levels for the fast and for the slow component, and t1h and t2h are hydration times for fast and slow component (tightly and loosely bound water fraction), respectively.

The averaged (over the all hydration courses) value of A0h=0.04±0.02. It is a very tightly bound fraction of water, which is not removed by dehydration over silica gel surface. Its relative mass is close to that obtained for other species of Antarctic lichens, e.g. for fruticose Usnea antarctica Du Rietz A0h=0.040±0.011, [6], and for foliose Umbilicaria aprina Nyl. A0h=0.054±0.011 [7].

The amplitude of the tightly bound water fraction equals to A1h=0.037±0.005, with the hydration time t1h=(0.43±0.10) h, whereas the loosely bound water fraction appearing for the courses performed from the values of relative humidity higher than 63% is characterized by the hydration time equal to t2h=(14.2±6.2) h. The saturation level of the loosely bound water component, A2h, increases gradually with increased humidity.

Loosely and tightly bound fractions may be distinguished by their proximity to the surface of solid matrix and thus by their mobility.

The total saturation hydration level, Ch=A0h+A1h+A2h, obtained at a given relative humidity, h, was taken for construction of sorption isotherm.

4.2Sorption isotherm

For Cetraria aculeata thalli the sorption isotherm is sigmoidal in form (Fig. 4), which usually is well fitted by the multilayer sorption models distinguishing two types of water binding sites on the surfaces of the investigated system, namely, (i) ‘primary’ water binding sites (directly bound to the adsorbent surface); and (ii) ‘secondary’, usually weaker, water binding sites (to the previously bound water molecules, or to the weaker binding sites on a solid matrix surface). The most commonly applied ones are the classic BET model [36], and GAB (Dent) model [37]. BET model takes a fixed value of the ratio of binding sites covered by n water molecules expressed in units of binding sites covered by n-1 water molecules b=Sn/Sn−1|h=1=1, which is to some extent an artificial assumption, whereas in Dent model this number may be varied between 0 and 1 (somehow simulating the clustering effect), and is equal b=Sn/Sn−1|h=1. Sorption isotherm for both models is expressed by:(14)Ch(h)=ΔMm0b1h(1−bh)⋅(1+b1h−bh)where h is relative humidity expressed in absolute units, ΔM/m₀ is the mass of water saturating primary binding sites, where Si is the number of binding sites covered by i water molecules, and the contribution of empty primary binding sites, S₀, on the surface in units of sites with one water molecule, S₁, at h=1 is expressed by the reciprocal of b₁: S0/S1|h=1=1/b1.

Fig. 4

The sorption isotherm for Cetraria aculeata (Schreb.) Fr. The values of h represent the relative humidity, and the values of relative mass increase, ∆m/m_0, are taken as the saturation values, Ch, from hydration kinetics (open circles = experimental data). Fitted Dent multilayer sorption model (Eq. (2)) (solid line = fitted Dent model), and ‘primary’ water binding sites monolayer coverage for Dent isotherm equal Δmm0=ΔMm0⋅b1h1+(b1−b)h[36] (dotted line = fitted ‘primary’ water binding sites population).

Fig. 4.

To test the relevance of the model used the sorption isotherm is usually presented in parabolic form (see Fig. 5), which for BET model is expressed as:(15a)hΔm/m0=A+Bh−(A+B)h2,whereas for Dent model as:(15b)hΔm/m0=A+Bh−Ch2,where parameters ΔMm0, b, b1 are connected with A, B, C by the formulas:(16a)b=B2+4AC−B2A(16b)b1=BA+2b(16c)ΔMm0=1Ab1

Fig. 5

Parabolic form of Dent and BET model (open circles – experimental data, solid line – fitted Dent model (Eq. (3)b), dotted line – fitted BET model (Eq. (3)a)).

Fig. 5.

For Cetraria aculeata thalli the sorption isotherm is much better described by Dent model, as the parameter equals b=0.958±0.003. The sorption isotherm for other Antarctic lichens is also better fitted using Dent model, e.g. for Leptogium puberulum b=0.955 [8], for Umbilicaria aprina b=0.941 [7], and for Usnea antarctica b=0.913 [6].

The mass of water saturating primary water binding sites is equal ΔM/m₀=0.046±0.003, which is very close to the value obtained from hydration kinetics for very tightly bound water pool.

The proportion of unoccupied water binding sites at h=1 equals for Cetraria aculeata 1/b₁=0.925%. This result is comparable with values reported for other lichens, e.g. for U. aprina, 1/b1=0.02% [7], Himantormia lugubris 1/b₁=1.11%, and for Caloplaca regalis 1/b₁=1.93% [5].

4.3¹H NMR spectra

Fig. 6 shows ¹H NMR spectra for C. aculeata thalli recorded at 300 MHz at two different hydration levels. For C. aculeata at the lower hydration level (Δm/m₀=0.112) the spectrum was the superposition of the broad component coming from the solid matrix of the lichen thallus, which may be successfully fitted by Gaussian function, and two narrow components coming from water bound in thallus, both fitted by Lorentzian functions (Eq. (17)):(17)A(ν)=ASΔνG⋅ln4⋅π/2exp[−2ln4⋅((ν−νG)ΔνG)2]+2AL1π[ΔνL14⋅(ν−νL1)2+ΔνL12]+……+2AL2π[ΔνL24⋅(ν−νL2)2+ΔνL22],where Δν_G, ΔνL1, and ΔνL2 are the half-widths of the NMR line,; ν_G, νL1, and νL2 are peak positions,; and finally A_S, AL1, and AL2 are the amplitudes of the Gaussian and two Lorentzian peaks, respectively.

Fig. 6

¹H NMR spectra for Cetraria aculeata (Schreb.) Fr. thalli recorded at 300 MHz; the pulse length π/2=1.5 μs. The relative mass increase was (a) ∆m/m₀=0.112, and (b) ∆m/m₀=0.218.

Fig. 6.

With the increased hydration level (at ∆m/m₀=0.218) the tightly bound water fraction is no longer fitted, and the spectrum is fitted by the superposition of one Gaussian component, and only one Lorentzian component coming from loosely bound water fraction, according to:(18)A(ν)=ASΔνG⋅ln4⋅π/2exp[−2ln4⋅(ν−νGΔνG)2]+2ALπ[ΔνL24⋅(ν−νL2)2+ΔνL22],

The halfwidths of the solid line component in the Gaussian approximation decrease from Δν_G=(46.4±0.2) kHz down to (43.4±0.7) kHz, with the hydration level increased from Δm/m₀=0.112 up to 0.218. Tightly bound water (broader) line component is detected only for the sample at the lower hydration level, with ΔνL1=2.238 kHz, whereas for the higher hydration level it is not detected by a fitting procedure. For the loosely bound water liquid component, the halfwidth is equal ca.ΔνL2=1.422 kHz, which is close to the value γ·ΔB₀ of the linewidths broadened by the B₀ inhomogeneities. Table 2 sets up the parameters of the recorded spectra.

Table 2

The parameters fitted to the ¹H NMR spectra for Cetraria aculeata thalli recorded at room temperature for two different hydration levels.

Table 2.

Δm/m0	νG[Hz]	ΔνG[kHz]	νL1Z [Hz]	ΔνL1Z [kHz]	νL2Z [Hz]	ΔνL2Z [kHz]	L/S*	L/S (FID)
0.112	2450±80	46.4±0.2	1535±23	2.24±0⊡2	821±7	1.27±0.02	0.463	0.338
0.218	3680±250	43.4±0.7	–	–	1151±2	1.57±0.01	1.348	0.719

4.4Proton free induction decays

Free induction decay for protons of the Cetraria aculeata thallus may be sufficiently well fitted by the superposition of one Gaussian component, with the amplitude S, coming from the solid matrix of the thallus; and one with the amplitude equal to L₁ or for higher hydration levels two, L₁ and L₂, exponential components coming from water tightly and loosely bound on the surfaces of the thallus, respectively [3], [6], [7]:(19)FID(t)=S⋅exp(−(tT2S*)2)+L1⋅exp(−tT2L1*)+L2⋅exp(−tTSL2*)where T2S* is the proton relaxation time of the solid component taken for 1/e-value of the Gaussian solid signal, and T2L1* and T2L2* are the relaxation times of proton liquid fractions L₁ and L₂ respectively (Fig. 7).

Fig. 7

The proton FID function recorded at 30 MHz for Cetraria aculeata (Schreb.) Fr. thalli hydrated to the relative mass increase ∆m/m₀=0.109; the superposition of Gaussian and two exponents (Eq. (14)) is fitted (dotted line); if the superposition of Abragam function ((16a), (16b), (16c)) and two exponentials (solid line)fits the characteristic “beat” pattern (better seen on the inset).

Fig. 7.

However, a more precise analysis reveals the presence of a “beat pattern” [38], [39] and in such a case the solid component of FID signal is better approximated by the Abragam function [40]:(20)S(t)=exp[−(tT2S*)2]sinatat.

To estimate the error made by the Gaussian approximation, the moment expansion of the solid component of the FID signal was performed:(21)S(t)=S(1−M22!t2+M44!t4−M66!t6+…),where M_i is the i-th moment of the absorption line, and t is the time starting from the final slope of /2 pulse.

For Abragam function the moment expansion equals [40]:(22)S(t)=1−(2(T2S*)2+13a2)t22!+(12(T2S*)4+4a2(T2S*)2+15a4)t44!−(120(T2S*)6+60a2(T2S*)4+6a4(T2S*)2+a67)t66!+…where for second, fourth, and for sixth moment we get, respectively:(23a)M2=2(T2S*)2+13a2and:(23b)M4=12(T2S*)4+4a2(T2S*)2+15a2and:(23c)M6=120(T2S*)6+60a2(T2S*)4+6a4(T2S*)2+a67.

Hence, for the Gaussian approximation of the solid signal (a→0) we get:(24a)M4M22=3,and(24b)M6M23=15.

For the Cetraria aculeata thalli the moment expansion of the solid component of FID signal gives M₂=3.9·10⁹ s⁻², M₄=4.0·10¹⁹ s⁻⁴, and M₆=2.5·10²⁹ s⁻⁶, with the magnitude S=106.9. The solid line was nearly Gaussian (with M4M22=2.46), thus for the next routine measurements the solid line was approximated by a Gaussian function (Eq. (6)). In fact, the solid signal for lichen thalli is often nearly Gaussian in form [6], [7], [8].

Relaxation time for the solid component, T2S*≈21 μs, is close to the value obtained for other lichens, and different solid biological systems as lyophilized photosynthetic membranes [1], [2], lyophilized model membranes from DGDG [23], or lyophilized DNA [41], which suggests similar distribution of local magnetic fields for limited number of chemical groups forming the biological system at a submolecular scale [42].

The observed water fractions are differentiated by their mobility and thus by their binding and/or proximity to the solid thallus surfaces, which means that intracellular water as well as extracellular water fractions usually contributes to both these water fractions. The T2L1*≈120 μs of the L₁ component is characteristic for tightly bound water of lichen thalli as well as of many other biological systems [e.g. 5, and the references therein]. The L₂ signal with T2L2*≈560 μs, shortened by B0 inhomogeneities, comes from water loosely bound on the thallus surface and for the higher hydration level from free water fraction (Fig. 7).

The spin-spin relaxation times measured in FID experiment are shortened by B₀ inhomogeneities according to [43]:(25)1T2*=1T2+γΔB02where T₂ is the spin-spin relaxation time, γ is the gyromagnetic ratio, and ΔB₀ is a change of magnetic field B₀ within the sample. The solid and short exponential components are detected and not changed (as compared to CPMG echo train), but L₂ may be an average of several loosely bound water pools, eg. extra- and intramolecular loosely bound water fraction.

4.5The hydration dependence of NMR signal

Fig. 8 shows the a parameter obtained from the Abragam function (Eq. (2)0) fitted to FIDs recorded for Cetraria aculeata thalli as a function of the hydration level. The parameter decreases slightly with the hydration level, which remains in correspondence with the change of the solid component halfwidth (a decrease from 32 kHz down to 25 kHz with hydration level increase from ∆m/m₀=0.04 up to 0.2). For the higher hydration level the ‘beat pattern’ characteristic for Abragam function is no longer detected.

Fig. 8

The parameter a for Abragam function fitted to solid component signal (thallus solid matrix line widths) calculated for Cetraria aculeata (Schreb.) Fr. thalli as a function of hydration level recorded as relative mass increase expressed in units of dry mass Δm/m₀.

Fig. 8.

For the hydration level below Δm/m₀=0.18 a tightly bound water, L₁, and a loosely bound water component, L₂, are detected (Eq. (19)). For the tightly bound water component, the relaxation time T2L1*≈100 μs, whereas for the loosely bound water component the effective relaxation time equals T2L2*≈100 μs (see Fig. 9).

Fig. 9

The hydration dependence of proton FID relaxation times for Cetraria aculeata (Schreb.) Fr. Solid Gaussian, S, component – closed circles, tightly bound water, L₁, component – open triangles, and loosely bound water, L₂, fraction – open squares.

Fig. 9.

For Cetraria aculeata the relaxation time T2S* of the solid component (in a Gaussian approximation) does not change much with the thallus hydration level (Fig. 9), which suggests that the structure and molecular dynamics of the thallus solid matrix are not much modified by the hydration process. Thus, we used a solid signal amplitude, S, as an unit to scale the amplitudes of liquid components.

Fig. 10 shows the hydration dependence of the tightly and of loosely bound water signal amplitude in units of solid amplitude, L₁/S and L₂/S, respectively. The hydration level is expressed as the relative mass increase in units of dry mass of the thallus, Δm/m₀. The tightly bound water signal, L₁/S, does not change within the measured hydration range, which suggests that it has already saturated for a lower hydration level. The hydration dependence of mobile (loosely bound) water signal, L₂/S, increases with the relative mass increase and is well fitted by a rational function.

Fig. 10

Tightly bound water component, L₁/S, and loosely bound water component, L₂/S, hydration dependence for Cetraria aculeata (Schreb.) Fr. The solid line is a rational function fitted (See text).

Fig. 10.

That the rational function successfully models the dependence, suggests the application of Eq. (6) for L₁/L hydration dependence, and Eq. (11) for the L/S hydration dependence for the Cetraria aculeata thallus. The combined fit of Eq. (6) and Eq. (11) performed for γ=0.63±0.08 is presented on Fig. 11. The lowest hydration level at which L₂ signal component is detected (and thus L₁/L becomes less than 1) is fitted as ΔM₁/m₀=0.075±0.006 (see Fig. 10 for comparison). The saturation concentration of the water-soluble solid fraction is c_p=0.57±0.12.

Fig. 11

The tightly bound water in units of total liquid signal, L₁/L, (open triangles) and the total liquid signal in units of solid, L/S, (open circles) hydration dependence for Cetraria aculeata (Schreb.) Fr. The solid lines were calculated from Eq. (6) and from Eq. (11), respectively.

Fig. 11.

5Discussion

Although the model proposed by the Authors yields the averaged saturation concentration for water-soluble solid fraction as equal to c_s=(57.3±12.0)%, which is close to the value for averaged carbohydrate and polyol saturation concentration calculated for Cetraria aculeata thallus (65.7±8.3)%, the actual concentration of carbohydrates and polyols in the dry thallus is much too small to be responsible for the nonlinearity in L/S hydration dependence.

Thus, it can be suggested than other cellular compounds are responsible for C. aculeata thallus hydration. Lichen thalli are known to contain considerable amounts of water-soluble glucans: lichenin, a hot-water-soluble linear (1→3) (1→4)-β-glucan and isolichenin, a cold-water-soluble (1→3) (1→4)-α-glucan, both with different proportions of linkage groups in diverse taxa of lichens. Honegger and Haisch [44] revealed that in fully hydrated and desiccated thalli of C. islandica, both the cortical extracellular matrix and the outer wall layer of medullary hyphae shrink dramatically during drought stress, indicating that these are major sites of water storage. Moreover authors suggested that a mycobiont derived, hydrophilic wall surface layer prevents the thallus interior from becoming waterlogged at full hydration. Similar properties, such as solubility and rheological behaviour in the solution and gel states, indicate cereal β-glucans, occurring in grasses (among them also cereals), lower plants and fungi [45], [46]. In seeds glucans play a role of storage material, which is degraded during seed germination [47]. Thus, it can be suggested that lichenin and isolichenin are also hydrolyzed in hydrated lichen thalli, according to the metabolic/physiological demand of cells. However, the content of glucans in C. aculeata was not analyzed in our study. Therefore we can only speculated that during thallus rehydration, causing the return of lichenized fungus from the cryptobiotic state, the additional pool of monosaccharides is supplied by the enzymatic action of endoglucanases. In such situation, our results could be a confirmation of the effect of the enzymatic action on lichen polysaccharides resulted in increase in monosaccharide concentration in lichen thallus returning from cryptobyosis state during rehydration. That could be a reason why L/S hydration dependence is well fitted by our model. Moreover, the increase in the levels of monosaccharides (releasing from glucans) and polyols (as a product of following algal and fungal cells metabolism) can support osmotic potential of cells leading to further water uptake [48].

In horse chestnut Aesculus hippocastantum most of water-soluble solid fraction consists mainly of sucrose. The dissolution process occurs without enzymatic action. The hydration level at which the total portion of the water-soluble solid fraction totally dissolves is directly observed in horse chestnut bast. It equals ΔM/m₀=0.180, which responds for the total amount of water-soluble solid fraction equal to ΔM_c/m₀=0.325. Also the NMR signal from tightly bound water fraction, L₁, is detected in wide range of hydration [3].

Thus, we recalculated the data for horse chestnut bast and fitted Eq. (6) to the hydration dependency of L₁/L, and Eq. (11) to the hydration dependence of L/S (see Fig. 12(a)). The combined fit of these two Equations yields c_s=(74.5±5.1)%, which is a little higher than 64.3% obtained using more precise model applied by them [3], and higher than those for sucrose (c_s=68%). For this fitting we used γ=0.59, which is the averaged value for biological carbohydrates, as the polyol fraction for horse chestnut bast is negligible. The difference in fitted values between our and their model is a measure of real accuracy of our model. However, it may be applied in case the limit of total water-soluble solid fraction dissolution is beyond the measured hydration level range or if it does not exist at all as it is in case of germinating seeds.

Fig. 12

a. The tightly bound water in units of total liquid signal, L₁/L, (open triangles) and the total liquid signal in units of solid, L/S, (open circles) hydration dependence for horse chestnut (Aesculus hippocastanum L.) bast. The solid lines were calculated from Eq. (6) and from Eq. (11), respectively. b. The tightly bound water in units of total liquid signal, L₁/L, (open triangles) and the total liquid signal in units of solid, L/S, (open circles) hydration dependence for Usnea antarctica. The solid lines were calculated from Eq. (6) and from Eq. (11), respectively.

Fig. 12

In addition, we recalculated (for γ=0.63±0.08) some already published data for other species of fruticose lichenized fungi, eg. for Usnea antarctica[6]. The combined fit to Eq. (6) and to Eq. (11), respectively, yields c_s=0.81±0.04, which is close to the value obtained by us for Cetraria aculeata (Fig. 12b). This suggests a similar way of recovery of the lichen metabolism from cryptobiotic form at dehydration, i.e. by the solid carbohydrate and polyol dissolution.

6Conclusions

The hydration dependency of high power ¹H NMR spectra for a dehydrated thallus of Cetraria aculeata shows two fractions of residual bound water which differ in mobility, i.e. tightly and loosely bound water, and the presence of the water soluble solid fraction in the thallus. The proposed simple model allows the quantitative description of that effect and yields the saturation concentration of the discovered water soluble solid fraction. The applicability of proposed model may also extend to other tissues/organisms at drastically low hydration level, as e.g. horse chestnut bast, other species of lichenized fungi or germinating seeds.